Addis Ababa, May 2015 
EVALUATION OF COMMUNITY￾BASED HEALTH INSURANCE 
PILOT SCHEMES IN ETHIOPIA:
FINAL REPORT
Federal Democratic Republic of Ethiopia 
Ethiopian Health Insurance Agency
Recommended Citation: Ethiopian Health Insurance Agency. May 2015. Evaluation of Community-Based 
Health Insurance Pilot Schemes in Ethiopia: Final Report. Addis Ababa, Ethiopia. 
Program Management and support, funding for the CBHI pilot evaluation, and printing of this evaluation report 
was provided by the United States Agency for International Development (USAID) through Health Sector 
Financing Reform/Health Finance & Governance (HSFR/HFG) Project which is part of the USAID’s global 
Health Finance & Governance Project (HFG) Implemented by Abt Associates Inc. under cooperative 
agreement -AID-OAA-A-12-00080. 
EVALUATION OF 
COMMUNITY-BASED 
HEALTH INSURANCE PILOT 
SCHEMES IN ETHIOPIA: 
FINAL REPORT
Federal Democratic Republic of Ethiopia 
Ethiopian Health Insurance Agency

FOREWORD
Ethiopia is committed to achieving universal health coverage (UHC) – expanding high-quality health 
care services that are equitable and accessible to all. Because financial risk protection is a critical 
component of UHC, Ethiopia has begun establishing a comprehensive and sustainable risk protection 
system with health financing mechanisms adapted to our country’s needs. The legal framework for 
the formal sector social health insurance (SHI) scheme has been put in place and final preparations 
are being made to fully implement the initiative. In the informal sector – which comprises over 85 
percent of Ethiopians – community-based health insurance (CBHI) is being expanded. To coordinate 
and regulate health insurance in the country and lead implementation of SHI and CBHI, the Ethiopian 
government established the Ethiopian Health Insurance Agency (EHIA) in 2011. Although the agency 
is still relatively new, the attention given to it by the government has enabled it to open a 
headquarters in Addis Ababa and 20 branch offices throughout the country, and to hire a staff of 
more than 500. 
The EHIA in collaboration with the regional governments of Amhara, Oromia, SNNP, and Tigray 
successfully implemented CBHI pilots in 13 districts. The agency now is working with the four 
regional governments to expand CBHI to an additional 185 districts. 
The evaluation that is the subject of this report generated in-depth qualitative and quantitative 
evidence that is critical to making the policy decisions needed for scale-up of this high-priority 
national initiative and to addressing the challenges that were faced in pilot implementation. The EHIA 
has been mining the lessons generated by this evaluation, starting with the preliminary findings and 
continuing to this final report. Even in the draft phase, we used the evidence extensively to refine 
key CBHI design parameters for initial CBHI expansion and we will continue to use the latest 
information to scale up CBHI to the entire country. We encourage all our stakeholders to review 
the report and forward their practical comments on CBHI scale-up in the informal sector. We are 
glad to share the evidence with our peer countries in Africa and elsewhere so that they can 
collaborate in filling our gaps and can take lessons from our endeavor. 
The successful completion of this exercise would not have been possible without the committed 
efforts and vital contributions of a wide range of stakeholders. Special thanks go to the United States 
Agency for International Development (USAID). I would also like to commend the staff of the 
Breakthrough International Consultancy firm for their professional work. This evaluation, similar to 
all other our interventions, materialized from the strong professional leadership, follow-up, and 
continued support of agency staff as well as the USAID Health Sector Financing Reform/Health 
Finance and Governance project country team and project backstopping provided by the Abt home 
office. 
Mengistu Bekele (MD, MPH) 
Acting Director General, Ethiopia Health Insurance Agency 

i 
CONTENTS
Acronyms.................................................................................................................. v
Acknowledgments ................................................................................................. vii
Executive Summary ............................................................................................... ix
SECTION 1. INTRODUCTION ........................................................................ xix
1. Introduction .................................................................................................... 21
1.1 Background ..........................................................................................................................21
1.2 Objectives and Deliverables of the Study ...................................................................22
1.3 Study Methodology ...........................................................................................................22
1.4 Data Collection Instruments ..........................................................................................26
1.5 Recruitment and Deployment of Enumerators and Supervisors .........................26
1.6 Models Specification ..........................................................................................................26
1.7 Organization of the Report ............................................................................................29
SECTION 2. LITERATURE REVIEW: OTHER COUNTRY 
EXPERIENCES ON CBHI .................................................................................... 31
2. Literature Review: Country Experiences on CBHI .................................. 33
2.1 Theoretical/Conceptual Framework of CBHI ...........................................................33
2.2 Empirical Evidence on the Impact of CBHI ................................................................34
2.3 Success Factors for CBHI Schemes ..............................................................................35
SECTION 3. CBHI Evaluation Findings .............................................................. 37
3. Ethiopian CBHI Pilot Schemes: Design and Current Status .................. 39
3.1 Scheme Design and Parameters ....................................................................................39
3.2 Establishment and Functioning of CBHI Schemes ....................................................43
3.3 Current Status of CBHI Pilots .......................................................................................45
4. Basic Household Characteristics ................................................................. 49
4.1 Basic Household Characteristics ...................................................................................49
5. CBHI Impact on Financial Risk Protection, Health Care Utilization, 
and Resource Mobilization ............................................................................... 55
5.1 Determinants of Enrollment ...........................................................................................55
5.2 Increased Utilization of Health Care Services ...........................................................61
5.3 Improving Financial Access/Financial Risk Protection..............................................72
5.4 Roles of Different Parties on CBHI Design, Implementation and 
Management ........................................................................................................................77
5.5 Increased Resource Mobilization ..................................................................................79
5.6 Resource mobilization potential during the scale-up...............................................86
SECTION 4. POLICY RECOMMENDATIONS ................................................. 91
6. Policy and Programmatic Implications of Evaluation Findings on 
Scale-up............................................................................................................... 93
References .............................................................................................................. 99
ii
ANNEXES ............................................................................................................ 101
Annex A: Literature reviewed: Evidence on CBHI ..................................... 103
Annex B: Household Survey Questionnaire ................................................. 107
Annex C: Sample Kebele/Tabia selection minute of one woreda ............. 141
Annex D: Sample Got selection minute ....................................................... 143
Annex E: Fiscal Implication of CBHI scale up, 50 percent participation.. 145
Annex F: Fiscal Implication of CBHI scale up, 75 percent participation .. 147
Annex G: Fiscal Implication of CBHI scale up, 100 percent 
participation ..................................................................................................... 149
Annex H: Patient Exit Interview Questionnaire .......................................... 151
Annex I: Key Informant Interview ................................................................. 163
Annex J: Focus Group Discussion Guide....................................................... 179
Annex K: People Interviewed ......................................................................... 183
 
List of Tables 
Table ES1: Projected Annual Financial Implications of CBHI Scale-up to Federal, 
Regional, and Woreda Governments ('000 Birr) ..................................................... xv
Table 1.1: List of Pilot and Control Woredas ..................................................................... 21
Table 1.3: Distribution of Planned and Surveyed Households by Woreda ................. 23
Table 1.4: Distribution of Planned and Actual Patients Exit Interviews ....................... 24
Table 1.5: Distribution of Key Informant Interviews by Region..................................... 25
Table 1.6: Regional Distribution of FDGs Planned and Carried Out ........................... 25
Table 1.7: Number and Distribution of Deployed Enumerators and 
Supervisors by Region ..................................................................................................... 26
Table 1.8: Definition of Catastrophic Spending and Its Thresholds .............................. 28
Table 2.1: Reviewed papers on CBHI impact on access to care.................................... 34
Table 2.2: Reviewed papers on impact CBHI on protecting impoverishment ........... 35
Table 3.1: Main Parameters of the Ethiopian CBHI scheme ........................................... 41
Table 3.2: Financial Status of CBHI Schemes, June 2013 (Birr) ...................................... 47
Table 4.1 (a): Characteristics of Sampled Households by Region ................................. 49
Table 4.2 (b): Characteristics of household members by region ................................... 51
Table 4.2: Educational Characteristics of Household Members Disaggregated by 
Pilot and Control Woredas ........................................................................................... 52
Table 5.1: Sources of information about CBHI (%) ........................................................... 55
Table 5.2: Responses about the Role and Concept of CBHI (%) .................................. 56
Table 5.3: Reasons for Joining the CBHI Scheme (Multiple Response) (%) ................ 56
Table 5.4: Reasons for not enrolling in the CBHI schemes 
(multiple response) (%) ................................................................................................... 57
Table 5.5: Sources of finance for enrolling into CBHI pilots ........................................... 57
Table 5.6: Determinants of enrollment in CBHI ................................................................ 60
Table 5.7: Referral system ......................................................................................................... 62
Table 5.8: Seeking Health Care and Number of Visits...................................................... 63
Table 5.9: Patients’ Perception of Improvement in Quality of Health Service 
Provision (%) ...................................................................................................................... 65
Table 5.10: Patients’ level of satisfaction with service quality (%) ................................. 66
Table 5.11: Outpatients’ levels of satisfaction ..................................................................... 67
Table 5.12: Inpatients’ level of satisfaction ........................................................................... 68
Table 5.13: Regression results of health care utilization .................................................. 70
Table 5.14: Regression results of frequency of health care utilization ......................... 71
Table 5.15: Incidence of and Average OOP Payments by CBHI Members, 
from PEI Survey ................................................................................................................. 73
Table 5.16: Incidence of and average OOP payments made by CBHI members and 
non-members, household survey ................................................................................. 73
Table 5.17: Poverty head count and overshoot between members and non￾members due to OOP payments at 15% and 25% non-food expenditure 
thresholds ............................................................................................................................ 74
iii
Table 5.18: Estimated Impact of OOP Payments on Poverty Headcount and 
Poverty Gap on CBHI Members and Non-members ............................................. 75
Table 5.19: Regression result of the impact of CBHI membership on levels of 
impoverishment ................................................................................................................. 75
Table 5.20: Community contributions/premiums (‘000 Birr) ......................................... 79
Table 5.21: Membership rate, contributions, and non-paying membership rate 
(June 2013) .......................................................................................................................... 80
Table 5.22: Reimbursement to health facilities ................................................................... 81
Table 5.23: Share of CBHI Reimbursements Paid to Health Centers 
and Hospitals ...................................................................................................................... 82
Table 5.24: Government Subsidy to CBHI Pilot Schemes up to June 2013 
(‘000 Birr)............................................................................................................................ 83
Table 5.25: Balance, all CBHI pilot schemes, June 2013 (‘000 Birr).............................. 84
Table 5.26: Assumptions for Projecting Resource Mobilization During Scale-up ..... 87
Table 5.27: Projected Income, Subsidies, Expenditure, and Balance (‘000 Birr)........ 88
Table 5.28: Projected Government Subsidy Requirements under 
Three Scenarios (‘000 Birr)............................................................................................ 89
Pilot Case Participation Rate .................................................................................................... 89
Table 6.1: CBHI Issues, and Evaluation Recommendations for Action and 
Responsible Party .............................................................................................................. 93
Annex A.1: Positive and Negative Factors Influencing Uptake of CBHI and 
SHI in Sub-Saharan Africa, and Implications for Policymakers ...........................103
Annex A.2: CBHI development model: potential and pitfalls for scaling up .............105
Projected Government Subsidy Requirements under Three Scenarios 
(‘000 Birr) 50% Participation Rate ..............................................................................145
Projected Government Subsidy Requirements under Three Scenarios 
(‘000 Birr) 75% Participation Rate ..............................................................................147
List of Figures 
Figure 2.1: CBHI Development Framework ........................................................................ 33
Figure 3.1: Enrollment Rate (%) and % of Indigent HHs Registered as Members, 
 June 2013 ........................................................................................................................... 45
Figure 3.2 Regional Enrollment Rate and % of HHs Registered as Indigents, 
 June 2013 ........................................................................................................................... 46
Figure 3.3: Per Capita Health Service Utilization Rate for Year 2012/13 .................... 46
Figure 5.1: Distribution of Targeted Subsidy by Expenditure Quintiles (in %) .......... 58
Figure 5.2: Affordability of Premiums .................................................................................... 59
Figure 5.3: Affordability of Registration Fee ....................................................................... 59
Figure 5.4: Reported Benefits of CBHI Membership ......................................................... 72
Figure 5.5: Utilization and CBHI Reimbursement in Health Centers and 
Hospitals, 2012/13 ............................................................................................................ 82
Figure 5.6: Ability of Schemes to Finance Health Service Costs (in %) ........................ 85

v
ACRONYMS
BOFED Bureau of Finance and Economic Development 
CSO Civil Society Organization 
CBHI Community-Based Health Insurance 
EFY Ethiopian Fiscal Year 
EHIA Ethiopian Health Insurance Agency 
FMOH Federal Ministry of Health 
FAMS Financial Administration and Management System 
FMHACA Food, Medicines, Health Care Administration and Control Authority 
FGD Focus Group Discussion 
GA General Assembly 
HC Health Center 
HH Household 
HSFR/HFG Health Sector Finance Reform/Health Finance and Governance 
ICT Information, Communication and Technology 
IPD Inpatient Department 
KHIIC Kebele/Tabia Health Insurance Initiative Committee 
KII Key Informant Interview 
M&E Monitoring and Evaluation 
OOP Out-of-Pocket 
OPD Out-patient Department 
PEI Patient Exit Interview 
PFSA Pharmaceutical Fund and Supply Agency 
RHB Regional Health Bureau 
SHI Social Health Insurance 
SNNP Southern Nation and Nationalities Peoples 
TA Technical Assistance 
Tabia It is equivalent to Kebele in Tigray region 
UHC Universal Health Coverage 
WHISC Woreda Cabinet and Woreda Health Insurance Steering Committee 
WHO World Health Organization 
WOFED Woreda Finance and Economic Development 
WorHO Woreda Health Office 
ZHD Zonal Health Department 

vii 
ACKNOWLEDGMENTS
The Ethiopia Health Insurance Agency (EHIA) would like to extend its gratitude to all involved in the 
CBHI evaluation data collection and analysis, and the writing of this evaluation report. The evaluation 
involved extensive data collection from both primary and secondary sources including creating of a 
standard questionnaire used for the household interviews, a literature review of international 
experiences with health insurance, a review of documents from the Federal Ministry of Health, 
EHIA, Regional Health Bureaus, the Health Sector Financing Reform/Health Finance and Governance 
(HSFR/HFG) project, and other sources. 
The EHIA acknowledges in particular the financial and technical support of the United States Agency 
for International Development (USAID), which was made through the HSFR/HFG project, led by Abt 
Associates. Together, the agency team and HSFR/HFG selected the consulting firm that would do 
the day-to-day work, and they provided written and verbal feedback at critical junctures such as 
during development of the survey methodology, sampling techniques, and survey instruments and on 
different versions of the evaluation report. 
The EHIA is extremely thankful for the professional work done by the Breakthrough International 
Consultancy (BIC) Plc., which conducted the qualitative and quantitative data collection and analysis, 
and drafted this report. We particularly appreciate the team for their high-quality work, patience, 
and flexibility in receiving and addressing extensive and at times uncoordinated comments from the 
agency and HSFR/HFG project. 
The joint technical leadership, management support, and quality assurance provided by the EHIA and 
Abt Associates has been superb, as usual. The agency congratulates HSFR/HFG as well as its own 
experts who led and contributed at various stages of the data collection and analysis, and report 
writing. 
Professional review, editing, and formatting by the Abt Associates home office was critical in giving 
the document its final shape, and contributed greatly to its readability. 

ix
EXECUTIVE SUMMARY
Background, Objectives, Scope, and Methodology 
As part of its health care financing strategy in general and its health insurance strategy in particular, 
the Government of Ethiopia endorsed and launched community-based health insurance (CBHI) 
schemes in 13 pilot woredas in Amhara, Oromia, Southern Nations, Nationalities, and Peoples 
(SNNP), and Tigray regions in 2010/11 to provide risk protection mechanisms for those employed in 
the rural and the informal sectors. Three years on, the government has decided to scale up CBHI, 
with schemes in 161 woredas. This evaluation of the impact of the pilot schemes was intended to 
inform the scale-up process. 
More specifically, the evaluation was to generate evidence on the impact of CBHI pilot schemes in 
terms of providing access to and utilization of health services; improving quality of health care; 
reducing financial risks for members and increasing resource mobilization in the health sector; and 
strengthening of community participation in the management of health services. The evaluation was 
also expected to provide policy recommendations for scaling up the schemes. 
The evaluation used several complementary methods of data collection to generate evidence. 
Review of relevant documents on the design and status of CBHI schemes and lessons from other 
countries was done. Two quantitative surveys (a household survey and patient exit interviews [PEI]) 
were also conducted. Information was collected from 2,987 households (target: 200 in each pilot 
woreda and 100 in each control woreda) using a random sampling method at woreda and 
kebele/tabia levels and systematic sampling at the household level. PEIs were conducted with 462 
patients at the facility level. Two qualitative data collection techniques (key informant interviews 
[KII] and focus group discussions [FGD]) also were used to generate information. To this end, 144 
KIIs with CBHI stakeholders and 52 FGDs with CBHI members, non-CBHI members, and health 
providers were conducted. Administrative data from CBHI schemes and reports from the Health 
Sector Financing Reform/Health Finance and Governance (HSFR/HFG) project provided information 
for the analysis on resource mobilization and financial status of the pilot schemes. The evaluation 
used both descriptive and econometric analysis methods to generate the findings. 
Findings 
Design and Current Status 
1. Overall there is no "one size fits all" strategy for implementing risk-pooling mechanism. Some 
countries have used top-down public financing (tax based) and social health insurance (SHI) 
without CBHI, while others have used CBHI as the main model of reaching the informal sector. 
As a result of these differences in design, country experiences show huge variation in the 
breadth, depth, and height of coverage achieved. Successful CBHI models show that there are 
important conditions for CBHI to grow and develop, including: (i) existence of a minimal level of 
(perceived) quality of care and gradual improvement of quality at the supply side; (ii) instituting 
adequate organizational practice and design including responsiveness to people's felt needs by 
the scheme management; (iii) government commitment and political will with clear action plans, 
national scope of implementation, existence of regulatory frameworks, and - last but not least - 
the unequivocal commitment to subsidize and finance the premium for the poorest in society; 
and (iv) the need for CBHI schemes to join forces to expand risk pooling and ensure financial 
sustainability. 
x 
2. Evidence from the experience of countries worldwide informed Ethiopia's CBHI design, which 
eschewed small-scale, voluntary membership and included mechanisms to finance the 
membership of the poor. The willingness and ability to pay of the pilot woreda population 
conducted by HSFR during the design phase as well as the assessment of readiness of facilities in 
the pilot woredas to provide health services informed the design parameters. Certain major 
features are unique to Ethiopia. Enrollment in a CBHI scheme is decided collectively at the 
kebele/tabia level as opposed to the household level. Associated kebeles/tabias form a larger 
woreda-wide scheme. Scheme management is integrated and works within the woreda 
administration office. A general subsidy from the federal government is provided for all scheme 
members and a targeted subsidy from the regional and woreda governments is provided for the 
very poor who cannot afford to pay the contribution. Scheme staff are employed through the 
government payroll. The federal government also provides resources to health facilities 
contracted to provide services to CBHI members, so that the providers maintain an acceptable 
quality of care. Although there is overall guidance from the federal level, scheme parameters are 
decided at the regional level based on the regional CBHI directives; as a result, there are minor 
variations in schemes in different regions in terms of registration and enrollment rates, 
membership caps, and service provider contracting. 
3. The CBHI routine monitoring data from HSFR/HFG show that as of June 2013 the overall 
enrollment rate in pilot CBHI schemes is 48 percent of households (range: 25 percent in Deder 
to 100 percent - universal enrollment - in Yirgalem). Of the total eligible households, 7 percent 
registered as indigents (range: 1 percent in Deder to 15 percent in Tehuledere and Yirgalem). 
The average enrollment of the very poor is greater in Tigray and Amhara than in SNNP and 
Oromia. The per capita health service utilization rate of CBHI members was 0.7 outpatient visits 
in 2005 EFY (2012/13), more than twice the nationwide average of 0.3 visits. Ten of the 13 
schemes are financially sound, while the other three (one in Amhara and two in SNNP) are 
having financial difficulties. 
On Knowledge, Enrollment, and Affordability 
4. Knowledge about CBHI: 95 percent of both members and non-members in pilot woredas are 
aware of the CBHI schemes. The main sources of information are a neighbor, a CBHI official, or 
a house-to-house sensitization program; these three represent 100 percent of information 
sources in Amhara, 96 percent in Oromia, 86 percent in SNNP, and 81 percent in Tigray. More 
than 96 percent of member households and 87 percent of non-member households know that it 
is not only those who are sick who should enroll in CBHI. This clearly shows the value of the 
intensive sensitization work done by government and especially by the HSFR/HFG project. 
5. Reasons for enrollment: 37 percent of CBHI members joined primarily to reduce out-of-pocket 
(OOP) expenditure when seeking health care, 35 percent joined to more frequently seek care in 
order to improve their health status, and 18 percent joined because their premium is less than 
their OOP payments; only 4 percent said they joined because government paid their registration 
fees and premiums. Although the design of CBHI in Ethiopia states that the decision on whether 
to join the scheme is taken at the kebele/tabia level, there was no pressure by either the 
community or the kebele/tabia administration during the enrollment process. According to the 
household survey results, 84 percent of CBHI members feel that premiums are either easily or 
somewhat affordable and 83 percent feel that the registration fee is easily or somewhat 
affordable. The affordability of premiums and registration fee is therefore an issue for only 16 
percent of registered members. All of the FGDs with CBHI members found that the payment is 
affordable if the services in the benefit package are indeed available. Though the schemes 
reported a low membership renewal rate, 97 percent of households enrolled as CBHI members 
and included in the evaluation survey indicated their intention to renew their membership when 
the current one expires. Eighty percent of non-members plan to join a scheme in the future. 
xi
6. The evaluation looked at the determinants of enrollment in CBHI and found the major factors to 
be household size; age, education, and sex of the head of household; and size of cultivated land 
(not only owned land). Larger families and those with older household heads are more likely to 
enroll than those with the opposite characteristics. Household heads who are literate are also 
more likely to enroll in to the scheme than those with no education. Household heads who 
completed primary education are more likely to join a scheme than those with no education. 
Households headed by a female are more likely to join than are male-headed households. There 
is no significant difference in the enrollment rate in Amhara, Oromia, and Tigray regions, while 
the rate in SNNP is lower than in Amhara (the reference region for this comparison). 
7. Of the 1,287 CBHI member households interviewed for the household survey, 1,169 (91 
percent) paid their own registration. The remaining 9 percent were enrolled either through the 
local government subsidy (5 percent) or, in Tigray, through their own contribution that was 
deducted from their "safety net" payment (4 percent). Amhara and Tigray seem to perform 
better in financing the poor. The assessment explored how far the targeted subsidy for CBHI 
was reaching the very poor and documented that 60 percent and 83 percent of the households 
benefitting from the targeted subsidy came from the lowest and second-lowest expenditure 
quintiles, respectively; only 5 percent of the targeted subsidy might have reached the richest two 
quintiles. This clearly shows that the selection of beneficiaries for the targeted subsidy is by and 
large fair though stopping the leakage should be tackled promptly. 
8. The survey also asked non-members their reasons for not being enrolled in the CBHI scheme. 
For 49 percent of the respondents, the registration fees and the premium are either not 
affordable (39 percent) or the payment schedule is not appropriate for them (10 percent); 
another 17 percent did not have adequate knowledge of and information about CBHI, and yet 
another 12 percent wanted to see the CBHI scheme in operation before they would join. The 
affordability of premiums and registration fee is therefore an issue for 39 percent of non￾members. This percentage, added to the 16 percent of non-members taking the "wait and see" 
approach, shows that there is need to further explore how the fiscal space can be expanded to 
ensure the full coverage of indigents by CBHI and to ensure the financial viability of the schemes. 
On Health Service Utilization by CBHI Members and Non-members: 
9. Most health centers in the pilot woredas are contracted by the CBHI scheme to provide 
services to scheme members. Even some health centers that had gaps in their readiness to 
provide quality care were contracted based on the demand from the community. In some 
regions physical access overrides woreda boundaries (Amhara and SNNP) and/or readiness of 
health facilities. All pilot woredas have also signed service contracts with their referral hospitals. 
Amhara, Oromia, and Tigray (not SNNP) also entered contracts to ensure the possibility of 
referrals within the region. 
10. Almost all health facilities visited by the evaluation team asserted that they provide all of the 
services that are expected at their level. All KII respondents said that the introduction of CBHI 
has increased health service utilization, for the following reasons: better financial protection for 
members when seeking care, the sensitization program when schemes were established, and 
continued health promotion efforts of the facilities, health extension workers, and the Health 
Development Army. FDGs with CBHI members in all pilot woredas confirmed that they visit 
health facilities immediately when they feel sick, which was not the case before. Of those 
reporting illness in the reference period, 1,049 individuals (71.7 percent) reported that they 
visited health facilities. Overall, 72.3 percent of CBHI members visited health facilities, similar to 
the aforementioned per capita visit rate of 0.7. Eighty-two percent reported using their 
membership card when utilizing health care. The econometric analysis documented that the 
likelihood of CBHI members visiting a health facility when feeling sick in the pilot woreda was 
26.3 percentage points higher than that of non-members. This is clear evidence that scaling up 
CBHI is likely to increase utilization and intensity of care in Ethiopia. 
xii 
11. The establishment of CBHI schemes provided health professionals a degree of freedom to 
prescribe the appropriate diagnostic tests and drugs without worrying about the CBHI member's 
ability to pay. This, according to the providers, is a great relief and key to improvement in 
service quality. By increasing utilization, CBHI is also generating more resources for contracted 
health facilities, resources that they can invest in improving quality of care. The additional in-cash 
and in-kind resources provided to contracted health facilities by the woreda, regional, and 
federal governments also are contributing to quality. Both CBHI members and non-members PEI 
respondents said they have witnessed improvements since the establishment of CBHI. In all 
measures of perceived quality of care, a greater percentage of CBHI members than non￾members feel that there is improvement; about 90 percent of these respondents reported to 
have either been satisfied or very satisfied with cleanness of the facility, courteousness of health 
professionals, and waiting time. 
12. There are major challenges in the quality of services provided. First, contracted providers differ 
in their readiness in terms such as pharmacy services, laboratory facilities, reception, and 
outpatient services. Second, health facilities, especially hospitals, are frequently short of drugs 
and patients must buy items from outside (private) retailers. Health facilities attribute the 
shortages mainly to shortages at Pharmaceutical Fund and Supply Agency (PFSA) hubs. The 
practice of moral hazard by pharmacists at contracted health facilities was reported to 
exacerbate the shortages.1 Third, there are frequent breakdowns of medical equipment, due 
mainly to lack of preventive maintenance but also to health worker negligence and mishandling. 
Fourth, there are complaints about availability and capacity of staff. Finally, no pilot woreda has 
systematic, regular, and standard mechanisms for collecting and properly addressing complaints. 
13. When contracted services are not available, members have to pay out of pocket for drugs, and 
for diagnostic and other health services in non-contracted health facilities. This affects CBHI 
members more than it does the general public. Those who have no money, in part because they 
spent it on their CBHI fee and premium, are not able to access care. Second, whereas CBHI 
members should be reimbursed by the scheme for full or partial payments, they often are not - 
their expenditures might be disallowed by scheme bylaws, the pharmacist at the contracted 
health facility fails to stamp the back of the member's prescription, or the reception might not 
record/register it in the patient's file. Even when private drug retailers provide a member with 
prescribed drugs, the members might need to travel to the woreda town, sometimes repeatedly, 
to process the reimbursement. 
14. There are reported cases of moral hazard and inappropriate practice both by providers and 
CBHI members. CBHI members have become assertive in demanding their rights and sometimes 
making unreasonable demands, including jumping the queue; demanding a prescription for a 
particular type of drug; returning to the facility for another consultation or treatment before 
finishing the prescribed drug regimen, demanding unnecessary diagnostics and prescriptions; 
requesting drugs for their children without bringing them to the facility; and demanding 
immediate referral. Types of moral hazard and other inappropriate provider practices reported 
include lack of courtesy and mistreating patients, such as displaying partiality toward some 
patients; overprescribing services including drugs and diagnostics; unnecessarily referring patients 
to private wings and private clinics; and making claims for reimbursement without backing them 
up with necessary evidence. 
 
1 This was suggested in one referral hospital in Amhara and in one of the pilot woredas in Tigray
xiii 
On Financial Risk Protection 
15. The analysis of PEI responses clearly shows that very few CBHI members incurred OOP 
expenses during their visit. Of the 184 members interviewed, only 2.1 percent paid for 
consultations, 1.6 percent for diagnostics, and 2.7 percent for drugs. Of the 259 non-members 
surveyed in the pilot and control woredas, 88 percent paid for consultations, 38 percent for 
diagnostics, and 90 percent for drugs. The average per person payment for members was half of 
the average paid by non-members. The findings of the household survey also showed that both 
the incidence and amount of OOP payment were slightly higher for non-members for all three 
types of services mentioned above. The evaluation estimated the extent to which households 
face the risk of being impoverished (i.e., falling below a defined poverty line) by OOP health 
expenditure using the non-food consumption expenditure. It used two alternative thresholds: 25 
percent and 15 percent (i.e., if a household's OOP expenditure on health is equal to or greater 
than either 25 percent or 15 percent of its non-food expenditures). The findings showed that 
the risk of being impoverished by OOP health expenditure is 7 percent for CBHI members and 
19 percent for non-members at the 15 percent threshold and is 3 percent for members and 9 
percent for non-members at the 25 percent threshold. This shows that CBHI members have a 
lesser risk of being impoverished as a result of OOP payments than non-members. We also 
attempted to estimate the head count, poverty gap, and normalized poverty gap for members 
and non-members by adopting Soumitra Ghosh's model (Ghosh 2010). The estimate shows that 
an additional 1 percent of CBHI member households and 5 percent of non-members dropped 
below the threshold poverty as result of OOP payments. The poverty gap was Birr 58 for 
members and Birr 143 for non-members. A regression analysis, based on a dichotomous choice 
(logistical regression) model, controlling for other factors, also showed that being a member is 
negatively related to impoverishment due to OOP payments. The evidence in Ethiopia therefore 
shows OOP payments in general have an impoverishing impact on households, but the impact on 
CBHI members is much less than on non-members. The analysis thus provides evidence that 
scaling up CBHI schemes will have a beneficial pay-off by reducing the incident and severity of 
poverty for CBHI members. 
On Mobilizing Additional Resources to Health Providers 
16. The CBHI schemes have been able to collect a total of Birr 22.7 million in premium payments. 
Amhara and Oromia regions have mobilized more than the other two regions. Revenue 
collected was greater in the first year of CBHI than in subsequent years, evidence of a decline in 
new and renewing membership. Inadequate contribution/premium collection from the 
community affects the total available resources. However, resources available to the schemes in 
the form of targeted subsidy are not affected because the woreda and regional governments are 
paying these premiums. Hence the lower the level of enrollment rate, the lower is the collection 
from the community and thus the total available resources. Actual premium collected compared 
with potential contribution/premium calculated based on number of eligible households shows 
that there are serious challenges in enrollment and collection of premiums in the pilot woredas. 
This has to be rectified in the scale-up phase because the effectiveness and continued relevance 
of CBHI scheme as one of the paths to universal health coverage (UHC) depends on CBHI 
collecting a reasonable share from the community. 
xiv 
17. CBHI schemes reimbursed about Birr 16.9 million to contracted health centers and hospitals for 
services rendered. Key informants and FGDs with health facility staff confirmed that CBHI has 
increased the utilization of services and retained revenues in health facilities, particularly health 
centers. More than 90 percent of service utilization by members takes place in, and more than 
90 percent of reimbursement is made to, health centers in Amhara, SNNP, and Tigray. In 
Oromia, hospitals account for only about 5 percent of CBHI member utilization but they took in 
about 31 percent of the total reimbursement paid out. Overall CBHI schemes seem to provide 
the correct utilization pattern (using lower-level facilities more) and payment trends also follow 
this pattern. The reasons for a large proportion of payment going to hospitals in Oromia needs 
to be further explored to reduce the undesirable impact on scheme financial status. 
18. According to the CBHI design, there are three types of government subsidies to the schemes: 
targeted and general subsidies, and financing the scheme management costs (salaries, office 
space, and operational costs). The regional and woreda governments finance premiums of 
indigents using different arrangements. In Tigray, the regional government finances 70 percent of 
indigents' premium contribution and the woreda finance 30 percent. In Amhara, the split is 90/ 
10. In SNNP and Oromia, woredas finance all the costs of indigents. Woreda governments also 
finance the salaries and operational costs of all schemes. The federal government subsidizes 25 
percent of the CBHI premiums, for both paying and non-paying members. Through the end of 
June 2013, the total amount of subsidy paid to the 13 pilot woredas was Birr 16.5 million: Birr 
9.7 million from the general subsidy and Birr 6.8 million from the targeted subsidy. The total 
subsidy constituted about 42 percent of the total revenue generated by schemes. Of this, the 
general subsidy accounts for 25 percent and the targeted subsidy for 17 percent. Members' 
contribution/premium accounts for 58 percent of total revenue collected by the schemes. Of the 
total revenue generated from all pilot schemes, 35.3 percent was from Amhara, 27.3 percent in 
Oromia, 24.2 percent in Tigray, and 13.1 percent in SNNP. However, there is still a very strong 
government commitment to CBHI. If the pilot arrangement continues during scale-up, the 
financial implication for the government at all levels will be enormous. While government 
commitment and financing is one ingredient of a successful CBHI scheme, too much subsidy also 
raises questions about the relevance and added value of CBHI as against different tax-financed 
health services and SHI. Scale-up should therefore balance effective premium mobilization and 
government financing. 
19. Overall, Ethiopia's CBHI schemes were able to finance the health service costs using financial 
resources generated from contributions. The reimbursement made to the health facilities stands 
at about 75 percent of contributions from paying and non-paying members (without any 
subsidy). However, CBHI schemes in three woredas (Fogera, Yirgalem, and Damot Woyde) 
would not have been able to finance their health service costs without the subsidy (owe more 
money from health facilities). When the targeted subsidy was included, all woredas except 
Yirgalem had a positive balance and a claims ratio of 57.4 percent. When the general subsidy 
was included in this total (contribution and targeted subsidy), all woredas remained in good 
financial situation and the claims ratio became 43.2 percent. When we look at regional 
performance, schemes in Amhara, Tigray, and Oromia are financially healthy while those in the 
SNNP are the ones whose financial status is flagging, jeopardizing sustainability. 
xv
20. The resource mobilization potential of CBHI schemes during scale-up has been explored based 
on three sets of assumptions.2 Thus, the total amount of resources that can be mobilized from 
community in the form of premiums in pilot and scale-up woredas in all four regions will be 
approximately Birr 345 million per year. The projected total annual reimbursable for health 
facilities in all four regions will be Birr 316.5 million per year. The fiscal implications for the total 
government subsidy at all levels of the scale-up range from Birr 93,875 million per year under 
the pilot case (see Table ES1) to close Birr 400 million if all estimated households living below 
poverty line are to be covered through the targeted subsidy. The average financial burden for a 
woreda is Birr 121,963.07 per year (first scenario) to Birr 221,925.76 per year (third scenario). If 
the government considers scaling up the schemes in non-pastoralist areas (first scenario), the 
fiscal implications will increase to Birr 817.9 million per year, of which 616.7 million per year will 
be from the Federal Ministry of Health (FMOH) as a general subsidy. 
Table ES1: Projected Annual Financial Implications of CBHI Scale-up to Federal, Regional, and 
Woreda Governments ('000 Birr) 
161 expansion woredas All non-pastoralist woredas
10% Pilot 29% 10% Pilot 29%
Total Budget 148,346 146,466 203,087 817,962 836,801 1,196,815
Federal Budget 93,875 93,875 93,875 616,680 616,680 616,680
Regional 
Budget
32,090 31,212 68,241 118,883 140,501 342,423
Average woreda budget (161 woredas) Total woredas' subsidy budget, all non￾pastoralist woredas
Total 121.96 117.74 221.93 82,398 79,260 237,712
 
2 The first set of assumptions are the policy decisions already made by each regional government for scaling up, 
particularly with regard to the number of expansion woredas, premiums and registration fees, scheme 
organization, indigents’ coverage rate, and government subsidy structures. The second set of assumptions are 
related to the performance of schemes, that is, that service utilization rates and reimbursements for health 
facilities will be similar to the averages of regional pilot woredas. The third set of assumptions relates to the 
number of indigents to be financed. There are three scenarios: regional averages of pilot performance in each 
region; 10% of eligible households in woredas; and full coverage of people (29% of CBHI eligible population) 
who are expected to be below poverty line were considered. The projection is made only for a year and thus 
does not take inflation into account.
xvi 
Policy and Programmatic Implications of Findings on Scale-up 
Overall Policy Recommendation 
The plan to scale up CBHI to other woredas is timely. Given the evidence generated in this 
evaluation, CBHI scale-up as a path to UHC needs to be pursued with commitment of governments 
at all levels. However, internationally acknowledged CBHI pillars should guide the scale-up to UHC 
(promoting equitable access by removing financial barriers, especially direct payments; making 
prepayments mechanisms compulsory; creating larger risk pools for financial sustainability, and 
governments covering the premiums of very poor). There is a need for a comprehensive CBHI 
scale-up strategy whose development considers different design parameters and implementation 
arrangements, such as urban vs rural. It is recommended that implementation be phased. 
Explore the Relevance of CBHI to Pastoralist Context 
The way of life in pastoralist communities differs from that of urban and rural contexts. The service 
delivery mechanism remains weak and the FMOH is working to strengthen these systems. CBHI has 
been piloted only in rural and to a limited context in urban settings. Its relevance to the pastoralist 
context should be explored during the scale-up. 
Legislation and Operational Frameworks 
As learned from the CBHI pilot, there is a need to provide a legal basis for CBHI scale-up and this 
requires either the development of a CBHI legislative framework or revisiting the health insurance 
proclamation. This legislative framework should clearly define the CBHI scale-up roles and mandates 
of the FMOH/Ethiopian Health Insurance Agency (EHIA) vis-a-vis regional health bureaus and 
woreda administrations and/or health offices. It must also provide a legal basis for networking of 
CBHI schemes to create larger risk pools for the purposes of reinsurance and future integration of 
CBHI into SHI as a vehicle for UHC. Finally, CBHI bylaws and the CBHI Financial and Administrative 
Management System (FAMS) need to be reviewed and modified. 
Strengthen Institutional Arrangements 
If CBHI is to be a vehicle for moving toward UHC, it requires strong and functional institutional 
arrangements. This requires reexamining the structure of the EHIA and its branch offices, as well as 
regional and woreda government structures and their relationship in CBHI implementation. It is also 
necessary to review and establish kebele/tabia-level structures for CBHI. These structures could be 
new or an incentive and sanction mechanism to get kebele/tabia executives to play a role in CBHI 
enrollment, and collection and depositing of premiums. There also must be strengthening strategies 
to recruit and retain the CBHI team: clarifying the structure of the team, revising the salary scale, 
defining the career structure, and creating an enabling working environment by allocating an 
adequate operational budget. 
Revisit CBHI Parameters 
The disease burden, utilization pattern, and ability of the households to pay for the premiums differ 
between rural and urban areas. There is therefore a need to adjust CBHI parameters to fit the urban 
and rural contexts. In revising the parameters, it is recommended that (i) premium levels be revised 
according to willingness and ability to pay analysis in each area, include introducing a stratified 
(sliding) premium schedule especially in urban woredas; (ii) co-payment per outpatient visit be 
introduced, especially in urban areas; (iii) the feasibility of including ambulance services in the benefit 
package be considered; and (iv) the provider payment mechanism be reviewed. 
xvii 
Strengthen Scheme and Provider Interface 
The long-term sustainability of the CBHI schemes will depend greatly on their healthy and 
transparent working arrangement with providers. During scale-up, it will be necessary to invest time 
and resources on sensitizing health facility personnel to an extent greater than has been done in the 
pilots. There is also a need to strengthen health facility audits to control for moral hazard on the 
part of providers. Establishing/strengthening scheme and provider consultation forums will help to 
avoid misunderstanding and to clarify working arrangements and issues around claims management 
and processing of reimbursements. SNNP's facility-community forum might be considered for 
replication as it gives providers and CBHI members a venue in which to raise complaints and agree 
on solutions. In urban areas, there is a need to consider contracting private health facilities that are 
nearby, popular with residents, meet accreditation/certification requirements, and are willing to 
negotiate on fees to control costs. 
Strengthen CBHI Management and Coordination 
The financial sustainability of schemes will depend largely on the strength of CBHI management. The 
scaling-up process should consider strengthening several aspects of the management systems. There 
is a need to review the operational guidelines developed during the pilots to fix current weaknesses. 
Revising renewal mechanisms and enforcing the scheme bylaws on membership renewal and 
collection of premiums is a priority. The sensitization strategy needs to be strengthened; this 
includes developing sensitization tools, documenting pilot successes, and incorporating lessons 
learned could help increase enrollments. There is also a need to strengthen CBHI coordination 
structures and make them functional. Applying strategies that the government is using in other 
aspects of health service delivery, it is suggested that kebele/tabia/community sensitization and 
follow-up be linked with the Health Extension Program and the Health Development Army. 
Enhancing the involvement of zone and woreda health offices in the management of the CBHI and 
providing better orientation to zonal, woreda, and kebele/tabia officials could help strengthen 
leadership and commitment at all levels. Best practices from the enrollment-leading woredas of 
Yirgalem and Tehuledere - making enrollment part of the overall woreda cabinet evaluation criteria 
included in the checklist - should be considered for replication. 
Strengthen Monitoring and Evaluation System 
Timely generation and use of evidence on what works and what doesn't is critical for good scheme 
management and for policy decision making. There is a need to review and strengthen the CBHI 
monitoring and evaluation (M&E) and management information system by taking actions that include 
defining key data analysis and use requirements; defining responsibilities at the federal, regional, 
zonal, and woreda levels; strengthening record keeping at scheme and kebele/tabia levels; and 
strengthening periodic reporting and performance reviews. There is also a need to improve the 
information system to ensure that data on health service utilization by CBHI beneficiaries are 
properly documented and shared with the schemes. The routine M&E system should include regular 
supportive supervision by government authorities at all levels. 
Enhance Financial Sustainability of Schemes 
As noted above, the financial sustainability of some of the pilot CBHI schemes is in jeopardy. There 
is a need to increase enrollment for better risk pooling at the scheme level. Premium collection 
mechanisms should be strengthened and premium levels made congruent with benefit packages and 
service utilization. There is also a need to establish larger risk pools to have reinsurance and risk 
pooling among schemes. To do this, zonal and regional CBHI risk pools with clear resource 
contribution and expenditure assignment criteria should be established. This could gradually evolve 
into a national CBHI risk pool. 
xviii 
Enhance Protection of Poor Households through Targeted Subsidy 
The evaluation found different levels of commitment to financing indigents among regions. There is a 
need to enhance commitment of certain regional governments for this. In this connection, there is a 
need not only to increase resource allocation to cover the very poor - in all regions but especially in 
regions where coverage is currently lower - but also to improve the identification process to reduce 
leakage. If the fiscal space allows, all the households living below poverty line should be considered 
eligible for targeting. The flat 10 percent target should be adopted based on the conditions of 
woredas, for instance, highest percentage for food-insecure and lesser percentage for food-secure 
woredas. 
Meet financial and Technical Resource Requirements for Scale-up 
Regions and the FMOH need to be aware of the financial and technical resources needed to scale up 
CBHI and they should take appropriate measures to mobilize the additional resources. Regions need 
to mobilize about Birr 48.3 million and the FMOH about Birr 126 million per year. There is also a 
need to support the schemes with per diem, fuel, vehicle, and other travel-related expenses. When 
viewed from the perspective of movement toward UHC, these fiscal implications are not huge. 
Considerable technical input also is required to establish and manage the schemes. 
Remove Overall Health Systems Constraints 
The ultimate goal of health insurance is to improve health status through increased access to quality 
health services. CBHI schemes can only be successful if members have access to good quality care, 
which depends on the performance of the overall health system. The findings of this evaluation imply 
that government should continue investment in removing health system constraints. There is also a 
need to improve the availability of drugs and medical supplies by closely working with PFSA. Finally, 
it is clear that CBHI increases health services utilization by beneficiaries; thus, there is a need to 
design mechanisms to address complaints by and create incentives for health workers in reference 
to the additional work burden that increased utilization will mean for them. 
SECTION 1. 
INTRODUCTION 

21
1. INTRODUCTION 
1.1 Background
Prior to 1998, the per capita expenditure for health in Ethiopia was very low: in the years from 1980 
into the mid-1990s, it fluctuated between US$1.00 and US$1.20, which was far below the sub￾Saharan African per capita average of US$6.70 (FMOH, 1998). The resource allocation was also 
skewed in favor of hospitals and urban areas. The user fees charged in health facilities did not reflect 
the cost of health care and all revenue collected was remitted to the government treasury. There 
was no or little insurance coverage in the country. The involvement of the private sector in health 
was limited. All these circumstances made accessing health care a challenge for many households. To 
resolve these issues, a Health Care Financing strategy was endorsed by the Council of Ministers in 
1998. Its objectives were to mobilize additional resources from both domestic and external sources, 
improve efficiency especially by shifting resources to primary care, and ensure sustainability of quality 
health care services. It also aimed at enhancing community participation and ownership of health 
services. 
The Health Care Financing strategy, implemented over the past 15 years, is meeting its objectives. It 
has provided greater autonomy to health care facilities and mobilized revenue for facilities, including 
primary care facilities, by reforms such as establishing facility boards, retaining user fees, outsourcing 
non-clinical services, opening private wings in public hospitals as an income-generating and -retaining 
mechanism for medical professionals and health facilities, and introducing a third-party payer for fee 
waived patients. 
As these supply-side reforms took root in the regions, the government also initiated demand-side 
financing reforms. The Federal Ministry of Health (FMOH) developed a two-pronged health 
insurance strategy of social health insurance (SHI) and community-based health insurance (CBHI) 
schemes. These demand-side initiatives are designed to pool risks and protect households from out￾of-pocket (OOP) expenditures when receiving health facility services. The SHI scheme will provide 
financial coverage for the formal sector while CBHI targets those employed in the rural and the 
informal sectors. CBHI implementation began in Ethiopian fiscal year (EFY) 2003 (2010/11) as pilot 
schemes in 13 woredas of Amhara, Oromia, Southern Nations Nationalities and Peoples (SNNP), 
and Tigray regional states. The large Oromia Region has four pilot woredas, and the other three 
regions each have three pilot woredas. Each region has also identified a control woreda to ensure 
proper impact evaluation. Table 1.1 presents the list of pilot and control woredas by region. 
Table 1.1: List of Pilot and Control Woredas 
Amhara Oromia SNNP Tigray
A. Pilot Woreda
South Achefer Gimbichu Yirgalem town Kilte Awlaelo
Fogera Kuyu Damboya Ahferom
Tehuledere Deder Damot Woyde Tahitay Adiabo
Limmu Kossa
B. Control Woreda
Dembia Merti Wonago Raya Azebo
22
After three years of piloting, the government decided to expand CBHI schemes to 161 woredas of 
the country. So that the scale-up process would benefit from the experiences of the pilot schemes, it 
was decided to evaluate the impact of CBHI in the pilot woredas with respect to the intended 
objectives. The United States Agency for International Development (USAID)-financed Health 
Sector Financing Reform/Health Finance and Governance (HSFR/HFG) Project engaged 
Breakthrough International Consultancy (BIC) to carry out the evaluation. 
1.2 Objectives and Deliverables of the Study
As clearly indicated in the terms of reference, the overall objective of the study was to evaluate the 
impact of the CBHI pilot schemes on equitable access to sustainable quality health care, increased 
financial protection, and effective social inclusion among the communities in which the pilot 
programs have been implemented. The specific objectives of the study included evaluating the extent 
to which CBHI has impacted on, and document the lessons learned, with respect to the following: 
 Improving financial access to health care services; 
 Improving quality of health care services; 
 Increase resource mobilization in the health sector; and 
 Strengthening of community participation in the management of health services. 
Furthermore, based on the findings, study was also expected to provide recommendations that will 
inform policymakers on optimal actions and options for scale-up of the schemes to the national level. 
This report synthesizes CBHI evaluation reports on each of the four regions with the CBHI pilot 
schemes. The regional reports reported on qualitative and secondary information sources. The 
household survey and the patient exit interview (PEI) results were analyzed nationally. Any finding 
specific to any regions for any issue is highlighted on the relevant sections of this report. 
1.3 Study Methodology
1.3.1 Methods
The evaluation used a variety of complementary methods of data collection for generating evidence 
for the study (see Box 1.1). In addition to the 
review of relevant documents, the evaluation is 
based on two quantitative surveys (a household 
survey and PEI) and two qualitative data 
collection techniques (key informant interviews 
(KII) and focus group discussions (FGD)). The 
combination of data collection methods provided 
a wealth of information in order to assess not 
only the process of CBHI design and 
implementation, but also factors contributing to 
enrollment in CBHI schemes and the impact of 
CBHI schemes on access to health services and 
financial protection. The analysis of the effect of 
CBHI on resource mobilization and financial 
sustainability was based on administrative data 
from CBHI schemes and reports from the health 
sector reform project. 
Box 1.1 Reports and evidences that help 
generate the evidence for this Report
This report was informed by a number of 
evidences generated during the evaluation 
process. These evidences include:
 Design and progress reports from 
HSFR/HFG project
 Four regional reports written to provide 
input to this evaluation based on qualitative 
information (KIIs and FGDs) and secondary 
data at facility and woreda levels
 Analysis of the patient exit interview
 Analysis of the household survey data 
23
Box 1.2: Sampling Procedure for 
Household Survey in Pilot Woredas
 Five kebeles/Tabias from each woreda were 
selected randomly (through a lottery 
method).
 In each kebele/Tabia, two gots/kushets were 
selected randomly in consultation with the 
kebele/Tabia administration.
 A fresh list of households was taken for each 
of the 10 selected gots/kushets, from which 
sample households were randomly selected.
 Systematic sampling was used to select the 
10 CBHI member and the 10 non-CBHI 
member household (nth household) in each 
got
 A reserve list was prepared for replacement 
if and when some of the selected households 
are not available. The size of the 
replacement was in total 10 percent of the 
households.
1.3.2 Sampling Design 
Sampling for household survey: A household 
survey was conducted using the questionnaire 
attached as Annex B. The objective of the 
household survey was to generate data and 
evidence on the impact of CBHI on improving 
financial access and health service seeking 
behavior of members. 
The sampling design was a combination of 
purposive, simple random and systematic 
sampling. We have four regional states where 
each region has three pilot woredas and one 
control woreda (Oromia has four pilot woredas). 
After getting the list of Kebeles/ Tabias in pilot 
and control woredas, most logistically 
inaccessible Kebeles/ Tabias were excluded from 
sample selection for cost reasons. Five Kebeles/ 
Tabias were selected randomly (lottery method) 
with the presence of each woreda administration 
and CBHI management team (see sample minutes 
of one woreda as Annex C). Once the five 
Kebeles/ Tabias in each woreda were selected, 
the survey was carried out in two gots/ 
kushets/development committees in each 
kebele/tabia.3 Gots are selected randomly (lottery method) with the presence of kebele/tabia 
leaders (see sample got selection minutes, Annex D). The survey team selected a given number of 
households from each kebele/tabia with some reserves using systematic sampling (see Box 1.2 for 
details). Of the total planned 3,000 households, data were collected and analyzed from 2,987 
households (99 percent). 
Table 1.3: Distribution of Planned and Surveyed Households by Woreda 
Woreda # of HHs planned for survey # of HHs surveyed
South Achefer 200 200
Fogera 200 200
Tehuledere 200 200
Dembia (control woreda) 100 100
Kilte Awlaelo, 200 200
Ahferom 200 200
Tahitay Adiabo 200 200
Raya (control woreda) 100 100
Gimbichu 200 200
Kuyu 200 200
Deder 200 200
Limmu Kossa 200 198
Merti (control woreda) 100 100
 
3 Gots and kushets are sub-kebele entities. 
24
Woreda # of HHs planned for survey # of HHs surveyed
Yirgalem City 200 197
Damboya 200 195
Damot Woyde 200 200
Wonago (control woreda) 100 97
Total 3000 2,987 (99%)
Source: HH survey 
Note: HH=household 
Patient exit interviews: The PEI was intended to generate information from members and non￾members of CBHI on their perceptions of the impact of CBHI on improving access to health care 
services, quality of care, and what lessons can be learned for scale-up of the scheme. Three facilities 
were randomly selected (two health centers and one hospital) from all contracted health facilities in 
each pilot woreda. At least 30 patients (eight outpatients from each hospital and 10 outpatients from 
each health center plus two inpatients from hospitals) were selected for the interview (see Table 
1.4). PEI was conducted using the tool attached as Annex H. Of the total planned 510 PEI, the total 
collected was 462 (91 percent). The shortfall is the result of inadequate availability of patients during 
the data collection process at the facility level. 
Table 1.4: Distribution of Planned and Actual Patients Exit Interviews 
Woreda Planned PEIs Actual PEIs conducted Percent
South Achefer 30 24 80%
Fogera 30 25 83%
Tehuledere 30 27 90%
Dembia (Control woreda) 30 20 67%
Kilte Awlaelo 30 28 93%
Ahferom 30 28 93%
Tahitay Adiabo 30 27 90%
Raya (Control woreda) 30 30 100%
Gimbichu 30 30 100%
Kuyu 30 27 90%
Deder 30 30 100%
Limmu Kossa 30 30 100%
Merti (Control woreda) 30 29 97%
Yirgalem city 30 21 70%
Damboya 30 28 93%
Damot Woyde 30 31 103%
Wonago (control woreda) 30 27 90%
Total 510 462 91%
Source: PEI survey 
25
Key informant interviews: As part of the qualitative survey, KIIs were carried out using a guide 
developed for this purpose and attached as Annex F. The KIIs were intended to gather information 
from people who have better understanding of health care financing in general and CBHI 
implementation in particular on key issues such as management and governance, regulatory 
framework, sensitization, capacity building, and CBHI parameters (premium, benefit packages, 
general and targeted subsidies, etc.) with the purpose of drawing lessons for scale-up of the 
schemes. The interviews were conducted with government authorities and HSFR/HFG project 
managers at federal, regional, and woreda levels. Table 1.5 shows the distribution of KII by region. 
All the planned KIIs were carried out. 
Table 1.5: Distribution of Key Informant Interviews by Region 
Key 
Informant 
interviews
Planned Accomplished 
Amhara Tigray SNNP Oromia Total Amhara Tigray SNNP Oromia Total 
Head of 
woreda
3 3 3 4 13 3 3 3 4 13
Regional level 
(RHB, BOFED; 
HSFR/HFG)
3 3 3 3 12 3 3 3 3 12
Heads at 
woreda level 
(WorHO; 
WOFED or 
two CBHI 
board 
members), 
CBHI 
management 
team)
16 16 16 20 68 16 16 16 20 68
Facility levels 12 12 12 15 51 12 12 12 15 51
Total KIIs 31 31 31 39 144 31 31 31 39 144
Source: Regional reports 
Note: RHB=regional health bureau, BOFED=bureau of finance and economic development, WorHO=woreda health office, WOFED= woreda office of finance 
and economic development 
Focus group discussions: FGDs were carried out with three types of groups to gather qualitative 
information in each pilot woreda. These target groups were members of CBHI schemes (one in each 
pilot woreda), non-members of CBHI schemes (one in each pilot woredas), and the staff of the 
contracted facilities (one in each health center). The objective was to obtain the perceptions of the 
FGD participants on the benefits of CBHI and its perceived positive and negative impacts for 
members and facility staff. All the planned FDGs were carried out. The FGD guide used is presented 
in Annex G. Table 1.6 shows the distribution of FGDs carried out by region. 
Table 1.6: Regional Distribution of FDGs Planned and Carried Out 
FGDs Amhara Tigray SNNP Oromia Sum
Members of CBHI 3 3 3 4 13
Non-members of CBHI 3 3 3 4 13
Health center staff 6 6 6 8 26
Total 12 12 12 16 52
Source: Regional reports 
26
1.4 Data Collection Instruments
The evaluation used structured questionnaires both for household and patient exit interviews, while 
guidelines for key informant interviews (KIIs) and focus group discussions (FGDs). The household 
questionnaire gathered information on households and individuals socio-economic and demographic 
characteristics including household expenditure for consumable goods, health care and their 
participation in the CBHI program. The patient exit interview questionnaire on the other hand 
collected information on patient’s demographic characteristics, health services utilization and 
perceived quality of health care services. 
1.5 Recruitment and Deployment of Enumerators and 
Supervisors
Data collectors were recruited based on their experience in household data collection in general 
and in CBHI in particular. The household survey data collection was conducted independently for 
research by the Ethiopian Economics Association (EEA) and was carried out in three rounds. The 
data collectors recruited and trained for this evaluation survey knew the pilot woredas, and had 
experience on how to work with schemes and the population. The plan was to recruit and deploy a 
total of 51 enumerators (household survey and PEI) and eight supervisors. In the end, 55 
enumerators and 17 supervisors were recruited and deployed, 8 percent and 113 percent more than 
the respective targets. The number and distribution of enumerators and supervisors deployed are 
presented in Table 1.7. 
Table 1.7: Number and Distribution of Deployed Enumerators and Supervisors by Region 
Amhara Tigray SNNP Oromia Total
Data Collectors 
PEI 2 2 2 3 9
HH survey 11 11 11 13 46
Total 13 13 13 16 55
Supervisors 4 4 4 5 17
Total 17 17 17 21 72
All the enumerators and supervisors deployed for the household survey were trained for one and 
half days (December 20 and 21, 2013). PEI enumerators were trained for half a day. 
1.6 Models Specification
The evaluation used the basic model of regression that was used to estimate the impact of CBHI in 
the prepayment scheme in Rwanda (Schneider and Diop 2001), the country that managed to achieve 
almost universal coverage of its population with these schemes. The Rwanda model was used 
because its scheme design elements, such as government commitment by issuing policies and 
strategies and legalizing schemes, scheme establishment at the district level, and government’s 
commitment to subsidize the poor, are similar to what Ethiopia used. 
Accordingly four models were used to investigate the impact of CBHI schemes in the pilot woredas 
based on a set of explanatory variables. These are: 
a. The probability to buy health insurance for specific population groups in the pilot 
woredas; 
b. The probability of using basic health care services by the insured and non-insured 
households; 
27
c. The intensity of using basic health care services by the insured and non-insured 
households; and 
d. The impoverishing effect of OOP health expenditures per episode of illness for all sick 
individuals and for those who sought professional care. 
1.6.1 Demand for Health Insurance
The logit regression model was used to estimate the probability of CBHI enrollment for households 
in pilot woredas to determine the factors that affect enrollment and also assess if the poorest among 
the poor is covered by a CBHI scheme. The model estimated households’ CBHI enrollment 
probability, and the extent to which this decision is influenced by specific socio-demographic and 
economic characteristics. The hypothesis tested was that the CBHI member and non-member 
households do not differ in their socio-economic characteristics. In a logit regression, the dependent 
variable “demand for insurance” Di, will equal 1 if individuals buy insurance, or zero otherwise. 
Formally, the logit model can be written as a linear function of the explanatory variables: 
Li =b1 +b2 X2i +....+bk Xki (1) 
Pi (D for CBHI membership) = 1/(1+1/eLi) (2) 
The second equation shows that the conditional probability to register into CBHI Pi is a non-linear 
function of the explanatory variables Xi, which represents a series of attributes that are assumed to 
have caused a household to buy health insurance membership in the pilot woredas. Learning from 
the experiences of other countries, the explanatory variables used in determining the estimates in 
the model were health status (as proxies by illness), household size, age and sex of the family head, 
education of the family head, total livestock unit, land size covered by crops, per capita expenditure, 
and regional dummies. These variables are by and large relevant to the characteristics of the 
population in the four regions. 
We estimated the unknown coefficients bi which are the weights to each of the households’ socio￾demographic and economic characteristics in the probability that Di = 1 for given Xi. Insurance was 
only an option for those households who live in pilot districts and the regression was performed 
with household survey data from pilot woredas only. The household head is the unit of analysis. 
1.6.2 Access to the Modern Health Care System
Outpatients’ health care utilization was measured for those individuals who reported sickness during 
the four weeks preceding the interview in the household survey, and has responded to the curative 
care questionnaire. 
The probability of CBHI in increasing health care utilization was examined at three different levels: 
outpatient, inpatient, and maternal and child health care. However, the observations for inpatient 
and maternal and child health services were too limited to warrant a regression analysis. The 
evaluation team suggests that the impact of CBHI on maternal and child health be undertaken during 
the scaling-up process. For the outpatient health care utilization, we examined the effect of CBHI 
membership on the probability of a person visiting health facilities when he or she feels ill and on the 
intensity/frequency of health care utilization (number of visits to a health facility). 
To examine the probability of CBHI members using health care facilities when they feel ill relative to 
non-members, we estimated equation 3 using logit regression. The specification of the equation is 
presented below. 
 (3) 
Where, indicates the probability of using health care for individual i (individual visiting health 
facilities when they feel ill), is a dummy variable indicating whether individual i is enrolled in 
Yi
=CBHIi
g + Xi
d +ei
Yi
CBHIi
28
the scheme, X consists of controlling variables such as size of livestock, cultivated land size, sex of 
head of household, distance to health facility, age of head household, household size, education of 
head of household, per capita total expenditure, expenditure/wealth quintile, and region. These 
variables are theoretically and contextually relevant to influence decision to seek care in the 
Ethiopian context. 
To examine whether CBHI members are more likely to frequently visit health facilities when they 
feel ill relative to non-members, we estimated equation 4. Since the dependent variable is a count 
data for which the value zero cannot occur, we can only use either zero-truncated negative binomial 
or zero-truncated Poisson regression. Exploring the data revealed that there is evidence for over 
dispersion and in this case the proper model is zero-truncated negative binomial regression 
(Cameron and Trivedi 2009). The specification of the equation is presented below. 
, (4) 
Where, indicates the intensity/frequency of health care utilization for individual i, is a 
dummy variable indicating whether individual i is enrolled in the scheme, X consists of controlling 
variables such as size of livestock, cultivated land size, sex of head of household, distance to health 
facility, age of head of household, household size, education of head of household, per capita total 
expenditure, expenditure, wealth as measured by tropical livestock unit and land size, economic 
quintile, and region. 
1.6.3 Financial impact of household OOP health expenditures
The methodology employed to measure catastrophic payments for health care has been discussed by 
Wagstaff, Adam and Eddy van Doorslaer (2003). An OOP payment for health care is considered 
catastrophic when the payment exceeds some threshold (Z cat), defined as a fraction of total 
household consumption or non-food consumption. If T represents OOP payments for health care, x 
represents total household expenditure and f(x) stands for food expenditure, then a household is 
said to have incurred catastrophic payments when T/x or T/[x-f(x)] exceeds a specified threshold. 
One of the approaches used to measure catastrophic payments for health care involves analyzing the 
incidence of catastrophic payments, that is, the percentage of households that spend more on health 
care than the threshold, which can be measured by the headcount and ‘catastrophic overshoot.’ 
Table 1.8 shows the definition of catastrophic and their associated thresholds. 
Table 1.8: Definition of Catastrophic Spending and Its Thresholds 
Indicator Definition Source of information
Incidence of catastrophic health 
expenditure due to OOP 
payments
Percentage of population whose health 
expenditures exceeds 10% of total 
expenditures
Estimations using household 
expenditure surveys
Percentage of population whose health 
expenditures exceeds 40% of non-food 
expenditures
Estimations using household 
expenditure surveys
Incidence of impoverishment due 
to OOP payments
Percentage of population whose health 
expenditures put them below the 
poverty line
Estimations using household 
expenditure surveys
Mean positive overshoot of 
catastrophic payments
Average amount by which OOP spending 
exceeds threshold, for those with 
catastrophic payments
Estimations using household 
expenditure surveys
Poverty gap due to OOP 
payments
Average amount by which expenditures 
fall below the poverty line, for those 
impoverished by OOP payment
Estimations using household 
expenditure surveys
Source: Evans et al. (2013) 
yi
=CBHIia + Xib +ei
yi CBHIi
29
In estimating the financial protection impact of CBHI, we tried to measure impoverishment using 
three dependent variables and two thresholds. The dependent variables were total expenditure, 
food expenditure and non-food expenditure. The thresholds used as catastrophic was 104 percent 
for total spending, 40 percent for food and 15 and 25 percent for non-food spending. 
The poverty headcount and the poverty gap were estimated to measure the extent to which 
households have become poorer by making OOP payments for health care. While the poverty 
headcount measures the number of households living below the poverty line as a percentage of total 
households, the poverty gap captures the depth of poverty or the amount by which poor 
households fall short of reaching the poverty line. In so doing we estimated the average prepayment, 
post-payment, and normalized poverty gap for CBHI and non-CBHI households’ using 5-10% 
consumption expenditure used by Soumitra Ghosh, 2010 to our context to the level that it can 
become significant and see if membership to CBHI has an impact in reducing the impoverishing effect 
of OOP payments. 
We also used a logit regression model to explore if there is a link between the CBHI membership 
and OOP caused impoverishment. The explanatory variables used to estimate this include 
membership in CBHI, household size, sex and age of household head, education, total livestock unit 
as a proxy for wealth, land size covered by crops, per capital expenditure, and dummy for regional 
variation. 
1.7 Organization of the Report
This first chapter has presented the background, objectives, methodology, sampling, and data 
collection instruments of the CBHI evaluation. Chapter 2 reviews evidence on the impact of CBHI in 
other countries in terms of increased utilization of health services and financial protection measures. 
Chapter 3 documents the design and implementation of CBHI in Ethiopia and the current status of 
the pilot schemes. Chapter 4 presents the major characteristics of household survey and PEI 
respondents. Chapter 5 presents the findings of the evaluation as per the terms of reference. 
Chapter 6 outlines the lessons learned and policy implication of the findings. 
 
4 These thresholds are internally accepted as can be seen from the SPAAN et al. 2012. 

SECTION 2. 
LITERATURE REVIEW: 
OTHER COUNTRY 
EXPERIENCES ON CBHI 

33
2. LITERATURE REVIEW: 
COUNTRY EXPERIENCES ON CBHI 
2.1 Theoretical/Conceptual Framework of CBHI
CBHI is a risk-pooling approach that tries to spread health costs across households with different 
health profiles to prevent catastrophic expenditures that come with unexpected health events or 
chronic diseases, and enables cross-subsidies from rich to poor populations. In theory CBHI 
schemes have five characteristics: 
 Dynamic risk pooling, where the schemes are organized by and for individuals who share 
common characteristics (geographical, occupational, ethnic, religious, gender, etc.); 
 Solidarity, where risk sharing is as inclusive as possible within a given community and 
membership premiums are independent of individual health risks; 
 Participatory decision making and management; 
 Nonprofit character; and 
 Voluntary affiliation (Soors et al. 2010). 
There is no “one size fits all” strategy for implementing universal health coverage (UHC) in 
developing countries. Colombia, Mexico, and the Republic of Korea used top-down public (tax￾based) financing and SHI without CBHI; Rwanda used CBHI as the main model for reaching the 
informal sector. Wang and Pielemeier (2012) argued that a bottom-up approach may better fit low￾income counties and accelerate progress toward UHC. They provided a three-model framework for 
the development of CBHI schemes in low-income countries as a way to achieve UHC. The 
characteristic of these models, their potential, and their limitations are presented in Figure 2.1 below 
and Annex A.2. 
Figure 2.1: CBHI Development Framework 
Adapted from: Hong Wang and Nancy Pielemeier (2012) 
Level of supports to community
Tax-based/social insurance characteristics
 Government political commitment, 
stewardship, legislation, and funding 
support.
 Regional level professional 
management
 Community level mobilization, abuse 
and fraud control
 Government practical endorsement
 Government subsidy to the poor and to 
catastrophic risk
 Network for management and service delivery
 Community initiation and operation
 Voluntary participation
 Membership contribution
Generic Model
Enhanced Model
Nationwide Model
34
2.2 Empirical Evidence on the Impact of CBHI
In recent years, interest in using CBHI as a vehicle for moving toward UHC has grown. Studies done 
in many countries provide evidence on the determinants of enrollment in CBHI and the impact of 
CBHI membership on health service utilization and financial protection. Recent publications on the 
experience of CBHI and UHC around the world include Acharya et al. (2013), Giedion et al. (2013), 
Lagomarsino et al. (2012), and Spaan et al. (2012), and Lu et al. (2012) studied the impact of CBHI in 
Rwanda. This section relies heavily on these studies to summarize the major findings of CBHI impact 
and provide a better understanding and perspective of the successes and challenge of CBHI schemes 
in other countries. 
The studies reviewed by Acharya (2013) found differences in the determinants of insurance uptake 
but there are some shared findings, including the following: 
 The sex of the head of the household seems not to matter, although there are some cases in 
which female-headed households are more likely to join; 
 There is no clear pattern in other demographic variables, although families with young children 
and families headed by the elderly seem more likely to join; 
 Better-educated households are consistently more likely to join, particularly if household 
member(s) have secondary or higher education degrees; 
 Participation in an insurance program is also consistently correlated with per capita expenditure: 
richer households are more likely to join; 
 Initial conditions, such as chronic illnesses, seem not to influence the decision to join either way; 
and 
 Residence in rural areas and distance from health facilities do not seem to deter households 
from joining insurance programs. Isaac Odeyemi (2014) summarized the positive and negative 
factors that affect CBHI uptake rates in Africa (see Annex B). 
Assessments have been conducted on the impact of CBHI on access to health services. According to 
Gideon et al. (2012), 25 of the 29 studies that analyzed the impact of CBHI schemes on universal 
health coverage (UHC) found favorable and statistically significant impacts on household access (see 
Table 2.1). 
Table 2.1: Reviewed papers on CBHI impact on access to care 
Author/ Reviewer # papers 
reviewed
# of papers 
with access 
indicators
Impact findings
Existence of 
heterogeneous 
impact
Overall 
positive
Varies 
across 
services
Greater 
impact on 
worse-off
Giedion et al. (2012) 41 29 25 14 15 23
Achaya (2013) 14 14 9
This provides international evidence that financial protection schemes like CBHI do improve access 
to and utilization of services. Several papers also suggest that financial protection schemes affect the 
type of care used, moving people from use of self-medication or alternative medicine to formal care. 
And in some cases, schemes have an impact on the type of provider chosen rather than on 
utilization levels (Gideon et al. 2012). 
Yip, W. and W.C.Hsiao (2008) documented that the Rural Mutual Health Care scheme in the 
western and central regions of China has increased the probability of an outpatient visit by 70 
percent and reduced the probability of self-medication by similar percentages, which, according to 
the authors, suggests that enrollees substitute self-medication for formal health care in the absence 
of the scheme. 
35
There is also evidence that suggests that financial protection schemes that focus on price subsidies 
sometimes improve access to and utilization of health services through mechanisms that go beyond 
affordability. In Colombia, for instance, uptake of some services such as prenatal care and 
immunization increased after people enrolled in CBHI schemes although these services were free for 
everyone irrespective of a person’s insurance membership status. The evaluation of Rwanda’s CBHI 
also documented that at the individual level, mutuelles (CBHI schemes) improved utilization among 
the general population, children under five, and women delivering a child. At the provincial level, the 
study found a positive effect of mutuelle coverage on child and maternal care. 
Twenty-three of the 29 studies in the Gideon et al. (2012) review found differences in how CBHI 
affected access and utilization across different population groups (demographic and socio-economic), 
regions, countries, and/or the particular design of the scheme. Nevertheless, the evidence indicates 
that, overall, UHC schemes improve access and utilization of services. 
The positive results of the CBHI schemes in promoting medical care utilization suggests that CBHI 
can be an effective tool for achieving UHC, together with other policy instruments (Lu et al. 2012). 
Shimelis Abebe (2010) also concluded that CBHI membership has a potential to increase health care 
utilization by about 15 percent following an illness episode. The effect is slightly higher for poor 
households than for the non-poor. 
Many studies have also assessed the impact of CBHI on financial protection. According to Giedion et 
al. (2012), SHI and CBHI schemes frequently reduce OOP expenditures, and sometimes prevent 
catastrophic expenditures and impoverishment. Furthermore, the worse-off seem to reap greater 
benefits from SHI and CBHI schemes in terms of lowering their OOP expenditures and improving 
their financial protection. 
Table 2.2: Reviewed papers on impact CBHI on protecting impoverishment 
Author # papers 
reviewed
Impact findings
Overall 
positive 
impact
Positive 
impact on 
reducing 
catastrophic 
health 
expenditure
Impact on 
protecting 
impoverish
ment
Positive 
impact on 
other financial 
protection 
indicators
Greater 
impact 
among the 
worse-off
Giedion et al. 
(2012)
41 15 14 15 2 7
2.3 Success Factors for CBHI Schemes
Soors et al. (2010) reviewed the design and implementation of 16 African and six Asian countries’ 
CBHI experiences and concluded that there is great heterogeneity in institutional designs and 
organizational models for implementing CBHI on both continents. They also documented the huge 
variation in coverage achieved, in terms of breadth, depth, and height. Except in Rwanda and Ghana, 
CBHI in sub-Saharan Africa remains a relatively marginal, although growing phenomenon that 
currently occupies only a minor role in the wider endeavor of achieving UHC. Coverage at the 
country level rarely exceeds a few percent. They also note that most schemes still are – again with 
the notable exceptions of Rwanda and Ghana – the result of scattered local project initiatives heavily 
dependent on support from external organizations. They concluded that the important conditions 
for CBHI to develop and grow are: 
 The existence of a minimal level of (perceived) quality of care in health facilities. Development of 
CBHI must go hand in hand with gradually improving quality on the supply side, with the 
necessary institutional and regulatory environment to steer and control provider behavior. 
36
 The need for adequate organizational structure and practice. The design of the schemes should 
include responsiveness to people’s felt needs, financially soundness, and rationality. 
 The need for political will, clear action plans, national scope of implementation beyond pilot 
project settings, existence of regulatory frameworks, and – last but not least – the unequivocal 
acceptance of the need of subsidies to finance partly or totally the premium for the poorest in 
society. 
 The need for CBHI schemes to join forces as has been shown in Mali and is in progress in 
several African countries. Otherwise, they can make affiliation mandatory, as did Ghana and 
Rwanda. Failing to join forces – as experienced in Senegal – raises a question about the feasibility 
of voluntary solidarity. 
 Affordability is important but may not be enough to positively influence health status. Improving 
affordability is a necessary condition, but careful consideration should be given to other 
dimensions of access. There is thus a need to design complementary interventions for other 
access parameters as financial protection measures are implemented. The impact of the CBHI 
schemes is often greater on the poor than on the non-poor implying that they could be used as 
a vehicle to improve welfare of the poor. 
 The CBHI needs to be an integral part of a national health financing strategy; this strategy should 
ensure that small and independent CBHI schemes must gradually evolve through the three 
stages, from basic model, to enhanced model, and eventually to a nationwide model, along the 
way addressing the inherent limitations of CBHI (Wang and Pielemeier 2012). 
SECTION 3. 
CBHI EVALUATION 
FINDINGS 

3. ETHIOPIAN CBHI 
PILOT SCHEMES: DESIGN AND 
CURRENT STATUS 
3.1 Scheme Design and Parameters
3.1.1 Influence of International Experience on Ethiopia’s CBHI 
Design 
The design of the Ethiopian CBHI pilot learned from the experiences of other countries, especially 
those that are believed to have been successful. The best practices around avoiding small-scale, 
voluntary membership and ensuring the membership of the very poor were considered and 
incorporated in the design the schemes design. The feasibility study that provided the evidence on 
the willingness and ability to pay of the pilot woreda population as well as the readiness of facilities in 
the pilot woredas informed the design parameters (HSFR 2009). The major features of the Ethiopian 
CBHI pilot scheme include: 
 The decision to join the CBHI scheme is made collectively at the kebele/tabia level with the 
direct participation of the kebele/tabia population. However, actual enrollment to scheme 
happens when the household decide to pay contribution. 
 CBHI will have sections in each kebele/tabia. The association of these kebele/tabia sections will 
form the woreda-level CBHI scheme. 
 Woreda CBHIs are integrated and work within the woreda administration office, and will be 
responsible for pooling and administrating CBHI funds, contracting with and processing 
reimbursements for health service providers. To ensure schemes are accountable to their 
members, woreda-level boards will be established that will comprise members of the CBHI. 
 The bylaws, endorsed by each CBHI scheme, provide the legal status supported by the federal 
health insurance strategy and regional CBHI directives. 
 General and targeted subsidies are an integral part of the design of the pilot scheme. A general 
subsidy is provided to all members of the CBHI by the federal government, while a targeted 
subsidy from the regional and woreda governments is provided to the very poor who cannot 
afford to pay the contribution. 
 The federal government has also provided resources for investing in CBHI contracted facilities 
to make sure that the service providers have some acceptable quality of care when services to 
members of the CBHI starts. 
The design of the CBHI pilot, as described above, was informed not only by reviewing best practices 
around the world, but also by study tours to Ghana, Mexico, Rwanda, and Senegal that exposed the 
design team to the different perspectives of what works and what does not and an understanding of 
the comparative CBHI models. This enabled the design team to develop different policy options for 
consideration by the top management of the FMOH. The decision to introduce CBHI pilots in 12 
woredas (later expanded to 13) as well as the different parameters was taken by FMOH. This initially 
was supported by the development and endorsement as well as proclamation of the Ethiopian 
Health Insurance Strategy. The CBHI Directive developed at the federal level served as “Prototype 
Directive” for the regions to adapt and endorse their own directives and guidelines. This helped the 
40
establishment of the pilot CBHI schemes in the 13 woredas. The design also included the definition 
of the four control districts to help see the counterfactual during the evaluation of the pilot 
schemes. 
3.1.2 Major Parameters of Ethiopian CBHI Pilot Schemes
As noted above the federal ministry of health (FMOH) provided the necessary guidance on defining 
the CBHI major parameters based on the findings of the regional feasibility studies and the 
respective regional context. The major CBHI parameters and their regional adaptations are 
presented in Table 3.1. 
Table 3.1: Main Parameters of the Ethiopian CBHI scheme 
No Major 
Parameters Federal Guidelines
Regional Adaptations
Amhara SNNP Oromia Tigray
1 Membership Decision to enroll in 
scheme to be taken by 
kebele/tabia collectively
Membership determined 
collectively at 
kebele/tabia level based 
on majority vote
Membership determined 
collectively at 
kebele/tabia level 
Membership determined 
collectively at kebele/tabia 
level based on majority vote
Membership determined 
collectively at kebele/tabia 
level based on majority vote
Membership also based 
on household level 
Membership limited to 5 HH 
members 
All core family members 
(father, mother, and children 
under 18); households who 
wish to enroll children over 18 
years of age pay Birr 30 per 
child.
2 Registration fee 
and premiums
Registration fee set by 
regions
Fee of Birr 3 per HH Fee of Birr 5 per HH Fee of Birr 5 per HH Fee of Birr 5 per HH
Premium set by regions Premium payment of 144 
per family per year
Premium payment of 
10.50 per month per 
HH= Birr 126 per year
Premium payment of Birr 180/ 
core family members and Birr 
36 /additional noncore family 
member
Birr 132 per household per 
year plus In addition Birr 30 
per person/ year for 
dependents more than18 years 
of age. 
3 Financing the very 
poor (indigents)
Regions and woredas 
pay registration fee and 
premiums of the very 
poor. Payment of this 
contribution covered 
30% by the woreda and 
70% by region.
90% of targeted subsidy 
from the region and 10% 
from the woreda. 
Beneficiaries selected 
with participation of the 
community.
No clear guideline by 
the region. Payment of 
contribution covered 
100% by woreda. 
Beneficiaries selected 
with participation of the 
community.
No clear guideline by the 
region. Payment of 
contribution will be covered 
100% by the woreda. 
Beneficiaries will be selected 
with participation of the 
community to ensure fairness 
and transparency.
70% of targeted subsidy from 
region and 30% from the 
woreda. A maximum of 10% of 
the eligible household will be 
selected.
4 General subsidy Federal government 
finances 25% of 
enrollment 
contributions per year
Federal government 
finances 25% of overall 
enrollment contributions 
per year
Federal government 
finances 25% of overall 
enrollment contributions
per year
Federal government finances 
25% of overall enrollment 
contributions per year
Federal government finances 
25% of overall enrollment 
contributions per year
5 Provider payment 
mechanism
Fee-for-service Fee-for-service Fee-for-service Fee-for-service Fee-for-service.
42
No Major 
Parameters Federal Guidelines
Regional Adaptations
Amhara SNNP Oromia Tigray
6 Housing of CBHI 
scheme
Woreda administration Woreda administration Woreda administration Woreda administration
7 Benefit package All services available in 
health centers and 
hospitals, excluding 
tooth implantation and 
eyeglasses
All services available in 
health centers and 
hospitals, excluding tooth 
implantation and 
eyeglasses
All services available in 
health centers and 
hospitals, excluding 
tooth implantation and 
eyeglasses
All services available in health 
centers and hospitals, 
excluding tooth implantation 
and eyeglasses
All services available in health 
centers and hospitals, 
excluding tooth implantation 
and eyeglasses.
8 Governance and 
management 
Woreda CBHI board 
oversees the initiative. 
The scheme staff is 
employed by woreda 
administration.
Woreda CBHI board 
oversees the initiative. 
The scheme staff is 
employed by woreda 
administration.
Woreda CBHI board 
oversees the initiative. The 
scheme staff is employed by 
woreda administration.
Woreda CBHI board oversees 
the initiative. The WorHO 
assigns the curative core 
process owner as a 
coordinator in addition to his 
duties in the health office. 
Other CBHI executive staffs 
are employed by the woreda 
administration.
Before launching the pilot, extensive training and sensitization was given to policymakers, planners, 
regional and woreda officials, kebele/tabia officials, CBHI executive staffs, community workers, 
community members, and other CBHI stakeholders. Below are listed some of the stakeholders who 
were trained/sensitized before or immediately after the establishment of the schemes: 
 Woreda cabinet and woreda health insurance steering committee (WHISC) members in the 
pilot woredas. 
 Kebele/tabia cabinet and kebele/tabia health insurance initiative committee (KHIIC) members 
(two each from each pilot kebele/tabia), supervisors of health extension workers, WorHO 
representatives, health facility staff, and amateur artists. Woreda cabinet and WHISC members 
were part of the team of trainers. 
 Kebele/tabia- and got-level executive staff, who received training on the CBHI financial 
administration and management system (FAMS). 
 Community sensitization and awareness creation activities were conducted in collaboration with 
implementing partners both at woreda and kebele/tabia levels, using local amateur artists. 
Additional awareness creation activities were also conducted, including kebele/tabia-level 
consultation with community members; production and distribution of posters and leaflets in 
local languages; and organization of a one-day orientation workshop for zonal cabinet members 
in some pilot regions. 
In addition, health providers in the four regions were trained on selected topics including basics of 
health insurance; the rationale for the CBHI program; legal framework documents; the roles and 
responsibilities of health facilities; and contents of contract agreements to be signed between health 
facilities and schemes. Advocacy events were organized for health workers, newly hired health 
extension workers, development agents, schoolteachers, CBHI executive staff, woreda sector 
offices, newly assigned woreda cabinet members, kebele/tabia leaders/managers, and influential 
community members. 
3.2 Establishment and Functioning of CBHI Schemes
3.2.1 Endorsement of bylaws
Stakeholders including RHBs, woreda cabinet and WHISCs members, kebele/tabia cabinet and 
KHIIC members, and HSFR/HFG project staff organized kebele/tabia-level CBHI consultative 
meetings in the four pilot regions to get kebele/tabias’ decisions on whether they wanted to join the 
CBHI schemes and if so to enable them select their representatives for the CBHI General Assembly 
organized at the woreda level. After a thorough discussion, the community in each pilot kebele/tabia 
and woreda unanimously decided to join the scheme and designated his or her delegates for the 
General Assembly. Each General Assembly endorsed the CBHI bylaws decided on the amount of the 
registration fee and annual premium, and fixed the timetable for the collection of premiums. It also 
established a Board of Directors, and officially established the scheme. 
44
3.2.2 CBHI operational templates preparation
CBHI identification cards, voucher pads, and other supporting documents and materials were 
printed by the HSFR project at central level and distributed to the pilot regions/woredas. The 
HSFR/HFG project facilitated the production of CBHI seals and distributed them to each pilot 
woreda. It also prepared a detailed list of equipment and furniture to be purchased for each CBHI 
scheme office (housed in the respective woreda administration offices). 
3.2.3 Bank accounts
Schemes bank accounts were opened with Omo Micro-finance for SNNP and with commercial bank 
of Ethiopia for schemes in other three pilot regions. 
3.2.4 Contracts with health facility
Each region adapted for its own use a prototype agreement that had been developed centrally and 
each scheme entered into contracts with health facilities. 
3.2.5 Monitoring and evaluation system
The RHB in each CBHI pilot regions selected one control woreda, based on socio-economic and 
other factors. Analytical framework and data collection and reporting formats were developed to 
analyze routine monitoring data collected from pilot woredas and produce regional CBHI 
performance reports. 
In all of these preparatory stages as well as in the day-to-day implementation of CBHI schemes to 
date, the technical and financial support role played by HSFR/HFG project was essential – the 
project has been at the forefront of CBHI conceptualization, design, piloting, and pilot expansion. 
3.3 Current Status of CBHI Pilots
3.3.1 Enrollment 
According to the data from the HSFR/HFG project information system, the overall enrollment in the 
pilot schemes is 48 percent, with wide variation by woreda. Enrollment ranges from a low of 25 
percent in Deder to 100 percent (universal enrollment) in Yirgalem5 (see Figure 3.1). The overall 
percentage of households registered as indigents is 7 percent, and ranges from 1 percent in Deder 
to 15 percent in Tehuledere and Yirgalem. 
Figure 3.1: Enrollment Rate (%) and % of Indigent HHs Registered as Members, June 2013 
Source: HSFR/HFG 
 
5 The data collected from pilot woredas during the assessment differs somewhat from the information obtained from 
HSFR/HFG. In Yirgalem, for instance, the enrollment rate based on woreda data included civil servants, pensioners, and 
drop-outs; the data had not been updated for lack of an ICT (information communication technology) person since EFY 
2005 (2012/13). 
62
38
91
25
67
28
41
100
49
67
43
57
45 48
0
20
40
60
80
100
120
Enrollment rate % Indigent HHs
46
Looked at from the regional perspective, households enrollment rate ranges from 61 percent in 
SNNP to 36 percent in Oromia (see Figure 3.2). 
Figure 3.2 Regional Enrollment Rate and % of HHs Registered as Indigents, June 2013 
Source: HSFR/HFG Project 
3.3.2 Health Service Utilization of CBHI Members
The per capita health service utilization rate for CBHI members was 0.7 outpatient visits in EFY 
2005 (2012/13). This is more than double compared with the national per capita utilization rate of 
0.3 visits. The rates in Amhara and SNNP pilot woredas again exceeded those in Oromia and Tigray 
(see Figure 3.3). 
Figure 3.3: Per Capita Health Service Utilization Rate for Year 2012/13 
 Source: HSFR/HFG 
47.4
57.9
36.1
61.2
12.0 11.0
3.0 4.0
0
10
20
30
40
50
60
70
Tigray Amhara Oromia SNNP
Enrollment rate % Indigent HHs covered
1.1 1.1
1.0
0.2
0.4
0.2
0.5
1.2
0.9
1.1
0.3
0.2
0.7 0.7
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
3.3.3 Scheme Finances
The financial condition of the schemes when taking into consideration all financial resources i.e., 
contribution from paying members, general subsidy and targeted subsidy; the financial condition of all 
schemes (13) is so far positive. When contribution collected from paying members was taken into 
account alone, the three schemes namely Fogera, Yirgalem and Damboya can’t cover their 
expenditures for health care services. Yirgalem, for instance, has a current negative balance and 
outstanding liabilities of more than six months. As can be expected, those schemes with higher 
utilization per capita also are financially weaker. 
Table 3.2: Financial Status of CBHI Schemes, June 2013 (Birr) 
Premiums 
collected
Reimbursements 
to health facilities Balance
Amhara S. Achefer 2,881,681.32 2,080,113.95 801,567
Fogera 1,670,205.00 2,451,874.50 -781,670
Tehuledere 3,339,650.00 3,033,377.99 306,272
Total 7,891,536 7,565,366.44 326,170
Oromia Deder 1,293,318.00 647,954.00 645,364
Gimbichu 2,901,874.00 633,014.00 2,268,860
Kuyu 607,019.00 202,990.00 404,029
L. Kossa 1,610,987.00 1,404,014.00 206,973
Total 6,413,198 2,887,972.00 3,525,226
SNNP Yirgalem 406,050.70 635,705.52 -229,655
D. Woyde 1,219,089.40 1,240,137.27 -21,048
Damboya 1,859,072.25 1,713,059.34 146,013
Total 3,484,212 3,588,902.13 -104,690
Tigray Ahferom 2,056,577.00 1,134,207.87 922,369
K/Awlaelo 1,637,561.00 687,016.92 950,544
T/Adiabo 1,184,091.00 1,038,618.49 145,473
Total 4,878,229 2,859,843.28 2,018,386
Grand Total 22,667,175.67 16,902,083.85 5,765,092
Source: HSFR/HFG Project 

49
4. BASIC HOUSEHOLD 
CHARACTERISTICS 
4.1 Basic Household Characteristics
The quantitative data gathered for this evaluation, as described in Chapter 1, came from a survey of 
2,987 households and 462 patient exit interviews. Basic characteristics of sampled households’ are 
presented in table 4.1 (a). Of total sampled households, 81 percent were men and 19 women; and 
89.3 percent were in the productive age group (15-65 years) and nearly 11 percent aged (> 65 
years). Though there are regional variations, nearly 80 percent of HH heads were married, 12.4 
percent widowed, 6.5 percent divorced and nearly 2 percent were single. In regard to religion, about 
61.3 percent of household heads were Orthodox Christians; 23.1 percent Muslims; 14.9 percent 
Protestants; and Catholic, no religion, and others account less than one percent each. In regard to 
ethnic composition, 30.2 percent were Amhara, 26.2 percent Oromo, 23.2 percent Tigryan, 6.6 
percent Kambata, 7 percent Wolayta, Sidama 2.3 and Gedeo 2.6 percent each, and Guraghie 1.5 
percent. Hadiya accounts less than one percent. 
Table 4.1 (a): Characteristics of Sampled Households by Region 
Distribution
Pilot woredas Control 
woreda All Region
Amhara Tigray Oromia SNNP
Gender distribution
Male 518 86.33 415 69.17 703 88.10 462 78.04 332 83.63 2430 81.35
Female 82 13.67 184 30.67 95 11.90 128 21.62 65 16.37 554 18.55
Missing 0 0 1 0.17 0 0 2 0.34 0 0 3 0.10
Total 600 100 600 100 798 100 592 100 397 100 2987 100
Household size
<=5 378 63 370 61.67 460 57.64 283 47.25 202 50.63 1,693 56.51
>5 222 37 230 38.33 337 42.23 309 51.59 197 49.37 1,295 43.22
Missing 0.00 0.00 0.00 0.00 1 0.13 7 1.17 0.00 0.00 8 0.27
Total 600 100 600 100 798 100 599 100 399 100 2996 100
Age Distribution 
Age 1-15 0 0 0 0 0 0 0 0 0 0 0 0
Age 16-65 524 87.33 520 86.67 712 89.22 543 91.72 368 92.70 2667 89.29
Age >65 76 12.67 80 13.33 84 10.53 45 7.60 29 7.30 314 10.51
Missing 0 0 0 0 2 0.25 4 0.68 0 0 6 0.20
Total 600 100 600 100 798 100 592 100 397 100 2987 100
Marital status 
Married 499 83.17 410 68.33 677 84.84 444 75.00 326 82.12 2356 78.88
Single 7 1.17 15 2.50 15 1.88 15 2.53 4 1.01 56 1.87
Divorced 31 5.17 85 14.17 24 3.01 26 4.39 28 7.05 194 6.49
Widow 61 10.17 88 14.67 79 9.90 106 17.91 36 9.07 370 12.39
Missing 2 0.33 2 0.33 3 0.38 1 0.17 3 0.76 11 0.37
50
Distribution
Pilot woredas Control 
woreda All Region
Amhara Tigray Oromia SNNP
Total 600 100 600 600 798 100 592 100 397 100 2987 100
Religion 
Orthodox 399 65.50 564 94.00 447 56.06 173 29.22 254 63.98 1831 61.30
Catholic 0 0 0 0 0 0 10 1.69 0 0 10 0.33
Protestant 1 0.17 2 0.33 14 1.75 364 61.49 64 16.12 445 14.90
Muslim 206 34.33 34 5.67 334 41.85 37 6.25 78 19.65 689 23.07
No religion 0 0 0 0 0 0 1 0.17 0 0 1 0.03
Other 0 0 0 0 2 0.25 6 1.01 1 0.25 9 0.30
Missing 0 0 0 0 1 0.13 1 0.17 0 0 2 0.07
Total 600 100 600 100 798 100 592 100 397 `100 2987 100
Ethnicity
Amhara 600 100 4 0.67 103 12.19 53 8.95 141 35.52 901 30.16
Oromo 0 0 0 685 85.84 16 2.70 80 20.15 781 26.15
Tigryan 0 0 595 99.17 0 0 3 0.51 95 23.93 693 23.20
Gurage 0 0 0 0 2 0.25 42 7.09 0 0 44 1.47
Kambata 0 0 0 0 2 0.25 195 32.94 0 0 197 6.60
Hadiya 0 0 0 0 0 0 0 0 2 0.50 2 0.07
Wolayta 0 0 0 0 0 0 207 34.97 1 0.25 208 6.96
Sidama 0 0 0 0 0 0 69 11.66 0 0 69 2.31
Gedio 0 0 0 0 0 0 0 0 77 19.40 77 2.58
Other 0 0 0 0 5 0.63 5 0.84 1 0.25 11 0.37
Missing 0 0 1 0.17 1 0.13 2 0.34 0 0 4 0.13
Total 600 100 600 100 798 100 592 100 397 100 2987 100
Source: Household Survey 
51
The 2,987 households had 15,633 members; and of these 50.2 percent were men and 49.4 percent 
women. Over half of household members are in the productive age group (16-65 years), 46.1 
percent young and 2.7 percent in the old age group (65 years and above). Of all household members, 
31.1 percent are married, 23 percent single, 3.1 percent widowed, and divorced 2.1 percent. Over 
half of household members are Orthodox Christians, 22.5 percent Muslims, and 17.3 percent 
Protestant and catholic 0.42 percent. In terms of ethnicity, Amhara, Oromo, Tigryan and Kembata 
have highest proportions (4.1 (b)). 
Table 4.2 (b): Characteristics of household members by region 
Distribution
Pilot woredas Control 
woreda All Region
Amhara Tigray Oromia SNNP
Gender distribution:
Male 1516 51.46 1429 49.53 2180 51.74 1610 47.46 1119 50.93 7854 50.24
Female 1430 48.54 1450 50.26 2031 48.21 1739 51.27 1076 48.93 7726 49.42
Missing 0.00 0.00 6 0.21 2 0.05 43 1.27 2 0.09 53 0.34
Total 2946 100 2885 100 4213 100 3392 100 2197 100 15633 100
Household size:
<=5 378 63 370 61.67 460 57.64 283 47.25 202 50.63 1,693 56.51
>5 222 37 230 38.33 337 42.23 309 51.59 197 49.37 1,295 43.22
Missing 0.00 0.00 0.00 0.00 1 0.13 7 1.17 0.00 0.00 8 0.27
Total 600 100 600 100 798 100 599 100 399 100 2996 100
Age Distribution: 
Age 1-15 1,296 4399 1,330 46.10 2,008 47.66 1,495 44.17 1,084 49.34 7,213 46.14
Age 16-65 1,545 52.44 1,457 50.50 2,084 49.47 1,829 54.03 1,066 48.52 7,981 51.05
Age >65 102 3.46 96 3.33 121 2.87 61 1.8 41 1.87 421 2.69
Missing 3 0.10 2 0.07 0.00 0.00 7 0.21 6 0.27 18 0.12
Total 2,946 100 2,885 100 4,213 100 3,392 100 2,197 100 15,615 100
Marital status: 
Married 1,038 35.23 831 28.80 1,386 32.90 924 27.24 681 31.00 4,860 31.09
Single 752 25.53 606 21.01 912 21.65 917 27.03 407 18.53 3,594 22.99
Divorced 67 2.27 114 3.95 53 1.26 44 1.30 42 1.91 320 2.05
Widow 79 2.68 99 3.43 121 2.87 133 3.92 48 2.18 480 3.07
Other 0.00 0.00 4 0.14 0.00 0.00 0.00 0.00 0.00 0.00 4 0.03
Missing 1010 34.28 1231 42.67 1741 41.32 1374 40.51 1019 46.38 6375 40.78
Total 2946 100 2885 100 4213 100 2,020 100 1,173 100 15633 100
Religion:
Orthodox 2123 72.19 2687 93.14 2234 53.03 871 25.68 1,341 61.04 9256 59.21
Catholic 0.00 0.00 0.00 0.00 7 0.17 58 1.71 0.00 0.00 65 0.42
Protestant 7 0.24 2 0.07 73 1.73 2217 65.36 402 18.30 2701 17.28
Muslim 813 27.57 183 6.34 1879 44.60 203 5.98 443 20.16 3524 22.54
No religion 0.00 0.00 0.00 0.00 4 0.09 3 0.09 0.00 0.00 7 0.04
Other 0.00 0.00 0.00 0.00 10 0.24 28 0.83 6 0.27 44 0.28
Missing 0.00 0.00 13 0.45 6 0.14 12 0.35 5 0.23 36 0.23
Total 2946 100 2885 100 4213 100 3392 100 2197 100 15,633 100
52
Distribution
Pilot woredas Control 
woreda All Region
Amhara Tigray Oromia SNNP
Ethnicity:
Amhara 2945 99.97 7 0.24 489 11.61 228 6.72 718 32.68 4387 28.06
Oromo 1 0.03 0.00 0.00 3,663 87.18 82 2.42 458 20.85 4214 26.96
Tigryan 0.00 0.00 2,875 99.76 0.00 0.00 5 0.15 511 23.26 3391 21.69
Gurage 0.00 0.00 0.00 0.00 10 0.24 224 6.60 2 0.09 236 1.51
Kambata 0.00 0.00 0.00 0.00 14 0.33 1272 37.50 0.00 0.00 1286 8.23
Hadiya 0.00 0.00 0.00 0.00 0.00 0.00 2 0.06 6 0.27 8 0.05
Wolayta 0.00 0.00 0.00 0.00 0.00 0.00 1139 33.58 7 0.32 1146 7.33
Sidama 0.00 0.00 0.00 0.00 0.00 0.00 394 11.62 0 0 394 2.52
Gedio 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 482 21.94 482 3.08
Other 0.00 0.00 0.00 0.00 21 0.50 30 0.88 8 0.36 59 0.38
Missing 0.00 0.00 3 0.10 6 0.14 16 0.47 5 0.23 30 0.19
Total 2946 100 2885 100 4204 100 3381 100 2,197 100 15633 100
Source: Household Survey 
Among those who were above school age, 52.2 percent were reported to be literate (able to read 
and write); 40 percent were not able to read or write (Table 4.2). And 51.3 percent have ever 
attended school and 2.1 percent never attended school. Nearly 30 percent completed primary 
education (1-6), 16.4 percent secondary education (7-10), 1.5 percent preparatory education (11-
12), and less than 1.2 percent college diploma and degree. And 41.3 percent were members of the 
CBHI scheme while 56 percent were non-members living both in pilot and control woredas. 
Table 4.2: Educational Characteristics of Household Members Disaggregated by Pilot and 
Control Woredas 
Distribution
Pilot woredas Control 
woreda All Regions
Amhara Tigray Oromia SNNP
Educational status and ever attended formal education
Literate 1,502 52.12 1,463 52.42 2,099 50.75 1,820 54.82 1,097 50.88 7,981 52.21
Illiterate 1,162 40.32 1,104 49.56 1,716 4149 1,231 37.08 879 40.77 6,092 39.86
Missing 218 7.56 224 8.03 321 7.76 269 8.10 180 8.35 1212 7.93
Total 2882 100 2791 100 4126 100 3320 100 1,976 100 15285 100
Ever 
attended 
school?
1,437 49.86 1,453 52.06 2089 50.51 1,812 54.58 1,085 50.32 7,876 51.53
Never 
attended?
109 3.78 43 1.54 56 1.35 72 2.17 43 1.99 323 2.11
Missing 1336 46.36 1295 46.40 1991 48.14 1436 43.25 1028 47.68 7086 46.36
Total 2882 100 1,496 100 4136 100 3320 100 2156 100 15285 100
Highest grade ever completed 
No education 65 2.26 55 1.97 97 2.35 73 2.20 54 2.50 344 2.25
Primary 
education (G 
1-6)
851 29.53 845 30.28 1,213 29.33 994 29.94 613 28.43 4,516 29.55
53
Distribution
Pilot woredas Control 
woreda All Regions
Amhara Tigray Oromia SNNP
Secondary 
education 
(G7-10)
438 15.20 477 17.09 657 15.88 587 17.68 354 16.42 2,513 16.44
Preparatory 
education 
(G11/12)
33 1.15 42 1.50 50 1.21 76 2.29 31 1.44 232 1.52
Certificate 
from 
VET/TTC
9 0.31 4 0.14 7 0.17 12 0.36 3 0.14 35 0.23
College 
diploma and 
degree
29 1.01 29 1.04 36 0.87 62 1.87 24 1.11 180 1.18
Non-formal 
education 
94 3.26 17 0.61 42 1.02 23 0.69 31 1.44 207 1.35
Missing 1363 47.29 1322 47.37 2034 49.18 1493 44.97 1046 48.52 7256 47.48
Total 2882 100 1,469 100 4126 100 3320 100 2156 100 15285 100
Currently attending formal education
Currently 
attending
831 28.83 799 28.63 1,207 29.18 1,029 30.99 655 30.38 4,521 29.58
Not currently 
attending?
685 23.77 665 23.83 909 21.98 798 24.04 462 21.43 3,519 23.02
Missing 1366 47.40 1327 47.55 2020 48.84 1493 44.97 1039 48.19 7245 47.40
Total 2882 100 2791 100 2,116 100 3320 100 2156 100 15285 100
Highest grade currently completed 
No education 14 0.49 13 0.47 12 0.29 13 0.39 7 0.32 59 0.39
Primary
education 
(G 1-6)
530 18.39 503 18.02 756 18.28 618 18.61 399 18.51 2,806 18.36
Secondary 
education 
(G7-10)
259 8.99 251 8.99 379 9.16 331 9.97 228 10.58 1,448 9.47
Preparatory 
education 
(G11/12)
23 0.80 22 0.79 43 1.04 32 0.96 18 0.83 138 0.90
Certificate 
from 
VET/TTC
7 0.24 7 0.25 7 0.17 6 0.18 2 0.09 29 0.19
College 
diploma and 
degree
11 0.38 17 0.61 19 0.46 28 0.84 12 0.56 87 0.57
Non-formal 
education 
1 0.03 2 0.07 3 0.07 10 0.30 2 0.09 18 0.12
Missing 2037 70.68 1976 70.80 2917 70.53 2282 68.73 1488 69.02 10700 70.00
Total 2882 100 2791 100 4136 100 3382 100 2156 100 4,585 100
54
Distribution
Pilot woredas Control 
woreda All Regions
Amhara Tigray Oromia SNNP
CBHI membership status 
CBHI 
member 
1,248 43.30 1,188 42.57 1,762 42.60 1,387 41.78 0 0 6,319 41.34
Non-CBHI 
member 
1,559 54.09 1,534 54.96 2,268 54.84 1,825 54.97 2087 95.80 8,539 55.87
Missing 75 2.60 69 2.47 106 2.56 108 3.25 69 3.20 427 2.79
Total 2882 100 2791 100 4136 100 3320 100 2156 100 15285 100
55
5. CBHI IMPACT ON FINANCIAL 
RISK PROTECTION, HEALTH CARE 
UTILIZATION, AND RESOURCE 
MOBILIZATION 
5.1 Determinants of Enrollment
Respondents in the pilot woredas were asked about their knowledge of the CBHI schemes. Ninety￾five percent of both members and non-members said that they were aware that schemes had been 
established. Households that responded to the question ‘from where did you get information about 
CBHI?’ said they got the information from a neighbor, a CBHI/official, or a house-to-house 
sensitization program. These three sources constitute 100 percent of awareness in Amhara, 81 
percent in Tigray, 96 percent in Oromia, and 86 percent in SNNPR (Table 5.1). Eight-one percent of 
members and 76 percent of non-members responded that they had attended CBHI-related meetings 
before CBHI was implemented in their kebele/tabia. This clearly shows that the intensive 
sensitization work done by government and especially by HSFR/HFG project, as described in the 
preceding section, was effective. 
Table 5.1: Sources of information about CBHI (%) 
Amhara Tigray Oromia SNNP
Neighbor 52 44 25 31
CBHI officials 37 25 62 42
CBHI house-to-house awareness 11 12 9 13
All other sources 0 19 4 14
Source: Household Survey 
The household survey showed that most households in the pilot woredas (both member and non￾member) correctly understand the role and concept of CBHI. As shown in Table 5.2, more than 96 
percent of member and 87 percent of non-member households know that not only the sick should 
enroll in CBHI. A larger percentage of members than non-members also know that both the poor 
and the non-poor should enroll, and that CBHI is not like saving scheme, that is, they will not earn 
interest on their premium payment nor will the premium be returned even if they do not use health 
services, but rather that the premium is a payment to finance future health costs. 
56
Table 5.2: Responses about the Role and Concept of CBHI (%) 
Issues of `understanding CBHI
Correct understanding in pilot 
woredas (in %)
CBHI 
members
CBHI non￾members
Only those who fall sick should consider enrollment in CBHI. (Incorrect) 96 87
Only the very poor who cannot afford to pay for health care need to join 
the schemes. (Incorrect)
90 83
Under the CBHI program, you pay money (premiums) in order for the 
CBHI to finance your future health care needs, if need arise. (Correct)
82 76
The CBHI program is like a savings scheme; you will receive interest and 
get your money back. (Incorrect)
81 65
If you do not make claims through CBHI, your premium will be returned. 
(Incorrect)
81 65
Source: Household survey 
Members were asked why they joined the CBHI scheme. Out of the 1,287 member households 
interviewed, 1,282 responded to the questions for a total of 2,625 responses (a household could 
provide more than one answer to a question). Regarding reasons for enrolling, 37 percent said to 
reduce the OOP payment when seeking care, 35 percent to seek more care so as to improve their 
health status, 18 percent because the premium is lower than the OOP payments, and 4 percent 
because government paid their registration fees and premiums (Table 5.3). The CBHI design in 
Ethiopia states that the decision on whether to join the scheme is taken at the kebele/tabia level, and 
households confirmed that there was no pressure to enroll either the community or the 
kebele/tabia administration during the enrollment process. 
Table 5.3: Reasons for Joining the CBHI Scheme (Multiple Response) (%) 
Amhara Tigray Oromia SNNP Total
Frequent health hazards 5 3 3 2 3
To meet health services requirements 29 36 35 41 35
Cover health expenditure 34 33 41 39 37
Government financed registration and premiums 4 7 1 2 4
Premium is less than OOP payments 23 19 17 13 18
Pressure from the community members 1 0 0 1 1
Pressure from the kebele/tabia administration 4 1 0 1 2
Total (in percent) 100 100 100 100 100
Total number of responses 686 659 744 536 2,625
 Source: Household survey 
57
The survey also asked non-members in the pilot woredas why they did not enroll in the CBHI 
scheme. Out of the total 1,303 non-member households, only 1,252 responded to the question, for 
a total of 1,681 responses (again, a household could provide more than one answer to the question). 
Regarding reasons for not enrolling, 39 percent stated that the registration fees and the premium 
are not affordable; 17 percent didn’t have adequate knowledge and information; 12 percent 
preferred to see the CBHI scheme in action first; and 10 percent stated that the payment schedule is 
not appropriate (Table 5.4). Availability and quality of health services did not seem to greatly affect 
their decision to enroll. Given that nearly 40 percent of eligible households did not enroll due to the 
affordability issue, regions and woredas should explore the feasibility of expanding the fiscal space for 
coverage of indigents to ensure that those without ability to pay are enrolled. 
Table 5.4: Reasons for not enrolling in the CBHI schemes (multiple response) (%) 
Reason for not enrolling (%)
No frequent health hazards 4
Registration and premiums are unaffordable 39
CBHI payment schedule is not appropriate 10
Wanted to see the experience from others 12
Lack of information and knowledge 17
Inadequate availability of health services 1
Low quality of health services 2
Inadequate benefit package 1
Lack of confidence on the CBHI management 2
Other 11
Total (in %) 100
Total number of responses 1,681
Source: Household survey 
Of the 1,287 CBHI member households covered in this survey, 1,169 (91 percent) enrolled in the 
scheme through own contribution. The remaining 9 percent are enrolled either through the local 
government subsidy (5 percent) or a contribution deducted from their “safety net” payment (4 
percent). As can be seen in Table 5.5, there is variation among regions on the extent to which local 
government finance the membership premiums of indigents. From this household survey, Amhara 
and SNNP do seem to perform better. 
Table 5.5: Sources of finance for enrolling into CBHI pilots 
CBHI members in pilot districts: only members
Amhara Tigray Oromia SNNP ALL CBHI
Freq. (%) Freq. (%) Freq. (%) Freq. (%) Freq. (%)
Own contribution 283 89 276 96 374 96 236 81 1,169 91
Deduction from their 
safety net
6 2 2 1 6 2 40 14 54 4
Local government 30 9 11 4 9 2 14 5 64 5
Total 319 100 289 100 389 100 290 100 1,287 100
Source: Household survey 
58
We explored how far the targeted subsidy for CBHI is reaching the very poor when compared with 
the expenditure data we collected during the survey. Figure 5.1 shows that 83 percent of the 
targeted subsidy beneficiary households (indigents/non-paying members) came from the lowest two 
expenditure quintiles. Only 5 percent of the subsidy might have reached the richest two quintiles. 
This shows that beneficiary selection for the targeted subsidy is by and large fair. The major issue 
that emerged during the KIIs and FDGs was that not all indigent households received membership 
cards because they did not provide photographs of each family member or group photograph. 
Figure 5.1: Distribution of Targeted Subsidy by Expenditure Quintiles (in %)
Source: Household survey 
The household survey asked respondents whether they would renew their CBHI membership. 
About 97 percent confirmed that they would renew. This survey asked non-members if they planned 
to join the scheme in the future given what they have seen going on in the woreda; 80 percent of 
non-members said they did have plans to join the scheme in the future. However, some of the non￾members gave inappropriate payment schedule as one of the reasons for not being enrolled. On the 
contrary, 88 percent of the CBHI members confirmed that the payment timing and schedule is 
acceptable (Table 5.4). 
The survey also asked about the affordability of the registration fee and premium. As can be seen in 
Figures 5.2 and 5.3, 84 percent and 83 percent of CBHI members feel that the premiums and 
registration fee are either easily or somewhat affordable; affordability is an issue for only 16 percent 
of registered members. In contrast (as was presented in Table 5.4), affordability is an issue for 39 
percent of non-members. This needs to be explored further to ensure that financial barriers are not 
preventing the very poor from joining. 
60
23
12
3 2
0
10
20
30
40
50
60
70
Expenditure Quintile
1 (the poorest)
Expenditure Quintile
2
Expenditure Quintile
3
Expenditure Quintile
4
Expenditure Quintile
5 (the richest)
59
Figure 5.2: Affordability of Premiums 
Figure 5.3: Affordability of Registration Fee 
Source: Household survey 
The household survey findings about the affordability of fees and premiums are supported by the 
FGDs. All four regional qualitative reports documented that all FGDs with CBHI members found 
that the premium is affordable if they get good services. Two FGD respondents in Damboya woreda 
stated the following: 
 “Some time ago I was admitted to the hospital (Durame) because I had TB, high blood 
pressure, and gastritis. When I was discharged, the total amount due was Birr 1,780, which 
would have been much beyond my means to pay. However, I was discharged without paying a 
single penny thanks to CBHI.” 
“This is something that God has brought about for the poor. I am relieved of a huge expense, 
entering into debt, and paying interest. I have found a shield (protection).” 
25
58 57
53
49
46
33
27
37 35
29
9
16
10
16
Amhara Tigray Oromia SNNP All Pilot Woredas
Easily affordable Somewhat affordable Unaffordable
35
60 62
52 53
41
25
21
36
30
23
15 17
11
17
Amhara Tigray Oromia SNNP All Pilot Woredas
Easily affordable Somewhat affordable Unaffordable
60
The FGDs with non-members had mixed results. Some participants stated that contribution is 
affordable but explained that the services offered at facilities are not good and the services for the 
most common cause of sickness (e.g., malaria) are free and therefore there is no need to pay for 
CBHI. Others say that CBHI is very good initiative but they have other more pressing needs on 
which to spend their money. 
After collecting data on the reasons for enrollment and non-enrollment, sources of financing for 
enrollment, and affordability of registration fees and premiums, the evaluation team used an 
econometric model to explore the factors that determined enrollment. (See Section 1 for a 
description of the model.) 
As was described in the literature review, many factors, including gender of head of the household, 
having families with young children, families headed by the elderly, education, households income, 
and health status of household, influence the decision of households to join CBHI schemes. 
The most significant variables that have positive association with enrollment in the CBHI pilot 
schemes in Ethiopia were found to be size of household, age of the head of the household, 
education, sex and cultivated land size (not own land) (Table 5.6). These determinants are 
statistically significant at a 1 percent degree of significance. Families with more members and with 
older heads are more likely to enroll in the pilot scheme than those who have fewer members and 
older heads. Also more likely to join are those who are literate and especially those who completed 
primary school. Female-headed households are more likely to join than male-headed ones. Another 
important determinant for membership is ownership of livestock as expressed in terms of total 
livestock units, which is statistically significant at 5 percent degree of confidence. The evaluation 
team looked into the regional variation in the enrollment determinants. Taking Amhara region as a 
reference group, the evidence suggests that there is no significant difference in enrollment between 
Amhara, Tigray and Oromia, while enrollment in SNNP is lower than in Amhara. 
 Table 5.6: Determinants of enrollment in CBHI 
VARIABLES (Enrollment in the pilot woreda) 
Model 1
Illness 0.0549
(0.0697)
Household size 0.0533***
(0.0116)
Age of head of family 0.0198***
(0.00115)
Head sex (1/0) -0.259***
(0.0671)
Head literate (1/0) 0.161***
(0.0447)
TLU 0.0127**
(0.00584)
Land size covered by crops 0.0974***
(0.0244)
Per capita expenditure -0.000139
(9.34e-05)
Tigray 0.00234
(0.0641)
Oromia -0.0465
61
VARIABLES (Enrollment in the pilot woreda) 
Model 1
(0.0589)
SNNP -0.119*
(0.0659)
Constant -1.420***
(0.102)
Observations 10,537
Source: Household survey 
Standard errors are in parentheses. *** p<0.01, ** p<0.05, * p<0.1 
5.2 Increased Utilization of Health Care Services
5.2.1 Health Service Utilization by CBHI Members and Non￾members 
The CBHI benefit package includes outpatient and inpatient services, laboratory services, imaging 
services, supply of drugs and related services with the exception of eyeglasses, dental implant, 
dialysis, etc. All government health centers that are situated in the woreda and fulfill the minimum 
standard of service delivery are contracted to provide services to members. When members in the 
pilot woredas live closer to public health centers in neighboring woredas, there are instances where 
these facilities are contracted based on community demand,6 despite the fact that – as stated in KIIs 
with representatives from the World Health Organization (WHO) and CBHI offices – these health 
centers have some gaps in their readiness to provide quality care. This shows that geographic access 
overrides woreda boundaries and/or readiness of health facilities and quality of care. 
All pilot woredas have also signed service contracts with their region’s referral hospitals. Three 
regions (Amhara, Oromia, and Tigray) entered contracts to ensure the possibility of interregional 
referrals. CBHI operational guidelines dictate that members can only get inpatient services free of 
charge if they have a proper written referral paper from a contracted health center and hospitals. 
The only exception to this is Yirgalem hospital, which residents in two Kebeles/Tabias are allowed to 
access directly because of the distance to Yirgalem health center. Members who do not follow the 
referral system are charged 50 percent of the hospital user fee as a bypassing fee. Members also can 
only be referred to the closest referral hospital. The only exceptions are members in (all) pilot 
woredas in Tigray region and Kuyu woreda in Oromia. In Tigray, all members can be referred to 
Mekelle and Ider referral hospitals. In Oromia, Kuyu pilot woreda allows its members to get services 
from Black Lion and Saint Paulos referral hospitals in Addis Ababa (Table 5.7). 
 
6 For example, in Amhara region, CBHI members in Tehuledere woreda can seek services at five health centers 
and Boru Meda hospital, all outside Tehuledere, in addition to the five health centers in Tehuledere. Residents 
of Debub Achefer can use two health centers outside the woreda. Similarly, residents of Damboya in SNNP can 
use one health center outside the woreda. 
62
Table 5.7: Referral system 
Region Pilot woredas Referral hospitals
Amhara South Achefer Felege Hiwot
Fogera Felege Hiwot and Debre Tabor
Tehuledere Dessie
Tigray Kilte Awlaelo Wukro, Mekelle and Ider
Ahferom Axum, Adwa, Mekelle and Ider
Tahitay Adiabo Axum, Suhul, Mekelle and Ider
Oromia Gimbichu Bishoftu and Adama
Kuyu Fichie, Adama, Tikur Anbesa, Saint Paulos
Deder Deder and Dire Dawa
Limmu Kossa Limmu Genet and Jimma
SNNP Yirgalem town Yirgalem 
Damboya Durame
Damot Woyde Wolayta Sodo
Source: KIIs 
Almost all health facilities visited asserted that they provide all services that are expected to be 
delivered at their level. However, some CBHI members are referred to higher-level government or 
private health facilities to access missing services such as CT scan, ultrasound, and in some cases X￾ray. The retained revenue (from fees and CBHI reimbursements) generated by the facilities, 
particularly by the health centers, has enabled facilities to buy some of the required medical 
equipment. 
All KII respondents (heads of visited health facilities, CBHI scheme managers, WorHOs, etc.) felt 
that CBHI has increased health service utilization because of the financial protection it offers, the 
awareness creation efforts made during scheme establishment, and the ongoing health-related 
lessons provided by the facilities, health extension workers, and the Health Development Army. 
FGDs with CBHI members in all pilot woredas confirmed that their health-seeking behavior has 
changed significantly: they not only seek services more, but they also seek them immediately when 
they feel sick. 
The evaluation team tried to confirm this using secondary data collected from contracted heath 
facilities by comparing before and after CBHI. However, lack of consistent and reliable facility-level 
data made the comparison impossible. 
Despite the increase in service utilization, the data from the household survey does not show a 
major difference in service utilization between members and non-members. This needs further 
investigation whether there is a response bias. Out of the 1,464 individuals who reported illness in 
the reference period (the four weeks preceding the survey), 689 were from CBHI member 
households and 583 were from non-member households (Table 5.8). Of those who reported illness, 
1,049 individuals (71.7 percent) reported visiting health facilities. When we disaggregate this to 
members and non-members, 72.3 percent of members had visited health facilities while 69.3 percent 
of non-members from the pilot woreda and 76.6 percent from the control woreda also visited 
health facilities. The higher average from control woredas might have been a result of the smaller 
number of households sampled. The 72 percent average health facility rate by members is very close 
to the number of per capita visits (0.7) reported in Section 2. (We will further explore empirically 
the determinants of health seeking behavior and whether CBHI has an impact on changing the health 
seeking behavior of members relative to non-members (Table 5.13 and 5.14).) 
63
Table 5.8: Seeking Health Care and Number of Visits 
Distribution
Pilot woreda
CBHI Control woreda
Member
Non-CBHI 
member
# (%) # (%) # (%)
Visiting health facilities 
for the illness felt in the 
last 4 weeks
Yes 498 72.3 404 69.3 147 76.6
No 191 27.7 179 30.7 45 23.4
Total 689 100 583 100 192 100
If yes, what type of 
service received
Outpatient 487 97.2 398 98.5 143 97.3
Inpatient 14 2.8 6 1.5 4 2.7
Total 501 100 404 100 147 100
Major reason for not 
visiting health facility
Did not feel it was necessary 90 47.4 75 42.61 22 48.9
Facility too far 9 4.7 8 4.6 0 0
Lack of money 30 15.8 54 30.7 16 35.6
Problem of quality care 14 7.4 12 6.8 2 4.4
Others 46 24.2 25 14.2 5 11.1
Total 190 100 176 100 45 100
Number of visits One time 368 73.5 298 73.0 107 73.8
Two times 73 14.6 64 15.7 12 8.3
Three times 60 12.0 46 11.3 26 17.9
Total 501 100 408 100 145 100
Type of facility visited HP 30 6.0 7 1.2 9 6.2
HC 310 61.9 227 56.5 80 54.8
Private clinic 64 12.8 95 23.6 27 18.5
Public hospital 86 17.2 64 15.9 23 15.6
Total 501 100 402 100 146 100
Source: Household survey 
5.2.2 Service quality and patient satisfaction
We also explored the benefits that CBHI has brought to the quality of care, from the perspectives of 
health professionals, patients, and households. 
In KIIs, health professionals in the visited health facilities asserted that establishment of CBHI 
schemes provided them with some degree of freedom to prescribe the appropriate diagnostic test 
and drugs without any worries about the ability of the CBHI member to pay. They explained that 
prior to CBHI they would either discuss with patients the patient’s financial capacity or write 
prescriptions based on what they thought the patients could afford. CBHI has given them the 
freedom to order the appropriate prescription, which is a relief professionally and key to improve 
the quality of services. 
The other factor that contributed to quality improvement in the health service is the ability of the 
CBHI schemes to mobilize financial resources and hence investment in the facility. As all contracted 
health facilities are implementers of health care financing reform, the resultant increase in patient 
flow due to CBHI scheme has increased the amount of resources in the hands of health facilities, 
allowing them to acquire more drugs, diagnostic facilities, and medical equipment. Furthermore, to 
64
prepare for CBHI, additional resources were provided to contracted health facilities by the woreda, 
regional, and federal governments in cash and in kind to improve the quality of services. 
There are, however, major challenges to the quality of services provided: 
i. Contracted providers differ in their preparedness to deliver contracted services such as 
pharmacy services, laboratory facilities, reception, and other outpatient services, and this 
affects quality. There are also complaints about availability and quality of existing staff. As per 
the “Ethiopian Standard Agency” health centers (both rural and urban) are required to have 
19 professionals and 13 support staff. However, Sulula health center in Tehuledere woreda 
(Rural HC) had 17 health professionals and eight support staff; while Woreta health center 
in Fogera woreda (Urban HC) had 33 health professionals and 24 support staff. This shows 
that variations on the availability of both professional and support staffs between these 
facilities. 
ii. Contracted health facilities, especially in hospitals, experience drug stock-outs and so 
patients often must buy items from outside private retailers. Stock-outs in Pharmaceutical 
Fund and Supply Agency (PFSA) hubs have been the major reason that health facilities give 
for their own drug shortages. Fraud on the part of pharmacists at contracted health facilities 
– the pharmacists refer patients to private drug dealers who give them a “kickback” – is 
reported to exacerbate the shortage.7
These problems pose a major challenge to CBHI 
members, greater than to the general community. 
Paying out of pocket is a problem for members 
who have already paid a registration fee and 
premium and may have no household budget left 
for these contracted services. Even if a member 
has cash on hand, there are cases in SNNP where 
scheme bylaws prevent members from being 
reimbursed; in the other three regions, some 
members have received reimbursement at best and 
some none at all depending on whether they 
submit supporting documents. System-related 
challenges like the following are major problems: 
Reimbursement first requires the pharmacist to 
put a stamp on the back of the prescription and 
record/register it in the member patient’s file. 
Usually one of these procedures is not done. As a 
result, the member’s reimbursement claim is 
rejected, leaving the member highly dissatisfied.8
Members patronize private drug dealers without 
full information on what is required for 
reimbursement (essentially, getting a receipt and 
submitting it with the prescription to the CBHI 
scheme), and again failure to follow procedures 
precludes reimbursement by the scheme. 
 
7
This has been reported about one referral hospital in Amhara and one pilot woreda in Tigray. 
8
 In KIIs, representatives of CBHI woreda offices and WorHOs and heads of contracted health facilities reported 
that health facilities failed to respect the contract, which requires health facilities to provide certain services. If 
a facility cannot do this, it supposed to contract with a third party to provide the service/s so that the CBHI 
member does not need to spend out of pocket. 
Box 5.1: Best practices in handling CBHI 
member complaints
Amhara Region
In Amhara, woreda CBHI coordinators are trying 
to use CBHI General Assembly, board and 
annual meetings as public fora in which to hear 
and discuss members’ comments and complaints. 
The FGDs participants in Tehuledere woreda 
also witnessed that these forums served this 
purpose and resulted in significant improvement 
in health service delivery both in the HCs and 
Dessie hospital.
SNNP Region
Although not specifically for CBHI, quarterly 
health facility forums chaired by WorHO Heads 
and Chief Administrators and attended by health 
facility managers and staffs, hear community 
feedback about the quality of the services 
provided in health facilities. Issues that can be 
resolved during the meeting are addressed 
immediately while those that need time are 
recorded, action taken later, and feedback 
provided to the community at the next forum. 
This has significantly improved service availability 
and quality in health facilities.
65
Reimbursement requires members to travel to the woreda town – sometimes multiple times – on 
workdays to process and resolve issues regarding the reimbursement request. 
iii. Medical equipment frequently breaks down, due mainly to absence of timely and proper 
maintenance but also negligence and mishandling by health workers. 
iv. Neither CBHI offices nor WorHOs have a systematic mechanism by which to measure 
CBHI member satisfaction and collect and properly address complaints about the services 
provided by the schemes or the contracted health facilities. There are, however, best 
practices from which the woredas and regions can learn. Amhara (particularly in Tehuledere 
woreda) and SNNP regions used general assembly, board and annual meetings; and 
community fora respectively to treat this issue in an informal, non-standard way in the 
presence of health providers.
This assessment used PEIs to evaluate patients’ perceptions about improvements in service quality 
since the CBHI schemes were established. Both CBHI members and non-members who were 
interviewed said they have seen improvements. As shown in Table 5.9, more CBHI members than 
non-members perceive improvement in all measures of quality of care. 
Table 5.9: Patients’ Perception of Improvement in Quality of Health Service Provision (%) 
Improvement area
CBHI members Non-CBHI non-members
Yes No Don’t 
know
Yes No Don’t 
know
Improvement in the overall quality 73.9 12.5 13.6 54.4 10.7 34.9
Availability of drugs/medical supplies 71.2 14.7 14.1 56.8 7.5 35.5
Availability of diagnostic facilities 73.2 12.8 14.0 52.7 10.3 37.0
Cleanliness of the facility 75.0 11.4 13.6 52.7 12.4 34.9
Short waiting time (from the time of arrival 
in the health facility up to seeing the health 
professional)
71.6 12.6 15.8 48.5 16.6 34.9
Short waiting time between services 
(between consultation and diagnostics)
72.1 12.0 15.8 48.5 17.2 34.3
Friendliness of staff 71.6 15.3 13.1 52.4 13.1 34.5
Attentiveness and adequate follow-up by the 
nursing staff (inpatient only)
87.5 0 12.5
Quality of food and other inpatient facilities 
(inpatient only)
87.5 0 12.5
Source: PEIs 
66
The PEIs also asked patients to rate their level of satisfaction, from very satisfied to very dissatisfied, 
in the above areas. About 90 percent of respondents were either satisfied or very satisfied in all 
quality improvement areas. As seen in Table 5.10, there was no significant difference between 
members and non-members. 
Table 5.10: Patients’ level of satisfaction with service quality (%)
Quality 
measures
CBHI members Non-CBHI non-members
Very 
satisfied Satisfied Neutral
Dissatisfied 
and very 
dissatisfied
Very 
satisfied Satisfied Neutral
Dissatisfied 
and very 
dissatisfied
Improvement in the 
overall quality of 
service
54.9 36.4 5.4 3.3 50.4 38.5 6.8 4.4
Availability of 
drugs/medical 
supplies
54.1 37.7 4.9 3.3 50.9 40.8 6.5 1.8
Availability of 
diagnostic facilities
54.3 36.7 4.5 4.5 48.3 42.1 7.0 2.6
Cleanliness of the 
facility
52.7 38.6 6.5 2.2 48.9 41.3 6.5 3.3
Short waiting time 
(from the time of 
arrival in the health 
facility up to seeing 
the health 
professional)
52.5 36.1 6.5 4.9 44.6 39.6 8.3 7.5
Short waiting time 
between services 
(between 
consultation and 
diagnostics)
56.0 35.3 6.0 2.7 46.0 39.9 7.6 6.5
Friendliness of staff 56.8 35.5 5.5 2.2 50.9 41.2 4.3 3.6
Attentiveness and 
adequate follow-up 
by the nursing staff 
(inpatient only)
75.0 18.7 6.3 0 87.5 0 0 12.5
Quality of food and 
other inpatient 
facilities (inpatient 
only)
68.7 12.5 6.3 12.5 62.5 12.5 25.0 0
Source: PEIs 
67
The household survey also assessed patients’ perceptions about improvements in service quality. In 
regard to outpatient visits, more than 80 percent of the CBHI members were satisfied or very 
satisfied by the diagnosis, the cleanliness of the facility, and the courtesy of the staff. Non-CBHI 
members reported similar satisfaction levels (see Table 5.11). 
Table 5.11: Outpatients’ levels of satisfaction
Distribution
Pilot woreda
Control 
CBHI member Non woreda -CBHI 
member
Freq. (%) Freq. (%) Freq. (%)
Level of 
satisfaction
Diagnosis Very satisfied 148 30.1 140 35.1 51 35.2
Satisfied 260 52.8 199 49.9 70 48.3
Indifferent 51 10.4 43 10.8 15 10.3
Dissatisfied 28 5.7 13 3.2 8 5.5
Very dissatisfied 5 1.0 4 1.0 1 0.7
Total 492 100 399 100 145 100
Cleanliness Very satisfied 133 27.0 121 30.3 46 31.5
Satisfied 291 59.2 232 58.2 85 58.2
Indifferent 53 10.8 38 9.5 14 9.6
Dissatisfied 11 2.2 6 1.5 1 0.7
Very dissatisfied 4 0.8 2 0.5 0 0
Total 492 100 399 100 146 100
Courteousness Very satisfied 157 31.9 149 37.3 46 31.5
Satisfied 249 50.6 189 47.4 72 49.3
Indifferent 50 10.2 45 11.3 17 11.6
Dissatisfied 28 5.7 12 3.0 9 6.2
Very dissatisfied 8 1.6 4 1.0 2 1.4
Total 492 100 399 100 146 100
Waiting time Very satisfied 138 28.1 126 31.6 47 32.2
Satisfied 192 39.0 163 40.8 59 40.4
Indifferent 86 17.5 55 13.8 24 16.4
Dissatisfied 63 12.8 45 11.3 14 9.6
Very dissatisfied 13 2.6 10 2.5 2 1.4
Total 492 100 399 100 146 100
Availability of 
drug
Very satisfied 133 27.3 120 30.2 41 27.9
Satisfied 251 51.5 192 48.2 74 50.3
Indifferent 50 10.3 53 13.3 15 10.2
Dissatisfied 39 8.0 28 7.0 15 10.2
Very dissatisfied 14 2.9 5 1.3 2 1.4
Total 487 100 398 100 147 100
68
Distribution
Pilot woreda
Control 
CBHI member Non woreda -CBHI 
member
Freq. (%) Freq. (%) Freq. (%)
Waiting time
Less than 30 min 262 54.2 250 62.6 91 61.9
30-60 min 143 29.6 96 24.0 30 20.4
1-3 hours 52 10.8 31 7.8 19 12.9
3-6 hours 20 4.2 11 2.8 4 2.7
>6 hours 6 1.2 11 2.8 3 2.1
Total 483 100 399 100 147 100
Availability of drugs/supplies
Not available 32 6.9 12 3.0 4 2.9
Rarely 171 36.6 131 33.1 54 38.8
Always 264 56.5 253 63.9 81 58.3
Total 467 100 396 100 139 100
Source: Household survey 
The household survey also queried respondents about their level of satisfaction on services they 
received during their admission some time ago. As seen in Table 5.12, there is no significant 
difference between the responses of CBHI members and non-members. Indeed, most non-members 
were more satisfied than members. This might be related to higher expectations of members or 
response bias of non-members (the tendency to provide positive responses even if that is not true 
response). 
Table 5.12: Inpatients’ level of satisfaction
Distribution
Pilot woreda
Control
CBHI members woreda Non-CBHI 
members
Freq. (%) Freq. (%) Freq. (%)
Diagnosis Very satisfied 20 50 12 48 8 61.54
Satisfied 11 27.5 9 36 3 23.08
Indifferent 5 12.5 3 12 0 0
Dissatisfied 2 5 1 4.00 1 7.69
Very dissatisfied 2 5 0 0.00 1 7.69
Total 40 100 25 100.00 13 100
Cleanliness Very satisfied 24 39.34 15 36.59 10 58.82
Satisfied 22 36.07 22 53.66 6 35.29
Indifferent 11 18.03 2 4.88 1 5.88
Dissatisfied 4 6.56 1 2.44 0 0.00
Very dissatisfied 0 0 1 2.44 0 0.00
Total 61 100 41 100 17 100
69
Distribution
Pilot woreda
Control
CBHI members woreda Non-CBHI 
members
Freq. (%) Freq. (%) Freq. (%)
Courteousness Very satisfied 22 36.67 13 31.71 10 58.82
Satisfied 25 41.67 24 58.54 7 41.18
Indifferent 8 13.33 3 7.32 0 0
Dissatisfied 3 5 1 2.44 0 0
Very dissatisfied 2 3.33 0 0 0 0
Total 60 100 41 100 17 100
Waiting time
Very satisfied 21 34.43 15 36.59 10 58.82
Satisfied 27 44.26 22 53.66 6 35.29
Indifferent 9 14.75 3 7.32 0 0
Dissatisfied 2 3.28 0 0 1 5.88
Very dissatisfied 2 3.28 1 2.44 0 0
Total 61 100.00 41 100 17 100
Drug supplies
Very satisfied 21 34.43 11 26.83 8 47.06
Satisfied 20 32.79 25 60.98 6 35.29
Indifferent 14 22.95 1 2.44 3 17.65
Dissatisfied 5 8.20 4 9.76 0 0
Very dissatisfied 1 1.64 0 0 0 0
Total 61 100 41 100 17 100
Bed cleanliness
Very satisfied 16 26.23 13 31.71 8 47.06
Satisfied 22 36.07 22 53.66 6 35.29
Indifferent 17 27.87 4 9.76 3 17.65
Dissatisfied 6 9.84 2 4.88 0 0
Very dissatisfied 0 0 0 0 0 0
Total 61 100 41 100 17 100
Staff care
Very satisfied 18 29.51 12 29.27 9 52.94
Satisfied 30 49.18 26 63.41 6 35.29
Indifferent 9 14.75 2 4.88 2 11.76
Dissatisfied 2 3.28 1 2.44 0 0
Very dissatisfied 2 3.28 0 0 0 0
Total 61 100 41 100 17 100
Source: Household survey 
70
The above points are anecdotal. In addition to collecting such information, the assessment also tried 
to estimate the impact of CBHI on health care utilization by employing an econometric method as 
discussed in Section 1.3.2. Equation 3 is estimated using three different control groups (non￾members in pilot, control and both pilot and control woredas), individuals: model 1 compares CBHI 
members against non-members in pilot woredas, model 2 compares members in pilot woredas 
against non-members in the control woreda; and model 3 compares members in pilot woredas 
against all non-members in both pilot and control woredas using a logistic regression model. As 
shown in Table 5.13, CBHI members are more likely to seek health care when they feel ill. It is 
estimated that a CBHI member visiting a health facility when falling sick is 26.3 percent more likely 
than a non-members doing so in the pilot woredas and 20 percent more likely than overall non￾members (in pilot and control woredas). This is consistent with the findings reported in the 
descriptive statistics. 
Table 5.13: Regression results of health care utilization 
Variables Model 1 Model 2 Model 3
CBHI status 0.263* -0.286 0.201
(0.144) (0.255) (0.140)
Household size 0.0858** 0.120** 0.0987***
(0.0351) (0.0565) (0.0337)
Head age -0.0111** 0.00288 -0.00905**
(0.00481) (0.00746) (0.00461)
Tropical Livestock unit -0.0431** -0.00312 -0.0364**
(0.0185) (0.0291) (0.0181)
Constant 0.874** 0.126 0.606*
(0.381) (0.603) (0.358)
Observations 1,051 519 1,173
Source: Household survey 
Note: For readability purposes, we did not report the values of the coefficient of the following explanatory variables: household head sex and literacy, per capita 
expenditure, land size covered by crops, and regional variations. All are statistically insignificant.
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 
71
The frequency of health care utilization (Equation 4) is estimated using same three control groups, 
using zero truncated negative binomial model (see Section 1.3.2 for the explanation). As observed in 
Table 5.14, CBHI membership has significantly increased the intensity/frequency of health care 
utilization. This does not hold true when CBHI members are compared against non-members from 
the control woreda. The coefficient for CBHI, 0.31, is statistically significant at 5 percent level and 
indicates that the log count of the intensity of health facility visits for CBHI members is 0.31 more 
than for non-members in the pilot woreda. 
Table 5.14: Regression results of frequency of health care utilization 
Variables (Model 1) (Model 2) (Model 3)
CBHI status 0.310** -0.0790 0.230*
(0.135) (0.178) (0.130)
Head literate 0.191 0.346** 0.257*
(0.146) (0.171) (0.137)
Land size covered by crops -0.174* -0.214* -0.123
(0.0957) (0.120) (0.0831)
Tigray -0.117 -0.387 -0.262
(0.220) (0.267) (0.213)
Oromia 0.425** 0.311 0.424**
(0.194) (0.227) (0.184)
SNNP 0.208 0.0125 0.246
(0.210) (0.243) (0.196)
Constant -0.455 -0.114 -0.536
(0.417) (0.475) (0.397)
Observations
Alpha 
(LR test against Poisson, chi2(1)
738
0.40
667.55***
382
0.11
308.64***
832
.47
739.33***
Source: Household survey 
Note: For readability purposes, we did not report the values of the coefficient of the following explanatory variables: Per capita expenditure, household size, sex 
and age of household head, and TLU. All are statistically insignificant. 
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 
5.2.3 Service provider and patients’ moral hazard
A major issue that emerged from KIIs with heads of contracted health facilities, WorHOs, and CBHI 
offices and FGDs with CBHI members and contracted health facility workers was moral hazard by 
both providers and CBHI members. Provider staffs raised a number of ways in which some CBHI 
members, who no longer need to pay at point of service, misperceive their CBHI entitlements and 
overuse health care services: 
 Members are rude in demanding their rights; for example, some expect to be seen without 
waiting their turn in the queue; 
 They demand to be prescribed a particular drug without waiting for the decision of the health 
worker; 
 They return to the facility before finishing the prescribed drug regimen and demand another 
treatment; 
 They demand unnecessary diagnostics, injections, and prescriptions; 
72
 They ask for drugs for their children without bringing the child in for a consultation and 
diagnosis; 
 They demand referral before using services of a lower-level facility; 
 They demand services without presenting a valid membership card. 
Providers also pointed out that CBHI has increased their workload without increasing their financial 
(salary) and/or non-financial compensation. 
There also have been cases of reported moral hazard on the part of providers and their staff, such as 
the following: 
 Lack of courtesy and poor treatment by some staff, especially at the health center level; 
 Over-prescription of services including drugs and diagnostics; prescribing drugs with similar 
effect/function at the same time; prescribing the same medicine repeatedly although it is not 
helping the patient; 
 Unnecessary referral to private wings and private clinics or else providing long appointments etc. 
at the hospital level; and 
 Irregularities during medical audit. 
5.3 Improving Financial Access/Financial Risk Protection
Of the total CBHI members surveyed in this evaluation, 82 percent have used their membership to 
access health care. When asked what benefits they gained as a result of membership, 40 percent 
stated that CBHI reduced their OOP payments; another 36 percent stated that it will reduce the 
risk of future spending on health, and another 24 percent responded that it increased their access to 
care (see Figure 5.4). 
Figure 5.4: Reported Benefits of CBHI Membership 
Source: Household survey 
21
28 26
19
24
38 35
42 44
40 40
35 32
36 36
1 1 0 0 0
0
10
20
30
40
50
Amhara Tigray Oromiya SNNP ALL CBHI pilot
woredas
For Accessin health care Reducing OOPs Reducing future financial risk Other
73
The analysis of the PEIs shows that very few CBHI members incurred OOP expenses during their 
visit to a facility. Of the total 184 members interviewed, only four paid for the consultation, three 
for diagnostics, and five for drugs. In contrast, of the 278 non-CBHI members interviewed in the 
pilot and control woredas, 244 paid for the consultation, 110 for diagnostics, and 249 for drugs (see 
Table 5.15). The average per person payment of members (Birr 13) is half that of non-members (Birr 
26). This is likely due to the fact that there are few observations relating to drugs and inpatient 
services with average payments of ETB 224.9 and 766 respectively that biased the mean in the case 
of CBHI members. 
Table 5.15: Incidence of and Average OOP Payments by CBHI Members, from PEI Survey 
Service for which OOP 
payment incurred
CBHI members Non-CBHI members
# cases 
with 
OOP 
payment
Average 
payment 
(Birr)
% of 
total 
observat
ion
# cases 
with 
OOP 
payment
Average 
payment 
(Birr)
% of 
total 
observat
ion
Consultation 4 9.3 2.2% 244 6.4 87.8%
Diagnosis 3 45.0 1.6% 110 23.4 39.6%
Drugs 5 225.0 2.7% 249 42.5 89.6%
Inpatient 1 766.0 0.5% 5 132.0 1.8%
Transport 7 41.4 3.8% 16 31.0 5.8%
Total PEI 
respondents/observations
184 12.8 278 25.8
Source: PEIs 
Similarly, we estimated and compared the incidence of OOP payments and average OOP payments 
between members and non-members using descriptive statistics (see Table 5.16). The incidence of 
OOP payment was more for non-members for all service categories. The difference here is that in 
all categories, the average OOP payment was lower for members than for non-members. Members 
pay on the average about half of non-member OOP payments when accessing care. 
Table 5.16: Incidence of and average OOP payments made by CBHI members and non￾members, household survey 
Service for which 
OOP payment 
incurred
CBHI members CBHI non-member
# HHs reported 
OOP
Average 
payment (Birr)
# HHs reported 
OOP
Average 
payment 
(in Birr)
Transport 59 20.02 68 25.72
Consultation 36 1.25 59 9.6
Diagnosis 35 19.86 51 50.96
Drug 40 41.03 58 80.02
Total HH members 
visited facility 
159 22.40 124 44.65
Source: Household survey 
74
Given the above descriptive statistics, it is necessary to go beyond the averages. For that, we tried 
to explore the extent to which households risk falling below a defined poverty threshold due to 
OOP health expenditure and the changes in the magnitude and depth of impoverishment. 
By using the non-food consumption expenditure, payment for health (including transport) showed a 
significant impact on household impoverishment. At a 15 percent threshold, about 7 percent of 
CBHI members are impoverished by OOP payments on health. The corresponding impoverishment 
level for non-members is 19 percent. The figures at a 25 percent threshold are 3 percent for 
members and 9 percent for non-members. This shows that CBHI members have less chance to be 
impoverished as a result of OOP payments than non-members. The same analysis goes for poverty 
overshoot, the intensity of poverty (see Table 5.17). 
Table 5.17: Poverty head count and overshoot between members and non-members due to 
OOP payments at 15% and 25% non-food expenditure thresholds 
As a share of non-food expenditure
15% 25%
Mean Std. Err. Mean Std. Err.
CBHI members
Head count 6.83% 0.019 3.11% 0.0039
Overshoot 1.37% 0.007 0.93% 0.006
Mean positive overshoot 40.09% 0.147 30.09% 0.147
Non-CBHI members
Head count 19.38% 0.040 9.18% 0.029
Overshoot 3.80% 0.0135 2.48% 0.011
Mean positive overshoot 37.09% 0.091 27.09% 0.091
Source: Household survey 
We also attempted to estimate the head count, poverty gap, and normalized poverty gap for 
members and non-members by adopting Soumitra Ghosh’s 2010 model (Ghosh 2010). In this model, 
while the poverty headcount measures the number of households living below the poverty line as a 
percentage of total households, the poverty gap captures the depth of poverty or the amount by 
which poor households fall short of reaching the poverty line. Since there are no universally 
accepted cut-off values or thresholds for defining the catastrophic nature of health care payments, 
generally, the catastrophic headcount has been defined as the percentage of households spending 
more than 5–25 percent of their total consumption expenditure on health care. However, it is 
evident from other empirical studies that 10 percent of total expenditure is widely accepted as the 
standard, because this represents an approximate threshold at which the household is forced to cut 
down on subsistence needs, sell productive assets, incur debts, or be impoverished (Van Doorslaer 
et al. 2006). In this evaluation, we used half and two-thirds of the median adult equivalent 
consumption expenditure as a cut-off point to calculate the head counts and poverty gaps. 
75
Our estimate clearly shows that the poverty impact of OOP payments is less for CBHI members 
than for non-members. The estimate shows that an additional 1 percent of CBHI member 
households fell below the threshold poverty as a result of OOP payments. For CBHI non-members, 
it was 5.5 percent. The poverty gap for members was Birr 58, for non-members Birr 143. Table 5.18 
presents the estimated head counts and poverty gaps for members and non-members. 
Table 5.18: Estimated Impact of OOP Payments on Poverty Headcount and Poverty Gap on 
CBHI Members and Non-members 
Source: Household survey 
Note: HP=poverty head count, G=poverty gap, NG=normalized poverty gap 
1 Members and non-members at household level in the pilot woreda 
2 Half of the median of the respective per adult equivalent consumption expenditure 
3 Two-thirds of the median of the respective per adult equivalent consumption expenditure
To analyze catastrophic health expenditures in more detail, a regression analysis, based on a 
dichotomous choice (logistical regression) model, was developed to predict the probability of 
catastrophic health expenditures in households. The share of health care expenditure in non-food 
expenditure is equal to 1 if the share is greater than a 15 percent (model 1) and 25 percent (model 
2) threshold level and 0 otherwise. Our estimate clearly shows that, controlling for other factors, 
being a member is negatively related to impoverishment due to OOP payments. The estimate is 
significant at 1 percent degree of confidence (see Table 5.19). 
Table 5.19: Regression result of the impact of CBHI membership on levels of impoverishment 
Variable (Cut off 15%) (Cut off 25%)
Model 1 Model 2
CBHI membership status (1=CBHI member) -1.405*** -1.347*
(0.532) (0.736)
Household size 0.0650 0.0474
(0.133) (0.180)
Sex of head of HH (1=Male) -0.282 -0.686
(0.622) (0.798)
Poverty measures CBHI member Non- member All
Poverty Headcount (in %)
Poverty line 1/2 2
2/33
1/2 2/3 1/2 2/3
Pre-payment headcount (pre-HP) 13.75 28.14 17.35 27.55 14.73 28.68
Post-payment headcount (post-HP) 14.94 26.62 22.83 31.52 18.29 31.30
Poverty impact-head count (post-HP- pre-HP) 1.19 -1.52 5.48 3.97 3.56 2.62
Poverty gaps (in birr)
Prepayment gap (Pre-G) 77.14 220.15 127.63 328.11 90.70 259.18
Post payment gap (Post-G) 134.81 360.93 270.55 647.49 167.08 452.65
Poverty impact-gap (Post-G- Pre-G) 57.67 140.78 142.92 319.38 76.38 193.47
Normalized poverty gaps (in %)
Prepayment normalized gap (pre-NG) 3.57 7.634 4.77 9.20 3.89 8.34
Post payment normalized gap (Post-NG) 4.34 8.71 6.54 11.74 4.86 9.89
Normalized poverty impact 
(Post-NG- pre-NG)
0.77 1.076 1.77 2.54 0.97 1.55
76
Box 5.2: Best experience in CBHI scheme ownership and leadership, Damot Woyde, SNNP
Damot Woyde woreda is the best performer in terms of giving guidance to and ensuring the sustainability of its 
CBHI scheme. There is good ownership at the woreda level. There are regular Board of Directors and General 
Assembly meetings and quarterly reviews of performance at which regional, woreda, kebele/Tabia, and 
community representatives discuss progress and challenges with the CBHI scheme. In addition, the various 
officials at woreda level get up-to-date information about CBHI performance as confirmed during KIIs. All KII 
respondents agreed that the woreda chief administrator is quite strong and experienced in leading CBHI issues. 
They also stated that there is good follow-up on CBHI, for example: 1) each woreda cabinet member is given 
CBHI as a major priority area to closely follow-up; and 2) there is a monthly review of premium collection and 
the deposit from each kebele.
Such strong ownership and commitment by woreda officials has translated into immediate results. Accordingly, 
the Damot Woyde scheme is the only scheme in SNNP that is operating at a surplus. The scheme also pays 
health facilities on time. In addition, a visit to CBHI office shows that it is well organized, with good record 
keeping and compilation capacity. All required data were readily available and compiled into six-month and 
annual performance reports, something not seen in any other pilot woreda in the SNNP. 
Variable (Cut off 15%) (Cut off 25%)
Model 1 Model 2
Age of head of HH 0.0357** 0.0221
(0.0174) (0.0227)
Literacy level of HH head (1=literate) -0.572 -0.659
(0.622) (0.910)
TLU -0.0804 -0.180
(0.124) (0.214)
Land size covered by crops -0.918* -1.338
(0.543) (1.004)
Per capita expenditure 0.000838 0.000115
(0.000995) (0.00141)
Tigray -0.429 -0.687
(0.841) (1.338)
Oromia 0.400 0.530
(0.718) (1.036)
SNNP 0.671 0.874
(0.720) (0.985)
Constant -2.713** -1.853
Source: Household survey 
Standard errors are in parentheses. *** p<0.01, ** p<0.05, * p<0.1 
All three estimates clearly show that OOP payments in general have an impoverishing impact on 
households. However, the impact on CBHI members is much less than that of the non-members. 
With the likelihood that user fees will be revised upward to reflect the changes in the cost of care, 
this impact will likely grow. The analysis, therefore, provides evidence that scaling up CBHI schemes 
in other woredas will have a beneficial pay-off by reducing the incidence and severity of poverty in 
those woredas. 
77
5.4 Roles of Different Parties on CBHI Design, 
Implementation and Management
5.4.1 Governance and Management of CBHI Schemes
CBHI regional steering committees have been functional lately. Zones had little or no role in the 
CBHI implementation, though this is changed in the pilot expansion woredas. Woreda steering 
committees were established, but they are also not functional – currently the woreda cabinet is 
involved in CBHI implementation. At the kebele/tabia level, provisional KHIICs were established, but 
now there are only the regular kebele/tabia management structures, which are involved in premium 
collection on voluntary basis. 
Woreda schemes are housed within the woreda administration; the scheme thus gets free office 
space, along with utilities, transport, and communication services, and stationery supplies. Two 
schemes in Tigray namely Ahferom and Tahitay Adiabo are located in the premises rented from 
private landlords. This has an impact on the moral of staffs and limited their day-to-day personal 
interactions with woreda officials. The woreda administration pays salaries of the CBHI executive 
staffs. Furthermore, CBHI schemes receive operational budget from woreda administration though it 
is usually tight. 
CBHI directives task woreda administrators and their offices with providing leadership to the 
scheme including setting up the General Assembly and the Board of Directors, and ensuring that 
these governance bodies meet regularly and often, and provide the required leadership to the 
scheme. However, in all pilot woredas except Damot Woyde, this is barely happening. General 
Assemblies meet annually in all pilot regions. In most schemes, the board does not meet regularly, 
meaning that important decision making is delayed or does not happen. For example, the board of 
the Yirgalem scheme used to meet when CBHI started (in 2010/11 and 2011/12); it has not met 
since then. Boards and General Assemblies in pilot woredas in Amhara region (except in Fogera 
woreda) used to have quarterly meetings attended by members and health facility staff, to solicit 
feedback and then deliberate and make the necessary corrections. Regular meetings stopped when 
HSFR/HFG per diem support ended. In addition, the number of representatives from the community 
has been reduced from nine to four. The board in Tehuledere woreda (Amhara) no longer has 
representatives from the community; the remaining three board members are from the woreda 
government. Frequent turnover of woreda administrators is another reason for irregular board 
meetings. For example, since CBHI was launched two and half years ago, the mayor of Yirgalem City 
has been changed three times. 
Recently, there have been attempts to close this gap in leadership. For example, the Yirgalem City 
mayor has assigned each sub-city with closely following up and reporting weekly on CBHI progress. 
In Damboya (SNNP), CBHI is a regular agenda item in weekly Cabinet meetings. Each Cabinet 
member is given list of kebele/tabias to follow up – membership renewal, contribution collection, 
and so forth are monitored and reported on. Tehuledere woreda has adopted a similar practice. 
The weak leadership and commitment at the woreda level is affecting progress at the kebele/tabia 
level. Kebeles/Tabias are responsible for member registration, follow-up, and collection and deposit 
of premiums. However, major gaps have been documented. Almost all regional assessments 
reported that Kebeles/Tabias have too many other workload demands and lack incentive to carry 
out their CBHI tasks. There is no staff assigned to carry out CBHI tasks at the kebele/tabia level; 
hence, busy staff members do it on voluntary basis. Also, kebele/tabias lack a budget to cover travel 
expenses of officials who travel to woredas to deposit CBHI funds or to do other tasks. Thus, 
collected funds are not often deposited promptly. Kebeles/Tabias do not continually remind 
members to renew membership. When households do pay, payment information on their member 
ID cards is not updated on a timely basis – there have been reports of ID cards not being stamped 
or receipts not issued for several months. Thus, health facilities cannot serve them without charge. 
This has discouraged membership renewal in some woredas. In Tigray and Oromia regions, 
78
kebele/tabia administrations have been accused of renewing delayed memberships without enforcing 
the one-month waiting time and without collecting back payment; this practice encourages 
households to renew only when health service is needed immediately, pushing the scheme toward 
financial difficulty. In addition, financial mismanagement has been reported in a few woredas, for 
example: 
 In 2004 EFY (2011/12) in Damboya woreda (SNNP), one kebele/tabia general manager was 
caught collecting a premium without issuing a receipt. Another general manager, also in 
Damboya, was caught depositing Birr 40,000 into his personal account. The money was 
immediately returned and the general manager terminated. 
 In Fogera woreda (Amhara), one CBHI scheme employee stole Birr 600,000 and disappeared. 
After long discussion, the regional government replaced the money stolen by the employee. This 
practice of the region can be used to convince the community that their money is secured and 
the government is playing the role of the last resort insurer. 
 In Kuyu woreda (Oromia), Birr 50,000 was collected but not deposited in the CBHI account and 
103 receipts are in the hands of kebele/tabia chairpersons. As a result, the scheme didn’t 
undertake financial audits. 
One important way in which the government is showing its commitment to CBHI is the provision of 
general and targeted subsidies. As has been described, regions and woreda administrations provide 
the targeted subsidy to cover the membership of indigent households. The FMOH and lately EHIA 
through the general subsidy subsidize 25 percent of premiums based on enrollment. When CBHI 
started, the FMOH provided Birr 40,000 for each contracted health center to support procurement 
of drugs, medical supplies, and/or equipment. In a few cases, the regional government sent resource 
to health centers to fulfill some essential missing infrastructure such as electricity/generator, water 
supply, shelves and tables, and doors and windows. For example in SNNP, health centers in Damot 
Woyde woreda received Birr 360,000 and those in Damboya received Birr 301,000. 
Findings show that the woreda subsidy is deposited into the CBHI account on time but the federal 
subsidy is not. At the time of this assessment, the SNNP subsidy from the federal government had 
been delayed for almost six months. Closer investigation of the indigent financing process at the 
woreda level highlighted major gaps. Woredas’ budget constraints in all four regions have resulted in 
under coverage of deserving indigent families. In Tigray and Amhara, the situation is aggravated by 
the assumption that 10 percent of the population of each woreda is indigent, which is far from the 
reality. Recently, Amhara changed this approach to assume that 10 percent is indigent only in food￾insecure woredas, and 5 percent in food-secure woredas. FGDs in Tigray documented the 
agreement of almost all participants that the process of indigent selection in each kebele/tabia is fair 
and transparent. 
Apart from government bodies, an important participant in CBHI design and implementation is the 
HSFR/HFG project, which provides the schemes with major technical, financial, and in-kind support. 
The project assisted CBHI from the beginning, conducting the feasibility study, drafting the 
directive/operational manual, designing the model for the schemes, and participating in the training 
and sensitization of regional and woreda managing bodies, community elders/leaders, and the 
community. In the early days, the project assigned a full-time CBHI coordinator to each pilot scheme 
and in the larger woredas an additional technical/field officer. The project supported CBHI offices 
with computers, furniture, and consumable items. It is still following the operation of the schemes, 
encouraging regular meetings of board and General Assemblies. 
Community groups are other key stakeholders in CBHI. Community members have permanent seats 
on the high-level decision-making bodies of CBHI. In the CBHI General Assembly, in addition to the 
two community representatives per kebele/tabia, the kebele/tabia chairman and got leader are 
community representatives as they are community members elected by the community. Community 
representatives from selected Kebeles/Tabias, also represent the community on the Board of Board 
of Directors. 
79
Also, one community member (two or more in Amhara) is represented on the Health Facility 
Management Board/Committee, a high-level management body for HFs. The board/committee is 
responsible for reviewing and approving the facility’s annual operational and financial work plan, 
reviewing and approving its utilization plan for the internal revenue, approving user fee revision 
proposals, and other tasks. 
In addition, a new quarterly health facility community forum was created in SNNP in 2005 EFY 
(2012/13). The WorHO and health facilities organize the forum, which is held in both health centers 
and hospitals. The WorHO /woreda administrator chairs the forum; community members, elders, 
religious leaders, and other community leaders and health facility staffs participate. The community 
raises issues/complaints they have with the health service in general and with the CBHI scheme in 
particular. Issues are discussed and, where possible, resolved. Those issues that cannot be resolved 
during the forum are registered by the WorHO for later action. The community is informed about 
the progress in the next forum. This forum has truly empowered the community and has gone a long 
way in improving the quality of health service. In Kuyu woreda (Oromia), CBHI members submitted 
complaints about the management of Kuyu hospital to the Zonal Health Desk, which resulted in a 
new hospital manager being appointed. Similarly, CBHI members in Limmu Kossa woreda call mobile 
numbers of the woreda administrator, WorHO head, and/or CBHI coordinator when they are 
dissatisfied with facility services. 
5.5 Increased Resource Mobilization
5.5.1 Community premium contributions
The CBHI schemes have been able to mobilize contributions in all the pilot woredas. The 
mobilization appears to have been stronger in the initial mobilization period than in the recent years. 
The total amount of community contribution from the 13 pilot woredas through June 2013 was 
approximately Birr 22.7 million (see Table 5.20). By region, Amhara and Oromia have mobilized 
appreciably more than the other two regions. The early initiation of CBHI in Amhara as well as its 
higher enrollment rate is the reason for its better collection. Oromia’s performance is explained by 
the fact that it has an additional (fourth) scheme and a higher premium than the other regions. 
Table 5.20: Community contributions/premiums (‘000 Birr)
Region/Woreda
Cumulative
Jun-11 Jun-12 Jun-13
Amhara Region
S. Achefer 346.15 2,772.68 2,881.68
Fogera 272.37 1,761.27 1,670.21
Tehuledere 635.7 1,973.72 3,339.65
Total Amhara 1,254.22 6,507.67 7,891.54
Oromia Region
Deder 287.28 1,496.00 1,293.32
Gimbichu 308.05 2,012.09 2,901.87
Kuyu 36.68 736.33 607.02
Limmu Kossa 12.28 1,200.00 1,610.99
Total Oromia 644.29 5,444.42 6,413.20
80
Region/Woreda
Cumulative
Jun-11 Jun-12 Jun-13
SNNP Region
Yirgalem 29.59 245.89 406.05
Damot Woyde 295.88 736.84 1,219.09
Damboya 105.43 584.62 1,859.07
Total SNNP 430.90 1,567.35 3,484.21
Tigray Region
Ahferom 54.65 1,949.44 2,056.58
Kilte Awlaelo 56.1 1,548.51 1,637.56
Tahitay Adiabo 24.29 840.91 1,184.09
Total Tigray 135.04 4,338.86 4,878.23
Grand Total 2,464.45 17,858.30 22,667.18
Source: HSFR/HFG Project 
The major drivers of the community contribution rate are the enrollment rate, the proportion of 
non-paying members in the total CBHI membership, and the contribution level. Community 
contribution levels are not affected by the proportion of non-paying members as their share is 
covered by the woreda and/or regional governments are paying indigents’ premiums. On the other 
hand, the lower the enrollment rate and/or contribution levels, the lower is the collection of 
contribution from the community. Table 5.21 shows the regional and national averages of the drivers 
of the premium collection from the community. 
Table 5.21: Membership rate, contributions, and non-paying membership rate (June 2013) 
Region Enrollment rate (%)
Contributions per 
household per year 
(Birr)
% non-paying members 
from total enrolled 
HHs
Amhara 57.86 144 19.18
SNNP 61.20 126 6.15
Oromia 36.09 180 7.14
Tigray 47.36 132 5.00 
National 47.88 15.44
Source: HSFR/HFG Project 
As has been noted, community contributions have declined over the years. The decline in 2013 over 
2012 suggests drop in both membership renewals and new registrations. Fogera woreda is one 
example of this: It was reported that about 33 percent of all its CBHI members in year 2013 were 
non-paying (subsidized) members and total membership declined by 23 percent from 2012. Such 
circumstances – where enrollment is low, membership is declining, and contributions are not paid by 
members of the community – jeopardizes the financial sustainability of a scheme. A large share of 
government-subsidized memberships raises the important question of the point of CBHI as 
compared with other forms of government financing (SHI and tax-based). The effectiveness and 
relevance of CBHI hinges on its ability to collect a reasonable share from the community. 
81
5.5.2 Reimbursement to health facilities (in 000 Birr) 
One of the expected benefits of CBHI is increasing the revenue that health facilities can generate and 
retain for their own use. Up to the end of June 2013, CBHI schemes reimbursed about Birr 21.4 
million to contracted health centers and hospitals. KIIs and FGDs with health facility staff confirmed 
this. 
Table 5.22: Reimbursement to health facilities 
Region/Woreda June 2012 June 2013
Amhara Region
S. Achefer 399.81 2,080.11
Fogera 890.41 2,451.87
Tehuledere 1,283.86 3,033.38
Total Amhara 2,574.08 7,565.37
Oromia Region
Deder 118 647.95
Gimbichu 60.65 633.01
Kuyu 73.37 202.99
Limmu Kossa 228.33 1,404.01
Total Oromia 480.35 2,887.97
SNNP Region
Yirgalem 169.34 636
Damot Woyde 461.89 1,240
Damboya 339.7 1,713
Total SNNP 970.93 3,588.90
Tigray Region
Ahferom 120.33 1,134.21
Kilte Awlaelo 146.57 687.02
Tahitay Adiabo 178.69 1,038.62
Total Tigray 445.59 2,859.84
Grand Total 4,470.95 16,902.08
Source: HSFR project 
82
In 2011/12 and 2012/13 fiscal years, the majority of the reimbursement amount goes to health 
centers – in Amhara, SNNP, and Tigray, about 91 and 88 percent respectively (Table 5.23). This 
obviously means that hospitals get 9 and 12 percent of the reimbursed amount. In the case of 
Oromia, hospitals earned 41 and 31 percent of reimbursement in 2011/12 and 2012/13, respectively. 
Table 5.23: Share of CBHI Reimbursements Paid to Health Centers and Hospitals 
Regions
2011/12 2012/13
HC Hospitals HC Hospitals
Amhara 95% 5% 93% 7%
Oromia 59% 41% 69% 31%
SNNP 96% 4% 90% 10%
Tigray 94% 6% 86% 14%
National 91% 9% 88% 12%
The data clearly show a link between utilization and reimbursement of health facilities in three 
regions (Amhara, SNNP, and Tigray) in 2012/13 (Figure 5.5). In Oromia, hospitals account only for 
about 5 percent of CBHI members’ utilization but received about 31 percent of reimbursement in 
that year. This shows that, overall, CBHI schemes seem to provide the intended utilization (at 
lower-level facilities) and payment patterns. The reasons for the large share of reimbursement going 
to hospitals in Oromia need to be identified so as to reduce the undesirable impact on scheme 
financial status. 
Figure 5.5: Utilization and CBHI Reimbursement in Health Centers and Hospitals, 2012/13 
86% 86%
14% 14%
91% 93%
9% 7%
95%
69%
5%
31%
91%
86%
9%
14%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
% of members visit % of reimbursement % of members visit % of reimbeuresment
Heath Centers Hospitals
Tigray Amhara Oromia SNNP
83
5.5.3 Government subsidies
As has been described, CBHI design in Ethiopia allows for three types of government subsidies to 
the schemes: targeted and general subsidies and financing the scheme management costs (salaries, 
office space, and operational costs). The regional and woreda governments finance premiums for 
indigents, using different arrangements. In Tigray, the region and woredas split targeted subsidy 
70/30 percent. In Amhara, it is a 90/10 percent split. In SNNP and Oromia, woredas pay the entire 
contribution for indigents. Woredas pay scheme salaries and operational costs in all pilots. The 
federal government pays the 25 percent general subsidy, which is applied to premiums of all CBHI 
members, paying and non-paying. 
Through the end of June 2013, the total amount of subsidy going to schemes in all 13 pilot woredas 
was Birr 16.5 million; the general subsidy was Birr 9.7 million and targeted subsidy was Birr 6.8 
million. This represents about 42 percent of the total revenue generated by the schemes. General 
and targeted subsidies constituted 25 and 17 percent of the total revenue collected, respectively. Of 
the total revenue generated from all pilot schemes, 35.3 percent was from Amhara, 27.3 percent in 
Oromia, 24.2 percent in Tigray and 13.1 percent in SNNP. This is due to appreciable variation in the 
proportion of contribution collected and the targeted subsidy received from regional government 
and woreda administration. The general subsidy on the other hand is almost comparable across 
regions (Table 5.24). 
Table 5.24: Government Subsidy to CBHI Pilot Schemes up to June 2013 (‘000 Birr)
Region
Sources of finance As % of total revenue generated by 
the schemes
Contribution General 
Subsidy
Targeted 
Subsidy Total
% of 
overall 
total
Contribution General 
Subsidy
Targeted 
Subsidy
Amhara 7,891.54 3,469.71 2,462.70 13,823.95 35.3 57.1 25% 18%
Oromia 6,413.20 2,802.54 1,487.52 10,703.26 27.3 59.9 26% 14%
SNNP 3,484.21 1,393.18 260 5,137.39 13.1 67.8 27% 5%
Tigray 4,878.23 2,019.83 2,590.47 9,488.53 24.2 51.4 21% 27%
Total 22,667.18 9,685.26 6,800.70 39,153.13 100 57.9 25% 17%
Source: HSFR/HFG Project 
In addition to the subsidies, government provides other support to schemes and providers. 
Woredas provide the schemes in-kind support in the form of office space, stationery, and 
transportation. They also pay salaries of the three CBHI staff. The FMOH has given each health 
center a subsidy of Birr 40,000 to improve availability of drugs and medical supplies. In addition, 
some regional governments gave grants to selected health centers to purchase missing infrastructure 
such as electricity/generator, water supply, shelves and tables, door/window, laboratory supplies. No 
hospital in any region reported CBHI-related support from the federal or regional government. The 
FMOH and RHBs provide technical assistance to CBHI-contracted health facilities. 
The above arrangement clearly shows that there is very strong government commitment to CBHI. If 
this support continues during scale-up, it will have great financial implications for the governments. 
While this government support is one feature of the CBHI design, too much subsidy also negates the 
relevance and added value of CBHI. Scale-up should therefore balance effective premium 
mobilization and government financing. 
84
5.5.4 Resource balance of the pilot program
As seen in Table 5.25, the pilot schemes have collected Birr 22.7 million as contributions from paying 
members and Birr 9.7 million and Birr 6.8 million from the general and targeted subsidy, respectively. 
They have reimbursed Birr 16.9 million to health facilities, which leaves a balance of Birr 22.3 million. 
Table 5.25: Balance, all CBHI pilot schemes, June 2013 (‘000 Birr)
Region/
Woreda
Premium 
collected
Targeted 
subsidy
General 
subsidy
Premiums 
+ subsidy
Reimburse
d to health 
facilities
Balance
Before 
general 
subsidy
After 
general 
subsidy
Amhara 
S. Achefer 2,881.68 609.02 1,116.55 4,607.24 2,080.11 1,410.59 2,527.14
Fogera 1,670.21 1,022.97 1,051.94 3,745.12 2,451.87 241.31 1,293.25
Tehuledere 3,339.65 830.72 1,301.22 5,471.58 3,033.38 1,136.99 2,438.21
Total 
Amhara
7,891.54 2,462.70 3,469.71 13,823.95 7,565.37 2,788.87 6,258.58
Oromia
Deder 1,293.32 350 852.9 2,496.22 647.95 995.37 1848.27
Gimbichu 2,901.87 243.75 763.14 3,908.76 633.01 2512.61 3275.75
Kuyu 607.02 527.47 427.74 1,562.23 202.99 931.5 1359.24
L. Kossa 1,610.99 366.3 758.76 2,736.05 1,404.01 573.28 1,332.04
Total 
Oromia
6,413.20 1,487.52 2,802.54 10,703.26 2,887.97 5,012.75 7,815.29
SNNP
Yirgalem 406.05 108 248.22 762.27 635.71 -121.66 126.56
D. Woyde 1,219.09 107 624.79 1,950.88 1,240.14 85.95 710.74
Damboya 1,859.07 45 520.17 2,424.24 1,713.06 191.01 711.18
Total SNNP 3,484.21 260 1,393.18 5,137.39 3,588.90 155.31 1,548.49
Tigray
Ahferom 2,056.58 1,249.50 887.19 4,193.26 1,134.21 2,171.87 3,059.06
Kilte Awlaelo 1,637.56 723.58 747.67 3,108.81 687.02 1674.12 2421.79
Tahitay Adiabo 1,184.09 617.39 384.98 2,186.46 1,038.62 762.86 1,147.84
Total Tigray 4,878.23 2,590.47 2,019.83 9,488.54 2,859.84 4,608.86 6,628.69
Overall Total 22,667.18 6,800.70 9,685.26 39,153.13 16,902.08 12,565.80 22,251.05
Source: HSFR/HFG Project 
85
As Figure 5.6 shows, overall, CBHI schemes were able to finance their costs from revenue generated 
from paying and non-paying members’ contributions, the reimbursement made to the health facilities 
stands at about 75% of contributions. However, scheme in Fogera, Yirgalem and Damot Woyde 
have had negative balance (schemes owe more money from health facilities). The general subsidy is 
important for these schemes (Fogera, Yirgalem and Damot Woyde) to keep them solvent. When 
targeted subsidy is included, all woredas expect Yirgalem have had positive balance and the claims 
ratio 57.4 percent. When general subsidy is included to this total (contribution & targeted subsidy), 
the claims ratio become 43.2 percent. The schemes account balances are healthy though those in the 
SNNP appear to be weaker than other schemes. 
Figure 5.6: Ability of Schemes to Finance Health Service Costs (in %) 
Source: HSFR/HFG Project 
It should be noted that in Oromia, despite a high level of hospital claims, the schemes have financial 
surpluses due mainly to lower utilization rates. In Tigray, woredas have done well in terms of 
resource mobilization over the entire period of CBHI but in 2013 there was a decline in members’ 
contributions, especially renewals. This is partly because CBHI members are not fully aware that 
they have to contribute every year; also, some members are deliberately not renewing until they 
need health services, which they often get without waiting for one month as CBHI rules require. 
This “free-rider culture” jeopardizes scheme financial sustainability by distorting community 
expectations with false promises on what CBHI can deliver. 
There are major but different reasons for the financial weakness of the three schemes named above. 
The financial status of Yirgalem suggests that the premiums and the benefit package are not 
appropriate for the urban context (see Box 5.3). In Fogera, absence of new enrollment, high dropout 
rate and increasing health services utilization rate are the major reasons for bad performance of the 
scheme during 2013. Deder woreda has seen a huge (83 percent) drop in member contributions, 
from Birr 1.2 million in 2012 to just over Birr 200,000 in 2013. The woreda is paying for members’ 
health care services in 2013 with revenue from contributions in 2012 and from government 
subsidies. This portends future insolvency; the dramatic drop in enrollment calls for further 
investigation. 
These findings suggest several points: First, the decline in membership must be reversed. Second, 
different design parameters may be needed for urban schemes, such as sliding-scale premiums that 
take into account differences in ability to pay. Third, a copayment may be needed, especially in urban 
woredas and others with high utilization rates. Finally, the current CBHI design has no provision for 
risk pooling among the pilot schemes – each scheme must cover all health service costs that accrue 
-40%
-20%
0%
20%
40%
60%
80%
100%
Without Subsidy With Targeted Subsidy
With Targeted + General subsidy -40%
-30%
-20%
-10%
0%
10%
20%
30%
Yirgalem
86
to it. As it scales up CBHI, Ethiopia should consider establishing regional and national risk pools with 
clear scheme contribution criteria and assignment of regional and federal pool responsibilities. 
5.6 Resource mobilization potential during the scale-up
5.6.1 Assumptions
The resource mobilization potential of CBHI schemes during scale-up has been explored based on 
three sets of assumptions. The first set is the policy decisions already made by each regional 
government for scaling up, particularly with regard to the number of expansion woredas, premiums 
and registration fees, scheme organization, indigents’ coverage rate, and government subsidy 
structures. The second set assumes that scheme performance in terms of service utilization and 
reimbursements to health facilities will be similar to the averages of their respective regional pilot 
woredas. The third is related to the number of indigents to be financed by government. Here, three 
scenarios were considered: regional averages of pilot performance in each region; 10 percent of 
eligible households in woredas; and full coverage (29 percent of CBHI-eligible population, the 
percentage based on the national average poverty line) of people who are expected to be below 
poverty line. The projection is made only for a year and does not take into account any change such 
as inflation. It is assumed that the challenge of declining premium collection after initial kick-off will 
be rectified as there is no point to scale-up unless this situation is not addressed. The assumptions 
considered for projecting the resource mobilization potentials are summarized in Table 5.26. 
Box 5.3: Lessons from Yirgalem woreda scheme
The financial status of the Yirgalem scheme is very weak and jeopardizes the existence of the scheme. The 
scheme has only Birr 68,000 in its bank account, and it owes health facilities Birr 352,283. It has not 
reimbursed health facilities for two quarters, which is affecting the quality and number of services the 
facilities are providing to scheme members. The main reasons for this dire financial situation are:
 As a city, health service utilization in Yirgalem is significantly higher than other pilot areas. The higher 
utilization is attributable to a high number of patients with
 Diabetes and other chronic conditions; these patients visit health facilities frequently and receive 
expensive treatment. 
 Some kebele/Tabias have been allowed to use Yirgalem hospital as the first point of contact, meaning the 
scheme incurs the higher hospital costs. 
 The 10.50/month/household premium seems very low for an urban context.
 There is no co-payment to rationalize utilization.
When scaling up to the urban context, the CBHI program should consider introducing higher premium rates 
per households, with sliding/stratified contributions, taking steps to ensure a large membership base and 
therefore a large risk pool, and developing a different payment agreement when hospitals are used as the 
entry point to care.
87
Table 5.26: Assumptions for Projecting Resource Mobilization During Scale-up
Amhara Oromia SNNP Tigray
Number of pilot woredas 3 4 3 3
Number of expansion woredas 39 59 47 15
Paying members in pilot woredas (June 2013) 40,513 35,750 18,556 26,962
Non-paying members in pilot woredas (June 2013) 9,612 2,750 1,217 2,067
Eligible households in expansion woredas 1,177,322 2,158,600 503,000 547,491
Participation rate (enrollment rate) 57.90% 36.10% 61.20% 47.40%
Non-paying members/indigents (% of eligible HHs) 10% 10% 10% 10%
Average contributions per household per year 144 180 126 132
Membership fee 10 10 10 10
Service utilization rate health center 96% 30% 93% 38%
Service utilization rate hospitals 11% 13% 13% 6%
Average number of beneficiaries per household 3.54 4.72 4.75 4.32
Average cost per visit health center 28.39 37.97 28.19 38.31
Average cost per visit hospitals 100 100 90 140
Premiums subsidy
Woreda subsidy of indigents contributions 30% 30% 100% 30%
Regional subsidy of indigents contributions 70% 70% 0% 70%
Federal subsidy of all CBHI members 25% 25% 25% 25%
Woreda subsidy for CBHI management salary and 
operating expenses 
Number of employees per woreda 3 3 3 3
Average salary 2,300 2,300 2,300 2,300
Social security contribution (% of gross salary) 8% 8% 8% 8%
Average operating cost per month 4,000 4,000 4,000 4,000
Source: Performance during pilot period and policy decisions 
5.6.2 Projections
Table 5.27 summarizes the projections of community contributions, government subsidy, 
reimbursement to health facilities, and overall balance based on the above assumptions. The total 
amount of resources that can be mobilized from community premiums in all (pilot and scale-up) 
schemes will be approximately Birr 345 million per year. Oromia region is estimated to mobilize the 
largest amount (Birr 146 million), while SNNP region is estimated to mobilize the least (Birr 42.4 
million); Amhara has the potential to mobilize Birr 106 million, Tigray Birr 50 million. 
Our estimations show that the total reimbursed to health facilities will be Birr 316.5 million per 
year. In Oromia, reimbursement is projected to be Birr 110 million and the resource balance with 
subsidy Birr 135 million (i.e., a surplus). In Tigray, these amounts are projected to be Birr 38.2 
million and Birr 36.7 million. In Amhara, a high participation (enrollment) and high utilization rate 
project a surplus of Birr 57 million. The projected surplus in SNNP is just Birr 1.4 million suggesting 
there may be a need to increase the user fee from its assumed level of Birr 126 per household per 
month. 
88
Table 5.27: Projected Income, Subsidies, Expenditure, and Balance (‘000 Birr)
Amhara Oromia SNNP Tigray Sum
Income 
Income from registration fee 6,131.04 7,011.60 2,770.68 3,015.75 18,929.06
Annual Income from premium:
Pilot woredas 5,833.87 6,435.00 2,338.06 3,558.98 18,165.91
Scale-up woredas 94,120.78 132,643.71 37,248.63 43,366.89 307,380.00
Total income 106,085.69 146,090.31 42,357.37 49,941.62 344,474.98
Subsidy
General subsidy-federal gov’t 37,486.11 57,331.17 14,579.92 16,955.54 126,352.74
Targeted subsidy:
Regional government 13,660.42 29,055.88 - 5,633.05 48,349.35
Woreda 11,626.27 21,110.23 13,865.34 4,887.80 51,489.64
Total subsidy 62,772.81 107,497.28 28,445.27 27,476.38 226,191.74
Expenditure
Reimbursement to health facilities 105,746.53 109,999.51 62,528.21 38,220.60 316,494.85
Management salary and operating 
expenses
5,771.81 8,657.71 6,871.20 2,473.63 23,774.35
Total expenditure 111,518.34 118,657.22 69,399.41 40,694.23 340,269.20
Balance 57,340.16 134,930.36 1,403.22 36,723.77 364,043.56
Source: Based on assumptions in Table 5.26 
5.6.3 Fiscal implications of scale-up
For the first indigent coverage scenario, 10 percent of eligible households, the subsidy from the 
federal government is estimated at Birr 93.9 million per year for the four regions. Regional 
governments will have to be ready to allocate Birr 32.1 million in targeted subsidy, with each 
regional government allocating as follows: Oromia Birr 19.5 million, Amhara Birr 8.6 million, and 
Tigray nearly Birr 4.0 million. There is no regional subsidy budget in SNNP as the woreda is assumed 
to cover the entire targeted subsidy. 
Woreda subsidies, covering indigents and CBHI staff salaries and office running costs, total Birr 22.4 
million per year. On average, each pilot expansion woreda has to subsidize Birr 254,474.06 per 
annum. The highest average is estimated to be Birr 183,571.17/woreda in SNNP, and the lowest Birr 
113,655.21 in Tigray. Woreda administrations need to be aware of the fiscal commitment required 
when introducing a CBHI scheme. 
For the second scenario, coverage of indigents will be the average of the pilot woredas in the region: 
Amhara 11.1 percent, Oromia 2.6 percent, SNNP 3.8 percent, and Tigray 11.5 percent. The third 
scenario assumes that indigents will be 29 percent of eligible households. As seen in Table 5.28, the 
total subsidy jumps to Birr 203.1 million from Birr 148.3 million per year if we apply the 29 percent 
case; woreda subsidies in particular will increase on average to Birr 221,925.76 from Birr 121,963.07 
per woreda per year. It is clear that 29 percent is not feasible from the budgetary standpoint. Table 
5.28 shows the fiscal implication of the scaling up to the originally planned 161 woredas and to all 
non-pastoralist woredas of Ethiopia, assuming that pilot case participation continues, under the three 
scenarios. The fiscal implications of CBHI scale up if 50%, 75% and 100% participation take place are 
indicated under annex E, F and G. 
89
Table 5.28: Projected Government Subsidy Requirements under Three Scenarios (‘000 Birr)
Pilot Case Participation Rate 
161 Expansion Woredas All Non-Pastoralist Woredas
Region 10% Pilot Case 29% 10% Pilot Case 29%
Total Budget
Amhara 30,658.27 35,629.89 40,952.53 252,437.71 308,954.69 369,462.08
Oromia 87,851.16 82,485.50 123,532.82 325,609.55 303,086.47 475,388.08
SNNP 14,162.58 13,536.03 17,258.57 126,485.43 114,441.90 185,996.15
Tigray 15,674.42 14,815.21 21,343.68 44,690.52 41,579.91 65,215.20
Addis Ababa 50,500.67 50,500.67 73,883.75
Other Non-Pastoralist 
Regions
18,238.35 18,238.35 26,870.59
Total 148,346.43 146,466.63 203,087.61 817,962.23 836,801.98 1,196,815.84
Federal Budget
Amhara 18,334.21 18,334.21 18,334.21 189,711.72 189,711.72 189,711.72
Oromia 60,014.88 60,014.88 60,014.88 246,380.00 246,380.00 246,380.00
SNNP 5,534.74 5,534.74 5,534.74 95,036.83 95,036.83 95,036.83
Tigray 9,991.66 9,991.66 9,991.66 33,662.76 33,662.76 33,662.76
Addis Ababa 38,193.78 38,193.78 38,193.78
Other Non-Pastoralist 
Regions
13,695.07 13,695.07 13,695.07
Total 93,875.48 93,875.48 93,875.48 616,680.15 616,680.15 616,680.15
Regional Budget
Amhara 8,626.84 12,106.98 15,832.83 43,908.19 83,470.08 125,825.25
Oromia 19,485.40 15,729.43 44,462.56 55,460.69 39,694.53 160,305.65
SNNP 0.00 0.00 0.00 0.00 0.00 0.00
Tigray 3,977.93 3,376.49 7,946.42 7,719.44 5,542.01 22,086.71
Addis Ababa 0.00 0.00 0.00 8,614.82 8,614.82 24,982.98
Other Non-Pastoralist 
Regions
0.00 0.00 3,180.30 3,180.30 9,222.87
Total 32,090.17 31,212.90 68,241.81 118,883.43 140,501.74 342,423.46
Woreda Budget
Amhara 3,697.22 5,188.70 6,785.50 18,817.80 35,772.89 53,925.11
Oromia 8,350.88 6,741.19 19,055.38 23,768.87 17,011.94 68,702.42
SNNP 8,627.85 8,001.29 11,723.84 31,448.61 19,405.07 90,959.33
Tigray 1,704.83 1,447.07 3,405.61 3,308.33 2,375.15 9,465.73
Addis Ababa 3,692.07 3,692.07 10,706.99
Other Non-Pastoralist 
Regions
1,362.99 1,362.99 3,952.66
Total 22,380.78 21,378.25 40,970.32 82,398.65 79,620.10 237,712.24
90
161 Expansion Woredas All Non-Pastoralist Woredas
Region 10% Pilot Case 29% 10% Pilot Case 29%
Average Woreda Budget (161 Woredas)
Amhara 88.03 123.54 161.56 
Oromia 132.55 107.00 302.47 
SNNP 172.56 160.03 234.48 
Tigray 94.71 80.39 189.20 
Addis Ababa
Other Non-Pastoralist 
Regions
Total 121.96 117.74 221.93
91
SECTION 4. 
POLICY 
RECOMMENDATIONS 

93
6. POLICY AND PROGRAMMATIC 
IMPLICATIONS OF EVALUATION 
FINDINGS ON SCALE-UP
This evaluation of the CBHI pilot schemes in Ethiopia shows that CBHI members are using health 
services more (26 percent more) than the non-members. It also documented that the impoverishing 
effect of OOP payments (including non-facility expenditures) on CBHI members is much less than 
that of non-members. It found the premiums and registration fees are affordable for members. 
Although there are issues with reimbursement, adequacy/quality of services, and the financial status 
of CBHI schemes in a few woredas, CBHI was generally able to mobilize additional resources for the 
health sector. This clearly supports the government plan to scale up CBHI to other woredas. 
This scale-up of CBHI in Ethiopia as a path to achieving UHC should be guided by best practices 
from around the world and the lessons learned from the current pilots. Internationally, best practice 
revolves around the four basic actions that WHO recommended as ingredients or pillars toward 
UHC: 
 Promoting equitable access by removing financial barriers, especially direct payments; 
 Making prepayments compulsory; 
 Creating large risk pools for financial sustainability; and 
 Having governments cover the health costs of people who cannot afford to contribute. 
Table 6.1 presents the major issues of CBHI pilot schemes, the major recommendations for scale-up, 
and the institution(s) suggested to take the recommendations forward. 
Table 6.1: CBHI Issues, and Evaluation Recommendations for Action and Responsible Party 
Issue Recommendation Responsible 
party
In Amhara, Tigray, and SNNP regions, 
political and administrative leaders 
engaged actively in sensitizing the 
community about the benefits of CBHI. 
The establishment of KHIIs responsible 
for community sensitization played a 
crucial role in enhancing enrollment. 
Members of the woreda cabinet were 
assigned specific kebeles/tabias to 
sensitize and follow up and this was made 
one of their performance measurement 
criteria (e.g., Tehuledere and Yirgalem). 
Foster political commitment, government 
ownership, and leadership at regional and 
woreda levels, and better organization at the 
community level. Before scaling up CBHI, there is 
thus a huge need for extensive orientation of zonal, 
woreda, and kebele/tabia officials. Consider Tehuledere 
woreda’s experience of introducing CBHI activity as part 
of the overall woreda checklist that the woreda cabinet 
is using as a follow-up and support mechanism. 
Regional 
Government, 
Woreda Chief 
Administrator
Although there has been success in 
communicating CBHI principles to 
communities, gaps remain. Some better￾off individuals like those we met in 
Damboya and Damot Woyde consider 
CBHI an intervention for indigents/poor 
households only. In contrast, there are 
instances of CBHI covering high medical 
Give ample time and effort to community 
sensitization. Experiences like that of Ider and Mekelle 
hospitals should be well documented and used to 
promote CBHI enrollment and renewal via media such 
as leaflets, signboards, or regional radio and TV. There is 
a need to review the current sensitization strategy and 
tools to make sure that the lessons are incorporated 
and used during scale-up. Given that the sensitization in 
RHB, Woreda 
Chief 
Administrator, 
Woreda 
Health Office, 
Woreda 
Cabinet, 
Kebele/Tabia 
94
Issue Recommendation Responsible 
party
costs at Ider and Mekelle hospitals, which 
would have been catastrophic had it not 
been for CBHI scheme.
the pilot woredas was daunting, as KIIs at all levels 
revealed, a major effort, even more than during the 
pilot, is required for scale-up. It is crucial to link 
kebele/tabia/community sensitization and follow-up with 
the Health Extension Program and the Health 
Development Army for a more effective result.
Administration
In some pilot areas, health providers did 
not have adequate awareness of CBHI 
principles and of what is expected of 
them. This is one reason for reports of 
provider’s discourteous treatment of 
members during health service utilization.
Do targeted sensitization of health providers.
This is needed to improve health provider understanding 
of CBHI functioning, reduce confrontations with clients, 
and improve courtesy toward clients. There is a huge 
need for extensive orientation of health facility staff 
(health centers and hospitals) before scaling up CBHI. 
Regions need to factor in the financial and technical 
implications of orientation programs.
RHB, WorHO, 
CBHI scheme
Currently, all regional CBHI schemes 
charge similar premiums and registration 
fees regardless of household income and 
location (town, rural area) of scheme. 
This in the long term will not be fair and 
will compromise the financial 
sustainability of schemes as evidenced in 
Yirgalem. As scale-up proceeds and 
enters urban areas, the risk of moral 
hazards is likely to increase.
Consider different design parameters and test 
them during scale-up. The FMOH and RHBs could 
explore the feasibility of introducing different and 
stratified premiums for more equity and sustainability. It 
is also important to consider introducing some sort of 
co-payment per incidence or visit, especially a certain 
percentage of drug costs (say, 5-10%). The most 
successful CBHI country, Rwanda, has moved from 
uniform to stratified payment. This could be piloted in a 
few woredas to judge feasibility. 
EHIA, FMOH, 
and RHBs
There were reports that the schemes 
provided services to members who had 
not paid their premiums. In most pilot 
woredas, there is a tendency to 
accommodate the interest of members 
rather than the interest of the scheme. 
The practice of overlooking timely 
renewal has created apathy on the part 
of the community; there is a tendency to 
hold off renewal until there is a health 
need. It will be difficult to un-do such 
practices, which damage membership and 
in particular the sustainability of scheme 
in the long run.
Enforce scheme regulations on renewal.
Membership should be considered active when, and only 
when, it is renewed. There should be a clear system and 
enforcement mechanism to enforce prompt membership 
renewal; for example, those who do not renew their 
membership within three months should be considered 
as having quit the scheme and having to register as new 
members, with a waiting period for services. 
RHB, WorHO, 
CBHI scheme
Household survey results, KIIs, and 
community FGDs showed that the 
current premium rate is affordable. 
However, the financial sustainability of 
some schemes is questionable. Some 
schemes have deficits after just two 
years. Although other mechanisms can 
also be considered, such as risk pooling 
among schemes at different levels, it is 
necessary to review and revise the 
premium so that it is commensurate with 
reimbursements that the schemes pay to 
the health facilities at a rate that allows 
the facilities to improve service 
availability and quality. 
Review and revise premium levels during scale￾up. This may need an analysis of the willingness and 
ability to pay of the scaled-up woredas. 
EHIA, FMOH, 
and RHBs
95
Issue Recommendation Responsible 
party
The CBHI strategy and its operational 
manual clearly reflect government’s 
commitment to protect the very poor in 
the form of the indigent payment. 
Government financing of the premiums is 
considered one of the best proactive and 
major pillars if CBHI is to contribute 
toward UHC. The implementation so far 
is mixed: Amhara and Tigray are doing 
well while Oromia and SNNP need to 
enhance their effort. The household 
survey clearly shows that targeting of 
indigent is generally fair. The percentage 
of people considered as indigent even in 
well-performing regions and woredas is 
still far below the estimated households 
that are below poverty line. However, it 
is necessary to base renewal of indigents 
on clear criteria or a poverty assessment. 
Subsidy for indigents is the right strategy to 
protect the very poor from impoverishment. The 
need to allocate a budget for this at the regional/woreda 
level is critical. This strategy needs to be enhanced and 
replicated in the scale-up woredas. The fiscal impact on 
these woredas will be similar to that in the pilot 
woredas and they may be able to afford it. However, 
regions need to mobilize more resources for this 
subsidy, as the numbers of indigents will increase. Poor 
households that just fail to meet indigent criteria are 
falling through the cracks because of the budget 
limitation and hence there is a need to increase 
resources to ensure these households are enrolled in 
CBHI. 
The cost of photographs of indigent households for the 
membership ID is one barrier that needs to be 
addressed during scale-up.
FMOH, 
Regional 
government, 
Woreda 
government
Currently, each pilot scheme covers all 
the health service costs that accrue to it 
and there is no risk pooling among pilot 
schemes. As documented in other 
countries, CBHI can only be successful 
when there is risk pooling at regional and 
national levels. This is recommended by 
WHO and implemented in the most 
successful country, Rwanda.
Enhance scheme solidarity at the woreda and 
regional levels. Regional and national risk pools should 
be established during scale-up. Woreda schemes could 
contribute a share of their premium to a regional 
scheme. The regional scheme could pay for accessing 
regional referral hospital for health care adhering the 
referral system and subsidize schemes that are under 
financial stress. A national pool would play a role similar 
to the regional one. 
EHIA, 
Regional 
government, 
Woreda 
government
Currently, with the exception of Kuyu 
(Oromia) and Kilte Awlaelo (Tigray), 
referral is limited to the nearby hospital 
unless the nearby hospital is a regional 
hospital. This is a good practice that 
other regions could learn from.
Patients tend to self-refer to hospitals or 
to health centers in woreda towns, 
bypassing nearby facilities. To curtail this 
practice, a bypass fee was introduced in 
some regions (Amhara); patients who 
visit a hospital without referral slip are 
required to pay 50% of their expenses 
(the scheme reimburses only 50%). To 
avoid this penalty, patients visit a nearby 
health center and follow the referral 
system. However, it appears that 
unnecessary referrals take place due to 
patients’ pressure on health personnel.
Strengthen the referral system to provide 
effective health services for members. Together 
with the establishment of regional and national CBHI 
risk pools, it may be necessary to include at least 
regional referral hospitals and one federal referral 
hospital in the referral system. Evidence from the 
financial feasibility assessment, however, should inform 
such expansion of the referral system. 
To avoid unnecessary referrals, a systematic review of 
utilization patterns and referral linkages should be done.
FMOH, RHB, 
ZHD, WorHO, 
CBHI scheme
96
Issue Recommendation Responsible 
party
There is general agreement that the 
defined benefit package is adequate. The 
only suggestion for expansion is 
improving quality of care in health 
facilities, specifically to make available 
ambulance services to transport patients 
from health centers to hospitals. It is also 
agreed that benefits are simple and easy 
to communicate so community 
expectations are clear. This should 
continue during scale-up; added detail 
could complicate understanding and 
generate suspicion. 
Continue investing in service providers to 
improve quality of care and manage benefit 
expectations. The facility-community forums launched 
in some regions last year, although not specifically 
related to CBHI, is an initiative that will improve quality 
and responsiveness of health providers. This is 
something that CBHI scale-up woredas may need to 
establish or strengthen. 
FMOH, RHB, 
ZHD, WorHO, 
CBHI scheme
Some health facilities are not complying 
with the contract they signed. They do 
not provide the complete benefit 
package. They lack drugs and medical 
supplies and so refer members to other 
facilities for certain services and 
commodities. Qualitative study results 
have shown that availability of drugs and 
required diagnostics at the facility of first 
contact is the most important factor for 
households to join and/or sustain CBHI 
membership. If these problems are not 
addressed, CBHI achievements in pilot 
woredas might not be maintained there, 
let alone successful scaled up. 
Address unavailability of drugs, supplies, and 
some diagnostics. FMOH and regional governments 
invested on facility preparedness when CBHI started and 
ensured member satisfaction with the services provided. 
This is a best practice that the FMOH and regional 
governments should replicate by assessing readiness of 
facilities and then mobilizing resources to finance the 
gaps. They also need to engage and work more with the 
PFSA and its hubs to address inadequate supply of drugs 
and supplies. 
FMOH, RHB, 
ZHD, WorHO, 
CBHI scheme
As noted above, some contracted health 
facilities lack drugs and medical supplies 
and so refer members outside the 
network for services and commodities. It 
is often difficult and sometimes 
impossible for these members to get 
reimbursed for the cost they incurred at 
out-of-network facilities. 
Consider contracting the private sector 
providers as part of regional and national risk pools, to 
ensure there is a participating provider nearby and 
preferred by the community, and willing to meet CBHI 
quality and other requirements including charging fees 
similar to the public providers. 
EHIA, RHB, 
ZHD, WorHO, 
CBHI scheme
Currently, the kebele/tabia 
administration plays a critical role in 
registering scheme members and 
collecting premiums. However, there are 
challenges in many woredas in following 
up and collecting premiums, enrolling 
new members, and promptly collecting 
and transferring and/or depositing the 
premiums collected. The existing 
structure (premium collection on a 
voluntary basis) is not adequate. 
Review and adjust the current registration and 
premium collection systems. It is necessary to 
consider alternative ways of providing incentives and 
accountability at the kebele/tabia level. Establishing a 
structure at that levels (premium collection by 
kebele/tabia management structures on voluntary basis) 
will enhance accountability in effectively registering new 
households, renewing membership, collecting 
contributions, and channeling them to the scheme. If this 
doesn’t work, consider introducing incentives like those 
of the revenue authority (2% of money collected), to 
motivate the kebele/tabia administration to promptly 
collect and transfer/deposit premiums.
Regional 
government, 
Woreda 
government
97
Issue Recommendation Responsible 
party
Resources required financing subsidies 
(both targeted and general) and covering 
administrative costs to run the scheme. 
Projected resource requirements are too 
large to be financed by the woreda or 
region alone. 
As the number of woredas and kebeles/ 
tabias implementing the CBHI program in 
four regions rapidly increased, the 
HSFR/HFG project downgraded its 
financial support for operational costs 
such as per diem, fuel, vehicle and other 
travel expenses. This has limited CBHI 
teams’ regular supportive supervision in 
sections/kebeles/tabias. There is no 
longer dedicated staff at each scheme to 
monitor member enrollments and 
renewals, and premium collection and 
deposits at the kebele/tabia level. 
Mobilize additional resources at the regional and 
federal levels to meet the resource requirements 
of scale-up. The adoption of current design elements 
for scale-up (30/70 share of financing contribution 
between woreda and regions; 25% subsidy for all 
members by FMOH, and current premium rates) will 
increase the fiscal burden for the regions and FMOH. 
The 30% indigent financing is projected to be an average 
Birr 300,000 per woreda per year and woredas might 
not be able to cover it.
Regions need to mobilize about Birr 30 million and the 
FMOH about Birr 120 million per year. When viewed as 
a move toward UHC, the fiscal implication is not that 
great; however, regions and the FMOH need to be 
aware of what is needed and mobilize these additional 
resources. 
Establish/strengthen CBHI coordination and 
technical support at federal, regional, zonal, and 
woreda levels to support operations and steer the 
CBHI schemes at woreda levels.
EHIA, 
Regional 
government, 
RHB, Woreda 
government, 
and WorHO
Recruitment and retention of CBHI team 
remains a challenge. According to the 
current structure, the CBHI executive 
team has 3 staff (team leader/health 
coordinator, accountant, and ICT 
person). In addition, there is one focal 
person for the team, which is assigned by 
the woreda administration. The staffing 
challenge remains due to delays in 
sharing an approved position and 
structure by the regional civil service 
office with woreda administrations. 
Moreover many existing staffs do not 
meet the academic and work experience 
required for these positions. In addition, 
the CBHI staffs have not been able to 
benefit from salary increases and other 
benefits that are part of the career 
structure. Currently most of them are 
also lent by other government offices, 
usually health centers. All these have 
created job insecurity for the staffs. As a 
result, many of them seek alternative 
employment; turnover is high. 
The structure of CBHI teams should be 
approved at the regional level and adequately 
budgeted in all the woredas. This should include 
revising the salary scale and career structure of CBHI 
staff (making them at least comparable to those of health 
center staff), allocating an adequate operational budget 
on a regular basis, properly pinning down the structure 
and give it the appropriate housing within the 
government structure but with semi-autonomous status 
(as part of the EHIA or learning from the experience of 
regional/sub-regional structures of the Food, Medicines, 
Health Care Administration and Control Authority, 
FMHACA). In addition, there is a need to revisit the 
staffing structure and assign adequate staff to properly 
run the schemes. 
Regional 
government, 
Regional Civil 
Service 
Commission, 
RHB, Woreda 
administration, 
WorHO
The involvement of the ZHD and 
WorHO in the CBHI implementation is 
limited. Most CBHI responsibility falls on 
the woreda administration. When there 
is turnover of the administrator or lack 
of political commitment, scheme 
functioning is seriously affected.
Enhance the involvement of the ZHD and 
WorHO in CBHI management. It is necessary to 
develop a sense of ownership of CBHI by zonal and 
woreda health officials. Such a push should come from 
above. Bringing CBHI under the WorHO should be 
considered given its direct relation with the health 
sector. The woreda administration should continue to 
do community sensitization and mobilization from 
initiation through the implementation of the CBHI 
program. 
Regional 
government, 
RHB, EHIA
98
Issue Recommendation Responsible 
party
Even though information management 
was given attention in the pilot design, 
there is still a significant gap in 
generating, storing, sharing, and using 
CBHI data at all levels. Basic information 
such as membership renewals, premium 
collection, reimbursement to health 
facilities, members’ service utilization is 
often inconsistent, out of date, and not 
available at the scheme levels.
Enhance CBHI information management system.
During scale-up, it is necessary to explore the major 
issues around database management and reporting, and 
to design and implement an appropriate financial 
administration and management system (FAMS). If CBHI 
schemes are going to lead toward UHC in the informal 
sector, an electronic database management system at 
scheme, region, and national levels is needed.
EHIA, RHB, 
WorHO
Except in Oromia, more than 85% of 
CBHI reimbursements are paid to health 
centers, reflecting the high level of 
services provided to CBHI members by 
health centers compared to hospitals. 
The routine monitoring data suggest that 
CBHI schemes are contributing to the 
strengthening of financial capacities of 
lower levels of the health delivery 
system. In addition, there is an overall 
increase in utilization of health services 
as a result of CBHI. 
Do further research to assess how CBHI changes 
patterns of utilization in different levels of the 
health delivery and referral system. This 
assessment could further document the effect of CBHI 
on efficiency in the use of resources in the health system 
in pilot woredas. Related to this is a need to investigate 
the impact of CBHI on the use of maternal health and 
other preventive services. 
FMOH, EHIA, 
HSFR/HFG
Box 6.1: Results of effective communication
Effective communication and good understanding resulted in community members effectively defining what 
CBHI is as evidenced by a member and a non-member:
“CBHI is like Iddir [insurance for funeral services]. You become a member of iddir to ensure 
that you get buried when you die, and you become a member of CBHI so that you do not die 
(to ensure that you accessed the required health service before it is too late”), FGD 
participant in Yirgalem.
Non-CBHI members in Damboya were appreciative of CBHI even though they could not afford to join. One 
FGD participant stated that 
“ማአጤመ/ ‘MA’ATEMA’ [Kambata/ Amharic for CBHI] mean a hidden wealth and we 
consider it like that. CBHI is like our bank account….It is the most important of government 
strategies introduced.
99
REFERENCES
Acharya A., S. Vellakkal, F. Taylor, E. Masset, A. Satija, M. Burke, and S. Ebrahim. (2013). The Impact 
of Health Insurance Schemes for the Informal Sector in Low- and Middle-Income Countries: 
A Systematic Review, Policy Research Working Paper 6324, the World Bank. 
———. (2012). Impact of national health insurance for the poor and the informal sector in low- and 
middle-income countries: a systematic review. London: EPPI-Centre, Social Science Research 
Unit, Institute of Education, University of London. 
Cameron, A. Colin and P.K. Trivedi, 2009. Micro econometrics using STATA. College Station, TX: Stata 
Press. 
———. (1998). Regression analysis of count data. Cambridge, UK: Cambridge University Press. 
Evans, David B. et al. 2013. Measuring Progress toward Universal Health Coverage. World Bank and 
World Health Organization.
FMOH. (1998). Health Care and Financing Strategy 
Ghosh, Soumitra. (2010). Catastrophic Payments and Impoverishment Due to Out-of-Pocket Health 
Spending: The Effects of Recent Health Sector Reforms in India. Asia Health Policy Program 
working paper #15. 
Giedion, Ursula, Eduardo Andrés Alfonso, and Yadira Díaz. (2013). The Impact of Universal 
Coverage Schemes in the Developing World: A Review of the Existing Evidence. UNICO 
Studies Series 25, World Bank. 
Health Sector Financing Reform Project. (September, 2008). Brief Note on the Design of 
Community Based Health Insurance (CBHI): technical briefing Number 1. Bethesda, MD: Abt 
Associates. 
———. (2011). User fee revision in Ethiopia. Bethesda, MD: Abt Associates. 
———. (2009). Feasibility Studies on Community-Based Health Insurance Schemes in Amhara, Oromia, 
SNNP, and Tigray. Bethesda, MD: Abt Associates. 
Lagomarsino, Gina, Alice Garabrant, Atikah Adyas, Richard Muga, and Nathaniel Otoo. (2012). 
Moving toward universal health coverage: health insurance reforms in nine developing 
countries in Africa and Asia. Lancet 380: 933–43. 
Lu C., B. Chin, J.L. Lewandowski, P. Basinga, L.R. Hirschhorn, K. Hill, M. Murray, and A. Binagwaho. 
(2012). Toward Universal Health Coverage: An Evaluation of Rwanda Mutuelles in Its First 
Eight Years. PLoS One 7(6): e39282. doi:10.1371/journal.pone.0039282. 
Odeyemi, Isaac. (2014). Community-based health insurance programmes and the national health 
insurance scheme of Nigeria: challenges to uptake and integration. International Journal for 
Equity and Health 13. 
Schneider, Pia and Francois Diop. (October 2001). Impact of Prepayment Pilot on Health Care 
Utilization and Financing in Rwanda: Findings from Final Household Survey. Bethesda, MD: The 
Partners for Health Reformplus Project, Abt Associates Inc. 
Shimeles Abebe, Abebe. (2010). Working Paper 120 - Community Based Health Insurance Schemes 
in Africa: The Case of Rwanda." Working Paper Series 257, African Development Bank. 
100
Soors, Werner, Narayanan Devadasan, Varatharajan Durairaj, and Bart Criel.(2010). Community 
Health Insurance and Universal Coverage: Multiple paths, many rivers to cross. World 
Health Report (2010) Background Paper, 48. 
Spaan, Ernst, Judith Mathijssen, Noor Tromp, Florence McBain, Arthur ten Have, and Rob Baltussen. 
(2012). The impact of health insurance in Africa and Asia: a systematic review. Bulletin of the 
World Health Organization 90:685-692. 
Van Doorslaer, E, O O’Donnell, R Rannan-Eliya et al. (2006). Effect of payments for health care on 
poverty estimates in 11 countries in Asia: an analysis of household survey data, The Lancet, 
2006, 368, 1357-1364
Wagstaff, Adam and Eddy van Doorslaer. (2003). “Catastrophe and impoverishment in paying for 
health care: with applications to Vietnam 1993–1998.” Health Economics 12(11, November): 
921-33. 
Wang, Hong and Nancy Pielemeier. (2012). Community-Based Health Insurance: An Evolutionary 
Approach to Achieving Universal Coverage in Low-Income Countries. Journal of Life Sciences
6: 320-329. 
Yip, W. and W.C. Hsiao. (2008). The Chinese System at a Crossroads.” Health Affairs 27 (2); 460-
68. 
101
ANNEXES

103
ANNEX A: LITERATURE REVIEWED: 
EVIDENCE ON CBHI
Annex A.1: Positive and Negative Factors Influencing Uptake of CBHI and SHI in Sub-Saharan 
Africa, and Implications for Policymakers 
Factor Examples 
of countries
Issues identified and policy implications References
Factors positively linked to uptake
Provision of 
uniform benefit 
packages 
Ghana Benefits should be predefined and comprehensive, with good 
coverage of the likely disease burden
NHIS [Ghana] [9]; 
Odeyemi & 
Nixon [6]
Provision of services at accredited facilities helps to ensure 
uniformity of benefits offered
Adequate public 
financing/realistic 
pricing
Ghana, 
Rwanda, 
Burkina Faso
Use of funds from taxation is necessary to allow funding to be 
progressive and to encourage/enable the less well-off to join 
through subsidies and fee exemptions
NHIS [Ghana]; 
Odeyemi & 
Nixon; Logie; 
Schmidt; De 
Allegri; Parmar; 
Souares
Targeted subsidies positively influenced enrollment in Nouna, 
BF, although there is also a danger of adverse selection
Elimination or 
minimization of 
copayments
Rwanda Increases in subsidies to the point where co-payments are 
eliminated could lead to as much as 100% coverage
Dhillon [30]; 
Schneider & 
Hanson [31] User fees were found to be linked to substantial inequality in 
utilization, with medical visits being more common among the 
better-off uninsured
Strong desire/ 
willingness to join
Cameroon, 
Nigeria
Greatest willingness noted among poorest households in Nigeria Donfouet [14]; 
Onwujekwe [26]
Policymakers should commission research to determine 
willingness to pay; social marketing can encourage participation
Avoidance of 
focus on 
maximization of 
health revenue
Rwanda CBHI participation and a focus on the generation of health care 
revenues are mutually exclusive
Schmidt [29]
Improvements in 
education and 
socio-economic 
status
Burkina Faso Enrollment in schemes may increase with social and economic 
progress and development over the long term
De Allegri [35]
Provision of 
maternal health 
care benefits
Senegal, Mali, 
Ghana, 
Rwanda, 
Nigeria
Inclusion of maternal health care benefits drives interest in CBHI 
on the demand side, and CBHI is a primary contributor to 
strong maternal health services
Smith [28]; 
Bucagu [19]; 
Adinma [21]
Scheme organizers should ensure that packages are 
comprehensive, as excessive limitation discourages uptake
Awareness of the 
limitations of 
traditional 
medicine
Burkina Faso Further research is needed, but this observation emphases the 
value of improved education and communication
De Allegri [35]
104
Factor Examples 
of countries
Issues identified and policy implications References
Negative factors that discourage or limit uptake
Excessive 
requirement for 
OOP 
expenditure, 
inability to pay
Uganda, 
Burkina Faso, 
Guinea, 
Senegal, 
Nigeria
Major determinant of enrollment; even where implementation 
has been predominantly successful, the very poorest populations 
may still find participation financially difficult
Basaza; Dong; 
Criel; 
Onwujekwe; 
Metiboba; Jütting OOP expenditure remains significant in health care systems in 
many countries (despite actions such as abolition of user fees in 
government institutions in Uganda)
Regressive flat-rate payments are a problem in Nigeria, and 
inability to pay premiums is the single biggest obstacle in Uganda. 
No mechanism is in place to help those who cannot afford to 
join
Ambiguous and contradictory health care funding policy is a 
significant problem that must be addressed
Social exclusion 
due to religion or 
ethnicity
Senegal Noted in Senegal, where the Roman Catholic Church supports 
the mutuelles, and where Christians were reported in 2003 to be 
more likely than Muslims to enroll. In interviews, Muslims were 
under the mistaken impression that CBHI was open to 
Christians only
Jütting [27]
Lack of legal 
framework 
Guinea, Benin Failure to provide proper governance or official framework for 
CBHI schemes is linked to low enrollment
Soors
Lack of 
government (or 
donor) support
Uganda, 
Burkina Faso, 
Nigeria
Small budgets, low enrollment, and lack of government support 
cause schemes to fail. Schemes need substantial support to build 
their sustainability; technical and policy decisions should account 
for this
Kyomugisha [34]; 
De Allegri [35]
Excessively rigid 
enrollment 
requirements or 
institutional 
rigidity
Uganda, 
Burkina Faso
Failure to recruit the required number of people in a village has 
been a key feature affecting schemes in Uganda (mandatory 60% 
of a group or 100 families per village)
Basaza [32,33]; 
De Allegri [35]; 
Onwujekwe [25]; 
Onwujekwe [23] Rules for group membership should reflect what is achievable
Mismatch of 
values expressed 
by promoters and 
subscribers; 
failure to match 
benefits with 
willingness to pay
Senegal, 
Burkina Faso, 
Nigeria
Need to align expectations/needs of promoters (focus on 
financial sustainability) and subscribers (who look for 
sustainability and solidarity)
Ouimet [12]; 
Dong [15-17]; 
Onwujekwe [23]; 
Onwujekwe [24]; 
Metiboba [7]
Increase participation of members in decision making; failure to 
engage beneficiary participation in Nigeria has been pinpointed 
as a major problem
Ensure that prospective members are willing to pay for the 
benefits offered, and that the market in any locality matches the 
theoretical one on which projections are based
Lack of 
information
Uganda, 
Burkina Faso, 
Nigeria
Governments and promoters must ensure that schemes are 
properly and accurately publicized, and the public properly 
informed; lack of knowledge can lead to skepticism
Basaza [13]; De 
Allegri [35]; 
Dienye [22]; 
Onwujekwe [23]; 
Metiboba [7]
Lack of information is a significant problem in Nigeria
Authorities must ensure that government and health officials are 
fully informed about the packages offered
Poor quality of 
health care
Uganda, 
Guinea
Concerns relate to cleanliness, long queues before being seen, 
and lack of some prescribed medicines
Basaza [32,33]; 
Criel [39]
Noted as the main reason for lack of interest in the Maliando 
Mutual Health Organization in Conakry, Guinea
105
Factor Examples 
of countries
Issues identified and policy implications References
Lack of trust; 
perception that 
schemes are 
unfair or even 
unnecessary; 
dislike of health 
care personnel 
and cultural 
resistance
Uganda Belief that non-members are treated better in hospital than 
scheme members
Basaza [32,33]; 
Kyomugisha [34]
Integrity of fund managers and transparency of operation: 
“Nothing is done to ensure that fund managers account to 
scheme members” (Ugandan interview respondent)
Some members pay premiums continuously but never fall sick
“I wasn’t bothered since I am young and not likely to fall sick”; 
“If I do not fall sick, I should not pay for someone else” 
(Ugandan survey respondents)
Schemes must be fair, well run, and affordable, and the public 
sufficiently well-informed to appreciate the need for coverage 
and mutuality
High drop-out 
rates
Burkina Faso Related to other factors noted in this table: affordability, health 
needs and demand, quality of care and household characteristics
Dong [36]
Improve perception of schemes by heads of households, ensure 
that large households are able to maintain contributions (e.g., 
flexibility in payment options); ensure that service offered meets 
expectations (e.g., in line with education, etc.)
Annex A.2: CBHI development model: potential and pitfalls for scaling up 
Model 
Type Characteristics Potential Challenges and pitfalls
Basic 
model
 Proto-type bottom-up 
financial protection 
model
 Voluntary-based 
membership 
 Main sources of 
revenue is membership 
premium
 Managed by community 
committee or health 
facility
 Services offered could 
be either or both 
outpatient and inpatient
 Payment mechanisms: 
Fee-for-service and 
capitation
 Make available risk protection 
mechanism at local level;
 Raise awareness and knowledge 
for health insurance;
 Reduce OOP payments; and
 Increase utilization of health 
services by members
 Not politically and financially 
supported by government and 
this undermines scheme’s 
stability;
 Voluntary membership has risk 
of adverse selection that may 
undermine financial 
sustainability; 
 Limited benefit package limiting 
schemes attractiveness;
 Coverage is also another issue. 
Most of the CBHI schemes in 
West Africa are voluntary, 
small, and with very low 
coverage;
 Limited risk pool (not large 
enough to make them 
sustainable).
Enhanced 
CBHI
 The adaptation to basic 
model basic above 
include: 
 Local government 
political endorsement, 
including the poorest of 
the poor through 
government subsidies, 
 Building networks for 
scheme management 
and service delivery 
 Government political endorsement 
increasing CBHI’s political 
stability/legitimacy; 
 Government and/or other donors’ 
subsidies increase scheme capacity 
to reach the poorest of poor, 
thereby increasing the equity of 
CBHI coverage; 
 Government also may provide 
financial support or a re-insurance 
mechanism to protect against 
 Lack of political, financial, and 
technical commitment and 
stewardship at the national level 
prevents isolated schemes from 
scaling up; 
 Fragmentation of a large number 
of separate small funders limits 
broader risk pooling; 
 Financial risk protection is still 
constrained by limited 
membership contributions from 
106 
Model 
Type Characteristics Potential Challenges and pitfalls
expenditure fluctuations and 
maintain financial sustainability;
 Small schemes form a regional or 
national network that can provide 
professional technical and 
managerial support on design and 
management; 
 A group enrolment requirement 
will reduce adverse selection; and
 Introduction of other payment 
mechanisms to control costs, and 
case management techniques, will 
limit expenditure fluctuations.
low-income informal sector 
members;
 Contribution capability and 
benefits may not be consistent 
across schemes, which leads to 
the inequity in health financing 
and service access; 
 Lack of professional and 
standardized management limits 
scheme efficiency and 
effectiveness; and 
 Ability to ensure the availability 
of the service, to improve the 
quality.
Nation￾wide 
Scheme
 New characteristics of 
a nationwide scheme 
include 
 Government political 
commitment, 
stewardship, legislation, 
 Funding support, 
 Regional-level 
professional 
management, and 
 Continuing community￾level support

 Political commitment, stewardship, 
and guidance, along with legislation 
and regulation backup; 
 Continuing strong community 
support in resource mobilization 
and fraud and abuse controls 
enable scale-up of CBHI nationally; 
 Increasing the size of risk pools at 
the regional or higher level and 
establishing risk equalization 
mechanisms to allocate resources 
across insurance schemes allows 
for cross-subsidy between high￾risk and low-risk regions; 
 Government regular budget 
support makes additional financial 
resource available and sustainable;
 Increased government leadership 
and financial and technical support 
enhances CBHI management 
capacity and helps control 
overhead costs; 
 Strategic service purchasing makes 
the scheme more effective and 
efficient; and 
 Government establishment and 
implementation of M&E systems 
ensures scheme more sustainable.
 Though has potential to scale up 
CBHI for achieving UHC, it 
faces challenges including:
 Continue to increase enrolment 
rate by reaching hard-to-reach 
populations without introducing 
mandatory mechanisms; 
 Control the costs of health 
services; 
 Ensure long-term financial 
sustainability; and 
 Integrate CBHI with other 
existing health insurance 
schemes, such as tax-based or 
SHI covering formal sector 
populations.

Source: Wang and Pielemeier (2012) 
107
ANNEX B: HOUSEHOLD SURVEY 
QUESTIONNAIRE
INSTRUCTIONS FOR HANDLING AND COMPLETING CBHI QUESTIONNAIRE 
 Please use only black ball points to complete questionnaire 
 Only eligible and trained data collectors should write on these questionnaire 
 Please fill the header in each page 
 Data correction: Cross-out the mistake (mistake has to remain readable), write the correction alongside 
together with your initials and date of correction. In case of non-self-explanatory mistakes, please add the 
reason for correction. Do not use type write correction fluid (Tipp-Ex) 
E.g. (DD/MM/YYYY) 
 
 1 MMK
 In open boxes/numeric fields please enter 
 Numbers Or ticks 
 Always enter digits right aligned and fill open spaces to the left with zeroes
 Please mark data which could not be recorded as follows: Cross out boxes and write “NOT DONE” on 
the side 
 Date: Day. Month. Year: (GC) 
 
 Date Month Year 
 Please enter initials in the following order: First letter of the first name, First letter of the middle name, First 
letter of the surname 
 Please do not omit to date and sign the pages where required. 
2 4 / 0 3 / 2 0 1 6
4 9
0 4 9
0 4 / 0 3 / 2 0 1 6
M M K
108
COMMUNITY BASED HEALTH INSURANCE (CBHI) EVALUATION 
HOUSEHOLD QUESTIONNAIRE (VERSION ONE) 
 DATA COLLECTION VISIT DATES: (DD/MM/YYYY) 
Start time (hh:mm) |__||__||__||__|Interview end time |__||__||__||__| 
FOR FIELD USE (DATA COLLECTOR AND SUPERVISOR) AND OFFICE USE (DATA ENTRY 
CLERKS) 
Collected by Supervised by Checked by Entered by 
Initials |__|__|__| |__|__|__| |__|__|__| |__|__|__|
Date Day |__||__|
Month |__||__|
Year|__|__|__|__|
Day |__||__|
Month |__||__|
Year|__|__|__|__|
Day |__||__|
Month |__||__|
Year|__|__|__|__|
Day |__||__|
Month |__||__|
Year|__|__|__|__|
Full 
name 
Signature
CBHI Survey: / / 2 0 1
109
FORM ONE: AREA IDENTIFICATION OF THE SELECTED HOUSEHOLD 
CRITERIA FOR RESPONDENT 
Only head of household or spouse can be used as respondents. The head of HH has to be a living member of the 
HH and determined by the HH members themselves. The head of HH can be female. (If the head of household or 
spouse cannot provide information the interviewer can ask the de facto head of HH (e.g. member who earns main 
income.) 
TICK THE ONE YOU INTERVIEW 
1.8 HOUSEHOLD HEAD 
1.9 SPOUSE 
1.10 DE FACTO HEAD 
1.1 1.2 1.3 1.4 1.5 1.6
Region Zone Woreda Kebele/Tabia Village/got Full name of Household Head
1.7 Household Participate 
CBHI 
(tick below)
Not in CBHI
(tick below)
Rural
(tick below)
Urban
(tick below)
110
FOR OFFICE USE ONLY (ADDIS ABABA)
Status Name Signature Date
Editor
Editor-Verifier
111
FORM TWO: HOUSEHOLD CHARACTERISTICS 
Serial numbers of the household members
2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8
Name of the 
household 
members
Relation to household head HH
Size
Sex Marital Status Age Religion Ethnicity
Please give me 
your full name and 
list names of each 
household 
member,
starting with 
household head 
followed by 
spouse
1=Head
2=Spouse
3=Son/daughter
4=Mother/Father of head/spouse
5=Sister/Brother of head/Spouse
6=Adopted/ Foster children
7=Other Relatives
8=Paid domestic worker
9=Non-Relatives
[number]
1=male
2=female
1=Married
2=Single
3=Divorced/Separated
4=Widowed
6= Other (specify)
[ years]
1=orthodox 
Christian
2=Catholic
3=Protestant
4=Other 
Christian
5=Muslim
6=None
7=Other 
(specify)
1=Amhara
2=Oromo
3=Tigryan
4=Gurage
5=Kambata
6=Hadya
7=Wolayta
8=Sidama
9=Agew
10=Gedio
11=Other (specify)
This is only for ages above 15 
otherwise fill n/a.
01 1=head
02
03
04
05
112
06
07
08
09
10
113
FORM THREE: EDUCATION OF HOUSEHOLD MEMBERS: For only household members age five and above otherwise fill N/A. 
Serial numbers of the household members
Serial numbers of the household members
3.1 3.2 3.3 3.4 3.5 3.6 3.7
Name of the 
household 
members
Can [name] 
read and 
write in any 
language?
Has [name] ever 
attended school 
(formal 
education)
What is the highest grade [name] completed? Is [name] 
currently 
attending 
school?
Which grade is [name] currently 
attending?
Is [name] 
currently 
member of 
CBHI 
scheme?
Copy all the 
household 
members from 
form two 
question number 
2.1.
Copy only 
household 
members age 
greater than 5 1=Yes [skip 
3.3]
2=No 
1=Yes
[skip to 3.4]
2=No
[skip to3.5]
0=none
1=Grade 1 
2=Grade 2
3=Grade 3
4=Grade 4
5=Grade 5
6=Grade 6
7=Grade 7
8=Grade 8
9=Grade 9
10=Grade 10
11= Grade 11
12=Grade 12
13=Certificate from 
TVET/TTC
14=College Diploma 
15=University 
16=Non-formal education 
17= Adult education
18=others (specify)
1=Yes
[skip to 
3.6 ]
2=No
0=none
1=Grade 1 
2=Grade 2
3=Grade 3
4=Grade 4
5=Grade 5
6=Grade 6
7=Grade 7
8=Grade 8
9=Grade 9
10=Grade 10
11= Grade 11
12=Grade 12
13=Certificate from 
TVET/TTC
14=College Diploma 
15=University 
16=Non-formal 
education
17= Adult education
18=others (specify)
1=Yes
2=No
114
01
02
03
04
05
06
07
08
09
10
115
FORM FOUR: HOUSEHOLD LIVESTOCK OWNERSHIP 
Types of livestock Number of livestock currently own
Ask the household if they have any of the following livestock 
assets Do you 
have these 
animals?
1=Yes
2=No
How many of these animals do this 
household currently own? 
[Number]
CATTLE 
4.1 Plough oxen
4.2 Fattened ox
4.3 Cows
4.4 Heifer 
4.5 Bull
4.6 Calf 
SMALL RUMINANTS 
4.7 Goat
4.8 Sheep
4.9 Fattened sheep
TRANSPORT ANIMALS
4.10 Donkey 
4.11 Horse 
4.12 Mule 
OTHER
4.13 Chicken 
4.14 Beehives 
4.15 Others (Specify) 
4.16 Others (Specify)
116
FORM FIVE: HOUSEHOLD ASSETS 
Types of assets Do you 
have these 
assets
Number of assets currently own
Ask the household they have any of the following assets 1=Yes
2=No
How many of these items do this household 
own? 
[Number]
5.1 Modern beds
5.2Chairs/bench 
5.3 Radio/TV (working)
5.4 Computer/laptop
5.5 Mobile/cell-phone/wireless
5.6 Modern stoves
5.7 Bicycle
5.8 Knapsack sprayer
5.9 Water pump
5.10 Motorbike
5.11 Other (specify) 
117
FORM SIX (1): LAND USE AND PRODUCTION IN THE LAST 12 MONTHS. 
CROP CODES NUMBER QUESTIONS ANSWERS
CEREALS 
1= Teff
2=Barely 
3=Wheat
4=Maize
5=Sorghum
6=Finger millet
7=Oats
8=Rice
PULSES 
9=Bean
10=Peas
11=Boleke
12=Chickpeas
13=Lentils
14=Grass peas
15= Haricot beans
16= Fenugreek
17= Soya beans
18=Gibto
OILSEEDS
19=Nueg
20= Linseed
21= Groundnuts
22= Sunflower
23= Sesame
24= Rapeseed
ROOT CROPS
25=Beet root 
26=Carrot
27= Red onion 
28= White onion
29= Potatoes
6.1
Did you rent/lease out land over the last 12 
months?
6.2
How much land did you rent/lease out in 
hectares?
6.3
Did you rent/lease in land last 12 months?
6.4
How much land did you rent/lease in hectares?
6.5
How much cropland was irrigated in the last 12 
months in hectares?
6.6 How much cropland does your household own 
in hectares?
6.7
How much cropland (own and rented/lease in) 
was cultivated by the household over the last 12 
months in hectares (annual and perennial)?
6.8
Which crops did you grow/cultivate in the last 
12month Production period? [write crop 
codes]
6.9 
How much land (in hectares) was cultivated with 
[crop] in the last 12 months
[listland size for each crops]
6.10
In the past 12 months, how many Kilo grams of 
each cropdid you produce/harvest? [ list 
production volume for each crop]
6.11
In the past 12 months how many kilo grams did 
you sell? [list amount sold for each crops]
6.12 
What was the price of your sell per kg? [list sell 
price for each crops]
118
30= Sweet 
potatoes
31=others 
VEGETABLES 
32=Lettuce 
33=Head Cabbage 
34= Tomatoes
35= Green 
peppers
36= Red peppers
37= Others
FUITS AND 
OTHER CASH 
CROPS
38=Avocado
39=Lemon
40=Zeytun
41=Mango
42=Orange
43=Papaya 
44=Pinapple
45=Gesho
46=Sugar cane
47=coffee
48=chat
49=Others
119
FORM SIX (2): HOUSING CHARACTERISTICS (URBAN ONLY) 
6.13 In what type of dwelling does the household head 
live?
1= Permanent building 
2= Semi-Permanent 
3= Temporary 
4= Traditional
Indicate code:|__|
6.14 Do you have a domestic worker living with you?
1=Yes
2=No
Indicate code:|__|
6.15 Is your dwelling owned by your household or 
rented, or do you reside here without payments?
1= Owned by family or one of its members 
2= Rented
3= Occupied without payment 
4= Other (specify________) 
Indicate code:|__|
6.16 What is the main type of material for the floor in 
your house?
1. Mud/crow dung 
2.Stone
3.Cement/bricks
4.Wood
5.Grass
6.Iron sheets
7.Tiles
8.Other (specify)
Indicate code:|__|
6.17 What is the main type of material that your walls 
are made of?
1. Mud/crow dung 
2.Stone
3.Cement/bricks
4.Wood
5.Grass
6.Iron sheets
7.Tiles
8.Other (specify)
Indicate code:|__|
6.18 What is the main type of material that your roof is 
made of?
1. Mud/crow dung 
2.Stone
3.Cement/bricks
Indicate code:|__|
120
4.Wood
5.Grass
6.Iron sheets
7.Tiles
8.Other (specify)
6.19 What is your main source of cooking fuel?
1 = Firewood
2 = Charcoal 
3 = Kerosene 
4 = LPG (Gas)
5 = Electricity 
6 = Other (specify) ___________
Indicate code:|__|
6.20 What is your main source of lighting? 
1.Electricity 
2.Kerosene
3.Gas
4.Candle 
5.Firewood
6.Solar
7.Other (specify) ___________
Indicate code:|__|
6.21 What is the main source of water for the 
household?
1. Piped into residence 
2. Piped into the compound or plot
3. Public well 
4. Public tap
5. Well/borehole with pump in the 
compound/well in the residence or compound 
or plot
6. Rainwater collection 
7. Well without hand pump 
8. Pond/River/Stream/Dam 
9. protected spring 
10. Unprotected spring 
11.Others (specify)
Indicate code:|__|
121
6.22 What type of toilet facility does the household use?
1.Own flush toilet with sewage/septic tank
2.Shared flush toilet in area
3.Traditional pit latrine
4.Ventilated Improved Pit Latrine
5.Bush or field
6.Bucket Latrine 
7.Other:____ (specify)
Indicate code:|__|
122
FORM SEVEN: HEALTH CARE UTILIZATION: Outpatient Treatment (when the patient is gets 
consultation and/or other services without being admitted to stay at health facility) Serial numbers of the household members Serial numbers of the household members
7.1 7.1 7.2 7.3 7.4 7.5 7.6 7.7
Name of the 
household 
members
How do you 
rate the 
health status 
of [name]?
Has [name] 
fallen ill in 
the last 4 
weeks?
What was the 
major cause of 
illness for 
[name]?
Has [name] 
visited any 
health facility 
for the illness 
felt in the last 
4 weeks?
If yes to 7.4, 
what type of 
service did 
[name] 
receive?
What was the 
major reason 
why [name] 
did not visit a 
health facility?
Number of visits 
Copy all the 
household 
members from 
form two 
question 
number 2.1 
Copy only 
household 
members age 
greater than 5 
1=very good
2=good
3=acceptable
4=poor
5=very poor
1=Yes 
2=No [ skip 
to the next 
household 
member]
1=fever
2=lungs & 
breathing 
(cough)
3=stomach & 
bowel ache
4=teeth & 
gums
5=headache
6=joint pain
7=pain in the 
chest
8=irritation of 
the eye
9=watery 
faeces
10=blood 
stained faeces
11=vomit
12=wound
13=other, 
specify
1=Yes [skip to 
7.6]
2=No [ skip to 
the next 
household 
member]
1=outpatient
2=inpatient
3=both
1= did not feel 
it was 
necessary
2=facility too 
far
3= lack of 
money
4=did not feel 
that I would 
get quality care 
e
5= no CBHI 
coverage
6=other, 
specify
[number]
1st time
2nd
time
3rd
time
01
02
123
03
04
05
06
07
08
09
10
124
FORM SEVEN: HEALTH CARE UTILIZATION -Outpatient Treatment (CONTD.)
Serial numbers of the household members
Serial numbers of the household members
Name of the 
household 
members
7.8 7.9 7.10 7.11 7.12
Type of facility visited Distance to 
the health 
Facility (in 
Kilo meter)
Travel time 
(in hours)
Reason for choice of 
this facility Is [name] 
enrolled in 
CBHI? 
Copy all the 
household members 
from part two 
question number 2.1 
1=Health post
2=Health center
3=Private clinic
4=Mission/NGO clinic
5=Public hospital
6=Private hospital
7=Mission/NGO hospital
8=Pharmacy/drug store
9=traditional healer
10: home service by 
HEWs
11=Other (specify)
[Number] [Number]
1=Proximity
2=Inexpensiveness
3=Staff are always 
available
4=Medicines are 
available
5=Short waiting time
6=Staff are more 
compassionate
7=Staff more capable
8=CBHI coverage
9=Referral from first 
visited facility
10=other, specify
1=Yes 
2=No 
01
02
03
04
05
06
07
08
09
10
125
Serial numbers of the household members
Serial numbers of the household members
Name of the 
household 
members
7.13 7.14 7.15 7.16 7.17
How much money 
has the household 
spent for 
transportation, 
consultation diagnosis, 
and medicine
Main source of 
financing How satisfied is 
[name] with 
Waiting 
time 
Availability of 
drugs/supplies 
Copy all the 
household members 
from part two 
question number 2.1 
Transport 
Consultation 
Diagnostics 
Medicine 
1=Own saving
2=Reduce HH 
food 
consumption (if 
the HH 
purchases food)
3. Reduce HH 
nonfood 
consumption
4=Sell of assets
5=Sell of food 
stocks
6=Borrowing
7=Remittance 
(in cash or in 
kind)
8= Community 
based health 
insurance 
(CBHI)
9= fee waivers 
from 
Kebele/Tabia/w
oreda
10= Increase 
sell of labor 
11=provided 
free for all
12= Other 
(specify)_____
1=Very satisfied
2=Satisfied
3=Indifferent 
4=Dissatisfied
5=very 
dissatisfied 
1= Less 
than 30 
minutes 
2=30 to 60 
minutes 
3= 1to 3 
hours 
4=3 to 6 
hours
5= 6 hours 
and more
5=More 
than a day 
1=Not 
available 
2=Rarely 
available
2=Usually 
available
3= always 
available
Diagnostics 
Cleanness of facility 
Courteousness of staffs
Waiting time 
Availability of drugs/supplies 
01
02
03
04
05
06
126
07
08
09
10
127
FORM SEVEN: HEALTH CARE UTILIZATION- maternal and child health services 
Serial numbers of the household members
Serial numbers of the household members
7.18 7.19 7.20 7.21 7.22 7.23 7.24 7.25
Name of 
the 
household 
members
Were 
there 
pregnant 
women in 
this 
household 
in the last 
12 months?
Did [name] 
have antenatal 
care visit to a 
health facility 
during her 
pregnancy?
Number of 
visits 
Type of 
facility 
visited for 
ANC
Did [name] 
deliver in a 
health 
facility?
Type of 
facility for 
delivery
Did [name] 
receive 
immunizatio
n/? ( to be 
asked only 
for 
household 
members 
who are 
under 5 
years of 
age)
If yes, state 
number of 
times 
immunization 
was received
Copy all 
the 
household 
members 
from form 
two 
question 
number 2.1 
Copy only 
household 
members 
age greater 
than 5 
1=Yes 
[indicate 
HH 
member 
code 
below]
2=No [ 
skip to 
7.24]
1=Yes 
2=No [ skip to 
7.22]
[number]
1=Health 
post
2=Health 
center
3=Private 
clinic
4=Mission
/NGO 
clinic
5=Public 
hospital
6=Private 
hospital
7=Mission
/NGO 
hospital
8=Pharma
cy/drug 
store
9=traditio
nal healer
10: home 
service by 
HEWs
11=Other 
(specify)
1=Yes 
2=No [ 
skip to 7.24 
]
1=Health 
post
2=Health 
center
3=Private 
clinic
4=Mission/N
GO clinic
5=Public 
hospital
6=Private 
hospital
7=Mission/N
GO hospital
8=Pharmacy/
drug store
9=traditional 
healer
10: home
service by 
HEWs
11=Other 
(specify)
1=Yes 
2=No [ skip 
to 7.26]
1st time
2nd time
3rd time
4th time
1st time
2nd time
3rd time
4th time
01
02
03
04
05
06
128
07
08
09
10
129
FORM SEVEN: HEALTH CARE UTILIZATION inpatient Treatment (when the patient is 
admitted by health professional to stay at health facility) Serial numbers of the household members
Serial numbers of the household members
7.26 7.27 7.28 7.29 7.30
Name of the 
household 
members
Number of 
hospitalization in each 
episode
Health care facility Distance to the 
facility
Travel time 
to the facility 
Reason for choice of this 
facility
Copy all the 
household 
members from 
part two 
question 
number 2.1 or 
3.2.
After copying, 
use for only 
household 
members' age 
greater than 5 
other wise fill 
n/a
[fill number of days]
1=Health post
2=Health center
3=Private clinic
4=Mission/NGO clinic
5=Public hospital
6=Private hospital
7=Mission/NGO 
hospital
8=Pharmacy/drug store
9=traditional healer
10=Other (specify)
[kilometers] [hours] 1=Proximity
2=Inexpensiveness
3=Staff are always available
4=Medicines are available
5=Short-waiting time
6=Staff are more 
compassionate
7=Staff more capable
8=CBHI coverage
9=Referral from first 
visited facility
10=other, specify 1st episode
2nd
episode
3rd
episode
01
02
03
04
05
06
07
08
09
10
130
FORM SEVEN: HEALTH CARE UTILIZATION inpatient Treatment (CONTD.) Serial numbers of the household members
Serial numbers of the household members
Name of 
the 
househol
d 
members
7.31 7.32 7.33 7.34 7.35
How much money 
has the household 
spent on
Main source of 
financing How satisfied is 
the [name] with 
Waiting time Availability of 
drugs/supplies 
Copy all 
the 
household 
members 
from part 
two 
question 
number 2.1 
Copy only 
household 
members' 
age greater 
than 5 
Transport 
Consultation 
Diagnostics 
Medicine 
Bed
1=Own saving
2=Reduce HH food 
consumption (if the 
HH purchases 
food)
3. Reduce HH 
nonfood 
consumption
4=Sell of assets
5=Sell of food 
stocks
6=Borrowing
7=Remittance (in 
cash or in kind)
8= Community 
based health 
insurance (CBHI)
9= fee waivers 
from 
Kebele/Tabia/wore
da
10=Increase sell of 
labor 
11=provided free 
for all
12= Other 
(specify)___
1=Very satisfied
2=Satisfied
3=Indifferent 
4=Dissatisfied
5=very dissatisfied 
1= Within a day 
2=within a week 
3= within two 
weeks 
4=Within a 
month
5=After a month 
1=Not 
available 
2=Rarely 
available
2=Usually 
available
3= always 
available
Diagnostics 
Cleanness of facility 
Courteousness of staffs
Waiting time 
Availability of drugs/supplies 
Cleanness of bed and food
Professional care
01
02
131
03
04
05
06
07
08
09
10
132
FORM EIGHT: CBHI MODULE 
Part I: Awareness of CBHI (including both CBHI and non-CBHI woreda
1. Have you ever heard about community based health insurance program (CBHI)? [non-CBHI members 
only] 
1=Yes
 2=No Indicate code:|__|If No skip to 2
2. From whom did you hear about CBHI? 
 1= neighbors/friends 
 2= CBHI officials in public meeting 
 3= CBHI house to house awareness creation campaigns Indicate code:|__|
 4= mass media: ETV, radio 
5= health professionals in health facilities 
 6=others, specify___________________________________ 
3. Perceptions about CBHI [put ticks on the space provided] 
Correct 
(1)
Not correct 
(2)
Do not know (3)
3.1 Only those who fall sick should consider enrollment in CBHI.
3.2 Only the very poor who cannot afford to pay for healthcare 
need to join the schemes.
3.3Under CBHI program, you pay money (premiums) in order for 
the CBHI to finance your future health care needs?
3.4 CBHI program are like savings scheme, you will receive interest 
and get your money back.
3.5 If you do not make claims through CBHI, your premium will be 
returned.
4. When did you enroll into CBHI? ___________ [CBHI members only] 
Part II: Participation (Only CBHI woredas) 
5. Have you or any of your HH members attended any CBHI related meetings/training before CBHI 
was implemented in your kebele/tabia? 
1 = Yes 
2 = No Indicate code: |__|
6. In these meetings, did you or any of your HH members discuss and agree the timing/time interval of 
the regular contribution (premium)? 
1=Yes 
 2=No (if no, go to 15) Indicate code: |__| 
133
Part III. (Re-) Enrollment (Only CBHI woreda) 
7. Is your HH member enrolled in the CBHI program? 
1 = Yes 
2 = No Indicate code: |__|
8. Why has your household decided not to enroll in the CBHI program (multiple responses allowed-list 
in order of importance)? (code 2) ____, _____, _____
 1=Illness and injury does not occur frequently in our HH 
 2=the registration fee and premiums are not affordable 
 3 = want to wait in order to confirm the benefits of the scheme from others Indicate code: |__|
 4 = we do not know enough about the CBHI scheme Indicate code: |__|
 5 = there is limited availability of health services Indicate code: |__| 
 6= the quality of health care services is low 
 7 = the benefit package does not meet our needs 
 8= Lack of confidence in scheme management 
 9 = other reasons, please specify 
9. Do you plan to enroll in the CBHI scheme in the future? 
1=Yes 
2=No Indicate code: |__| 
10. Is your household enrolled in any other solidarity group (e.g. iddir, equb, microfinance, other informal 
systems etc) active in your area? 
1=Yes 
2=No Indicate code: |__| 
11. Are all members of your household enrolled in the CBHI program? 
1=Yes 
2=No Indicate code: |__| 
12. If no to question 13, why are some members of the HH are not enrolled in CBHI? 
 1=members are healthy 
 2=do not have enough money to pay for all Indicate code: |__| 
 3=members are not nuclear family members 
4=others, specify 
134
13. Why has your household or any of your HH members decided to enroll in the CBHI program 
(multiple answers allowed-list in order of importance)? 
 1=Illness and/or injury occurs frequently in our HH 
 2= our HH members need health care 
 3= To finance health care expenses 
 4= CBHI registration and premium is paid by the government Indicate code: |__| 
 5=Premium is low compared to the user fee price to obtain medical treatment Indicate code: |__| 
 6= Pressure from other family members/community Indicate code: |__| 
 7= Pressure from the kebele/tabia administration 
 8= others please specify 
14. Who paid for the enrollment fee? 
1=HH contribution 
2= local government (coverage for Indigent HH) Indicate code: |__| 
15. How long does it take, after payment of registration fee and premium, to start utilizing health 
services? _________ days 
16. Where Do you pay the premiums 
 1=at the CBHI office 
 2=Kebele/Tabia administration 
3=Official comes and collects Indicate code: |__| 
 3=others, please specify---------------------- 
17. When you current membership expires would you renew your CBHI membership for the following 
year? 1=Yes 
 2=No (if yes, go to question19) Indicate code: |__| 
18. If yes, what is the highest amount you are willing to pay to renew your membership? _____ ETB 
19. If no, why do you plan not to renew your CBHI membership (multiple responses allowed-list in 
order of importance)? 
 1= Illness and injury does not occur frequently in our HH Indicate code: |__| 
 2= the registration fee and premiums are not affordable Indicate code: |__| 
 3= there is limited availability and poor quality of health services Indicate code: |__| 
 4 = The quality of service (waiting time, staff attitude, medicine, diagnostics) for CBHI members is not 
as good as for out of pocket paying patients (non-CBHI members) 
 5= other, specify 
135
20. Have you made use of your CBHI membership to cover health costs? 
 1=Yes 
2=No Indicate code: |__| 
21. Why has your household not benefitted from CBHI? 
1 = No one in my HH has visited health facilities 
2= We still pay other additional costs for treatment (specify) 
3= The quality of service (waiting time, staff attitude, medicine, diagnostics) for CBHI members is not 
as good as for out of pocket paying patients (non-CBHI members) 
4= Delays in issuance and distribution of CBHI ID cards 
5=others, specify
Part IV: Affordability and expectations 
Please respond to the following statements and indicate your level of agreement 
22. The timing/time interval of premium payment is convenient for my household. 
 1=disagree, 2=indifferent, 3=agree Indicate code: |__| 
23. The CBHI registration fee is 
1=easily affordable 2=somewhat affordable 3=unaffordable Indicate code: |__| 
24. The CBHI regular contribution (premium) is 
1=easily affordable 2=somewhat affordable 3=unaffordable Indicate code: |__| 
Part V: CBHI experience 
(The following questions should be asked only to households who are enrolled in the CBHI program) 
1. The local CBHI agent tries hard to solve CBHI implementation problems 
 
 1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
2. The community (CBHI members) have the right to guide and supervise the activities of the CBHI 
management 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
3. Health professionals treat patients of CBHI membership as much as out of pocket paying patients 
(none members) 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
4. The CBHI benefit package meets the requirements of my household. 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
136
5. The local CBHI management is trustworthy. 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
6. I am satisfied with the experience at the local CBHI office when I go to register? 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
7. I am satisfied with the local CBHI office when I go to pay the regular contribution (premium)? 
1= Disagree 2= indifferent 3=Agree Indicate code: |__| 
8. I am satisfied with quality of healthcare services provided by the contracted provider? 
 1=Excellent 2=Good 3=Fair 4=Poor 5=Very poor 6=I don’t know Indicate code: 
|__| 
9. Have you benefited from the CBHI scheme? 
1=Yes 2=No Indicate code: |__| 
10. If yes to question 9, how does your household benefit? (Multiple responses possible-list in order of 
importance) (code 5) 
1= Increased access to health care 
2 = Reduced costs of health care 
3 = Reduced concerns about expected health care costs Indicate code: |__| 
4 = Others, please specify 
11. If No to question 9, why do you think your household has not benefited? (multiple responses 
possible-list in order of importance) (code 6) 
1 = No one in my HH has visited health facilities 
2= We still pay other additional costs for treatment (specify) Indicate code: |__| 
3= The quality of service (waiting time, staff attitude, medicine, diagnostics) for CBHI members is not 
as good as for out of pocket paying patients (non-CBHI members) 
4= Delays in issuance and distribution of CBHI ID cards 
5=others, specify 
137
FORM NINE: FOOD EXPENDITURE AND FOOD CONSUMPTION IN THE LAST SEVEN 
DAYS: 
9.1 9.2 9.3 9.4 9.5
Food Types Did your 
household 
consume 
this food?
Did you purchase any in the 
last week? How much was 
purchased? 
How much was spent? If the 
household did not purchase 
any write zero.
Did you consume this 
food from your own 
harvest or your own 
stock? If the household 
did not use any from the 
stock write zero
Did you receive this food as a 
GIFT, as Wage IN KIND, loan? 
How much? If the household did 
not get any fill zero. Write the 
codes of source listed below on 
this page.
1=Yes
2=No
Amount Unit Total 
Expenditu
re
Amoun
t
Unit Est. 
cost
Amou
nt
Unit Est. 
cost
Source9
Teff
Barley (Gebis) 
Wheat/ Durahh 
(Sinde) 
Maize(Bekolo/Bahir
mashla) 
Sorghum (Mashila; 
dagusa)
Millet (Zengada)
Lentils (Misir)
Horse Beans 
(Bakela)
Cow Peas (Ater)
Chick Peas 
(Shimbra)
Milk/yoghourt 
(ergo)
Beef (yekebitsiga)
Mutton (yebeg)/goat
meat(yefiyelsiga)
Chicken
 
9 1= Relatives 2= neighbors 3= Government 4= NGO 5= Food for work 6= Daily labor 7= Loan 8= crop by crop exchange 
9=Other (specify) 
138
9.1 9.2 9.3 9.4 9.5
Food Types Did your 
household 
consume 
this food?
Did you purchase any in the 
last week? How much was 
purchased? 
How much was spent? If the 
household did not purchase 
any write zero.
Did you consume this 
food from your own 
harvest or your own 
stock? If the household 
did not use any from the 
stock write zero
Did you receive this food as a 
GIFT, as Wage IN KIND, loan? 
How much? If the household did 
not get any fill zero. Write the 
codes of source listed below on 
this page.
1=Yes
2=No
Amount Unit Total 
Expenditu
re
Amoun
t
Unit Est. 
cost
Amou
nt
Unit Est. 
cost
Source9
Eggs
butter/cheese
tella/tej
birra (bottled)
Chat
araqi/kathikala
soft drinks
Coffee
Sugar
Honey
Salt
cooking oil
spices/karia/berbere
bread (dabo)
macaroni/spaghetti
Potatoes
sweet potatoes
green leafy 
vegetables
139
FORM TEN: NON-FOOD EXPENDITURE: THE LAST 12 MONTH TOTAL EXPENDITURE IN BIRR. 
FORNON PURCHASED ITEM FILL ZERO.
10.1 10.2 10.3 10.1 10.2 10.3
TYPES OF EXPENDITURE Did you 
purchase?
EXPENSE TYPES OF EXPENDITURE Did you 
purchase?
EXPENSE
1=Yes
2=No
BIRR 1=Yes
2=No
BIRR
1 EXPENDITURE ON CLOTHES 
AND RELATED 
3.9 Rent for oxen
1.1 Clothes/shoes/fabric for MEN 3.10 Compensation, penalty and 
credit
1.2 Clothes/shoes/fabric for WOMEN 3.11 others
1.3 Clothes/shoes/fabric for BOYS 4 EXPENDITURE ON HEALTH 
1.4 Clothes/shoes/fabric for GIRLS 4.1 Modern medical treatment 
and medicines 
1.5 Cloth/shoes/fabric for babies 4.1.1 Drugs
1.6 Linens (sheets, towels, blankets) 4.1.2 Fees
1.7 Cosmetics (hair oil, butter, perfume) 4.1.3 Transportation
1.8 Others 4.1.4 Lodging/board 
2 EXPENDITURE ON HOUSEING 
AND RELATED 
4.1.5 Others 
2.1 Kitchen equipment (cooking pots, 
etc.) 
4.2 Traditional medicine and 
healers 
2.2 Furniture 4.2.1 Drugs
2.3 Lamp/torch 4.2.2 Fees
2.4 Sieve (wonfiet), gourd (kil), sefed, 
mesob, etc
4.2.3 Transportation
2.5 Transport (other than health 
purpose, crop production and crop 
sale)
4.2.4 Lodging/board 
2.6 Building materials 4.2.5 Others 
2.7 Repair and maintenance 5 EXPENDITURE ON 
EDUCATION
2.8 Matches 5.1 Education (school fees) 
2.9 Batteries 5.2 Others 
2.10 Candles (tua’af), incense 6 EXPENDITURE ON 
AGRICULTURAL 
INPUTS AND LIVESTOCKS
2.11 Hand soap, Laundry soap, 6.1 Fertilizer
140
10.1 10.2 10.3 10.1 10.2 10.3
TYPES OF EXPENDITURE Did you 
purchase?
EXPENSE TYPES OF EXPENDITURE Did you 
purchase?
EXPENSE
1=Yes
2=No
BIRR 1=Yes
2=No
BIRR
/OMO/endod/ besana leaves 
2.12 other personal care goods (including 
sendel, matentetc)
6.2 Pesticides (incl. fungicides and 
herbicides)
2.13 Others 6.3 Seeds and young plants 
(chigegn)
3 EXPENDITURE ON SOCIAL 
OBLIGATIONS 
6.4 Labor for crop production
3.1 Ceremonial expenses (gold, dowry 
for spouse)
6.5 Transport related to crop 
production and crop sale
3.2 Contributions to IDDIR 6.6 Labor for herding 
3.3 Donations to the church 6.7 Animal feed 
3.4 Taxes and levies 6.8 Veterinary services/medicines 
3.5 Compensation and penalty 6.9 other expenses 
3.6 Voluntary contributions (including 
ereban)
7 DEATH RELATED 
EXPENDITURE 
3.7 Rent (other than oxen and land 7.1 Funerals and Teskar
3.8 Rent for land
GRAND TOTAL
141
ANNEX C: SAMPLE KEBELE/TABIA 
SELECTION MINUTE OF ONE 
WOREDA

143
ANNEX D: SAMPLE GOT 
SELECTION MINUTE

145
ANNEX E: FISCAL IMPLICATION OF 
CBHI SCALE UP, 50 PERCENT 
PARTICIPATION
Projected Government Subsidy Requirements under Three Scenarios (‘000 Birr)
50% Participation Rate 
Region
161 Expansion Woredas All Non-Pastoralist Woredas
10% Pilot Case 29% 10% Pilot Case 29%
Total Budget
Amhara 36,848.62 43,222.50 50,046.40 252,437.71 308,954.69 369,462.08
Oromia 75,007.49 70,536.11 104,742.21 325,609.55 303,086.47 475,388.08
SNNP 18,171.75 17,161.17 23,165.28 126,485.43 114,441.90 185,996.15
Tigray 15,674.42 14,815.21 21,343.68 44,690.52 41,579.91 65,215.20
Addis Ababa 50,500.67 50,500.67 73,883.75
Other Non￾Pastoralist Regions
18,238.35 18,238.35 26,870.59
Total 145,702.28 145,734.99 199,297.57 817,962.23 836,801.98 1,196,815.84
Federal Budget
Amhara 22,996.40 22,996.40 22,996.40 189,711.72 189,711.72 189,711.72
Oromia 50,301.18 50,301.18 50,301.18 246,380.00 246,380.00 246,380.00
SNNP 8,545.19 8,545.19 8,545.19 95,036.83 95,036.83 95,036.83
Tigray 9,991.66 9,991.66 9,991.66 33,662.76 33,662.76 33,662.76
Addis Ababa 38,193.78 38,193.78 38,193.78
Other Non￾Pastoralist Regions
13,695.07 13,695.07 13,695.07
Total 91,834.43 91,834.43 91,834.43 616,680.15 616,680.15 616,680.15
Regional Budget
Amhara 9,696.56 14,158.27 18,935.00 43,908.19 83,470.08 125,825.25
Oromia 17,294.42 14,164.45 38,108.72 55,460.69 39,694.53 160,305.65
SNNP 0.00 0.00 0.00 0.00 0.00 0.00
Tigray 3,977.93 3,376.49 7,946.42 7,719.44 5,542.01 22,086.71
Addis Ababa 0.00 0.00 0.00 8,614.82 8,614.82 24,982.98
Other Non￾Pastoralist Regions
0.00 0.00 3,180.30 3,180.30 9,222.87
Total 30,968.91 31,699.21 64,990.14 118,883.43 140,501.74 342,423.46
Woreda Budget
Amhara 4,155.67 6,067.83 8,115.00 18,817.80 35,772.89 53,925.11
Oromia 7,411.89 6,070.48 16,332.31 23,768.87 17,011.94 68,702.42
146
Region
161 Expansion Woredas All Non-Pastoralist Woredas
10% Pilot Case 29% 10% Pilot Case 29%
SNNP 9,626.55 8,615.98 14,620.08 31,448.61 19,405.07 90,959.33
Tigray 1,704.83 1,447.07 3,405.61 3,308.33 2,375.15 9,465.73
Addis Ababa 3,692.07 3,692.07 10,706.99
Other Non￾Pastoralist Regions
1,362.99 1,362.99 3,952.66
Total 22,898.94 22,201.35 42,473.00 82,398.65 79,620.10 237,712.24
Average Woreda 
Budget (161 
Woredas)
Amhara 98.94 144.47 193.21 
Oromia 117.65 96.36 259.24 
SNNP 192.53 172.32 292.40 
Tigray 94.71 80.39 189.20 
Addis Ababa
Other Non￾Pastoralist Regions
Total 125.96 123.39 233.51
147
ANNEX F: FISCAL IMPLICATION OF 
CBHI SCALE UP, 75 PERCENT 
PARTICIPATION
Projected Government Subsidy Requirements under Three Scenarios (‘000 Birr)
75% Participation Rate 
Region
161 
Expansion 
Woredas
All Non￾Pastoralist 
Woredas
10% Pilot Case 29% 10% Pilot Case 29%
Total Budget
Amhara 50,917.62 60,478.43 70,714.29 252,437.71 308,954.69 369,462.08
Oromia 107,116.67 100,409.59 151,718.74 325,609.55 303,086.47 475,388.08
SNNP 23,446.96 21,931.10 30,937.26 126,485.43 114,441.90 185,996.15
Tigray 21,683.13 20,394.33 30,187.03 44,690.52 41,579.91 65,215.20
Addis Ababa 50,500.67 50,500.67 73,883.75
Other Non-Pastoralist 
Regions
18,238.35 18,238.35 26,870.59
Total 203,164.37 203,213.45 283,557.31 817,962.23 836,801.98 1,196,815.84
Federal Budget
Amhara 33,592.30 33,592.30 33,592.30 189,711.72 189,711.72 189,711.72
Oromia 74,585.43 74,585.43 74,585.43 246,380.00 246,380.00 246,380.00
SNNP 12,506.32 12,506.32 12,506.32 95,036.83 95,036.83 95,036.83
Tigray 14,508.46 14,508.46 14,508.46 33,662.76 33,662.76 33,662.76
Addis Ababa 38,193.78 38,193.78 38,193.78
Other Non-Pastoralist 
Regions
13,695.07 13,695.07 13,695.07
Total 135,192.50 135,192.50 135,192.50 616,680.15 616,680.15 616,680.15
Regional Budget
Amhara 12,127.73 18,820.29 25,985.39 43,908.19 83,470.08 125,825.25
Oromia 22,771.87 18,076.91 53,993.32 55,460.69 39,694.53 160,305.65
SNNP 0.00 0.00 0.00 0.00 0.00 0.00
Tigray 5,022.27 4,120.11 10,975.00 7,719.44 5,542.01 22,086.71
Addis Ababa 0.00 0.00 0.00 8,614.82 8,614.82 24,982.98
Other Non-Pastoralist 
Regions
0.00 0.00 3,180.30 3,180.30 9,222.87
Total 39,921.86 41,017.31 90,953.71 118,883.43 140,501.74 342,423.46
148
Region
161 
Expansion 
Woredas
All Non￾Pastoralist 
Woredas
10% Pilot Case 29% 10% Pilot Case 29%
Woreda Budget
Amhara 5,197.60 8,065.84 11,136.60 18,817.80 35,772.89 53,925.11
Oromia 9,759.37 7,747.25 23,139.99 23,768.87 17,011.94 68,702.42
SNNP 10,940.64 9,424.78 18,430.94 31,448.61 19,405.07 90,959.33
Tigray 2,152.40 1,765.76 4,703.57 3,308.33 2,375.15 9,465.73
Addis Ababa 3,692.07 3,692.07 10,706.99
Other Non-Pastoralist 
Regions
1,362.99 1,362.99 3,952.66
Total 28,050.01 27,003.63 57,411.10 82,398.65 79,620.10 237,712.24
Average Woreda 
Budget (161 Woredas)
Amhara 123.75 192.04 265.16 
Oromia 154.91 122.97 367.30 
SNNP 218.81 188.50 368.62 
Tigray 119.58 98.10 261.31 
Addis Ababa
Other Non-Pastoralist Regions
Total 154.26 150.40 315.60
149
ANNEX G: FISCAL IMPLICATION OF 
CBHI SCALE UP, 100 PERCENT 
PARTICIPATION
Region
161 Expansion Woredas All Non-Pastoralist Woredas
10% 29% 10% 29%
Total Budget
Amhara 64,986.62 91,382.17 252,437.71 369,462.08
Oromia 139,225.84 198,695.27 325,609.55 475,388.08
SNNP 28,722.17 38,709.24 126,485.43 185,996.15
Tigray 27,691.84 39,030.38 44,690.52 65,215.20
Addis Ababa 50,500.67 73,883.75
Other Non-Pastoralist Regions 18,238.35 26,870.59
Total 260,626.47 367,817.06 817,962.23 1,196,815.84
Federal Budget
Amhara 44,188.19 44,188.19 189,711.72 189,711.72
Oromia 98,869.68 98,869.68 246,380.00 246,380.00
SNNP 16,467.44 16,467.44 95,036.83 95,036.83
Tigray 19,025.26 19,025.26 33,662.76 33,662.76
Addis Ababa 38,193.78 38,193.78
Other Non-Pastoralist Regions 13,695.07 13,695.07
Total 178,550.57 178,550.57 616,680.15 616,680.15
Regional Budget
Amhara 14,558.90 33,035.78 43,908.19 125,825.25
Oromia 28,249.31 69,877.91 55,460.69 160,305.65
SNNP 0.00 0.00 0.00 0.00
Tigray 6,066.61 14,003.59 7,719.44 22,086.71
Addis Ababa 0.00 0.00 8,614.82 24,982.98
Other Non-Pastoralist Regions 0.00 3,180.30 9,222.87
Total 48,874.82 116,917.28 118,883.43 342,423.46
Woreda Budget
Amhara 6,239.53 14,158.19 18,817.80 53,925.11
Oromia 12,106.85 29,947.68 23,768.87 68,702.42
SNNP 12,254.73 22,241.79 31,448.61 90,959.33
Tigray 2,599.98 6,001.54 3,308.33 9,465.73
Addis Ababa 3,692.07 10,706.99
Other Non-Pastoralist Regions 1,362.99 3,952.66
Total 33,201.08 72,349.20 82,398.65 237,712.24
150
Region
161 Expansion Woredas All Non-Pastoralist Woredas
10% 29% 10% 29%
Average Woreda Budget (161 Woredas)
Amhara 148.56 337.10 
Oromia 192.17 475.36 
SNNP 245.09 444.84 
Tigray 144.44 333.42
Addis Ababa
Other Non-Pastoralist Regions
Total 182.57 397.68
151
ANNEX H: PATIENT EXIT 
INTERVIEW QUESTIONNAIRE
Cover Section A: Informed Consent
Instruction to Interviewer:
Meet with the HF Head or any other person in charge and explain the purpose of the study and 
inform them that you want to conduct patient exit interview at the outpatient level (at health center) 
and outpatient and inpatients levels(at a hospital) and request an agreement to conduct one. Have the 
HF Head/ Person in charge or any other assigned staff takes you to the pharmacy/inpatient 
department. Identify 5 CBHI members and 5 non-members (HC outpatient) and 4 CBHI members 
and 4 non-members (hospital outpatient). For inpatient interview, meet with Head Nurses of each 
Inpatient Department. Get a list of inpatients that are ready to be discharged on the day from each. 
Also, check the ones that are CBHI members and select 2 inpatients among CBHI members ready to 
be discharged on that day and conduct the interview. 
Once you meet with the interviewee use the consent for below:
የቅድመ መረጃ ፍቃድ
ጤና ይስጥልኝ! የማህበረሰብ አቀፍ የጤና መድህን መረጃ ሰብሳቢ ነኝ፡፡ በቅድሚያ ለምንጠይቅዎት ጥያቄዎች
መልስ ለመስጠት ጊዜዎን በመስጠትዎ አናመሰግናለን፡፡ መንግሥት ባለፉት ሁለት ዓመታት በወረዳው ውስጥ
የማኀበረሰብ አቀፍ የጤና መድህን መጀመሩ ይታወቃል፡፡ ይህ ፕሮግራም በኀብረተሰቡ የጤና
አጠቃቀም፤የአገልግሎት ጥራትና ወጭን በመቀነስ ረገድ ያሳደረውን ለውጥ በማጥናት ላይ እንገኛለን፡፡
የማኀበረሰብ አቀፍ የጤና መድን እንዴት መሻሻል እንዳለበት የእርስዎን አስተያየት እንጠይቃለን፡፡
እርሰዎ በአጋጣሚ ተመርጠዋል፡፡ መጠይቁ 20 ደቂቃ ይፈጃል፡፡ የሚሰጡት ምላሽ ሁሉ ሚስጢራዊ ሲሆን
ከአማካሪዎቹ እና ከቃለ መጠይቅ አድራጊው ውጪ ጥቅም ላይ አይውልም፡፡ በዚህ ጥናት የሚሳተፉት
በፈቃደኝነት ብቻ ነው፡፡ በዚህ ጥናት በፈቃደኝነት ለመሳተፍ መስማማትዎን በቃል እንዲያረጋግጡልኝ
በአክብሮት እጠይቃለሁ፡፡

153
Cover Section B: Background
1.Date: 
2.Name of Interviewer:
3.Name of Regional Coordinator:
4. Facility information
a. Name: 
b. Region: Indicate code: |__|
c. Woreda: Indicate code:|__|
1=Amhara
11=South Achefer 
12=Fogera
13=Tehuledere
14=Dembia (Control woreda)
2=Tigray
21=Kilte Awlaelo
22=Afeherom
23=Tahitay Adiabo
24=Raya (control woreda)
3=Oromia
31=Gimbichu
32=Kuyu
33=Deder 
34=Limukossa
35=Merti (control woreda)
4=SNNP
41=Yirgalem city administration
42=Damboya
43=Damot Woyde
44=Wonago (control woreda)
5. Level of Facility Indicate code: |__|
1=Regional referral Hospital 
 2=General Zonal Hospital 
3=Primary District Hospital 
4=Health Center 
6. Patient No.
Day Month Year

155
|__||__| years
1=Male 2=Female Indicate code: |__|
Indicate code: |__|
1=urban 2=rural Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
Section1: Patient’s Background
1‐ Age (in completed years) 
2‐ Sex (fill as observed) 
3‐ highest educational attainment 
1=Illiterate 
2=reading and writing 
3=primary education (Grade 1-6)
4=secondary education (Grade 7-12) 
5=vocational training 
6=tertiary education 
7=under age 
4‐ Residence
5‐ marital status 
1=never married 
2=married 
3=living together 
4=divorced/separated 
5=widowed 
6‐ Occupation/employment 
1=employed/self-employed (e.g. farming) 
2=non‐employed (above 18 years) 
3=student 
4=other, specify_____________
7 – Are you a member of CBHI? 
1=Yes 
2= No 
8 – Are you a fee waiver beneficiary? (Only for control 
Woreda)
1=Yes 
2= No 

157
Section2: Service utilization and quality
9‐ For how long have you used this health facility? (state number of 
months)
|__||__| |__|months
10- What were the reasons for you visit today? (multiple 
response is possible)
1=Diarrhea Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
Indicate code: |__|
2=Fever 
3=Respiratory problem
4=Cough
5=Nausea/vomiting 
6=Head ache
7=Stomachache
8=Toothache
9=Joint/muscle pain 
10=Delivery
11=Injury 
 12=eye problem
13=other(specify)_________________
11‐ Which services did you obtain during this visit/stay? (multiple 
response is possible)
1=Consultation/card Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
2= Diagnosis (lab and other)
3=Drugs and medical supplies 
4= Inpatient services (bed/food)
5=Delivery 
6=Surgical procedure
7=others, specify
12- If the answer to question 11 is 3, were you able to get the 
prescribed drugs/supplies? 
1= fully (all prescribed)
2= partially (only some of them)
3= none (go to question 14)
Indicate code: |__|
158
13. If the answer to question 12 is 1 or 2, where did you get it from? 
1= within the visited facility Indicate code: |__|
2= being sent to other CBHI contracted facilities (including 
drug retailers)
3= being sent to other facility/drug stores 
14. If the answer to question 12 is 3, what is the reason? 
1= unavailability within the facility visited Indicate code: |__|
2= unable to buy
15‐ How satisfied are you with the service you received during this 
visit/stay? ( Indicate the code for questions 15.1-15.9) 1=Very satisfied
2=satisfied
3=Neither satisfied 
nor dissatisfied
4=dissatisfied
5=Very dissatisfied
15.1= overall quality of service? Indicate code: |__|
15.2=Availability of drugs/medical supplies? Indicate code: |__|
15.3=Availability of diagnostic facilities? Indicate code: |__|
15.4=Cleanliness of the facility? Indicate code: |__|
15.5=Waiting time (from the time of arrival in the health 
facility up to seeing a health professional?
Indicate code: |__|
 15.6=Waiting time between services (e.g. between 
consultation and diagnosis)?
Indicate code: |__|
15.7=Friendliness of staff? Indicate code: |__|
15.8= Attentiveness and adequate follow up by the nursing 
staff? (inpatient only in hospitals)
Indicate code: |__|
15.9= Quality of food and other inpatient amenities? 
(inpatient only in hospitals)
Indicate code: |__|
16‐ Have you observed any improvement in service quality since this 
health facility has been contracted by the CBHI scheme (mention 
the year) (to be asked in CBHI pilot woreda only) – (put code 
as appropriate for questions 16.1 to 16.9)
1=Yes
2=No
3=Don’t observe
16.1= overall quality of service? Indicate code: |__|
 16.2= Availability of drugs/medical supplies? Indicate code: |__|
 16.3= Availability of diagnostic facilities? Indicate code: |__|
 16.4= Cleanliness of the facility? Indicate code: |__|
16.5= Waiting time (from the time of arrival in the health 
facility up to seeing a health professional?
Indicate code: |__|
 16.6= Waiting time between services (e.g. between 
consultation and diagnosis)
Indicate code: |__|
159
16.7=Friendliness of staff? Indicate code: |__|
16.8=Attentiveness and adequate follow up by the nursing 
staff? (inpatient only)
Indicate code: |__|
16.9=Quality of food and other inpatient amenities? 
(inpatient only)
Indicate code: |__|
17‐For how long did you wait before you had consultation with staff 
today or the last time you had consultation? 
1= Less than 30 minutes Indicate code: |__|
2=30 to 60 minutes 
3= 1 to 3 hours 
4=3 to 6 hours
5= 6 hours and more
18 - For how long did you wait between services (e.g. between 
consultation and diagnosis) today or the last time you had to utilize 
various services?
1= Less than 30 minutes Indicate code: |__|
2=30 to 60 minutes 
3= 1 to 3 hours 
4=3 to 6 hours
5= 6 hours and more
19 - Is this health facility your preferred service point for your 
future health care needs?
1=Yes because I am satisfied with the service quality
2= Yes because I do not have another option 
3= No
Indicate code: |__|
20 – Do health professionals treat CBHI members and non-CBHI 
members equally, with impartiality?
1= Yes
2= No
3 = No comment Indicate code:|__|
21 ‐ How much did you pay for the service you obtained (in ETB)?
(for non-members and members who utilized a service which is not 
covered by CBHI e.g. dental service, eye glass etc) (ask the patient 
the type of service they acquired check against the list of CBHI 
service package)
1=Consultation ____________ETB
2=Diagnosis ____________ETB
3=Drugs and medical supplies ____________ETB
4= Inpatient services ____________ETB
5= transport (for inpatient only) ____________ETB
6=others, specify ____________ETB
22 – How affordable is the fee you paid for the service?
160
1= affordable
Indicate code:|__|
2=somewhat affordable
3=not affordable
161
Section3: CBHI Related
23‐ Are you enrolled in a CBHI scheme in your woreda?
1=Yes 
2=No (go to 28)
Indicate code:|__|
24 – When did you become a CBHI member? Write the year |__|__|__|__|
25 – Why did you decide to enroll in CBHI? (multiple response 
is possible)
1=Illness and/or injury occurs frequently in our household
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
2= To finance health care expenses
3= household is exempt from registration fee and premium 
payment 
4=Premium is low compared to the user fee price to obtain 
medical treatment 
5= Pressure from other family members/community
6= Pressure from the CBHI office
7= Others, please specify
26 – Do you plan to renew your CBHI membership?
1= Yes 
2=No (go to 29)
Indicate code:|__|
27 - How do you feel about the adequacy of the benefit package?
1=very adequate
2= somewhat adequate
3= inadequate
4=I do not know
Indicate code:|__|
162
28 – If you are not a CBHI member, why did you decide not to 
enroll in CBHI? (multiple response is possible)
1=Illness and injury does not occur frequently in our 
household
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
2=The registration fee and premiums are not affordable 
3=Want to wait in order to confirm the benefits of the 
scheme from others
4= We do not know enough about the CBHI scheme
5= There is limited availability of health services
6= The quality of health care services is low
7= The benefit package does not meet our needs
8= CBHI management staff is not trustworthy
9= Waiting time to access services is longer for CBHI 
members
10=I am fee waiver beneficiary 
11= Other reasons, please specify
12= I don’t know
29 – If you have decided not to renew your CBHI membership, 
state the reason. (multiple response is possible)
1=Illness and injury does not occur frequently in our 
household
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
Indicate code:|__|
2=The registration fee and premiums are not affordable
3= There is limited availability of health services
4=The quality of health services is low
5= CBHI management staff is not trustworthy
6= The quality of service for CBHI members is worse than for 
non-CBHI members
7= Waiting time to access services is longer for CBHI 
members
8= Other, specify
163
ANNEX I: KEY INFORMANT 
INTERVIEW
3.1 Ethiopian Health Insurance Agency (EHIA), Resource Mobilization and Utilization 
Directorate and State Minister (Jointly) 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with a broad range of stakeholders. Find the manager or the most senior
person responsible at this level. After introducing yourself and greetings, eexplain briefly the 
purpose of the interview, the interview process and thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also provide 
recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Phone no:
Date:
Generally, can you talk about your organization and your responsibilities in relation to the CBHI piloting 
and its scaling up? 
1. What is your expectation from this CBHI evaluation? 
2. What are the primary policy purposes of CBHI in Ethiopia? 
3. What are the roles and responsibilities of various levels of government authorities with regards to 
CBHI policy making, design and implementation? 
a. FMOH 
b. EHIA 
c. Regional government/BOFED/RHB 
d. Zonal authorities 
e. Woreda government/WOFED/WorHO 
4. Can you please describe for us the role FMOH and EHIA played in supporting the CBHI schemes (viz. 
subsidizing the schemes, improving the quality of services by providers, training and skill upgrading 
plans for CBHI staff etc?) 
164
5. Are the CBHI schemes working? What do you think are the main achievements of CBHI in the pilot 
woredas, in terms of 
a. Access to care? 
b. Quality of care? 
c. Utilization of services? 
d. Equity? 
e. Mobilization of resources to health facilities? 
6. The achievement of the CBHI in terms of enrolment and coverage of the poor varies from region to 
region and woreda to woreda. What strategies have worked in successful regions and woredas? Any 
best practices you have identified? 
7. The pilot scheme design was implemented and tested for some time now. What are the major 
successes and challenges in the scheme parameters: 
a. Benefit package? 
b. Level of premiums?
c. Membership scenario (HH vs. individual basis) 
d. Reimbursement amounts? 
e. Referral mechanism? 
f. Payment to the health facilities on timely manner? 
g. CBHI management schemes? 
h. Institutional arrangement (staff size, etc.,) 
i. Staff skills and capacity? 
Do you think there is need to revisit some of these design parameters? 
8. Design Parameters (benefit package, level of premium, membership scenario (household vs. 
Individual), contribution frequency etc.) vary slight across regions. 
a. What is the plan of the FMOH/EHIA for future scale up? Do you plan to proceed with 
different design parameters? 
b. Have benefit packages been assessed in line with government policy priorities and needs of 
community members/eligible households, etc.? If not, is there plan to do so before scale up? 
9. The performance of schemes in terms of enrolment varies from region to region and from woreda to 
woreda? What do you think are the major factors for this regional and woreda variation in 
enrolments? Are you aware of any complaints of CBHI members in terms of service coverage, 
premiums and quality of service? What should be done to encourage non- members to join and 
members to be satisfied? 
10. Are there instances of fraud and abuse in CBHI financial resources management in the pilot woredas? 
If yes, what actions have been taken? 
11. Who are the major stakeholders in the initiation and scale-up of CBHI schemes? Can you please 
describe how you these stakeholders have reached the decision to scale up the CBHI schemes to 160 
Woredas? What were the criteria used to select the woredas? Have the fiscal implications of scale-up 
been calculated? 
12. What is the organizational and operational readiness for scale up? IS EHIA legally mandated to lead 
CBHI? What is EHIA’s organizational relation with regional governments? Does EHIA have the 
required organizational arrangement and capacity appropriate to support, lead and regulate CBHI 
schemes? 
13. FINANCING SCALEUP – 
a) What is status of readiness and willingness of Federal Government to finance general subsidy to scale 
up CBHI? 
b) Given your close working relationships with the regions, are the regional and woreda governments 
ready to take the fiscal implications of the scaling up? 
c) What is the plan regarding financing of the scale-up? What are sources of government subsidy 
(targeted and general subsidy), i.e. donor funding vs own/government revenue? What is the possibility 
of having regular budgeting and budget codes for general health insurance and CBHI? Is there a 
political and legal base to do so? How to make them legally obligated in the future? 
165
d) What is the plan to ensure financial sustainability of CBHI schemes? What about reinsurance to CBHI 
schemes? Risk pooling among different schemes in the region and then at national level among all 
schemes? Which authority will be mandated to protect CBHI schemes from problem of insolvency 
and complete collapse? 
14. What do you think are the major lessons learned to inform the scale up process? 
166
3.2. Regional and Woreda Health Office Heads 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with a broad range of stakeholders. Find the manager or the most senior
person responsible at this level. After introducing yourself and greetings, eexplain briefly the 
purpose of the interview, the interview process and thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also provide 
recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Region:
Zone:
Woreda:
Date:
Tel No.:
Roles and Responsibilities 
1. What are the roles and responsibilities of various levels of government authorities with regards to 
CBHI policy making, design and management? 
a. FMOH 
b. EHIA
c. Regional government/BOFED/RHB 
d. Regional steering committee 
e. Zonal authorities 
f. Woreda government/WOFED/WorHO 
Have all been fully engaged in the process? If not, what can be done to better engage these authorities 
during the future scale up? 
2. How do you assess the policy guidance, financial and technical support received from FMOH and 
EHIA? How about commitment and support from regional government in terms of budget allocation 
for general and targeted subsidy? 
3. Who else is providing the required technical support? 
Experience and Impact of the Schemes 
4. Given the experiences of the pilot woredas, what were the strength and weaknesses regarding the 
167
 Legal frameworks, 
 Directives, by-laws, manuals and guidelines, 
 Sensitizing the community
 The structures and recruitment of staff 
5. The pilot scheme design was implemented and tested for some time now. What are the major 
successes and challenges in the scheme parameters: 
a. Benefit package? 
b. Level of premiums?
c. Membership scenario (HH vs. individual basis) Reimbursement amounts? 
d. Referral mechanism? 
e. Payment to the health facilities on timely manner? 
f. Enabling the schemes to cover all health related expenses through its income? 
g. Affordability to member? 
h. Fairness compared to benefit package? 
i. CBHI management schemes? 
j. Institutional arrangement (staff size, dual assignment, etc.,)? 
k. Staff skills and capacity? 
What are the major complaints of members in this regard, if any? 
6. What are the successes and challenges in mobilizing the community to enrol/renew membership in 
CBHI? What proportion of the woreda population is currently enrolled? What strategy has worked 
and what hasn’t?
7. Has the targeted (and general) subsidy allowed pilot woredas to adequately include indigents in the 
CBHI? Are there indigents who are left out? And how significant are they in number? To what extent 
do you see CBHI as one of the mechanisms to increase access to health care equitably? Discuss 
challenges in this regard, if any. 
8. Could you tell us about the successes, challenges and the areas that need improvement regarding the 
defined benefit package and the views of CBHI members on its coverage and adequacy as well as 
availability of these services in the health facilities? 
9. How successful has the CBHI scheme been in negotiating agreeable terms and contract with service 
providers – in terms of service quality, fee, and reduction in unnecessary services/prescription (moral 
hazard) etc.? What are the successes and challenges in contract administration? 
10. Does the implementation of the CBHI scheme have any impact (positive or negative) on the health 
facilities in terms of increasing resources, improving quality of care, motivation of the staff? 
Health Service Utilization and Quality 
11. Have you seen any difference between CBHI woredas and non CBHI Woredas in terms of utilization 
of services? If there is increase in patient flow, how successful have facilities been in coping with this 
demand surge? How about coping with further demand increase with the scale up? Please provide 
evidence. 
12. How is the referral of the CBHI members being carried out? Any specific challenges given that they 
are likely to claim preferential treatment? Any specific measures introduced? Lessons learned for 
scaling up.
13. To what extent are health facilities providing quality health care services for CBHI schemes as well as 
other clients? How does your organization support facilities to make them respond to increased 
demand for quality care? 
168
Management 
14. The management of the schemes is heavily dependent on the scheme managers. How do you view 
CBHI management and staffing structure? Any successes and challenges regarding retention and 
motivation of CBHI management team? What should scaling up woreda learn in this regard? 
15. How and to what extent is the community involved in the management of CBHI scheme viz. CBHI 
design and administration? Discuss 
16. How and to what extent is the community involved in the management of health service delivery in 
the district? Discuss 
17. What is the role of the woreda administration in enrolment drive, allocation of resources and staff 
recruitment? What worked and what didn’t? What are the innovative strategies in successful woredas 
that should be scaled up? 
Financial Status 
18. Overall, what is the financial status of the CBHIs? 
19. If the surplus of the CBHIs is increasing over time, why is this so? Are the beneficiaries not utilizing 
services? Is the user fees paid too low? Or are the premiums higher? Can you explain this for us? 
20. Have the regional health bureau/Woreda health office invested any additional resources on health 
facilities (human resources, water and electricity, other equipment) to ensure that CBHI members get 
quality services? If yes, please describe the investments made
21. Can the health bureau/other woredas be able to do such investments in the scaling up woredas? 
22. What was the role of the FMOH in improving quality of care and CBHI scheme in the pilot woredas? 
Please describe the support you received from FMOH for the CBHI schemes?
Scale up 
23. What do you think are the challenges of the scaling up to other woredas for regional and woreda 
governments? 
24. In scaling up the CBHI to other woredas, what should be the criteria for selecting the woredas for 
scaling up in terms of 
a. Availability of health facilities 
b. Budget allocation 
c. Economic status of households 
d. Others 
169
Data to be collected from the RHB and Woreda offices from secondary Sources 
1. Outpatient and inpatient visits 
No Item CBHI Woredas Non-CBHI Woredas
1 Total population 
2 Total outpatient visits in 2005 EFY
3 Total inpatient visits in 2005 EFY
2. Financial support from the FMOH (in ETB)
No For the region 
(amount in ETB)
For the pilot woreda
(Amount in ETB)
1 25% subsidy for the premiums
2 Investment for health facilities 
3 Subsidy for the tertiary care
4 Any other, specify
3. Implication of the CBHI scaling up
No
1 # of woredas to be scale up
2 Total # of estimated households in the scaling up woredas
3 Estimated number of indigents in the scaling up woredas
4 Estimated regional subsidy for indigents in scaling up woredas
5 Estimated woreda subsidy for indigents in the scaling up woredas
6 25% subsidy for the premiums from FMOH for the scaling up woredas
7 Salaries for scheme managers in the scaling up woredas
8 Investment for health facilities in the scaling up woredas
9 Subsidy for the tertiary care, if any
10 Any other support, specify
170
3.3. Regional/Woreda Finance Bureau/Office 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with a broad range of stakeholders. Find the manager or the most senior
person responsible at this level. After introducing yourself and greetings, eexplain briefly the 
purpose of the interview, the interview process and thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also provide 
recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Phone no:
Region:
Zone:
Woreda:
Date:
General 
1. Are you aware of the CBHI schemes that are piloted in your region/woreda? 
2. Generally, can you talk about your BOFED/ WOFED’s role in the implementation of the CBHI in the
pilot woredas/ woreda and your relationship, if at all, with the scheme management? 
3. What do you think are the major achievements (impacts) of CBHI scheme in the Region/ woreda? 
4. Not all residents of the pilot woreda are enrolled into the CBHI scheme. What do you think are the 
major reasons for not enrolling in CBHI or renewing their membership? 
5.
Experience and Impact of the Schemes 
6. What is the perceived or actual contribution of the CBHI on the following: 
a. Access to care? 
b. Quality of care? 
c. Equity? 
d. Affordability? 
7. Given the experiences of the pilot woredas, what were the strength and weaknesses regarding the 
 Legal frameworks, 
 Directives, by-laws, manuals and guidelines, 
 Sensitizing the community
 The structures and recruitment of staff 
8. Looking at the operationalization of CBHI pilots, how successful have the schemes been in 
 Funding the indigent 
171
 Other design parameters (setting of premiums, benefit packages, premium payment 
frequency, enabling the schemes to cover all health related expenses through its income etc.) 
 What were the strengths and weaknesses? 
9. What are the major challenges and successes in the implementation of CBHI? 
Scale up 
10. FMOH and regional government have agreed to scale up the CBHI in new woredas. We would like to 
understand the budget implication of this scaling up. Please provide the following data. 
11. What is your opinion regarding the possibility of assigning budget code and appropriating budget for 
CBHI schemes (targeted subsidy) especially in line with the envisaged expansion to 160 woredas and 
eventually to scale-up to all woredas in the county? 
12. In scaling up the CBHI to other woredas, what should be the criteria for selecting the woredas for 
scaling up in terms of 
a. Availability of health facilities 
b. Budget allocation 
c. Economic status of households 
d. Others 
13. Learning from the pilots so far, what would you advise for the heads of WOFEDs that are going to 
scale up CBHI in their woreda? 
Data to be collected from Secondary Sources in BOFEDs/WOFEDs 
Budget implications of scaling up 
Number Major budget items for consideration Expenditure in 2005 
EFYIn ETB
1 Total Regional/ woreda population (# of households)
2 Number of indigent households supported by the region/woreda
3 Subsidy to the indigent members from the woreda allocation
4 Any other regional subsidy for CBHI schemes
5 Salary for CBHI scheme managers
6 Any investment in health centres to improve quality of care, 
specifically targeting CBHI woredas
Total 
172
3.4. CBHI Management Team 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with a broad range of stakeholders. Find the manager or the most senior
person responsible at this level. After introducing yourself and greetings, explain briefly the 
purpose of the interview, the interview process and thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also provide 
recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Region:
Zone:
Woreda:
Date:
Telephone No.
Tell us a little about your functions in the management of the CBHI scheme? 
Enrollment 
1. What is the status and progress of the woreda in enrolling its residents into CBHI? What are the 
successes and challenges? 
2. What do you think are the major reasons for some people not to enrol into the scheme or 
failure to renew membership? 
3. Does your scheme have partnership with microfinance institutions in your area? If so, how much 
of the CBHI contribution was mobilized through microfinance institutions? 
Service Utilization and Reimbursement 
4. How far CBHI members are using the health service in the contracted facilities? Do you think 
most of the members are using the services in the recommended (referral system) manner? 
5. How often do you reimburse health facilities for services used by CBHI members? What are the 
mechanisms by which you check whether the invoices sent from the health facilities are right? Do 
you have adequate capacity to check on health facilities? Are there instances by which health 
facilities tried to overstate the reimbursement request amount? 
6. Do you face a problem of unnecessary care seeking behaviour by CBHI members and 
unnecessary or over prescription of services including drugs, diagnostics etc. by health care 
providers (client and provider moral hazard)? Discuss how such circumstances, if they exist, affect 
the financial viability of the scheme? 
173
Service Quality and Patient Satisfaction
7. Are there mechanisms whereby you are able to check on the patient perceived quality of service 
in contracted health facilities i.e. waiting time, availability of staff, availability of services, drugs and 
supplies etc.? Discuss. If so, are these regular checks or in response to complaints from your 
members? Do your findings show any change in service quality? Cite examples 
8. Is there any other organization that conducts quality check up on health facilities? Do you get 
feedback from such an organization? 
9. How do you assess members’ satisfaction about the services provided by the schemes and also 
health service providers? Do you have a standard client compliant management mechanism? 
Describe how, if at all, action is taken based on feedbacks? What are the major complaints 
forwarded by your members? 
Financial Status of the Scheme 
10. Who is assisting you in collecting the premiums? Have the kebele/administration and/or saving 
and credit association collect and bring the contribution on time? What are the main successes 
and challenges in the collection of premiums? 
11. How healthy is the situation of the CBHI scheme in terms of its finances? Please describe the 
financial status of the scheme to us? Please give us evidence in the table below 
Organizational Status 
12. How successful has the CBHI scheme been in recruiting and retaining core staff? Discuss 
successes and challenges in this regard. 
13. How frequently does the Board of Management meets? What are the average attendance ratios 
of Board members? 
14. What is the support you are receiving from RHB, WorHO, woreda administration and Abt 
regional office (training, supervision, administrative support etc.)? Are you satisfied? What needs 
to improve? 
15. What is your overall assessment of the schemes? What do you recommend for the future in 
terms of organizational structure, staffing, budgeting, key design issues etc? 
Secondary Data to be collected from CBHI Management team: 
1. Status of enrolment 
No. Item Number of population
2004 EFY 2005 EFY
1 Total woreda population 
2 New Enrollment 
Premium payment
Government subsidy
3 Renewal of membership
 Premium payment
 Government subsidy
4 Total members in the woreda
174
2. Utilization of services by CBHI members in the woreda 
2004 EFY 2005 EFY
1 Total CBHI members
2 # of members that went for OPD visits in the health centers
3 # of members in the health centers using in-patients services (if 
any)
4 # of members that used hospital OPD services if any
5 # of members that used inpatient services in hospitals 
6 # of members that used private clinics and pharmacies
3. Financial status of the Woreda CBHI Scheme 
2004 EFY 2005 EFY
1 Total income of the CBHI scheme (ETB)
Total CBHI members
2.1 Reimbursement for the cost of OPD visits in the health centers (ETB)
2.2 Reimbursement for the cost of inpatient services in health centers (if 
any) (ETB)
2.3 Reimbursement for the cost of hospital OPD services if any (ETB)
2.4 Reimbursement of the cost of hospital inpatients services in hospitals 
(ETB
2.5 Reimbursement for cost of services in private clinics and pharmacies 
(ETB)
2.6 Administrative cost (ETB)
2 Total cost of the scheme (ETB)
3 Surplus/deficit of the scheme (ETB)
175
3.5. Health Facilities-Hospitals and Health Centers 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with a broad range of stakeholders. Find the manager or the most senior
person responsible at this level. After introducing yourself and greetings, eexplain briefly the 
purpose of the interview, the interview process and thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also 
provide recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Region:
Zone:
Woreda:
City:
Name of health facility:
Ownership:
Date:
Telephone No.
Tell us a little about your health center or hospital (beds, services, area of service, population covered, number 
of staff, size and makeup of the facility governance body,)?
1. Are you a provider of health services to CBHI scheme members in the woreda? If yes, in your 
opinion, what is the impact of the CBHI on utilization of services? How about impact on quality of 
services? Give evidence 
2. How do you perceive CBHI in terms of creating additional demand for health care – do you perceive 
it as creating additional workload and pressure to the health facility or as creating an opportunity to 
strengthen the capacity of your facility? 
3. What incentives and disincentives CBHI schemes created on the facility and your staff? 
4. After the establishment of CBHIs, are there differences in members and non-members in claiming 
their rights-i.e. requesting for better service? Please elaborate. 
5. The CBHIs have their benefit packages. Are you able to provide all the services listed in the benefit 
package (for your level) to members of the CBHIs? If not what are the major gaps? 
6. What are the major complaints from CBHI members on the quality of your services? 
7. What is payment modality you are using to get reimbursements from CBHI members (fee for service 
or capitation)? If there is a difference between the payment modes of CBHI schemes and Non￾members? Which mode of payment is advantageous for the health facility and why? 
176
8. How frequently do you request and collect reimbursement for the expenses you incurred for CBHI 
members? Do you face any challenge in the process? 
9. What is the impact of CBHIs in increasing your retained fees? 
10. What kind of support did you receive from RHB and FMOH to improve quality of care because you 
are a CBHI provider? 
11. What do you think RHB and FMOH should do for health facilities in the scaling up woredas to help 
them meet quality of care requirement? How about to cope with the surge in health service demand? 
Secondary Data to be collected at facility level 
CBHI members Non CBHI 
members
Total for the 
facility
Total catchment population
Outpatient visits for 2003 EFY
Outpatient visits for 2005 EFY
Total referral made to a hospital or received 
from a health center in 2005 EFY
Number of inpatients in 2003
Number of In-patients in 2005
Total retained fee in ETB
177
3.6. KII for HSFR/HFG project (National and Regional Levels) 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members 
during interviews with HSFR/HFG project staff at national and regional levels. After introducing
yourself and greetings, explain briefly the purpose of the interview, the interview process and 
thank them for seeing us.
Provide the following information and obtain verbal informed consent to take part in the 
interview.
The purpose of this assessment is to gather data to evaluate the impact of pilot CBHI schemes 
from different perspectives including improving financial access, quality of health services, 
increasing resource mobilization and community participation. This assessment will also provide 
recommendations for the scale-up of the pilot schemes at national level. 
If there are questions for which someone else is the most appropriate person to provide that
Information, I would appreciate if you introduce me to that person.
Any information you will provide as part of this interview will be held strictly confidential. Any 
reference to the information you provide in our analysis will be made without mentioning or 
implicating your name in any way. 
Interviewee (Name and Title):
Federal/ Region:
Date:
Telephone No.
1. Can you describe for us the processes under which the CBHI scheme design is developed, directives and 
byelaws developed? What was the role of: 
a. Federal FMOH and EHIA? 
b. Regional councils? 
c. RHB/BOFED? 
d. Woreda and members? 
e. Any role for health providers? 
2. What are the major issues around kick starting the CBHI schemes in the pilot woredas and what are the 
lessons that can be drawn on how to start in the scaling up process? 
3. Given the design parameters are now being tested, what do you think are the major successes and 
challenges in implementation? Which design parameters needs a re-look during the scaling up? 
4. What is the successes and challenges in the pilot woredas, in terms of: 
a. Mobilizing the community to prepay and enrol in CBHI? 
b. Regional/woreda efforts to support indigents? 
c. Renewing membership? 
d. Utilization of services (outpatients, inpatient services)? 
e. Quality of services provided by health facilities 
f. Perception of members on quality of services? 
g. Role of members in the management and oversight of schemes at woreda levels? 
h. Woreda administration and WOFED support in the management? 
i. Reimbursement of health facilities? 
j. Moral hazards from providers? 
k. Moral hazards from members? 
178
5. What do you think are the replicability of the current design to the identified scale up woredas? What are 
the best practices? Any issues around 
a. Readiness of health facilities? 
b. Fiscal implication to the federal, regional and woreda governments? 
c. Technical assistance to support the scale up? 
d. Sustainability of the schemes? 
e. Any need on enhancing pooling beyond the woreda levels? 
179
ANNEX J: FOCUS GROUP
DISCUSSION GUIDE
4.1 CBHI Members 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members to conduct 
focused group discussions with members of CBHI. Facilitators of the FGD should start by explaining briefly the 
purpose of the discussion, the evaluation process and thank them for coming to the meeting.
Woreda: __________ 
FGD No: __________
Composition of f FGD Participants:
1. Number of males:---------
2. Number of Females:__________
3. Number of youth_________________
4. Number of adults_____________________________
Date: _____________ 
1. How much do you pay for membership (registration +membership)? Is the amount affordable to you and 
your community? What about the payment scheduling? 
2. What types of households joined CBHI schemes and why? 
3. Indigent targeting - Some members of the CBHI get their contributions by the government as they are 
recognized as indigents. 
a. Were you involved in the identification of indigents? 
b. Do you think the selection process is transparent and fair? 
c. Are there some people who are not included and others inappropriately included in the targeted 
groups? 
d. What do you suggest to improve the process? 
4. Since you become member of the schemes, what are the major benefits you got? 
5. What is the community’s perception regarding the benefit package of CBHI?
6. What is the community’s perception about the quality of care received from CBHI contracted facility in 
terms of: 
b. Waiting time? 
c. Availability of staff? 
d. Attitude and motivation of staff? 
e. Availability of diagnostic facilities? 
f. Availability of essential medicines? 
g. Cleanliness of the facilities? 
h. The referral system? 
7. Are health service providers serving CBHI members better than non-CBHI members or vice versa, and 
why? 
8. What do you think should be done to enrol the non-members of your community into the CBHI in your 
community? What should be adjusted to keep current members including you as a member of the 
scheme? 
180
9. Do you participate in the management of the scheme? If so, what is your role in this regard? 
4.2: NON-CBHI Members 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members to conduct 
focused group discussions with non-members of CBHI. Facilitators of the FGD should start by explaining briefly 
the purpose of the discussion, the evaluation process and thank them for coming to the meeting.
Woreda: __________ 
FGD No: __________
Name of FGD Participants:
1. Number of males:---------
2. Number of Females:__________
3. Number of youth_________________
4. Number of adults_____________________________
Date: _____________ 
1. Are you aware of the existence of the CBHI Scheme in your woreda? 
2. Were you requested to become a member of the CBHI scheme? How were you communicated? 
3. Why did you decided not to join the CBHI scheme? 
4. Do you know how much is paid by the CBHI members? What do you think about the affordability of the 
pre-payment scheme (registration fee and membership Fee)? 
5. Do you think that members of the CBHI schemes in the community are benefiting? Please describe what 
you have seen and heard as their benefit? Do the benefits you heard make paying worth in your opinion?
6. Are health service providers serving CBHI members better than non-CBHI members or vice versa, and 
why? 
7. What should be changed in the current CBHI scheme set up (payment levels, payment scheduling, benefit 
package, service availability etc.) to make you a member of a CBHI scheme? 
181
4.3 Health Workers 
INSTRUCTION TO THE INTERVIEWER
This document is meant to be used as a general guide for the CBHI Evaluation team members to conduct 
focused group discussions with the staff of the CBHI contracted facility. Facilitators of the FGD should start by 
explaining briefly the purpose of the discussion, the evaluation process and thank them for coming to the 
meeting.
Woreda: __________ 
Facility Name___________ 
FGD No: __________
Name of FGD Participants:
1. Number of males:---------
2. Number of Females:__________
3. Number of youth_________________
4. Number of adults_____________________________
Date: _____________ 
1. What do you think is the impact of CBHI in changing the health seeking behaviour of its members? 
Compared to the non-CBHI members and also comparing the situation before the start of the CBHI 
pilot, is there a change in seeking care by members? Please discuss the changes (OPD visits, for 
exempted services like immunization and deliveries etc.)? 
2. Was your facility ready to provide quality care to members of CBHI? What were its strengths and 
weaknesses in service delivery when CBHI started? What kind of support have you received from the 
federal and regional level to improve health facility? Was it adequate? 
3. What is the change that CBHI has brought to your facility in terms of 
a. Increased financial mobilization from its members? 
b. Increased community participation in facility management? 
4. What is the impact of CBHI on the health staff in term of: 
a. Increased workload? 
b. Disagreement and conflict with members of CBHI when they claim their rights? 
5. Are there any efforts being made to motivate the health staff with any sort of incentives? If yes, please 
discuss 
6. What do you think should be done to motivate and encourage staff as well ensuring the readiness of 
facilities when scaling up CBHI scheme to other woredas? 

183
ANNEX K: PEOPLE INTERVIEWED
Annex K.1. Amhara 
Area KII Name/position Date
Tehuledere Woreda administration Chief Administrator 1 Jan 2014
WHO Head 1 Jan 2014
WoFED Deputy head 1 Jan 2014
CBHI office Ibrahim 2 Jan 2014
Sulula HC Head 1 Jan 2014
Haik HC Head 2 Jan 2014
Dessie hospital Executive manager 2 Jan 2014
Achefer WHO Head 15 Jan 2014
CBHI office Getu 10 Jan 2014
Lalibela HC Head 10 Jan 2014
Durbete HC Wondim Anley 10 Jan 2014
Fogera WHO Head 6 Jan 2014
CBHI office Goshu 9 Jan 2014
Alember HC Dejach 6 Jan 2014
Woreta HC Belayneh Birhanu, Head 9 Jan 2014
Dembia Woreda administration Amsalu, Chief administrator 14 Jan 2014
WHO Deputy 14 Jan 2014
WoFED Deputy 14 Jan 2014
Bahir Dar RHO Bayeh, HCF Officer 13 Jan 2014
BoFED Girma, Deputy head 13 Jan 2014
Regional CBHI office Genet, Abay 13 Jan 2014
Felege Hiwot hospital Chief Executive manager 13 Jan 2014
184
Annex K.2: Oromia 
Woreda Institution Name Position
Kuyu Woreda Administration Mamo Shiferaw D/Administrator
Woreda Administration Dereje Mengistu CBHI Coordinator
Wored Health Office Desta Legese D/Head
Woreda Finance Office Berhane Dejene Office Head
Kuyu Hospital Benyam G/Amlak Representative
Beriti Health Center Leulseged Tesfaye Office Head
Kareta Health Center Getu Worku Office Head
Gimbichu Woreda Administration Tadese Kasaye Administrator
Woreda Health Office Teketel Kebede CBHI Coordinator
Woreda Health Office Abebe Mamo Office Head
Woreda Finance Office Berehanu Kassaye Office Head
Chefedonsa Health Center Melkam Degaga Office Head
Aredagoro Health Center Abiye Mekonnen Office Head
Limmu Kossa Woreda Administration Habtamu Fufa Office Head
Woreda Administration Goji Gari CBHI Coordinator
Woreda Health Office Debesa Gobena Office Head
Woreda Finance Office Gezahegne Merja D/Head
Limugenet Health Hospital Nega Abajemal Office Head
Limugenet Health Center Abagiya Office Head
Embuye Health Center Seifedin Husien Representative
Deder Woreda Administration Dagnu Hailu Office Head
Woreda Health Office Abdo Alia CBHI Coordinator
Woreda Health Office Abyot Assefa Office Head 
Representative
Woreda Finance Office Meaza Bekele Core Process Owner
Deder Hospital Mohamed Abdurahman Manager 
Qufanzek Health Center Dagne Bodane Office Head
Kobo Health Center Mohamed Ibrahim Office Head
Merti Woreda Administration Ahmed Hamid Administrator
Woreda Health Office Desta Kuma Office Head
Woreda Finance Office Aman Koteb Office Head
Abomsa hospital Fanus Dechasa Manager
Abomsa Health Center Ejeta Waqtola Office Head
Gologota Health Center Tahir Gebi Office Head
185
Annex K.3: Tigray 
No. Woreda Institution Name of Interviewee Position
1 Raya and Azebo Bala Health Center Berhe Gashae Head Nurse
2 Raya and Azebo Lemlem Karl Henz Hospital Berhanu H/ Sellasie CEO
3 Raya and Azebo Mehoni Health Center Gebregziabher Tsadik Medical Director
4 Raya and Azebo WOFED Frewewini Tekeste Head
5 Raya and Azebo Woreda Health Office Woldemichael G/Medhin Head
6 Raya and Azebo Woreda Administration Haftu G/Kiros Administrator
7 Kilte Awlalo Wukro Hospital Dr. Mehari Desalegn Medical Director
8 Kilte Awlalo Wukro Hospital Alem G/Tsadik CEO
9 Kilte Awlalo Negash Health Center Birhane G/Aania Director
10 Kilte Awlalo Negash Health Center Hirit Asfa Finance 
11 Kilte Awlalo Agulae Health Center Tirete Zeleke Director
12 Kilte Awlalo CBHI Coordination Office Taeme G/Hiwot Coordinator
13 Kilte Awlalo Woreda Health Office Fitsum W/Aregay Head
14 Kilte Awlalo Woreda Administration Tsegaye Hadgu Administrator
15 Kilte Awlalo WOFED Kalayu G/Hiwot Head
16 Region HSFR/HFG Goitom Regional Coordinator
17 Region Health Bureau Berihu Core Process 
Owner/CBHI
18 Region BOFED Abebu Tadesse
19 Ahiferom Woreda Administration Gebremedihin Alemayehu Administrator
20 Ahiferom Woreda Health Office Tewelde Yigzaw Coordinator for 
CBHI
21 Ahiferom CBHI Coordination Office G/Kirstos Berha Coordinator
22 Ahiferom CBHI Coordination Office G/Medhin Asgedom Finance 
23 Ahiferom Enticho Health Center G/Egziabher Hailsellasie Director
24 Ahiferom WOFED Hailay Yohanes Head
25 Ahiferom Mezbir health Center Tsega Berhane Head
26 Adwa Town Adwa Hospital Meuz Abrha CEO
27 Tahitay Adiabo CBHI Coordination Office Haftom G/Giorgis Coordinator
28 Tahitay Adiabo WOFED Tsige Gesese Finance
29 Tahitay Adiabo Woreda Health Office Teklay Tesfay Head
30 Tahitay Adiabo Mai Kuhli Health Center Zeratsion Girmay Head
31 Tahitay Adiabo Sheraro Health Center Merid Mekonen Director
32 Shire Town Suhul Hospital Maasho Fiseha CEO
186
Annex K.4: SNNP 
No. Name Position Institution Type of Interview
Ashenafi Wagisso HSFR/HFG, SNNPR KII
Aklilu Tukela Development Plan 
Preparation, 
Monitoring and 
Evaluation Core 
Process Owner at the 
Vice Bureau Head 
Level
BOFED, SNNPR KII
Aschalew Ledetu CBHI Core Process 
Owner
CBHI Executive Team, 
Yirgalem City Administration
KII
Woyneshet Legesse Accountant KII
Sr. Meheret Arega Head Yirgalem Health Center KII
Fantaye Teshome Procurement and 
Payment Coordinator
KII
Bethlehem Wondimu Pharmacy Yirgalem Hospital FGD, staff
Dagem Desta OPD
Aregahegn B Laboratory
Tigist Yirgu Gyn/Obs
Ayalnesh Asegagn Medical
Meseret Bereda OPD Yirgalem HC FGD, staff
Amarech Annulo Emergency
Alemnesh Babu Pharmacy
Tamene Petros Head WOFED, Yirgalem KII
Admassu Arsicha General Manager Yirgalem Hospital KII
Tsegaye Gatiso Head WorHO, Yirgalem
Endashaw Esrael Mayor Yirgalem City Administration KII
Yohannes Nako Deputy Head WorHO, Wonago KII
Meselu Debela Medical and 
Renaissance Core 
Process Head
WorHO, Wonago KII
Tamrat Jilo Head Wonago HC KII
Aster Beshir Deputy Chief 
Administrator
Wonago Woreda KII
Asrat Tesfaye Head Hasse Haro HC, Wonago KII
Dr. Dagnachew 
Yohannes
Medical Director Dilla Hospital KII
GEbiso Hamito CBHI Non-Member Abosto kebele/tabia, Yirgalem Community FGD
Girma Dabusa
Fikru Yewa
Zenebe Senbeto
187
No. Name Position Institution Type of Interview
Mulunesh Darba CBHI Member Abosto kebele/tabia, Yirgalem Community FGD
Hameso Hasene
Getachew Woldie
Amarech Hussein
Mamo Gelana
Zerihun Yilma
Muluken Dubale Head Damboya HC, Damboya KII
Bediru Botoko CBHI Coordinator Damboya Woreda KII
Getachew Alemu Chief Administrator Damboya Woreda KII
Hasabe Hibso Head WorHO, Damboya KII
Dessalegn Mathewos Head Funto HC, Damboya KII
Tigist Amare Acting Head/OPD
Worknesh Wanna Emergency Funto HC, Damboya Staff FGD
Tigist Amare OPD
Belaynesh Kebebe Pharmacy
Mitiku Gebre Laboratory
Workie Ayele Pharmacy Damboya HC, Damboya Staff FGD
Kefle Abule OPD
Terefe Yacob Laboratory
Tadele Mekango OPD/Emergency
Degefu Kirgano Head WOFED, Damboya KII
Elias Anse CBHI Member Geramba kebele/tabia, 
Damboya
Community FGD
Asemo Abute
Da’o Egena
Markoye Wedlebo
Hemeto Banboro CBHI Member (Got 
Leader)
Mathewos Meharu CBHI Non-Member Geramba kebele/tabia, 
Damboya
Community FGD
Sayle Irgina
Habtamu Tadesse CEO Durame Hospital KII
Matusala Basa Head WorHO, Damot Weyde KII
Wondimu Gedebo Head WOFED, Damot Weyde KII
Meselech Shirko CBHI Members Bilu Bedesa kebele/tabia, 
Damot Weyde
Community FGD
Yohannes Tora
Ayelech Gebeyehu
Balcha Kora
Dawit Dana
Olana Kolcha
Matewos Minamo
Felekech Molla
188
No. Name Position Institution Type of Interview
Motote Bobe CBHI non-members Bilu Bedesa kebele/tabia, 
Damot Weyde
Community FGD
Labiso Lambebo
Tesfaye Biqamo
Azera Bine
Bereket Berhanu Head Koyo HC, Damot Weyde KII
Tegegn Badecho Head Bedesa HC, Damot Weyde KII
Mulugeta Under 5 OPD Bedesa HC, Damot Weyde Staff FGD
Alemnesh OPD
Aster Delivery
Zegene Woyesha Chief Administrator Damot Weyde Woreda KII
Abebayehu Likebo Pharmacy Koyo HC, Damboya Weyde Staff FGD
Tsedale Simeon Laboratory
Meseret Beyene MCH
Tesfanesh Bursamo OPD
Getu Mamo General Manger Wolayta Sodo Hospital KII
Zekariyas Zewde Health Coordinator Damot Weyde CBHI Team KII
Akale Gunta Accountant
 

***