Evaluation Research to 
Improve Newborn Health and Survival in the 
Integrated Nutrition and Health Program (INHP) II 
Area of CARE/India 
Endline Report 2006 
2
Acknowledgements 
The Project entitled “Evaluation Research to Improve Newborn Health and Survival in the 
Integrated Nutrition and Health Program (INHP) II area of CARE/India” is a partnership project 
of CARE/India, King George Medical University (KGMU) and the Johns Hopkins Bloomberg 
School of Public Health (JHSPH), USA. The project is funded by USAID/India. 
 
Collaborating Institutions: 
CARE-India, New Delhi, India 
King George Medical University (KGMU), Lucknow, India 
Johns Hopkins Bloomberg School of Public Health (JHBSPH), Baltimore, Maryland, USA 
Investigators: 
CARE-India: Usha Kiran, Sitaratna Devi, Dharmend Panwar, Deepika Chowdhury 
KGMC: Vinod Sreevashtav, Ramesh Ahuja 
JHU, USA: Mathuram Santosham, Abdullah H. Baqui, Gary L. Darmstadt, Robert E. Black, 
Peter Winch, Saifuddin Ahmed, Vishwajeet Kumar 
Data collection and processing: TNS-India 
Report prepared by: Abdullah Baqui, Praween Agarwal, Emma Williams, Saifuddin 
Ahmed, Mian B. Hossain, Gary L. Darmstadt, Sonia Walia and Amanda Rosecrans. 
Corresponding Author: 
Abdullah H. Baqui 
Johns Hopkins Bloomberg School of Public Health 
 615 N. Wolfe St., Ste. E8138 
Baltimore, MD 21205 
abaqui@jhsph.edu
3
 CONTENTS 
EXECUTIVE SUMMARY ..............................................................................................................................9
CHAPTER 1. OVERVIEW OF NEWBORN MORTALITY .......................................................................15
Magnitude of the newborn health problem at the global level ................................................. 15 
Newborn health in India and Uttar Pradesh state...................................................................... 15 
Models for community-based interventions to reduce neonatal mortality ............................... 16 
CHAPTER 2. THE INTEGRATED NUTRITION AND HEALTH PROGRAM OF CARE-INDIA ..............17
Longstanding collaboration with the government of India....................................................... 17 
INHP II interventions................................................................................................................ 18 
Elements of CARE-India’s newborn health care program ....................................................... 18 
Objectives of the evaluation research ....................................................................................... 19 
Collaborating partners............................................................................................................... 21 
CHAPTER 3. METHODOLOGY AND DATA COLLECTION ..................................................................22
Rationale for the study design................................................................................................... 22 
Study design.............................................................................................................................. 22 
Sampling design........................................................................................................................ 22 
Sample Size for Baseline and Endline Surveys .................................................................... 22 
Sample Size for Adequacy Surveys...................................................................................... 23 
Sample Selection................................................................................................................... 23 
Study instruments...................................................................................................................... 23 
Data Collection ......................................................................................................................... 25 
House Listing........................................................................................................................ 25 
Main Survey ......................................................................................................................... 25 
Service Provider Survey ....................................................................................................... 26 
Verbal Autopsy ..................................................................................................................... 26 
Sample coverage, Endline Survey ............................................................................................ 28 
Data quality assurance .............................................................................................................. 28 
Independent Data Quality Assurance by JHSPH Team........................................................ 28 
Key Variables........................................................................................................................ 29 
Data management...................................................................................................................... 29 
Statistical methods .................................................................................................................... 30 
Service Providers’ Knowledge ............................................................................................. 30 
Changes in Process Indicators and mortality between Study Arms ..................................... 30 
CHAPTER 4. SERVICE PROVIDERS’ TRAINING, KNOWLEDGE AND FACILITIES ..........................31
Training..................................................................................................................................... 31 
Auxiliary Nurse Midwives........................................................................................................ 32 
Anganwadi Workers ................................................................................................................. 34 
Change Agents.......................................................................................................................... 38 
CHAPTER 5. HOUSEHOLD CHARACTERISTICS ................................................................................41
CHAPTER 6. ANTENATAL CARE INDICATORS...................................................................................45
Coverage of home visits by service providers .......................................................................... 45 
Antenatal check-ups.................................................................................................................. 48 
Antenatal Check-ups............................................................................................................. 48 
Quality of antenatal check-ups and counseling ........................................................................ 51 
Tetanus Toxoid Immunization.............................................................................................. 51 
Iron-Folate Supplementation ................................................................................................ 52 
Birth Planning ....................................................................................................................... 53 
Components of Antenatal Counseling .................................................................................. 54 
4
CHAPTER 7. DELIVERY AND NEWBORN CARE.................................................................................58
Skilled birth attendance and institutional delivery care............................................................ 58 
Immediate Newborn Care Practices.......................................................................................... 60 
Postpartum Home Visits by Service Providers......................................................................... 65 
Care-seeking for Newborns ...................................................................................................... 68 
CHAPTER 8. EQUITY OF PROGRAM COVERAGE..............................................................................71
Comparison of Geographical Equity ........................................................................................ 71 
Comparison of Socio-cultural Equity ....................................................................................... 75 
Program Coverage by Standard of Living Index (SLI) in the Intervention District............. 75 
Comparison of Coverage by Scheduled Castes and Backward Classes vs. Other Groups... 81 
CHAPTER 9. TRENDS IN NEONATAL MORTALITY.............................................................................86
CHAPTER 10. CONCLUSIONS AND RECOMMENDATIONS ............................................................89
Program accomplishments and lessons learned........................................................................ 89 
Interpreting different findings within the study methods ......................................................... 92 
Lack of impact on neonatal mortality ....................................................................................... 93 
Recommendations:................................................................................................................ 94 
REFERENCES............................................................................................................................................97
APPENDIX: BEHAVIOR CHANGE MATRIX ..............................................................................................98
5
TABLES 
Table 1. Census data, Barabanki and Unnao .......................................................................... 21 
Table 2. Overview of the Study Design and Data Collected .................................................. 27 
Table 3. Endline Survey Coverage* ....................................................................................... 28 
Table 4. Changes in Auxiliary Nurse Midwives’ Knowledge................................................ 34 
Table 5. Anganwadi Workers’ Knowledge............................................................................. 37 
Table 6. Activities, Supplies and Equipment at Anganwadi Centers ..................................... 38 
Table 7. Characteristics of Change Agents at Endline............................................................ 39 
Table 8. Change Agents’ Knowledge Scores during the Intervention.................................... 40 
Table 9. Percentage distribution of selected household characteristics.................................. 42 
Table 10. Age distribution of the study population .................................................................. 43 
Table 11. Educational status of the study population ............................................................... 43 
Table 12. Characteristics of Recently Delivered Women......................................................... 44 
Table 13. Number and timing of antenatal check-ups ............................................................ 50 
Table 14. Antenatal check-ups by maternal characteristics*.................................................... 51 
Table 15. Services received during antenatal check-ups .......................................................... 53 
Table 16. Quality of Counseling during antenatal home visits by ANMs*.............................. 55 
Table 17. Quality of Counseling during antenatal home visits by AWWs*............................. 56 
Table 18. Quality of Counseling during antenatal home visits by Change Agents* ................ 56 
Table 19. Advice received during pregnancy from any source* .............................................. 57 
Table 20. Changes in Maternal Knowledge.............................................................................. 57 
Table 21. Institutional deliveries and birth attendance ............................................................. 58 
Table 22. Skilled birth attendance by Maternal Characteristics* ............................................. 59 
Table 23. Breastfeeding Initiation of First Day of Life, by Maternal Characteristics .............. 65 
Table 24. Care-seeking for Sick Newborns .............................................................................. 70 
Table 25. Mortality Rates.......................................................................................................... 87 
6
FIGURES 
Figure 1. Uttar Pradesh, India .................................................................................................. 20 
Figure 2. Barabanki (Intervention) and Unnao (Comparison) Districts................................... 20 
Figure 3. Coverage of CARE Training at Endline in Intervention District ............................. 31 
Figure 4. Training Coverage by Topic ..................................................................................... 32 
Figure 5. Correct Responses given by ANMs’ from Baseline to Endline ............................... 33 
Figure 6. Correct Responses given by Anganwadi Workers from Baseline to Endline .......... 36 
Figure 7. Antenatal Visits by Auxiliary Nurse Midwives........................................................ 46 
Figure 8. Antenatal home visits by Anganwadi Workers ........................................................ 46 
Figure 9. Antenatal home visits by ANMs, AWWs or Change Agents................................... 47 
Figure 10. Nutrition Health Day Attendance ......................................................................... 47 
Figure 11. Trends in coverage of antenatal home visits by Anganwadi Workers.................. 48 
Figure 12. Trends in Coverage of Antenatal Home Visits by Change Agents ...................... 48 
Figure 13. Received one or more antenatal check-ups........................................................... 49 
Figure 14. Trends in Coverage of Antenatal Check-ups........................................................ 50 
Figure 15. Received two or more Tetanus Toxoid Immunizations........................................ 52 
Figure 16. Iron-folic Acid Supplementation .......................................................................... 53 
Figure 17. Birth Planning Steps ............................................................................................. 54 
Figure 18. RDW with self-reported delivery complications that gave birth with a trained 
birth attendant ....................................................................................................................... 60 
Figure 19. Sterile cutting and tying practices at home deliveries .......................................... 61 
Figure 20. Timing of the newborn’s first bath ....................................................................... 62 
Figure 21. Newborn dried and wrapped immediately............................................................ 62 
Figure 22. Trends in prompt drying and wrapping ................................................................ 63 
Figure 23. Timing of Breastfeeding Initiation........................................................................ 63 
Figure 24. Colostrum Feeding................................................................................................ 64 
Figure 25. Trend line: breastfeeding initiation within one hour............................................. 64 
Figure 26. Timing of Postpartum Visits by ANMs................................................................ 66 
Figure 27. Timing of Postpartum Visits by AWWs............................................................... 66 
Figure 28. Timing of Postpartum Visits by Change Agents .................................................. 67 
Figure 29. Postpartum home visits by ANMs, AWWs or Change Agents ............................ 67 
Figure 30. Postpartum home visits by ANMs, AWWs or Change Agents ............................ 68 
Figure 31. Newborns that received a check-up from a trained provider within one month... 69 
Figure 32. Newborns with complications that received care from a trained provider.......... 69 
Figure 33. Auxiliary Nurse Midwife Visit Coverage by Sector............................................. 71 
Figure 34. Anganwadi Worker Visit Coverage by Sector...................................................... 72 
Figure 35. Change Agent Visit Coverage by Sector .............................................................. 72 
Figure 36. Antenatal Care Coverage by Sector...................................................................... 73 
Figure 37. Skilled Birth Attendance by Sector....................................................................... 73 
Figure 38. Clean cutting and tying practices by Sector.......................................................... 74 
Figure 39. Breastfeeding Initiation on Day of Birth by Sector .............................................. 74 
Figure 40. Received postpartum visit within three days by Sector........................................ 75 
Figure 41. Auxiliary Nurse Midwife Visit Coverage by SLI................................................. 76 
Figure 42. Anganwadi Worker Visit Coverage by SLI.......................................................... 77 
Figure 43. Change Agent Visit Coverage by SLI .................................................................. 77 
Figure 44. Coverage of one of more Antenatal Check-ups.................................................... 78 
7
Figure 45. Trained Birth Attendance (Home or Facility) by SLI........................................... 78 
Figure 46. Clean cutting and tying practices by SLI.............................................................. 79 
Figure 47. Breastfeeding Initiation on Day of Birth by SLI .................................................. 79 
Figure 48. Postpartum visit within three days by SLI............................................................ 79 
Figure 49. Auxiliary Nurse Midwife Antenatal Visits by Caste ............................................ 81 
Figure 50. Anganwadi Worker Antenatal Visits by Caste ..................................................... 82 
Figure 51. Change Agent Antenatal Visit Coverage by Caste............................................... 82 
Figure 52. Antenatal Check-up by Caste............................................................................... 83 
Figure 53. Trained Birth Attendance at Home or Facility by Caste....................................... 83 
Figure 54. Clean cutting and tying practices by Caste........................................................... 84 
Figure 55. Breastfeeding Initiation on Day of Birth by Caste................................................ 84 
Figure 56. Received postpartum visit within three days by Caste ......................................... 85 
Figure 57. Probability of Neonatal Death at Endline ............................................................. 87 
Figure 58. Probability of Neonatal Death, in Intervention district, Baseline vs. Endline...... 88 
Figure 59. Causes of Neonatal Deaths Established by Baseline Verbal Autopsy Study ....... 94 
8
LIST OF ACRONYMS 
ANC Antenatal check-up 
ANM Auxiliary nurse midwife 
AWC Anganwadi center 
AWW Anganwadi worker 
CA Change Agent (Community health volunteer) 
CBR Crude birth rate 
CEB Children ever born 
ICDS Integrated Child Development Service 
IFA Iron-folic acid (supplements) 
IMR Infant mortality rate 
INHP Integrated Nutrition and Health Project 
GOI Government of India 
JHSPH Johns Hopkins School of Public Health 
KGMU King George Medical University 
LBW Low birth weight 
NFHS National Family Health Survey of India 
NMR Neonatal mortality rate 
RDW Recently delivered women 
SLI Standard of living index 
TBA Traditional birth attendant 
TFR Total fertility rate 
TT Tetanus toxoid (immunization) 
9
EXECUTIVE SUMMARY 
Background: This report presents findings for the program evaluation of CARE-India’s newborn 
health intervention, one of the six technical interventions of the Integrated Nutrition and Health 
Project II project. Despite declines in global under-five and infant mortality rates during recent 
decades, newborn mortality has remained relatively high. An estimated 1.2 million neonates die 
each year in India alone, amounting to more than one-fourth of all global newborn deaths. Two￾thirds of infant deaths in India occur in the first month of life, and three-fourths of newborn 
deaths occur in the first week of life. As a result, neonatal health is increasingly viewed as a 
critical component of efforts to improve child survival, in India and globally.
CARE-India has been working with the government of India for decades to improve 
maternal and child health. Through USAID’s support, CARE-India began collaborating with the 
government of India to implement the Integrated Child Development Services (ICDS) program 
in 1984. In 1996, CARE-India, the government of India and USAID began the INHP I, which 
transformed ICDS from provision of supplementary food into a maternal and child survival 
initiative. INHP Phase I spanned from 1996 until 2001, at which point an evaluation concluded 
that the project was successful in facilitating joint problem solving, convergence of services at 
the community level, and community spirit and ownership to address health problems. Building 
on the successes of the INHP I, CARE-India began implementing the INHP II in 2001. The 
objective of INHP II was to achieve sustainable improvement in the nutrition and health status of 
vulnerable families in the project area by 2006. The project area included 7 million women and 
children in approximately 70 districts across 8 states in India. Specifically, the program planned 
to : 
• Improve the quality and coverage of maternal and child health services by community￾based health service providers, especially auxiliary nurse midwives (ANM) and 
anganwadi workers (AWW). 
• Strengthen key systems, including training, supply chain management and information 
management. 
• Sustain community activities for improved maternal and child survival. 
The INHP included 6 technical interventions: (1) Targeted supplemental feeding of pregnant and 
lactating women and children under 6 years of age; (2) Maternal and child immunization; 
(3) Antenatal care, including enhanced focus on nutrition, tetanus toxoid immunization, iron￾folate supplementation, and birth planning; (4) Infant feeding; (5) Vitamin A supplementation 
for children under the age of 3 years; and (6) Newborn care at the community level. 
Promoting newborn health in the context of Integrated Nutrition and Health Project II: The 
newborn care intervention was a new component of INHP. The newborn care package included 
technical interventions for both maternal and newborn care, including essential newborn care and 
strategies intended to improve care-seeking for newborn illnesses. 
The maternal care interventions included: promotion of early registration of pregnancy, antenatal 
care (ANC) visits and at least 3 visits; iron-folic acid supplementation; tetanus toxoid 
vaccination; maternal nutrition and rest; mebendazole during the second trimester of pregnancy; 
chloroquine prophylaxis during the second and third trimesters of pregnancy; birth planning for a 
clean and safe delivery. This includes identification of a trained birth attendant [i.e., auxillary 
nurse midwife (ANM) or a trained traditional birth attendant (TBA)] and a clean place for 
10
delivery; savings for emergency transportation; identification of emergency obstetric care 
facility; and identification of someone to accompany the mother, baby or both during an 
emergency; preparation for clean delivery (clean hands, clean surface, minimize vaginal 
examinations), including preparation (or purchase) and use of a clean delivery kit; recognition of 
danger signs during pregnancy, childbirth and the postpartum period and appropriate care￾seeking.The newborn care interventions were: immediate and exclusive breastfeeding; 
prevention of hypothermia through immediate drying, wrapping and delayed bathing; aseptic 
cord care; identification of small newborns and provision of extra attention, such as checking for 
hypothermia, providing skin-to-skin contact with the mother and giving extra feeding; 
identification of newborn danger signs, particularly inability to feed. 
Newborn Evluation Research Study Design: The purpose of this program evaluation was to 
demonstrate the level of impact that a basic maternal and newborn intervention package could 
achieve, in terms of use of improved maternal and newborn care practices and reduction of 
neonatal mortality. Although the project was implemented at scale in 8 states, its impact on 
neonatal mortality was evaluated under controlled conditions in one program district of Uttar 
Pradesh state. A quasi-experimental design was employed; one INHP II district, Barabanki, 
served as the intervention district, while a district receiving standard government health and 
ICDS services, Unnao, acted as the comparison. Nine rural blocks in the intervention district and 
eight rural blocks in the comparison district were randomly selected for the program evaluation. 
A baseline survey was conducted in 2003, before the launch of INHP II, and an endline survey 
was conducted in 2006. In addition, three smaller scale “adequacy” surveys were conducted to 
assess various process indicators related to coverage and adherence with the intervention..
Service Providers’ Training, Knowledge and Facilities: Capacity building was one of the critical 
components of the program. The project aimed to train all auxillary nurse midwives (ANMs) and 
anganwari workers (AWWs). As one component of the intervention 80% of AWWs and 43% of 
ANMs received training on pregnancy care, delivery care, and newborn care. In addition, a new 
category of volunteers from each village, called change agents, were recruited and trained. 
Changes in workers’ knowledge, from baseline to endline, were assessed in both the intervention 
and comparison districts. Both saw improvements in ANMs’ knowledge, but gains in the 
intervention area were more substantial. ANMs in the intervention area demonstrated 
improvements in all topic areas except knowledge of maternal post-partum danger signs, while in 
the comparison area ANMs demonstrated improvements in knowledge only of the five cleans of 
delivery and newborn danger signs. Reflecting the same trend, 38% of the AWWs in the 
intervention district received a knowledge score of 50% or higher at endline, compared with 28% 
at baseline. No change was observed in AWWs’ knowledge in the comparison district. The 
knowledge scores of CAs also increased from baseline to endline, suggesting that this group was 
able to master some key messages relayed during training. However, these findings also suggest 
that some persistent lapses remained in service providers’ knowledge. For example, only 27% of 
ANMs in the intervention area were able to list five or more delivery complications requiring 
immediate referral. Although topic-specific knowledge of CAs for newborn infection prevention 
and thermal care practices increased to11% and 13%, respectively, at endline, they were still 
low. 
11
Antenatal Care Indicators: In this report, an antenatal home visit refers to a health education 
session, while antenatal check-ups refer to sessions, usually conducted in a health facility, where 
pregnant women’s health was assessed by a doctor, nurse, ANM, or other qualified provider. The 
proportion of women visited during pregnancy at home increased in the intervention from 16% 
to 55%. The AWWs showed the greatest improvement in coverage with antenatal home visits, 
from 5% of women at baseline to 48% at endline survey in the intervention area. The auxiliary 
nurse midwives also showed improvement, from 12% at baseline to 36% at endline. There was 
no increase in antenatal home visits in the comparison area. Moreover, the percentage of women 
that received at least one antenatal check-up increased in the intervention area from 18% at 
baseline to 37% at endline, while no practical change was seen in the comparison area (26% to 
30%). However, only 13% of women in the intervention area received the three ANC visits 
recommended by the World Health Organization. Therefore, indicators moved in a positive 
direction but remained low by international standards. The program resulted in greatly improved 
coverage, but did not approach universal coverage. 
Behavior change communication (BCC) usually occurred in the context of home visits, or 
during Nutrition Health Days, designated days on which ANMs’ visits to a village coincided 
with ICDS food distribution. BCC promoted the use of antenatal care, tetanus toxoid 
immunization and iron-folate supplementation during pregnancy, birth planning, clean delivery 
and essential newborn care. The quality of antenatal counseling was assessed by asking women 
both unprompted and prompted questions regarding the information that was conveyed to them 
during pregnancy and also by assessing mother’s knowledge of danger signs. In general, most 
women in both the comparison and intervention areas received information about the importance 
of TT immunization and IFA supplementation at baseline, and small, equal increases in 
awareness were seen in both the intervention and comparison areas. About 15-30% of women in 
the intervention area indicated that the ANMs and AWWs had provided information about birth 
planning, the five cleans and essential newborn care practices, including immediate and 
exclusive breastfeeding, immediate drying and wrapping and delayed bathing, whereas few 
women (about 1-5%) in the comparison area had been given these messages during pregnancy. 
In contrast, messages about recognition of pregnancy or delivery complications or newborn 
danger signs were rarely recalled by mothers in the intervention or comparison areas. 
Furthermore, when maternal knowledge was assessed by asking women to list danger signs 
during pregnancy, childbirth, postpartum and the neonatal period newborns, knowledge scores of 
mothers in both arms increased only slightly. Therefore, it seems that both program coverage 
and quality improved during the antenatal period, but awareness of danger signs may have 
increased only marginally. 
Indicators related to the compliance with antenatal counseling varied. At baseline, about 
half of all women in the intervention area (47%) and just over half in the comparison area (58%) 
had received TTimmunization during their most recent pregnancy. This proportion increased to 
69% in the intervention area while remaining nearly the same in the comparison area at 62%. 
Receiving adequate supplementation of IFA (100 tablets) increased in both comparison and 
intervention areas, while consuming adequate supplementation increased only in the intervention 
arm, from 5% to 21%. While this indicator shows significant gain from baseline, only about one 
in five women took sufficient IFA. In addition, small gains have been observed in the proportion 
of women that complied with specific birth planning steps in both areas. 
12
Delivery Care and Essential Newborn Care: Most women in this setting give birth at home 
assisted by family members or others who have no formal midwifery training. Increasing 
institutional deliveries or birth attendance by doctors, nurses and ANMs was not a primary goal 
of this program, but patterns in birth attendance are presented because they are related to birth 
outcomes. Data suggest a modest trend toward more institutional births, from 11% to 15% in the 
comparison area and 12% to 17% in the intervention area. The cesarean section rate remained 
low at 2-3% in both communities, suggesting poor access to emergency obstetric care. 
Moreover, the proportion of women with self-reported delivery complications that received 
skilled birth attendance at home or in a facility remained unchanged. 
Behavior change communications in the intervention area focused on promoting essential 
newborn care practices, preventive care practices that have been shown to reduce risk of 
infection and hypothermia. Four key newborn care practices promoted through the intervention 
included: (a) cutting the umbilical cord with a blade that is new or boiled and tying it with boiled 
thread; (b) delaying the newborn’s first bath until for 3 or more days after delivery; (c) drying 
and wrapping the newborn immediately after delivery; (d) initiation of breastfeeding 
immediately (i.e., within 1 hour) after delivery. Almost of all these practices showed significant 
increase in the intervention area from baseline to endline. For example, the proportion of home 
deliveries in the intervention area for which the combined practice of sterile cord cutting and 
tying was used increased to 68% from 32% at baseline. The proportion of newborns bathed more 
than 6 hours after birth increased from 24% at baseline to 60% at endline, and the proportion 
bathed 3 or more days after birth also increased from 4% to 43%. The proportion of newborns 
that were dried and wrapped before the placenta was delivered more than doubled in the 
intervention area from 18% at baseline to 45% at endline. At baseline, most women in both study 
districts delayed the start of breastfeeding for several days after birth. Initiation of breastfeeding 
within 1 hour of delivery increased in the intervention area from 6% at baseline to 34% at 
endline. Initiation of breastfeeding during the first day of life rose from 20% to 64% in the 
intervention area. The comparison area saw no change in these essential newborn care indicators. 
Similarly, colostrum feeding increased from 20% to 88% in the intervention area, while little 
change (14% to 20%) was seen in the comparison area. These findings suggest that the program 
was successful in sharing information about essential newborn care with the communities. 
The program also emphasized postpartum home visitation to promote healthy newborn 
care practices and care-seeking for illnesses. Postpartum home visits by ANMs or AWWs were 
rare at baseline (< 2%) and remained that way in the comparison area at endline. In the 
intervention area, the greatest success in postpartum home visitation within 1 week of birth was 
achieved by the AWWs, which reached 23% at endline, followed by the change agents (17%) 
and the ANMs (9%). Rates of postpartum home visitations within 3 days were lower, 16% for 
AWWs, 13% for change agents and 6% for ANMs. These gains in coverage were notable and 
statistically significant, yet may have still represented lower than optimum coverage to impact 
mortality. Furthermore, in the intervention area only 1 in 3 sick newborns received care from a 
trained provider, and 27% received no care from outside the home. Reaching newborns during 
the first few days after birth clearly presents a logistical problem, and it remained challenging 
despite inroads made by the program. 
13
Equity of Program Coverage: The INHP II program was intended to cover all geographic sectors 
in the intervention district and to improve equity of coverage, regardless of socio-cultural factors. 
The program particularly sought to reach out to women from lower caste and lower 
socioeconomic groups, who might have been excluded by government programs in the past. 
Though some geographic variations were present, program coverage was relatively equal across 
sectors and improved within each sector over the course of the intervention. Most indicators 
showed relatively equal coverage among the low, medium, and high standard of living (SLI) 
households, and disparities that were present at baseline diminished at the endline. However, 
care-seeking outside of the program services continued to show striking disparities by wealth; for 
example, antenatal care coverage (low: 31%, high: 58%) and use of skilled attendance (low: 
16%, high: 42%) showed large differences by socioeconomic conditions, suggesting differential 
access to care. Similar patterns emerged when lower caste groups were compared with other 
groups with regard to care-seeking behavior. Anganwadi worker and change agent coverage for 
antenatal and postpartum visits showed slightly better coverage for lower caste households at the 
endline survey. One highly positive findings of this study is that improvements in coverage and 
practices were inclusive of all socioeconomic groups. However, it remains clear that barriers to 
care-seeking, though outside the scope of this project, remain significant.
Trends in Neonatal Mortality: Neonatal mortality rates were constant in both the comparison and 
intervention areas from baseline to endline. The neonatal mortality rate in the comparison area 
was 47.3 per 1,000 live births at baseline and 49.8 per 1,000 live births at endline. The neonatal 
mortality rate in the intervention area was 50.5 per 1,000 live births at baseline and 47.8 per 
1,000 live births at endline. No differences were seen in early neonatal, late neonatal or stillbirth 
rates. 
Conclusions and Recommendations: In summary, this program evaluation has identified a 
number of program strengths. The inputs that were provided were achieved at scale within a 
limited time frame. The increases in coverage were achieved mostly by improved management 
and training, without increasing the number of paid community-based workers. Greater gains in 
program coverage were seen in the antenatal period. However, improvements in essential 
newborn care practices were dramatic and suggest that the methods used to disseminate health 
information were effective. One of the critical lessons learned from this project is that it is 
possible to improve coverage of essential newborn care practices at scale through community￾based programs. Moreover, the improvements were made irrespective of socioeconomic status or 
caste distinction, which is an important achievement for a community-based program.
These findings also suggest some program limitations. The quality of home visits 
remains an issue of concern; specifically, very few mothers recalled receiving information about 
postpartum complications or newborn dangers signs. Having this information is one necessary 
prerequisite to obtaining health care for complications. Skilled attendance at birth and use of 
trained providers for complication remained low and differentials among the rich and poor 
remain striking. Skilled birth attendance and access to emergency obstetric care are critical to 
reducing maternal and newborn mortality. Increasing program coverage during the postpartum 
period proved challenging. Less than one-fourth of all newborns were checked by a trained 
provider. The proportion of newborns that received trained care increased from 20% to 30% in 
the intervention area. 
14
 Some program strategies were clearly effective, and some areas for improvement are 
suggested by the findings of this study. Ultimately, the improvements in process indicators were 
not reflected in a reduction in neonatal mortality in the intervention area. It is difficult to isolate 
specific factors that were responsible for the lack of impact. However, a few areas of concern 
emerged from these. The first is that extra care for small or low birth weight newborns – such as 
steps to prevent hypothermia and infection - may have been insufficient to impact deaths. 
Another problem clearly related to the timely identification of and treament for neonatal 
complications. Approximately one-fourth of deaths occurred on the first day of life, and three￾quarters of deaths took place during the first week. Findings suggest that significant barriers to 
accessing health care remain, particularly for poor and lower caste women. Developing effective 
strategies for community-based management of pneumonia and sepsis will be important to 
achieving reductions in neonatal mortality. Maintaining the knowledge and counseling abilities 
of community-based workers is one critical issue. In addition, it will be useful to develop 
strategies to help women and families retain the information conveyed in behavior change 
communication. Barriers to care, particularly skilled birth attendance and emergency obstetric 
care, mostly likely hindered the program’s impact on neonatal mortality. Overcoming these 
barriers should be a goal for future programs. 
Although proven interventions to reduce newborn deaths are known, much work needs to be 
done to determine how to deliver these interventions at scale without losing program 
effectiveness. This study was the first to examine the large-scale effectiveness of a neonatal 
health package implemented through existing governmental and nongovernmental organization 
services. Therefore, this study provides a significant contribution to researchers’ understanding 
of how to reduce newborn deaths in low-resource settings. 
15
CHAPTER 1. OVERVIEW OF NEWBORN MORTALITY 
Magnitude of the newborn health problem at the global level 
Despite declines in global under-five and infant mortality rates during recent decades, newborn 
mortality has remained relatively high (1, 2). Approximately two-thirds of infant deaths and 38% 
of under-five deaths now occur in the first month of life (2). An estimated 4 million newborns 
die globally each year, and approximately 98% of these deaths occur in developing countries 
(2). Another 3 million babies are stillborn during the last trimester of pregnancy (6). Perinatal 
mortality, which includes stillbirths of more than 28 weeks gestational age and newborn deaths 
during the first week of life, now nearly equals mortality during the entire first year of life (4, 7). 
The risk of dying per month is approximately 15 times higher in the first month of life than in 
each of the remaining 11 months of infancy. The risk of dying during the first month of life is 
also approximately 30 times greater than at any month during young childhood (months 12- 59) 
(8, 9). In South Asia, the loss of healthy life from newborn deaths represents 13.6% of the 
disease burden, or 53 million years of life lost (3). In order to achieve further gains in child 
survival, a new focus on improving newborn health is needed, particularly with regard to 
outcomes in the perinatal period (10-12). 
The three major causes of newborn death are infections (36%), birth asphyxia and other birth 
injuries (23%), and complications of prematurity (28%) (2, 4). Many of these deaths can be 
prevented through implementation of simple, cost-effective interventions that do not require 
highly technical equipment or highly trained personnel (13). The health of the mother during 
pregnancy, delivery and the postpartum period is intimately linked with the health of her 
newborn, which creates a need for integrated maternal and newborn health care strategies (14). 
Globally, two-thirds of births and most newborn deaths occur in the home, outside the formal 
health care system (13). Thus, in order to advance newborn health and survival, interventions 
must be introduced at the household and community levels, and they must include links with the 
health care system for treatment of life-threatening conditions. 
Newborn health in India and Uttar Pradesh state 
Infant and under-five mortality rates in India had declined over the past 2 to 3 decades until the 
mid-1990s, when rates stagnated. The newborn mortality rate, however, has declined little over 
recent decades, and remains 43 per 1,000 live births (5). Improvements in perinatal and newborn 
health are now seen as crucial to advancing child survival in India (15). Each year, an estimated 
1.2 million neonates die in India alone, amounting to one-fourth of all global newborn deaths (2-
4). Two-thirds of infant deaths in India occur in the first month of life, and three-fourths of 
newborn deaths occur in the first week of life (5). Moreover, India has the highest number of 
newborn deaths due to tetanus and the largest number of low birth weight (LBW) babies of any 
country in the world (4). An estimated 60 to 80% of perinatal and newborn mortality occurs 
among LBW infants (5, 10, 16). Two-thirds of all births and three-fourths of births in rural areas 
of India take place at home. Untrained providers attend nearly 60% of all births (17). 
Maternal and child health indicators of Uttar Pradesh state are poor when compared with India as 
a whole. The India National Family Health Survey of 1998-99 found that Uttar Pradesh state had 
16
the second-highest neonatal mortality rate in the country, at 53.6 per 1,000 live births and the 
second-highest infant mortality rate, 86.7 per 1,000 live births (5). Only 16% of births took 
places in health facilities and 22% of all births were attended by a health professional. Given that 
Uttar Pradesh is also the most populous state in India, addressing neonatal mortality in this state 
is critical to achieving national goals related to neonatal mortality reduction. 
Models for community-based interventions to reduce neonatal 
mortality 
The work of the nongovernmental organization, SEARCH has served as a demonstration for the 
potential to advance newborn health and survival through community-based health care. In 
Gadchiroli District of Maharashtra state, SEARCH trained female Village Health Workers 
(VHWs) to provide a package of home-based newborn care, including health education for 
pregnant women; recognition and management of birth asphyxia; identification of high-risk 
(premature and LBW) neonates for more intensive surveillance; temperature maintenance; 
promotion of breastfeeding; treatment of skin infections; identification of sick newborns 
suspected of having septicemia, meningitis and/or pneumonia using a simple algorithm; and 
administration of antibiotics (oral cotrimoxazole and intramuscular gentamicin) in the home (10, 
18). This community-based intervention reduced sepsis-related newborn mortality by 76%, 
asphyxia-related mortality by 48%, and overall newborn mortality by 62% in intervention 
villages, compared to comparison areas. The estimated cost was $5.30 per neonate. This cost 
estimate, however, is likely to be an under-estimate of the program’s full cost. 
In an earlier uncontrolled study in India, VHWs identified and managed high-risk neonates in the 
home, including preterm and LBW infants and those with feeding problems, illness or a history 
of prolonged and difficult labor. Interventions included: resuscitation of asphyxiated newborns, 
including cleaning of the mouth and the pharynx and mouth-to-mouth respiration; aseptic cord 
care; promotion of breastfeeding; minimal handling during delivery to reduce the risk of 
infection; postpartum home visitation and provision of guidance about routine newborn care; and 
referral of sick newborns and those with feeding problems to the hospital. Home care of preterm 
and LBW infants also included keeping the room warm and feeding with a dropper if necessary. 
Newborn mortality declined by 25% during the intervention year, compared to the year before 
the implementation of the program (19). Thus, a less intensive package of interventions than 
those used by SEARCH may save a significant proportion of newborn lives. 
17
CHAPTER 2. THE INTEGRATED NUTRITION AND HEALTH 
PROGRAM OF CARE-INDIA 
Longstanding collaboration with the government of India 
CARE has been collaborating with the Government of India (GOI) for more than 50 years in 
development assistance and relief projects. Support from USAID’s Title II Program has served 
as the foundation of these efforts, which began with food-for-work and school feeding projects 
and evolved into support for the government of India’s nutrition and health programs. In 1984, 
CARE-India, with USAID’s support, began collaborating with the Integrated Child Development 
Services (ICDS) program of the Department of Women and Child Development of the 
government of India in eight states to implement a nutrition and health program. In 1996, CARE￾India, the GOI and USAID began the Integrated Nutrition and Health Project (INHP), which 
transformed the Title II program support to the ICDS from provision of supplementary food into 
a maternal and child survival initiative. 
INHP was envisaged as a 10-year project to be implemented in two phases of 5 years each. The 
purpose of INHP was to work with governmental and non-governmental counterparts to improve 
the health and nutrition of women and children. Implementation strategies were designed to 
enable women to attain and maintain optimal health and nutrition for themselves and their 
children, and to enhance the capacity of communities and counterparts to achieve this goal. 
Working within the broader mandate and scope of the ICDS program and the Ministry of Health 
and Family Welfare, CARE-India focused on activities which they felt have the greatest potential 
to improve the health and nutritional status of pregnant and lactating women, adolescent girls, 
and children under the age of 2. INHP is significant in size and scope; it reached over 7 million 
women and children in 100,000 villages during its first phase. By innovating and then promoting 
replication of strategies that were found successful at small scale, INHP expanded its influence 
to an even larger population through the ICDS, which currently reaches almost 5 million women 
and 23 million children. 
INHP Phase I spanned from 1996 until 2001, at which point the INHP I evaluation team 
concluded that the project was successful in facilitating: joint problem solving, convergence of 
services at the community level, and community spirit and ownership to address health 
problems. The intervention emphasized the promotion of innovation and experimentation, 
increasing demand and coverage of prioritized interventions and continuously improving 
commodity management. The evaluation team also highlighted areas requiring improvement, 
which include uneven and lower than expected performance in certain outcome indicators. 
Factors related to low performance in some places were: too many interventions were 
undertaken, large geographic areas were covered with limited resources, sharing and promotion 
of best practices was inadequate, staff skills and technical support were uneven, and monitoring 
and information systems needed improvement. 
Building on the successes of the INHP I and solutions to past problems, CARE-India began 
implementing the INHP II in 2001. The objective of INHP II was to achieve sustainable 
improvement in the nutrition and health status of vulnerable families in the project area by 2006. 
The project area included 7 million women and children in approximately 70 districts across 8 
18
states in India. To achieve this objective, CARE and its partners were working toward the 
following sub-objectives: 
Improve the quality and coverage of maternal and child health services by service providers, 
especially auxiliary nurse midwives (ANM) and anganwadi workers (AWW). 
Strengthen key systems, including training, supply chain management and information 
management. 
Sustain community activities for improved maternal and child survival. 
INHP II sought to expand the maternal and child survival components to improve their potential 
for impact. In INHP II, CARE-India continued its role as an innovator and a catalyst with 
increasing efforts to promote replication of best practices. 
INHP II interventions 
INHP II focuses on the following 6 technical interventions: 
(1) Targeted supplemental feeding of pregnant and lactating women and children under age 6; 
(2) Maternal and child immunization; 
(3) Antenatal care, including enhanced focus on nutrition, tetanus toxoid immunization, iron￾folate supplementation, and birth planning; 
(4) Infant feeding; 
(5) Vitamin A supplementation for children under the age of 3; and 
(6) Newborn care at the community level. 
The newborn care intervention was a new component of INHP. It was a focused package of 
newborn care practices to be implemented at the community level to improve the health and 
survival of newborns. It included strategies intended to improve care-seeking for newborn 
illnesses. 
Key outcome indicators for INHP II are coverage rates for supplementary feeding, antenatal 
care, immunization, infant feeding, vitamin A supplementation, newborn care and neonatal 
mortality. 
Elements of CARE-India’s newborn health care program 
The elements of CARE-India’s newborn package were selected through wide consultation with 
CARE-India’s INHP team members, BASICS II, MEDS Project, CARE/USA, Johns Hopkins 
University, and a range of other stakeholders, including the government of India’s Ministry of 
Health and Family Welfare (MOH), Indian academicians, NGO leaders and bilateral 
organizations through an expert consultative meeting in the fall of 2001. The Newborn Health 
Care Program aimed to increase the practice of specific behaviors during the antenatal, delivery, 
and postnatal periods that have proven association with better rates of maternal and newborn 
survival. Enhanced community capacity to sustain gains in maternal and newborn health was 
also a major goal. 
The newborn care package included technical interventions for both maternal and newborn care. 
The maternal care interventions were as follows: 
(1) Promotion of early registration of pregnancy, antenatal care (ANC) visits and at least 3 
visits; 
19
(2) Iron-folic acid supplementation; 
(3) Tetanus toxoid vaccination; 
(4) Maternal nutrition and rest; 
(5) Mebendazole during the second trimester of pregnancy; 
(6) Chloroquine prophylaxis during the second and third trimesters of pregnancy; 
(7) Birth planning for a clean and safe delivery. This included identification of a “trained” birth 
attendant [i.e., trained traditional birth attendant (TBA) or ANM]; a clean place for delivery; 
savings for emergency transportation; identification of emergency obstetric care facility; and 
identification of someone to accompany the mother, baby or both during an emergency; 
(8) Preparation for clean delivery (clean hands, clean surface, minimize vaginal examinations), 
including preparation (or purchase) and use of a clean delivery kit; 
(9) Recognition of danger signs during pregnancy, childbirth and the postpartum period and 
appropriate care-seeking. 
The newborn care interventions were as follows: 
(1) Immediate and exclusive breastfeeding; 
(2) Prevention of hypothermia through immediate drying, wrapping and delayed bathing; 
(3) Aseptic cord care; 
(4) Identification of small newborns and provision of extra attention, such as checking for 
hypothermia, providing skin-to-skin contact with the mother and giving more frequent 
breastfeeding; 
(5) Identification of newborn danger signs, particularly inability to feed. 
The attached Appendix 1 on INHP II Nutrition and Health Behaviors Matrix provides more 
detailed information on key contacts, the target groups and the behaviors and services that were 
promoted. 
Objectives of the evaluation research 
The purpose of this program evaluation is to demonstrate the level of impact that the basic INHP 
II package can achieve, in terms of use of improved maternal and newborn care practices and 
reduction of neonatal mortality in a programmatic setting. INHP II included neonatal mortality as 
one of its impact indicators, but due to several evaluation-related constraints, it was not possible 
for INHP to evaluate this indicator for all of its project sites. Therefore, while the project was 
being implemented at scale in 8 states, its impact on neonatal mortality was evaluated using a 
quasi-experimental design in 2 districts of Uttar Pradesh state only. 
This report presents findings for the program evaluation of CARE-India’s newborn health 
intervention, one of the six technical interventions of the INHP II project. Findings are based on 
a household survey that was conducted in 2003, three adequacy surveys that were administered 
during the program implementation and the endline survey that was conducted in January – 
March 2006. The objectives of this analysis are to: 
Compare the intervention district (Barabanki) and the comparison district (Unnao) with regard to 
health indicators and socio-demographic characteristics. 
Describe the baseline coverage of key process indicators (coverage, knowledge, practices), and 
track changes in indicators over the course of the intervention. 
Describe changes in neonatal mortality from baseline to endline. 
20
Figure 1. Uttar Pradesh, India 
Figure 2. Barabanki (Intervention) and Unnao (Comparison) Districts 
Before selecting the study districts, available data from several districts were compiled and 
reviewed. Based on data available from the Reproductive and Child Health (RCH) rapid survey 
and the comparability of both districts, the CARE/INHP district of Barabanki was selected as the 
intervention district and the district of Unnao as the comparison. The following table provides 
data on the two districts from India’s 2001 census. 
21
Table 1. Census data, Barabanki and Unnao 
District 
Total 
population Males Females 
Intercensal 
growth 
1981-1991 
Intercensal 
growth 
1991-2001 
Barabanki 2,673,394 1,417,213 1,256,181 26.59% 26.40% 
Unnao 2,700,426 1,422,965 1,277,461 20.73% 22.72% 
The comparison district of Unnao continued to receive nutrition and health services provided by 
ICDS and Ministry of Health of government of India including antenatal care, tetanus toxoid 
immunization, and iron-folate supplementation. However, there were no inputs from CARE￾India in the district of Unnao. Therefore, Unnao did not have any of the CARE innovations 
including no change agents (volunteer community health workers) to educate women about birth 
preparedness and newborn care. Behavior change communications related to birth planning and 
newborn care were provided in the intervention district (Barabanki) only. Home visitation by 
ANMs and change agents was promoted in the intervention area, particularly during the first 
week of life, to identify and refer newborns with health problems. 
Formative research to identify routine newborn care practices and possible improvements in 
these practices was conducted by CARE-India as part of the design of INHP II. The formative 
research revealed that in INHP areas: 
Low literacy rates restricted awareness among women about health issues, including basic 
maternal and child health and the availability of services. (For example, tetanus toxoid 
immunization coverage was less than 50%). 
Decision-making rested largely with men and women had little influence about seeking maternal 
and newborn care outside the family. 
Newborn neglect was under-recognized, and under-reporting of births and deaths was common. 
A variety of potentially harmful traditional care practices persisted with regard to pregnancy, 
childbirth and the newborn period. 
Collaborating partners 
Johns Hopkins Bloomberg School of Public Health (JHSPH) in partnership with King George 
Medical University, Lucknow conducted the evaluation of the CARE-India newborn program. 
JHSPH’s and KGMU’s role was limited to assisting CARE with the design and evaluation of the 
Newborn Health Care Program. 
22
CHAPTER 3. METHODOLOGY AND DATA COLLECTION 
Rationale for the study design 
The objective of this study was to evaluate the impact of a package of community-based 
maternal and newborn care services on maternal and newborn care practices and neonatal 
mortality. The CARE program intended to reduce neonatal mortality by at least 20% in a rural 
district of India over a 2-year period. Existing community groups, health workers and change 
agents (volunteer health workers) were trained to provide counseling to pregnant women and 
their families through home visits so that they would seek necessary preventive and curative care 
from government and NGO facilities. In order to allow for two full years of program 
implementation, the study continued for 2 ½ years; the additional 6 months allowed for 
refinement and maturation of the intervention. In order to assess the effectiveness of the package 
of services, a quasi-experimental design with intervention and comparison areas was employed. 
In addition to baseline and Endline surveys, we assessed the adequacy of the intervention’s 
implementation every 6 months. These “adequacy surveys” were conducted so that the 
intervention strategies could be adjusted as appropriate. Three such adequacy surveys were 
conducted over the 2-year study, beginning 6 months after the launch of the intervention. 
Study design 
The district of Barabanki in Uttar Pradesh state, where CARE has been implementing the INHP, 
was selected as the intervention district. The nearby district of Unnao, which has the Government 
of India’s ICDS program, was selected as the comparison district. The evaluation components 
included surveys of households and health care providers to establish baseline levels for process 
and outcome indicators, including neonatal mortality, during the 2 years preceding the survey. 
To assess the impact of the intervention, the household survey was repeated after 30 months of 
program implementation. 
As the implementation of the interventions continued, the adequacy of the implementation of the 
interventions was assessed every 6 months by measuring selected process indicators in both 
intervention and comparison communities. The adequacy surveys were conducted in each study 
cluster three times during the 30 months of intervention: between 7-8 months, 13-14 months and 
19-20 months. Adequacy indicators assessed changes at the community-based provider level and 
at the community/household level, by collecting data from samples of women that gave birth 
during the 6 months preceding the survey and from community-based health workers. 
Sampling design 
Sample Size for Baseline and Endline Surveys 
The program aimed to decrease neonatal mortality in the intervention area by at least 20%. 
Sample size was calculated to detect expected differences in neonatal mortality rates between the 
intervention and comparison areas with 80% power and at the 95% confidence level. To account 
for clustering, the estimated sample size was multiplied by an assumed design effect of 1.2. This 
required a population that would provide information on at least 8,332 live births in each study 
arm, or a total of 16,664 live births. Since the study took place over 2 years, the number of births 
needed in the cohort per year is half the total number. The estimated birth rate in the area was 35 
23
live births per 1,000 people. Each household had an average of 5 members, so a population of 
1,000 would include about 200 households. Taking these factors into consideration, the number 
of households required to record 16,664 births was equal to: 
(16,664x200x0.5)/35= 47,611 
Therefore, approximately 24,000 households were needed in each of the 2 districts to record the 
outcomes of 16,664 live births over 2 years. Assuming a non-response rate of 10%, a minimum 
of 26,400 households were needed in each area. 
Sample Size for Adequacy Surveys 
Adequacy surveys measured coverage of the intervention, and selected process indicators of 
knowledge and practices. The sample size for the adequacy surveys was estimated on the basis of 
the following assumptions and considerations. To obtain the most conservative (largest) sample 
size, estimates were based on an assumption of 50% prevalence of a given care practice among 
households with a women who had given birth in the 6 months prior to the survey (i.e., recently 
delivered women, RDW). To measure each indicator at sector level with ± 10% precision, it was 
necessary to interview at least 100 RDW per sector.1
 This would result in at least 900 mothers 
from the intervention arm and 800 mothers from the comparison arm, 1700 women altogether. 
To detect an arm-level difference of 10%, we would need at least 538 mothers from each arm of 
the study. Therefore, this sample of 1700 mothers would allow us to detect a + 10% difference 
between study arms with 90% power. Assuming a 15% non-response rate, 115 RDW from each 
sector were sampled to participate in the study. The sample was randomly selected from each 
anganwadi center catchment area. The number of women selected from each anganwadi center 
was proportional to the number of births in the area. 
For the service providers’ survey, attempts were made to interview all ANMs, AWWs and 
change agents in the study areas of both districts. 
Sample Selection 
Rural blocks from the intervention and comparison districts were selected for the evaluation 
research. The district of Barabanki has 16 blocks, 15 of which are rural. At baseline, the Indian 
Council of Medical Research was conducting a pilot project in 2 of the 15 blocks, so they were 
excluded from this study. Nine of the remaining 13 blocks were randomly selected as 
intervention sites. One sector, an area with 15-25 anganwadi centers (AWC) and an estimated 
population of 20,000-25,000, was randomly selected from each of the 9 selected blocks. 
Similarly, in Unnao where there are 16 blocks, 15 of which are rural, 8 blocks were randomly 
selected as comparison sites and one sector was randomly selected from each block. 
Study instruments 
The following questionnaires were developed, pre-tested and administered in the field at 
baseline. 
1. House mapping and listing forms: This instrument was used to list all households, record 
the location of each household and assign the household a unique identification number. 
 
1
 Given that the average population of a sector is about 20,000 and the annual birth rate is about 25 per 1,000, the 
number of expected births per sector over 6 months period would be about 250. Therefore, it was feasible to sample 
100 women from each sector. 
24
Data collected included the structure number, household number, name of the head of the 
household, and number of women aged 13-49 who had been married. 
2. Household questionnaire: This instrument was used to list the usual members of all 
surveyed households and to collect limited socio-demographic information, including 
age, sex, marital status, relationship to the head of the household, level of education and 
occupation from each household member. Each person’s duration of stay in the village 
was recorded. The type of construction material used in the house’s roof, wall and floor 
were also recorded. 
3. Pregnancy history form: This form collected information from all eligible women, 
women between the ages of 13 and 49 years who were usual residents of the household 
and reported that they were married or had been married. Factors that are likely to 
influence demographic and reproductive health behavior - such as age, current marital 
status, education and age at first marriage - were collected. In addition, information about 
all pregnancies and pregnancy outcomes (live birth, still birth and spontaneous or induced 
abortion) was collected, including the date of each outcome. The status of each live born 
baby was noted, and if a baby was reported dead, the age at death was recorded. 
4. Recently Delivered Woman (RDW) questionnaire: All women whose pregnancy history 
included a live birth within the 24 months preceding the baseline survey were asked to 
complete this questionnaire. It included information on knowledge and practices 
regarding the antenatal period, delivery, the postpartum period and newborn care. It 
recorded levels of coverage for antenatal, delivery, postpartum and newborn services 
during the women’s last pregnancy. Antenatal questions asked about the essential care 
received during pregnancy, birth planning, birth complications and their management. 
Delivery care questions focused on safe delivery practices, problems experienced and 
their management. Questions about postpartum care included the appropriate timing and 
frequency of care. Questions about immediate newborn care included management of the 
umbilical cord, baby weighing, bathing and wrapping of the newborn, use of pre-lacteal 
feeds, and breastfeeding. Questions about newborn care during the first month of life 
asked whether the newborn received a check-up and who conducted it, cord care 
practices, breastfeeding practices and illnesses experienced by the newborns and 
treatments for them. 
5. Socio-economic survey questionnaire: This questionnaire collected information on the 
availability of basic amenities and facilities in the households that included a RDW. It 
recorded ownership of land and livestock, possession of durable goods and the 
household’s main source of income. 
6. Verbal autopsy questionnaire: Among RDWs who reported the death of a newborn or a 
stillbirth within the 24 months preceding the baseline survey, detailed information on the 
circumstances surrounding the death was collected so that cause of death could be 
inferred. The verbal autopsy instrument was adapted from a World Health Organization 
instrument that was developed by a team that included some of the investigators of this 
study, Abdullah Baqui and Gary Darmstadt. The verbal autopsy started with open-ended 
questions to elicit the mother’s narrative about the newborn’s death, followed by a series 
of structured questions for the mother or another family member who had firsthand 
experience of the incident. The verbal autopsy data was analyzed and reported separately. 
7. Service provider questionnaires: There were three separate instruments, one each for 
ANMs, AWWs and change agents. The questionnaires were designed so that each 
25
instrument would be appropriate to the workers’ responsibilities. Information on 
background characteristics, knowledge, practices, roles and responsibilities of the 
workers was gathered. 
Data Collection 
All survey data collection and initial processing were contracted to a survey research agency 
called TNS-MODE. Baseline and adequacy survey data collection procedures have been 
described previously in the baseline and adequacy reports. The same instruments were used in 
the baseline, adequacy and endline surveys to ensure comparability. A few changes were made 
over the study period. We carefully reviewed the changes and accounted for any change during 
analysis, to ensure comparability. 
House Listing 
Field investigators were recruited and trained by the survey research agency. Several CARE￾India and JHSPH investigators participated in the training. The investigators also participated in 
the assessment of the interviewers finally deployed in data collection. Training consisted of 2 
days of classroom training and 2 days of field practice. Investigators formed 20 teams for 
mapping and listing activities, each team consisting of one lister and one mapper. Five 
supervisors and 2 coordinators managed the listing and mapping operations under the guidance 
of field executives and a regional field manager. Listing and mapping of households in one 
anganwadi center (AWC) catchment area was completed in approximately 2 days. Completed 
listing-mapping forms were delivered regularly to the JHSPH office in Lucknow. These forms 
were returned to the survey agency after review. 
Coverage of the area was sequentially validated by two supervisors, field executives, and the 
JHSPH coordinator. The completed listing-mapping forms served as the sampling frame to 
identify eligible and recently delivered women. The listing and mapping operations were 
completed within two months. 
Main Survey (Household questionnaires, pregnancy histories, RDW questionnaire 
and socio-economic survey) 
The survey team consisted of interviewers, supervisors and field editors. The survey team was 
trained for 7 days, 5 in the classroom and 2 in the field. The researchers and field managers of 
the survey agency, evaluation and monitoring experts from CARE-India, and JHSPH 
investigators and coordinator provided the training. Field training was conducted in the areas that 
were not selected for the actual study, to avoid testing bias. After completing field practice, the 
survey team received a final briefing to ensure common understanding of the data collection 
instruments and related work responsibilities. Following the briefing, the interviewers’ 
knowledge and ability was evaluated. During the baseline data collection, 80 female 
interviewers, 20 supervisors and 20 field editors were selected to conduct the fieldwork. Twenty 
teams were formed, each consisting of 4 interviewers, one supervisor and one field editor. 
During the endline data collection, 120 female interviewers, 30 supervisors and 30 field editors 
were selected to conduct the fieldwork. Thirty teams were formed, each consisting of 4 
interviewers, one supervisor and one field editor like baseline. 
Fieldwork began simultaneously in both districts, covering one sector at a time. After office 
editing, questionnaires were sent weekly to the JHSPH coordinator for review and data quality 
26
assurance sampling. They were then returned to the survey agency for data entry and validation. 
Fieldwork for baseline was completed within 6 months whereas for endline it was completed 
within 3 months. 
Service Provider Survey 
Change agents, AWWS and ANMs were interviewed by field supervisors, either on the day of 
the main survey or a later date, depending on the service providers’ availability. 
Verbal Autopsy 
A separate team of senior level researchers, including medical doctors, was formed to conduct 
verbal autopsies. The verbal autopsy team revisited households that reported a neonatal death or 
stillbirth with a few weeks of the RDW survey. The recall period was the same as the recall 
period for the main survey, i.e., 24 months. Verbal autopsy findings will be presented separately. 
27
Table 2. Overview of the Study Design and Data Collected 
Data 
Collection 
Round 
Data Collection 
Instrument Respondents 
Recall 
Period 
Data 
Collection 
Period Sample Size 
 Comparison Intervention 
Baseline Household Listing All households in study area 26,595 29,560 
 Pregnancy History All ever-married women aged 13-49 Lifetime 22,845 26,420 
 Recently Delivered 
Women survey 
Women who had a live birth or still birth 6,490 9,220 
 Service Providers Auxiliary Nurse Midwives, Anganwadi workers, 
Change agents 
Jan. 
2001-
May 
2003 
Jan-Jun 
2003 
178 283 
Adequacy I Recently Delivered 
Women survey 
Women who had a live birth or still birth 832 927 
 Service Providers Auxiliary Nurse Midwives, Anganwadi workers, 
Change agents 
June 
2003-
March 
2004 
Jan–April 
2004 
182 717 
Adequacy II Recently Delivered 
Women survey 
Women who had a live birth or still birth 807 936 
 Service Providers Auxiliary Nurse Midwives, Anganwadi workers, 
Change agents 
March –
Sept. 
2004 
Sept-Oct 
2004 
170 819 
Adequacy III Recently Delivered 
Women survey 
Women who had a live birth or still birth 819 932 
Service Providers Auxiliary Nurse Midwives, Anganwadi workers, 
Change agents 
Sept. 
2004 – 
April 
2005 
March-May 
2005 
174 724 
Endline Household Listing All households in study area 29517 32319 
 Pregnancy History All ever-married women aged 13-49 Lifetime 27112 29263 
 Recently Delivered 
Women survey 
Women who had a live birth or still birth 7525 9792 
 Service Providers Auxiliary Nurse Midwives, Anganwadi workers, 
Change agents 
July 
2003- 
Feb. 
2006 
Jan – 
March 
2006 
182 727 
28
Sample coverage, Endline Survey 
The total number of AWCs and the target respondents listed in the study population and covered 
by the surveys are presented in Table 3. 
Table 3. Endline Survey Coverage* 
COMPARISON INTERVENTION TOTAL 
Sector 
Population 
Coverage
(%) 
Sector 
Population 
Coverage
(%) 
Sector 
Population 
Coverage 
(%) 
Sample 
Households 29,517 28,696 
(97.2%) 32,319 30,582 
(94.6%) 61836 59,278 
(95.9%) 
Ever-married 
women aged 
13-49 
27,112 25379 
(93.6 %) 29,263 26,808 
(91.6) 56,375 52,187 
(92.6%) 
Recently 
delivered 
women 
7,525 7,504 
(99.7%) 9,792 9,744 
(99.5%) 17,317 17,248 
(99.6%) 
* Based on final data including revisits 
Data quality assurance 
To ensure data quality, a hierarchy of continuous supervision and scrutiny was created by the 
survey agency, following the ISO 9000 procedures for data quality. There were 3 tiers of 
supervision and scrutiny as follows. 
First tier: There was a supervisor and an editor with every team of 4 interviewers. They 
instantly checked 33% of the questionnaires in the field. The Questionnaires in which 
inconsistencies were found were returned to the interviewers on the same day, whorevisited the 
households to rectify the inconsistencies. Finally, the data was cross-checked. 
Second tier: Each district had a district coordinator to monitor and supervise the progress of 
fieldwork by spot-checking the data quality. District coordinators checked selected items from 5-
10% of the questionnaires. 
Third tier: Two senior researchers from the survey agency were involved in overseeing the 
progress and quality of work in both districts. All discrepancies were addressed individually, and 
senior researchers took corrective measures after consulting with the JHSPH coordinator. 
Review and feedback meetings were held regularly with the field teams involving interviewers, 
supervisors, researchers, field managers and the JHSPH coordinator. The overall field operations 
and data quality were supervised by the JHSPH coordinator and JHSPH investigators. 
Independent Data Quality Assurance by JHSPH Team 
Data quality was a paramount requirement of this study. A data quality assurance (DQA) 
procedure was set up for concurrent validation throughout the data collection period. Commonly, 
the data quality in large studies is likely to be affected by sampling bias, non-response patterns, 
inconsistencies across the instruments used, field investigators’ lack of knowledge about key 
29
variables and the consequent effect on data, and field investigators’ deliberate attempts to 
misrepresent information. 
To address these issues, a sub-sample was selected from the main study, using a stratified 
sampling approach so that the sub-sample included cases for various categories of respondents, 
each type of instrument, and each field investigator. This method was implemented to locate and 
interview respondents for data validation, consistent with the data collection schedule of the 
main survey team and their data collection protocol. These households were re-interviewed by 
independent teams that had no access to the previous interview data. The main survey and DQA 
interviewers were trained together using the same modules. Some key variables from the main 
and DQA data were compared every week to identify the levels and patterns of disagreements. 
Furthermore, the distribution of these key variables was generated from the main data set every 
week. Unusual response patterns in some variables and inconsistent or conflicting responses 
were analyzed using a pre-designed algorithm for each interviewer in the main survey team. The 
findings were presented periodically before the main survey team, and appropriate corrective 
measures were taken. This process continued throughout the survey period. Cases of gross 
discrepancy were identified as examples of case studies and jointly visited by the leaders of the 
main survey and DQA teams. Results of such cases were shared with all the interviewers. 
Apart from the independent data collection, the DQA teams were also involved in spot-checking 
and back-checking of the TNS teams. When the TNS teams were in the field, the DQA team 
would observe their work and note any discrepancies and discuss them in the field on the same 
day, to avert future errors. The DQA would also come to the field unannounced and back-check 
some key variables in completed questionnaires, in order to maintain the quality of work. 
Key Variables 
Key variables included in the DQA process were the number of eligible women and RDW 
recorded in each household and that number’s concordance with the number of pregnancy 
history and RDW forms attached to the household form. Other key variables that were checked 
included: family size, number of surviving children under 3 years old, age of any woman near 
the beginning or end of the reproductive period, deaths at critical ages to distinguish between 
neonatal, infant, and child mortalities, birth year of the youngest child, registration with ANM 
for antenatal check-up, place of delivery, person conducting delivery, person who cut the 
umbilical cord, duration of breastfeeding, visits by various service providers after delivery, 
family’s main source of drinking water, main household source of light, type of fuel for cooking, 
and area of agricultural land owned. 
Data management 
For the baseline survey, completed questionnaires were sent to the data processing division at the 
survey agency’s headquarters in Delhi. For the endline survey, the data management was 
conducted in Lucknow. Data management consisted of office editing, coding, data entry, cross￾validation, and consistency checks. To ensure accuracy, the data were entered twice by 
independent operators, and then matching was done. Software for data entry, data validity and 
data processing was developed using Visual Fox-PRO, SPSS, C++, Fortran 77 and EXCEL. 
30
Statistical methods 
Data from completed interviews were used for computing descriptive statistics at the household 
level. Indicators of coverage with home visits by community-based workers, practices and care￾seeking for newborn conditions and delivery practices were based on information collected from 
RDWs. The computation of mortality levels utilized the life table approach. A summary measure 
for standard of living index (SLI) was calculated by the method used in NFHS-II (5). In this 
method, points are assigned based on the family’s house type, toilet facility, source of lighting, 
main cooking fuel, source of drinking water and possession of certain durable goods, such as car, 
tractor, bicycle, electric fan, television, sewing machine, etc. Points are added if the families 
have a separate room for cooking, own their home, own their agricultural land, own irrigated 
land or own livestock. Index scores are used to classify the family as having low, medium or 
high standard of living. In this report, denominators in various levels may not be the same 
throughout the analysis because all the questions were not applicable to all the respondents. 
Service Providers’ Knowledge 
Chapter 4 presents the proportion of correct responses for each of angwanwadi workers and 
auxiliary nurse midwives. In all of the project surveys, service providers were given a brief quiz 
to assess their knowledge related to maternal and newborn health. Most of the questions were 
unprompted (see questionnaires in the appendix for details), and respondents were given one 
point for each possible correct answer. For example, one question asked, “What advance birth 
planning must a family do to cope with any emergency during delivery?” Possible correct 
answers included: identify a clean place for delivery, identify a trained person to conduct the 
delivery, identify a hospital/clinic, identify a transport and save some money for emergiencies, 
arrange for new blade, soap and thread, or arrange for a disposable delivery kit (DDK). 
Therefore, a total of seven points could be received for this question. Percentages were then 
calculated based on the total number of points possible. Some indicators are also presented based 
on knowledge of specific topics. These indicators are described in Chapter 4. 
Changes in Process Indicators and mortality between Study Arms 
This report presents an intent-to-treat analysis. The analysis compares the rates of outcome 
indicators (e.g., behavior change, receipt of services, neonatal mortality) between the 
intervention and comparison areas between two time periods, using a difference of differences 
test. These indicators measure program coverage, quality of program implementation, and 
program effects on maternal and provider knowledge and behavior, in addition to effects on 
mortality. The baseline survey measured mortality during the 2 years preceding the survey, and 
the endline survey measured mortality during the 2 years of the intervention. Though reduction 
in mortality is the desired outcome, quality process data can help to distinguish elements of the 
program that were successful or not successful and provide insight for future modifications. 
31
CHAPTER 4. SERVICE PROVIDERS’ TRAINING, 
KNOWLEDGE AND FACILITIES 
Training 
 The project aimed to train all the community-based service providers during the course of 
the intervention. Figure 3 shows the proportion of auxiliary nurse midwives, anganwadi workers 
and change agents that reported participating in a CARE training during the intervention. More 
than 90% of AWWs and CAs participated in the training, while 43% of ANMs received training. 
By topic areas that are particularly relevant to this project – pregnancy care, delivery care and 
newborn care - the trends were similar (Figure 4). About 80% of AWWs participated in training 
for those topics, 80-90% of change agents and 43% of ANMs. Therefore, the high coverage of 
training for AWWs and change agents can be cited as one of the accomplishments of this project. 
The failure of more than half of ANMs to participate in the training should be investigated so 
that future trainings can improve their level of participation. 
Figure 3. Coverage of CARE Training at Endline in Intervention District 
43%
91%
98%
0%
20%
40%
60%
80%
100%
ANM AWW CA
Type of Service Provider
Proportion
32
Figure 4. Training Coverage by Topic 
43%
86%
92%
43%
76%
84%
43%
84%
80%
0%
20%
40%
60%
80%
100%
ANM AWW CA
PregnancyCare Delivery Care Newborn Care
Auxiliary Nurse Midwives 
 Data from the ANMs shows some improvement from baseline but also highlights some 
enduring problems with the government health system. The auxiliary nurse midwives are the 
community-based workers of the ministry of health. They are charged with providing 
contraceptive counseling and services, birth attendance, immunization and some basic curative 
health services. At endline in both districts, the ANMs’ average age was 44 in both areas, and 
they reported an average of 12 years education. Table 4 shows that ANMs in both districts 
showed improved knowledge scores from baseline to endline, but ANMs in the intervention area 
showed a greater degree of improvement. Nonetheless, no ANMs received more than 75% of the 
possible points. With regard to topic-specific indicators, ANMs in the intervention area improved 
in all indicators except knowledge of maternal post-partum danger signs, while ANMs in the 
comparison area improved only in knowledge of the five cleans of delivery and newborn danger 
signs. However, some knowledge levels, such as pregnancy and delivery dangers signs, are 
unexpectedly poor given that the ANMs are supposed to be educated as skilled birth attendants. 
Only 27% of ANMs in the intervention area were able to list five or more delivery complications 
requiring immediate referral. These findings suggest that some progress was made through this 
intervention, but there remains an urgent need to update the ANMs’ knowledge and skills. 
33
Figure 5. Changes in Proportion of Correct Responses given by ANMs’ from Baseline 
to Endline 
Comparison
13.3
83.3
3.3 0 0
73.8
26.2
0
0
20
40
60
80
<25% 25-50% 50-75% >75%
Proportion of Correct Responses
Proprotion of ANMs
Baseline Endline
Intervention
47.5 52.5
0 00
56.8
43.2
0
0
20
40
60
80
<25% 25-50% 50-75% >75%
Proportion of Correct Responses
Proporiton of ANMs
Baseline Endline
34
Table 4. Changes in Auxiliary Nurse Midwives’ Knowledge 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Received CARE 
Training 
 
Proportion of Correct Responses: 
<25% 13.3 0 -13.3 47.5 0 -47.5 
25-50% 83.3 73.8 -9.5 52.5 56.8 4.3 
50-75% 3.3 26.2 22.9 0 43.2 43.2 
>75% 0 0 0 0 0 0 
 
Topic-specific Indicators: 
Pregnancy Danger 
Signs1
30.0 23.8 -6.2 15.0 43.2 28.2 
Delivery Danger 
Signs2
36.7 7.1 -29.6 10.0 27.3 17.3 
5 Cleans3
 30.0 78.6 48.6 32.5 86.4 53.9 
Maternal Post￾partum Danger 
Signs4
10.0 0.0 -10 5.0 2.3 -2.7 
Newborn Danger 
Signs5
26.7 54.2 27.5 7.5 50.0 42.5 
Exclusive 
breastfeeding6
80.0 80.9 0.9 82.5 95.4 12.9 
 
Total Number 30 42 40 44 
1. Able to list 5 or more pregnancy complications (including “bad obstetric history,” severe headache, 
blurry vision, reduce or absent fetal movement, high blood pressure, edema of the face or hands, fever, 
unconsciousness, convulsions, excessive vaginal bleeding, lower abdominal pain, meconium and severe 
anemia. ) 
2. Able to list 5 or more delivery complications (including excessive vaginal bleeding, pre-term labor, high 
fever, breech presentation, “baby in bad position”, prolonged labor , meconium, retained placenta, torn 
uterus, prolapsed cord, cord around the neck, convulsions and premature rupture of the membrane.) 
3. Able to list the 5 cleans of delivery: clean surface, clean hands, clean blade, clean thread, clean cord. 
4. Able to list 5 or more postpartum complications (including excessive vaginal bleeding, foul-smelling 
discharge, high fever, inverted nipples, tetanus, retained placenta, severe abdominal pain, convulsions, 
engorged breasts and jaundice.) 
5. Able to list 5 or more newborn complications (including poor feeding, infant diarrhea, 
redness/discharge from around the cord, red/discharging eyes, persistent vomiting, newborn is cold to the 
touch, pneumonia, jaundice, convulsions, skin lesion/pustules, baby won’t cry, fever, unconsciousness, 
difficult/fast breathing, chest in-drawing, failure to pass urine, and failure to pass stool.) 
6. Knows that exclusive breastfeeding should continue for six months. 
Anganwadi Workers 
 Anganwadi workers are the community-based workers of the Integrated Child 
Development Services. Anganwadi workers’ demographic characteristics in the two study 
districts were comparable at the end of the project. Anganwadi workers had a mean age of 36 in 
both districts, and an average of 12 years of education. They reported an average of 8 years of 
35
service in the comparison (range: 1-23) and an average of 9 years of service in the comparison 
area (range: 2-28). 
Indicators related to anganwadi workers’ knowledge and to the activities, supplies and 
equipment at the anganwadi centers showed improvement from baseline to endline in the 
intervention area. Figure 6 shows that the anganwadi workers’ knowledge scores increased from 
baseline to endline. At baseline, 28% of the AWWs received a score of 50% or higher, compared 
with 38% at endline. Table 5 shows that anganwadi workers in the comparison area made little 
improvement in overall knowledge. Comparing the topic-specific indicators, some gains were 
measured in both districts but that the gains in the intervention district were generally more 
pronounced. For example, knowledge of the “5 cleans” of childbirth improved from 3% to 24% 
in the comparison area, but in the intervention area this indicator improved from 18% to 75%. 
Similarly, the proportion of AWWs able to list 3 or more maternal post-partum danger signs 
increased from 0% to 26% in the comparison area and from 0% to 58% in the intervention area. 
However, the proportion of AWWs that were able to list three or more newborn complications 
rose from 7% at baseline to 81% at endline in the comparison area, while gaining from 6% to 
82% in the intervention area. In other words, knowledge of newborn complications seemed to 
increase in a comparable fashion in both areas. This suggests that some capacity building took 
place in the comparison area during the program’s implementation. To provide some context for 
the health services in the communities, a few indicators of supplies and equipment at anganwadi 
centers are shown in Table 6. Some of the project’s best practices were to promote the formation 
of mahila mandal groups, to recruit and train change agents, to produce social maps with health 
indicators, to teach women to conduct individual monitoring and to hold nutritional health days 
monthly. These best practices were adopted by almost all anganwadi centers in the intervention 
area. Another project goal was to distribute iron-folate supplements and disposable delivery kits 
through anganwadi centers. In the intervention area, 72% of anganwadi centers were stocked 
with iron-folate supplements and 40% with disposable delivery kits. 
36
Figure 6. Changes in the Proportion of Correct Responses given by Anganwadi 
Workers from Baseline to Endline 
Comparison
6.9
82.6
9 7
0.7 3.6
90
6.4 0
0
20
40
60
80
100
<25% 25-50% 50-75% >75%
Proportion of Correct Responses
Proportion of AWWs
Baseline Endline
Intervention
9.1
63
27.9
0 0
61.7
35.8
2.5
0
20
40
60
80
100
<25% 25-50% 50-75% >75%
Proportion of Correct Responses
Proportion of AWWs
Baseline Endline
37
Table 5. Anganwadi Workers’ Knowledge 
 Comparison Intervention 
 Baseline Endline Change Baseline Endline Change 
 
Proportion of Correct Responses: 
<25% 6.9 3.6 -3.3 9.1 0.0 9.1 
25-50% 82.6 90.0 7.4 63.0 61.7 -1.3 
50-75% 9.7 6.4 -3.3 27.9 35.8 7.9 
>75% 0.7 0 -0.7 0 2.5 2.5 
 
Topic-specific Indicators: 
Pregnancy 
Danger Signs1
6.9 19.3 12.4 21.2 46.9 25.7 
Delivery Danger 
Signs2
3.5 56.4 52.9 6.7 73.5 66.8 
5 Cleans3
 2.8 24.3 21.5 17.6 75.3 57.7 
Maternal Post￾partum Danger 
Signs4
0 26.4 26.4 0 58.0 58.0 
Newborn Danger 
Signs5
6.9 80.7 73.8 6.1 82.1 76 
Exclusive 
breastfeeding6
87.5 91.4 3.9 84.8 98.8 14.0 
 
Total Number 144 140 165 162 
1. Able to list 5 or more pregnancy complications. (At baseline, symptoms included “bad obstetric history,” 
severe headache, blurry vision, reduce or absent fetal movement, high blood pressure, edema of the face 
or hands, unconsciousness, convulsions, excessive vaginal bleeding, lower abdominal pain, meconium, 
severe anemia and short stature. At endline, symptoms included all of the previous except short stature, 
plus fever and jaundice.) 
2. Able to list 3 or more delivery complications. (At baseline, correct responses included excessive vaginal 
bleeding, preterm labor, prolonged labor, and convulsions. At endline, correct responses included all of 
these plus high fever, “baby in abnormal position”, retained placenta and premature rupture of the 
membrane.) 
3. Able to list the 5 cleans of delivery: clean surface, clean hands, clean blade, clean thread, clean cord. 
4. Able to list 3 or more postpartum complications. (At baseline, correct responses included excessive 
vaginal bleeding, foul-smelling discharge and high fever. At endline, correct responses included all of 
these, plus severe abdominal pain, convulsions, and jaundice.) 
5. Able to list 3 or more newborn complications At baseline, correct responses included poor feeding (or 
unable to suckle), drowsiness or unconsciousness, newborn is cold to the touch, fast breathing and chest 
in-drawing. At endline, correct responses included all of these plus diarrhea, high fever, 
redness/discharge from around the cord, red/discharging eyes, persistent vomiting, jaundice, baby won’t 
cry, excessive crying, difficult/fast breathing, failure to pass urine, and failure to pass stool. 
6. Knows that exclusive breastfeeding should continue for six months. 
38
Table 6. Activities, Supplies and Equipment at Anganwadi Centers 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Activities 
AWC area with 
Mahila Mandal 
79.9 78.6 -1.3 89.1 95.1 6 
Change agents with 
assigned 
households 
0 2.1 2.1 7.3 93.2 85.9 
Social 
map/community 
based monitoring 
system 
2.8 18.6 15.8 15.8 91.4 75.6 
 
Supplies and 
Equipment 
 
IFA Small or Large 
Tablets 
20.8 30.7 9.9 47.9 71.6 23.7 
Disposable Delivery 
Kits 
4.9 10.7 5.8 13.3 39.5 26.2 
ORS packets 13.9 43.2 29.3 23.6 19.3 -4.3 
 
Total Number 144 140 165 162 
Change Agents 
 The change agents are a group of volunteers recruited and trained for this project. A small 
number of them were recruited and trained in the demonstration sites prior to the launch of this 
phase of the intervention. Most of them had little experience in health promotion before joining 
the project. Table7 presents an overview of their characteristics. Their median age was 39, and 
they had a median of 2 years of education. In fact 67% had no formal education. They were 
recruited at the start of the project, and thus reported a mean of 28 months of service as change 
agents. One of the program’s goals was to increase the INHP coverage of households that had 
previously been left out. As part of this aim, change agents were largely recruited from the 
scheduled castes (43%) and other backward classes (43%). Change agents reported being 
responsible for 34 households on average, and 41% had more than 35 households in their area of 
responsibility. 
An effort was made to contact change agents who had dropped out after the start of the 
intervention. To assess the change agents experience in the project, they were asked about the 
supportiveness of the community, the anganwadi workers and the auxiliary nurse midwives. 
Nearly all (94-97%) of change agents reported that these groups were supportive or very 
supportive. Table 8 shows changes in CAs knowledge scores for topic-specific indicators over 
the course of the intervention. 
39
Table 7. Characteristics of Change Agents at Endline 
 
Age, proportional distribution of 
<20 2.3 
20-29 16.8 
30-39 33.2 
40-49 26.7 
>50 20.9 
Mean Age 39 
Years of education, proportional distribution of 
None 66.8 
1-8 23.3 
>8 9.9 
Mean years of education 2.4 
Mean months of service as change agent 30.7 
Number of households in coverage area, proportional 
distribution 
≤15 9.1 
16-25 24.1 
26-35 25.1 
>35 41.7 
Mean number of households 33.7 
Sex 
Female 90.7 
Male 9.3 
Marital Status (% Distribution) 
Never married 3.7 
Married 83.4 
Divorced or separated 0.4 
Widowed 12.4 
Occupation (%)* 
Household work 81.4 
Agriculture 65.2 
Service/job 11.1 
Artisan or skilled worker 11.7 
Shop or business 8.1 
Traditional birth attendant 10.7 
Other 22.7 
Caste (% Distribution) 
Scheduled castes or tribes 43.5 
Other backward class 42.7 
Others 13.8 
 
Total Number 506 
* More than one response possible. 
40
Table 8. Change Agents’ Knowledge Scores during the Intervention 
Adequacy I Adequacy 
II 
Adequacy 
III 
Endline 
Topic-specific Indicators: 
Iron-folate supplements1
 66.9 84.2 71.3 81.6 
Complications of childbirth2
 32.3 52.0 39.4 50.7 
Thermal care practices3
 8.6 7.7 7.9 12.9 
Newborn infection prevention4
 3.1 6.8 2.1 10.9 
Early breastfeeding5 75.2 81.0 84.3 83.9 
 
Total Number 520 621 530 506 
1. Able to identify that iron tablets are taken to prevent anemia and that 90+ should be taken during 
pregnancy. 
2. Able to list at least two of the following childbirth complications: excessive bleeding, convulsions and 
prolonged or obstructed labor. 
3. Able to state that the newborn should be wiped with a dry cloth, kept in skin-to-skin contact with the 
mother, wrapped warmly at all times and bathed only three or more days after birth. 
4. Able to list at least three of the following steps to reduce newborns’ risk of infection: minimum 
handling, handwashing before handling the child, breastfeeding immediately after delivery nad 
avoiding prelacteal feeds and supplements. 
5. Able to state that mother should begin breastfeeding within one hour of delivery and to list at least 
one of the three following benefits of early breastfeeding: to protect child from infection, to help 
mother minimize feeding difficulties, and to minimize postpartum bleeding.
41
CHAPTER 5. HOUSEHOLD CHARACTERISTICS 
Table 9 presents the characteristics of all households included in the baseline and endline 
surveys by study districts. It demonstrates that the study districts were comparable at both the 
beginning and the end of the projects on several indicators of household characteristics and 
composition. The majority of houses, about 51% in the comparison district and 61% in the 
intervention district, were semi-pucca (made from a combination of more permanent and less 
permanent materials), and a small proportion had pucca (made from permanent materials). In 
both districts the mean household size was about 5.5, the majority of households where headed 
by males, the mean age of the head of household was low- to mid-40s, and there was an average 
of about 1 eligible woman per household. Table10 shows the age distribution and sex ratio of 
the study population, which is also comparable across study groups at baseline and endline. 
Educational attainment of the study population is presented in Table 11. Distribution is fairly 
comparable, but the intervention district has slightly higher illiteracy rates both at baseline and 
endline. 
Table 12 presents socioeconomic characteristics of the recently delivered women (RDW). 
These samples were also similar between study districts both at baseline and at endline. At 
endline 65% of RDW in the comparison district and 59% of RDW in the intervention district 
were between the ages of 20 to 29. The majority of RDW in both districts were illiterate, though 
this percentage was slightly higher in the intervention district. Most women were Hindu, were in 
either scheduled or other backwards castes, and had a low household standard of living. Future 
analyses will examine how program implementation, process indicators and neonatal mortality 
varied according to these characteristics. 
42
Table 9. Percentage distribution of selected household characteristics 
COMPARISON INTERVENTION
Baseline Endline Baseline Endline
House type* 
 Kachha 40.7 37.4 38.4 33.1
 Semi-Pucca 49.5 50.9 56.9 60.6
 Pucca 9.7 11.7 4.7 6.2
Number of members in HH 
 1 4.0 4.8 3.6 4.3
 2 8.9 9.7 8.8 9.5
 3 10.4 11.1 10.6 10.6
 4 14.9 15.2 14.0 14.0
 5 17.4 18.3 15.8 16.2
 6 15.8 15.7 15.4 15.9
 7 11.9 11.0 11.5 11.2
 8 7.0 6.2 7.7 7.4
 9+ 9.7 7.9 12.7 11.0
 
Mean HH size 5.40 5.16 5.63 5.43
Sex of head of household 
 Male 88.8 87.1 94.1 93.0
 Female 11.2 12.9 5.9 7.0
Age distribution of head of household
 <30 11.3 10.4 15.8 12.9
 30-44 37.2 37.4 40.2 39.4
 45-59 26.2 26.7 23.5 24.3
 60+ 25.3 25.4 20.5 23.4
 
Median age of head of HH 45 45 40 42
Eligible women in HH
 0 17.2 18.5 14.7 18.7
 1 70.4 70.6 70.1 69.5
 2 10.0 9.0 11.8 9.6
 3+ 2.4 1.8 3.4 2.2
 
Mean no. eligible women 0.98 1.16 1.08 1.18
Number of HH 26,595 28,696 29,560 30,582
*Kachha: less permanent dwelling with straw, mud/clay or tin roof and straw or mud/clay walls and 
mud/clay floor; Pucca: more permanent dwelling with cement/RCC roof and brick or cement/RCC walls 
and Cement/RCC, Tiles or Granite floor; Semi-pucca: mix of more permanent and less permanent with 
another combination of type of material used. 
43
Table 10. Age distribution of the study population 
 COMPARISON INTERVENTION 
 Baseline Endline Baseline Endline 
Age 
 <1 2.3 2.1 2.9 2.4 
 1-4 9.9 9.3 11.6 10.5 
 5-9 13.7 13.1 15.2 14.8 
 10-14 13.2 13.7 12.4. 13.3 
 15-19 10.5 11.0 9.7 10.3 
 20-24 7.9 7.8 8.1 7.6 
 25-29 7.3 6.8 7.6 7.1 
 30-34 6.5 6.5 6.7 6.6 
 35-39 5.9 6.2 5.8 6.2 
 40-44 4.6 4.9 4.4 4.5 
 45-49 4.0 4.4 3.7 3.7 
 50-54 2.8 2.6 2.4 2.4 
 55-59 2.5 2.5 2.1 2.2 
 60-64 3.3 3.2 2.8 3.2 
 65-69 2.1 2.3 1.7 2.1 
 70-74 1.9 1.8 1.6 1.6 
 75-79 0.7 0.8 0.5 0.6 
 80+ 1.0 0.9 1.0 0.9 
Sex 
 Male 51.9 51.4 52.3 52.3 
 Female 48.1 48.6 47.7 47.7 
 
Sex ratio 928 946 913 914 
Number of Person 143,539 147,978 166,517 166,275 
Table 11. Educational status of the study population 
COMPARISON INTERVENTION 
Baseline Endline Baseline Endline 
Education Level 
 Illiterate 37.5 34.5 51.0 46.8 
 Primary School < Middle School 35.1 35.4 30.7 32.8 
 Middle School 15.5 16.2 10.4 11.3 
 High School + 11.9 14.0 7.9 9.1 
 
Number of Person 121,714 126,916 136,485 138,947 
Note: Educational status is computed for the population of age > 5 years
44
Table 12. Characteristics of Recently Delivered Women 
COMPARISON INTERVENTION 
Baseline Endline Baseline Endline 
Mean Age at Most Recent Birth 24.8 25.4 24.9 25.6 
Age at Most Recent Birth 
 12-15 1.5 0.2 3.7 1.0 
 16-19 18.7 13.7 18.9 16.0 
 20-29 59.4 65.1 53.7 59.0 
 30-34 13.4 13.8 14.8 14.6 
 35-39 5.5 5.5 6.6 6.7 
 40-49 1.5 1.6 2.2 2.7 
 
Education Level 
 Illiterate 62.9 59.4 76.7 74.8 
 Primary School < Middle School 15.9 14.9 13.1 12.6 
 Middle School 12.7 15.0 6.5 7.8 
 High School + 8.4 10.7 3.7 4.8 
 
Parity 
 1 18.2 18.3 16.4 17.3 
 2-3 37.0 36.7 34.5 35.5 
 4-5 25.1 26.0 26.5 25.7 
 6+ 19.7 18.9 22.5 21.5 
 
Religion 
 Hindu 94.3 94.3 81.2 80.5 
 Muslim 5.6 5.5 18.7 19.0 
 Other 0.1 0.3 0.2 0.5 
 
Caste 
 Scheduled Castes - 42.8 - 40.2 
 Scheduled Tribes - 0.4 - 0.5 
 Other backward castes - 42.5 - 48.1 
 Others - 14.4 - 11.2 
 
Household Standard of Living 
 Low 53.1 57.5 46.5 51.6 
 Middle 32.0 26.6 39.8 34.0 
 High 14.9 15.8 13.7 14.4 
 
Number of Women 6,329 7,504 9,069 9,744 
- Not available in baseline data 
45
CHAPTER 6. ANTENATAL CARE INDICATORS 
This chapter presents indicators related to the coverage and quality of care during the 
antenatal period. One goal of the INHP was to increase the proportion of women that were 
visited at home during pregnancy by an auxiliary nurse midwife, anganwadi worker or change 
agent. Through these visits, women were encouraged to seek antenatal check-ups. Behavior 
change communication in the intervention district promoted the use of tetanus toxoid 
immunization and iron-folate supplementation during pregnancy. Service providers were taught 
to introduce the concepts of birth planning and clean delivery to women and their families. Some 
aspects of birth planning include counseling women to choose a birth attendant in advance, save 
money for the delivery, obtain a disposable delivery kit and plan for transport and a person to 
take them to an emergency obstetric care facility. Another element of prenatal counseling is to 
teach women to recognize signs of serious illness (“danger signs”) in themselves or their 
newborns, in hopes that this will encourage them to seek timely care. 
Coverage of home visits by service providers 
The proportion of pregnant women who were visited during pregnancy at home increased 
substantially in the intervention area from baseline to endline (Figures 7-9; Table 13). In general, 
no change was seen in the indicators of coverage in the comparison area. The proportion of 
women that were visited by any of the three types of workers increased from 16% at baseline to 
59% at endline in the intervention area (Figure 9). The anganwadi workers showed the greatest 
improvement in coverage, from 5% of women at baseline to 48% at endline survey (Figure 8). 
The auxiliary nurse midwives showed a smaller degree of improvement, from 12% at baseline to 
36% at endline in the intervention area (Figure 7). The recruitment and training of change agents 
was scaled up during the intervention, and this group reached 29% coverage of antenatal home 
visits by the endline survey. Attendance of recently delivered women at a Nutrition Health Day 
event rose from 2% at baseline to 31% at endline in the intervention area (Figure 10). 
Figures 11 and 12 (trend line) present coverage by four time periods: January – June 2004 
(the period corresponding to the first adequacy survey), July to December 2004 (the period 
corresponding to the second adequacy survey), January – June 2005 (the period corresponding to 
the third adequacy survey) and July to December 2005 (the period from the third adequacy 
survey to the endline survey). Figure 11 compares the reported coverage of antenatal home visits 
by anganwadi workers using two data sources, the adequacy surveys and the endline survey. It 
demonstrates that women who gave birth during each of the reference periods retrospectively 
reported higher coverage of visits during the endline survey than they had during the adequacy 
surveys. For example, during the adequacy 1 survey, 21% of women reported being visited by an 
AWW during pregnancy, while at the endline survey 41% of women who gave birth during the 
adequacy 1 recall period reported being visited by an AWW. This discrepancy is more marked 
earlier in the project. No discrepancy of this sort was observed in the comparison district (data 
not shown.) Figure 12 presents the same information using data for change agents’ visitation 
during the antenatal period. It also suggests a differential in reporting that was greater during the 
earliest months of the project. This finding is unexpected and cannot be attributed to different 
samples between the adequacy and end of report surveys. In fact, all women who participated in 
46
the adequacy surveys also were asked to participate in the endline survey. Therefore, this finding 
is unexpected and suggests a reporting bias. One explanation for this discrepant finding may be 
that the mothers report at endline was based on current knowledge and not based on actual 
experience. 
Figure 7. Antenatal Visits by Auxiliary Nurse Midwives 
Figure 8. Antenatal home visits by Anganwadi Workers 
14.9 15.6 12.3
34.0
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Difference of differences p<0.00
Baseline Endline
12.0 9.8
5.3
44.7
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline
Difference of differences p<0.00
47
Figure 9. Antenatal home visits by ANMs, AWWs or Change Agents 
 
Figure 10. Nutrition Health Day Attendance 
23.2 21.1
15.8
55 3
0
10
20
30
40
50
60
70
80
90
10
0
Proportion of RDW
Comparison Intervention
Baseline End line
2.2 1.6 1.8
28.3
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline
Difference of differences p<0.00
Difference of differences p<0.00
48
Figure 11.Trends in coverage of antenatal home visits by Anganwadi Workers, 
Intervention District Only 
Figure 12.Trends in Coverage of Antenatal Home Visits by Change Agents, 
Intervention District Only 
Antenatal check-ups 
Antenatal Check-ups 
Improving antenatal care is an important intervention for reducing neonatal mortality. A 
contact with a health worker was recorded as an antenatal check-up if the woman received, at a 
21%
31% 36%
46% 56% 45%
41%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Jan-June 2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Reference Period
Proportion of RDW
Adequacy Surveys Endline Survey
11%
22% 27%
36%
29% 31%
27%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Jan-June 2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Reference Period
Proportion of RDW
Adequacy Surveys Endline Survey
49
minimum, an abdominal examination or a blood pressure check. Nearly 40% of women in the 
intervention area received one or more ANC visits at endline, more than double the proportion 
that received one or more ANC visits at baseline (18%) (Figure 13). In the comparison area, the 
proportion of women who received one or more ANC visits increased slightly from 26% to 30% 
(Figure 13). The linear probability model suggests that the improvement from baseline to endline 
in the intervention district is significantly greater than in the comparison district (p<0.0). 
However, interpreting these data in light of the adequacy survey data suggests that the endline 
coverage in the intervention area is improbably high (Figure 14). Only 13% of women in the 
intervention area received the three ANC visits recommended by the World Health Organization 
(Table 13). As a result, indicators moved in a positive direction but remain low by international 
standards, or in comparison with southern India. 
Figure 13. Received one or more antenatal check-ups 
25.7
29.2
17.8
37.1
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline
Difference of differences p<0.00
50
Figure 14. Trends in Coverage of Antenatal Check-ups 
Table 13. Number and timing of antenatal check-ups 
 Comparison Intervention 
 Baseline Endline Change Baseline Endline Change 
Number of ANC* 
 0 74.3 70.1 -4.2 82.2 63.0 -19.2
 1 9.8 10.8 1 8.3 10.8 2.5
 2 8.6 9.1 0.5 5.8 13.6 7.8
 3+ 7.2 9.1 1.9 3.8 12.6 8.8
 
Timing of first ANC* 
 1st trimester 9.2 10.4 1.2 6.5 15.8 9.3
 2nd trimester 11.9 13.7 1.8 8.1 17.0 8.9
 3rd trimester 4.6 4.8 0.2 3.2 4.1 0.9
 
Total Number 6,329 7,504 9,069 9,744 
*Difference of differences are significant at p<0.00
29%
35%
37%
28%
31%
35%
37%
0%
5%
10%
15%
20%
25%
30%
35%
40%
Jan-June 2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Reference Period
Proportion of RDW
Adequacy Surveys Endline Survey
51
Table 14. Antenatal check-ups by maternal characteristics* 
Comparison Intervention 
Baseline Endline Baseline Endline 
Maternal Age 
<20 27.7 31.2 19.3 42.2 
20-34 26.3 29.9 18.1 37.5 
35-49 14.3 22.1 12.5 31.0 
 
Birth Order 
1 40.6 44.9 28.3 48.4 
2-3 28.3 30.6 20.6 37.9 
4-5 20.6 22.5 13.3 33.4 
6+ 13.5 14.8 11.2 28.7 
 
Mother’s Education 
Illiterate 16.9 18.9 12.7 31.5 
Primary School 27.9 31.9 25.3 46.3 
Middle School 41.9 39.3 41.1 53.9 
High School and Above 63.0 61.6 57.7 70.9 
 
Religion 
Hindu 25.6 29.3 18.9 38.3 
Muslim 27.1 27.2 13.0 32.2 
Others 33.3 31.6 18.7 40.0 
 
Standard of Living Index 
Low 17.9 20.8 11.9 30.9 
Middle 26.3 30.8 17.9 37.7 
High 52.1 56.6 38.0 57.8 
 
Total Number 6,329 7,504 9,069 9,744 
*Received one or more ANC from any source. 
Quality of antenatal check-ups and counseling 
Tetanus Toxoid Immunization 
Tetanus toxoid immunization is a critical intervention to prevent neonatal tetanus. At baseline, 
about half of all women in the intervention area (47%) and just over half in the comparison area 
(58%) had been immunized during their most recent pregnancy. This proportion increased to 
69% in the intervention area while remaining nearly the same in the comparison area, at 62% 
(Figure 15). The increase in the intervention area was statistically significant by a linear 
probability test. 
52
Figure 15. Received two or more Tetanus Toxoid Immunizations 
Iron-Folate Supplementation 
Iron-folate supplementation during pregnancy is a standard intervention for both maternal 
and neonatal health. Two aspects of iron-folate supplementation (IFA) have been tracked as 
program indicators: whether women received an adequate supply of iron-folate tablets (100 
tablets or more) and whether they consumed an adequate amount of tablets for a non-anemic 
woman (at least 100 tablets). The first reflects the adequacy of the health intervention, while the 
second may affect the quality of counseling and is necessary to impact health status. Adequacy 
reports showed improvements in the proportion of RDW in the intervention area that received an 
adequate amount of IFA supplements, but less than half of those women reported consuming an 
adequate amount of iron supplements. Findings from the endline survey present a similar but 
somewhat more encouraging picture: 32% of RDW in the intervention area received an adequate 
supply of IFA, and 22% consumed an adequate amount (Figure 16). In the comparison area, 
23% of RDW received adequate IFA and 8% consumed it; the latter figure represented no 
significant change from baseline. Thus, both these indicators showed significant gain from 
baseline, but only about one in five women took sufficient IFA. 
57.9
62.5
47.3
69.8
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline 
*Difference of differences are significant at p<0.00
53
Figure 16.Iron-folic Acid Supplementation 
Table 15. Services received during antenatal check-ups 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
 
Weight measured 22.5 24.8 2.3 37.1 52.7 15.6 
Blood pressure 
checked 
30.1 32.5 2.4 31.1 30.7 -0.4 
Urine test 31.4 35.5 4.1 36.5 35.7 -0.8 
Abdomen exam 85.8 91.0 5.2 87.4 89.7 2.3 
Ultrasound 16.1 20.7 4.6 19.5 17.2 -2.3 
Blood Test 30.1 30.1 0 32.9 27.5 -5.4 
 
Total Number that 
received ANC 
1,627 2,186 1,618 3,608 
Birth Planning 
The concept of birth planning is promoted as a strategy to prevent both maternal and 
newborn mortality, by encouraging families to prepare for a clean home delivery and to take 
action if an obstetric emergency occurs. Some components of birth planning include choosing a 
birth attendant, identifying a health facility in case of emergency, planning for emergency 
transport, saving money and obtaining a disposable delivery kit. Small gains have been observed 
in the proportion of women that took specific birth planning steps in both areas. In the 
intervention area, the largest gain has been in the proportion of women that reported saving 
money, from 15% at baseline to 50% at endline (Figure 17). A substantial increase in the 
proportion of RDW who saved money for an emergency, from 12% at baseline to 30% at 
endline, was also seen in the comparison area. Plans for transport to a health facility in case of 
emergency also increased in the intervention area, from 4% at baseline to 18% at the endline. A 
16.4
6.6
24.5
8.4
13.4
5.0
38.5
20.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Difference of differences: p<0.00
Received 100+ IFA Consumed 100+ IFA
54
seven percentage point increase was measured in the intervention area in the proportion of 
woman that obtained a clean delivery kit for use during the birth. 
Figure 17.Birth Planning Steps 
Components of Antenatal Counseling 
The quality of antenatal counseling was assessed by asking women both unprompted and 
prompted questions regarding the information that was conveyed to them during pregnancy. The 
data presented in Tables 16-18 are based on the unprompted responses. In general, most women 
in both the comparison and intervention areas had information about the importance of TT 
immunization and IFA supplementation at baseline, and small, equal increases in awareness were 
seen in both the intervention and comparison areas. The data indicate that messages about 
recognition of pregnancy or delivery complications or newborn danger signs were largely not 
received by RDW in the intervention or comparison areas. There was some improvement, 
however, in the proportions of women who reported that the ANMs (Table 16) and AWWs 
(Table 17) provided information about birth planning, the five cleans and essential newborn care 
practices, including immediate and exclusive breastfeeding, immediate drying and wrapping and 
delayed bathing. In general, about 15-30% of women in the intervention area indicated that the 
ANMs and AWWs had provided this information, whereas few women (about 1-5%) in the 
comparison area had been given these messages during pregnancy. A similar pattern was seen 
for counseling by change agents (Table 18), although the proportion of women who received 
messages about birth planning and essential newborn care practices from a change agent was 
lower (10-20%) than those from ANMs and AWWs. Tables 16-18 also include a mean total 
score, which indicates the number of messages each woman remembered receiving. For 
example, a woman who only recalled being told about TT immunizations and IFA tablets would 
have a total score of 2. For ANMs, the mean number of messages recalled was 1.6 at baseline 
and 2.9 at endline in the intervention area. For AWWs and change agents, the average total 
number of messages recalled was 1 at baseline and 3 at endline. As with the percentage coverage 
2 3 0.9
12.0
5.1 3.7
29.3
3.9
0.2
14.9
18.4
6.5
49.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Arranged Transport Obtained Delivery Kit
Saved Money
55
of specific messages, it suggests a minor change in quality of counseling. However, Table 19 
shows the proportion of all women who reported receiving specific types of counseling during 
pregnancy in response to prompted questions. In both the comparison and intervention groups, 
recognition of all messages increased from baseline to endline, but the intervention group had 
more dramatic increases for all messages. The difference in quality of counseling between 
comparison and intervention areas was more pronounced when the women were promted with 
the messages. 
Improvements in the quality of antenatal counseling would be expected to lead to an 
increase in maternal knowledge. Maternal knowledge was assessed by asking women to list the 
danger signs for women during pregnancy, childbirth, post-partum and for newborns. Table 20 
shows that there was no improvements of knowledge of danger signs in either group which is 
consistent with the low rate of counseling on the study subjects. As has already been described, 
women rarely recalled being counseled about maternal and newborn dangers signs. Although 
some women could recall being counseled about danger signs when prompted, they could not 
recall these signs sponteneously. The overall knowledge scores for both groups increased 
slightly and by the same magnitude, indicating no improved knowledge in the intervention group 
over the comparison group. 
Table 16. Quality of Counseling during antenatal home visits by ANMs* 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Topics discussed during 
antenatal home visits 
 
TT Immunizations 80.5 90.7 10.2 84.6 79.8 -4.8 
IFA Tablets 66.5 80.9 14.4 68.9 79.8 10.9 
Birth Planning 2.2 6.8 65.8 2.6 12.9 10.3 
Pregnancy Complications 6.6 3.2 -3.4 5.1 2.3 -2.8 
Delivery Complications 1.4 2.1 0.7 1.0 1.2 0.2 
Newborn 
Complications/Danger 
Signs 
0 1.6 1.6 0.3 3.0 2.7 
Five Cleans 0.7 1.2 0.5 1.4 19.3 17.9 
Immediate Breastfeeding 0.6 2.4 1.8 0.7 22.4 21.7 
Exclusive Breastfeeding 0.8 6.4 5.6 0.4 30.3 29.9 
Delayed bathing of newborn 0.1 0.6 0.5 0.2 19.7 19.5 
Drying and wrapping the 
newborn 
0.3 3.1 2.8 0.1 16.8 16.7 
 
Mean Total Score Out of 
11 (Median Total Score) 
1.6 (2) 1.9 (2) 0.3 (0) 1.6 (2) 2.9 (2) 1.3 (0) 
 
Total Number of Women 
that Received Visits 
941 1,170 1,120 3,309 
* Based on unprompted responses. 
56
Table 17. Quality of Counseling during antenatal home visits by AWWs* 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Topics discussed during 
antenatal home visits 
 
TT Immunizations 41.8 65.5 23.2 54.4 70.3 15.9 
IFA Tablets 36.1 57.4 21.3 43.7 77.6 33.9 
Birth Planning 2.1 6.8 4.7 2.1 15.3 13.2 
Pregnancy Complications 5.1 3.3 -1.8 5.0 2.6 -2.4 
Delivery Complications 0.7 1.6 0.9 1.0 1.7 0.7 
Newborn Complications/Danger 
Signs 
0.0 2.0 2.0 0.2 3.5 3.3 
Five Cleans 1.1 1.1 0 1.7 21.6 19.9 
Immediate Breastfeeding 1.1 2.7 1.6 2.5 29.4 28.9 
Exclusive Breastfeeding 1.2 5.6 4.4 2.3 34.8 32.5 
Delayed bathing of newborn 0.0 0.4 0.4 0.4 24.7 24.3 
Drying and wrapping the newborn 0.4 2.2 1.8 0.6 19.4 18.8 
 
Mean Total Score Out of 11 
(Median Total Score) 
0.9 (1) 1.1 (1) 0.2 (0) 1.5 (2) 3 (3) 1.5 (1) 
 
Total Number of Women that 
Received Visits 
758 734 478 4,352 
* Based on unprompted responses. 
Table 18. Quality of Counseling during antenatal home visits by Change Agents* 
Baseline Endline Change 
Topics discussed during antenatal home visits 
TT Immunizations 48.3 62.7 14.4 
IFA Tablets 34.5 63.7 29.2 
Birth Planning 13.8 15.9 2.1 
Pregnancy Complications 3.4 2.5 -0.9 
Delivery Complications 0.0 1.4 1.4 
Newborn Complications/Danger Signs 0.0 4.5 4.5 
Five Cleans 17.2 22.4 5.2 
Immediate Breastfeeding 6.9 26.6 19.7 
Exclusive Breastfeeding 13.8 35.6 21.8 
Delayed bathing of newborn 6.9 27.6 20.7 
Drying and wrapping the newborn 6.9 23.8 16.9 
 
Mean Total Score Out of 11 (Median Total Score) 1.5 (1) 2.9 (3) 1.4 (2) 
 
Total Number of Women that Received Visits 29 2833 
* Based on unprompted responses. 
57
Table 19. Advice received during pregnancy from any source* 
 Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
TT immunization 41.7 68.6 26.9 15.6 76.2 60.6 
 IFA tablets 38.4 58.0 19.6 14.2 71.8 57.6 
Birth Planning 
 Emergency 
transport 3.8 11.5 7.7 2.5 46.4 43.9 
 Saving money 5.1 24.0 18.9 3.4 57.3 53.9 
 Suitable location for 
delivery 9.2 20.3 11.1 4.5 45.9 41.4 
 Birth attendant 6.7 18.2 11.5 3.2 39.1 35.9 
 Emergency obstetric 
care 
9.4 12.3 2.9 3.7 33.5 29.8 
Danger Signs 9.4 24.7 15.3 5.1 60.8 55.7 
Aspects of Clean 
Delivery 
 Clean Surface 9.0 28.6 19.6 4.8 65.5 60.7 
 Clean Hands 9.1 34.7 25.6 5.1 70.4 65.3 
 New Blade 5.1 70.4 65.3 6.0 73.6 67.6 
 Boiled Thread 9.0 40.8 31.8 5.0 69.8 64.8 
 Clean towel to dry 
baby 7.9 19.5 11.6 4.6 55.0 50.4 
 Breastfeeding 9.4 24.7 15.3 5.1 60.8 55.7 
 
Total Number 6,329 7,504 9,069 9,744 
* Based on prompted responses. 
Table 20. Changes in Maternal Knowledge 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Percent Recalled: 
Pregnancy Danger Signs1
 12.9 15.7 2.9 15.7 20.0 4.3 
Delivery Danger Signs2 20.0 27.5 7.5 22.5 30.0 7.5 
Postpartum Danger Signs3
 13.3 21.7 8.3 15.0 23.3 8.3 
Newborn Danger Signs4
 15.0 33.3 18.3 13.3 28.3 15.0 
 
Total Danger Sign 
Knowledge 15.2 23.9 8.7 16.1 25.2 9.1 
1. Possible responses: severe headache, high fever, edema of the face, edema of the hands, 
unconsciousness, convulsions, vaginal bleeding. 
2. Possible responses: excessive vaginal bleeding, high fever, prolonged labor and convulsions. 
3. Possible responses: excessive vaginal bleeding, foul-smelling discharge, high fever, convulsions, 
swollen feet or hands and severe headache. 
4. Possible responses: poor feeding/unable to suckle, poor activity/can’t be awakened, not crying, cold to 
the touch, breathing difficulties and chest retraction. 
58
CHAPTER 7. DELIVERY AND NEWBORN CARE 
This chapter presents indicators related to quality of care and care-seeking during childbirth 
and describes newborn care practices. The baseline verbal autopsy found that 26% of neonatal 
deaths occurred on the day of birth and that 73% occured during the first week (CIT). In 
addition, the number of stillbirths almost equals the number of neonatal deaths, and evidence 
suggests that a large proportion of these deaths may be related to intrapartum events. Therefore, 
interventions during childbirth and the first few days after birth are particularly critical to 
reducing neonatal deaths in this population. 
Skilled birth attendance and institutional delivery care 
Most women in this setting give birth at home assisted by family members or others who 
have received no formal health care training. Increasing institutional deliveries or birth 
attendance by doctors, nurses and auxiliary was not a primary goal of this program, but patterns 
in birth attendance are presented because they are related to birth outcomes. These data suggest a 
modest increasing toward more institutional births, from 11% to 15% in the comparison area and 
12% to 17% in the intervention area (Table 21). The proportion of births attended by a trained 
traditional birth attendant (TBA) increased slightly in the intervention area from 1% at baseline 
to 6% at endline, perhaps because some TBAs were trained through the CARE intervention. The 
cesarean section rate remained very low, 2-3% in both communities, suggesting poor access to 
emergency obstetric care. The prevalence of self-reported delivery complications reduced 
dramatically from baseline to endline in both districts. The level of reported complications was 
65% in both districts at baseline and about 35% at endline, suggesting that differences in survey 
methodology led to divergent reporting. Figure 18 shows that among women with reported 
complications, the proportion that received skilled birth attendance at home or in a facility 
remained unchanged. However, given the inconsistency of the levels of reported complications, 
this finding should be interpreted with caution. 
Table 21. Institutional deliveries and birth attendance 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change 
Institutional deliveries 11.3 15.2 3.9 12.4 16.3 3.9 
Home delivery by 
 Trained health 
professional * 
6.8 6.8 0 4.3 6.1 1.8 
 Trained TBA 3.7 4.6 0.9 1.0 5.4 4.4 
 Others 78.2 73.4 -4.8 82.2 72.2 -10.0 
 
Cesarean section rate 2.1 2.7 0.6 1.4 2.5 0.8 
 
Total 6,329 7,504 9,069 9,744 
* MBBS Doctor, Nurse, Lady Health Visitor or Auxiliary Nurse Midwife. 
59
Table 22. Skilled birth attendance by Maternal Characteristics* 
Comparison Intervention 
Baseline Endline Baseline Endline 
Maternal Age 
<20 21.7 24.6 20.1 29.1 
20-34 18.0 22.6 15.8 22.8 
35-49 9.3 15.4 15.0 15.5 
 
Birth Order 
1 33.0 40.5 31.3 39.2 
2-3 19.9 20.7 16.3 21.1 
4-5 12.2 14.9 12.9 16.9 
6+ 8.5 11.2 11.3 14.9 
 
Mother’s Education 
Illiterate 10.3 12.8 11.7 17.4 
Primary School 20.4 21.5 23.5 28.9 
Middle School 30.0 30.4 37.0 37.7 
High School and Above 54.5 54.6 60.9 54.5 
 
Religion 
Hindu 18.0 22.3 17.7 23.6 
Muslim 20.1 17.3 12.3 17.4 
Others 22.2 15.8 18.7 26.7 
 
Standard of Living Index 
Low 11.3 13.4 11.0 15.8 
Middle 18.4 23.7 15.3 24.0 
High 41.9 50.2 40.4 42.3 
 
Total Number 6,329 7,504 9,069 9,744 
* Institutional delivery or MBBS doctor, nurse, lady health visitor or auxiliary nurse midwife at home 
delivery. 
60
Figure 18. Proportion of RDW with self-reported delivery complications that gave 
birth with a trained birth attendant 
 
Immediate Newborn Care Practices 
Behavior change communications in the intervention area focused on promoting specific 
newborn care practices that have been shown to reduce risk of infection and hypothermia. Four 
key newborn care practices promoted through the intervention included: (a) cutting the umbilical 
cord with a blade that is new or boiled and tying it with boiled thread; (b) delaying the newborn’s 
first bath for 3 or more days after delivery; (c) drying and wrapping the newborn immediately 
after delivery; (d) initiation of breastfeeding immediately (i.e., within 1 hour) after delivery. 
At the end of the project, the proportion of home deliveries in the intervention area for 
which the combined practice of sterile cord cutting and tying was followed increased to 68% 
from 32% at baseline (Figure 19). A smaller, 5 percentage point gain (from 36% to 42%) was 
measured in the comparison area. The practice of delaying the newborn’s first bath also showed 
noticeable increase in the intervention area. The proportion of newborns bathed more than 6 
hours after birth increased from 24% at the start of the project to 60% at the end of the project, 
and the proportion bathed 3 or more days after birth also increased from 4% to 43% (Figure 20). 
No changes of practical significance were measured in the comparison area. The proportion of 
newborns that were dried and wrapped before the placenta was delivered more than doubled in 
the intervention area from 18% at baseline to 45% at endline (Figure 21). However, when these 
indicators are plotted with the interim adequacy survey findings, they become more difficult to 
interpret. For example, the indicator related to drying and wrapping the newborn showed no 
improvement, or a slight decline, over the course of the adequacy surveys (Figure 22). The 
endline coverage of 45%, which includes the women who were surveyed during the adequacy 
surveys, is difficult to reconcile with these data. Perhaps there was some difficulty with the 
survey methodology for measuring this behavior. However it is unclear why such difficulty 
would be encountered only at endline, or only during the surveys prior to the endline and only in 
13.0 14.0 12.3 12.4
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline
61
the intervention district. An alternative explanation may be that perhaps there was diffusion of 
knowledge within the study area over the period of the intervention and the RDW answered the 
question in a way that more closely reflected their current knowledge than their actual practice at 
the time they gave birth, which was one month to two years before the endline survey. 
At baseline, most women in both study districts delayed the start of breastfeeding for 
several days after birth. In the intervention area, service providers have been encouraging women 
to start breastfeeding within one hour of birth, or as soon as possible. Initiation of breastfeeding 
within 1 hour of delivery increased in the intervention area from 6% at baseline to 34% at 
endline (Figure 23). Initiation of breastfeeding during the first day of life rose from 20% to 64% 
in the intervention area. The comparison area saw no change in these indicators. Similarly, 
colostrum feeding was reported to increase from 20% to 88% in the intervention area, while little 
change (14% to 20%) was seen in the comparison area (Figure 24). The endline data for this 
indicator are more consistent with data from the adequacy surveys throughout the intervention 
(Figure 25). However, at enline 44% of women reported starting breastfeeding within one hour, 
compared to 18%, 18%, and 22% for the previous three reference periods. Some caution should 
be taken with this indicator since this is an unlikely jump and could indicate increased 
knowledge rather than actual practices. Nevertheless, both adequacy survey data and endline data 
suggest that breastfeeding practices have improved markedly in the intervention area during the 
course of the program. 
Figure 19. Sterile cutting and tying practices at home deliveries 
36.1
42.1
32.1
68.0
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Difference of differences: p<0.00
Baseline Endline
62
Figure 20.Timing of the newborn’s first bath 
Figure 21. Newborn dried and wrapped immediately 
4.2 1.5
8.5
3.4
23.6
3.6
58.5
41.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Difference of differences: p<0.00
6+ hours 3+ days
8.8
4.4
17.7
44.1
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Difference of differences: p<0.00
Baseline Endline
63
Figure 22.Trends in prompt drying and wrapping 
Figure 23.Timing of Breastfeeding Initiation 
6%
12% 13%
44% 44% 42% 46%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Jan-June 2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Reference Period
Proportion of RDW
Adequacy Surveys Endline Survey
2.4
11.7
5.6
23.5
3.0
16.9
37.1
66.9
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Difference of differences: p<0.00
Within one hr. First day
64
Figure 24. Colostrum Feeding 
Figure 25.Trend line: breastfeeding initiation within one hour 
13.9
44.3
20.1
87.6
0
20
40
60
80Proportion of RDW
Comparison Intervention
Difference of differences: p<0.00
Baseline Endline
18%
22%
14%
44%
18%
17%
21%
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
Jan-June 2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Reference Period
Proportion of RDW
Adequacy Surveys Endline Survey
65
Table 23. Breastfeeding Initiation of First Day of Life, by Maternal Characteristics 
Comparison Intervention 
Baseline Endline Baseline Endline 
Maternal Age 
<20 10.7 21.3 16.2 16.2 
20-29 12.4 24.4 17.3 17.3 
30-39 6.3 18.7 15.0 15.0 
>39 
 
Birth Order 
1 14.0 22.8 18.9 67.6 
2-3 14.1 28.7 18.5 63.9 
4-5 10.7 23.8 16.1 64.7 
6+ 6.0 23.1 13.7 67.1 
 
Mother’s Education 
Illiterate 7.0 17.6 13.3 64.8 
Primary School 13.9 23.6 21.2 72.0 
Middle School 17.4 29.2 34.3 72.8 
High School and Above 33.1 44.1 44.4 75.6 
 
Religion 
Hindu 11.8 16.1 16.1 66.9 
Muslim 9.5 20.0 20.0 66.4 
Others 11.1 18.7 18.7 77.8 
 
Standard of Living Index 
Low 7.8 18.9 13.6 65.2 
Middle 12.0 24.6 16.8 66.7 
High 24.8 38.6 28.1 73.3 
 
Total Number 6,203 7,312 8,890 9,467 
Postpartum Home Visits by Service Providers 
Postpartum home visitation was an important part of the intervention strategy, to promote 
healthy newborn care practices and care-seeking for illness. Postpartum home visits by ANMs 
(Figure 26) or AWWs (Figure 27) were rare at baseline (< 2%) and remained that way in the 
comparison area at endline. In the intervention area, the greatest success in postpartum home 
visitation within 1 week of birth was achieved by the AWWs, which reached 23% of households 
at endline (Figure 27), followed by the Change Agents (17%, Figure 28) and the ANMs (9%, 
Figure 26). Values for postpartum home visitation within 3 days were lower, 16% for AWWs, 
13% for change agents and 6% for ANMs. Postpartum visits by ANWs, AWWs, or CAs within 
one week remained low in the comparison area (3.3%, Figure 29), but increased significantly in 
the intervention area from 1.9% to 28.9%. Visits within one day and within 3 days also 
increased markedly, suggesting an overall improvement in postpartum visits by any health 
worker in the intervention area. 
66
Figure 26.Timing of Postpartum Visits by ANMs 
Figure 27.Timing of Postpartum Visits by AWWs 
0.7 1.5 2.0 0.7 1.5 1.9 0.4 0.8 1.1 2.9 6.1 9.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Difference of differences: p<0.00
Within one day Within 3 days
Within 1 week
0.5 1.2 1.6 0.8 1.4 1.8 0.3 0.7 1.0
7.5
15.0
21.7
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Difference of differences: p<0.00
Within one day Within 3 days
Within 1 week
67
Figure 28.Timing of Postpartum Visits by Change Agents 
Figure 29. Postpartum home visits by ANMs, AWWs or Change Agents 
6.3
10.0 12.6
6.8
10.7 14.1
7.2
11.8
15.1
8.9
13.7
18.2
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Jan-June2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Within one day Within 3 days
Within 1 week
6.3
10.0 12.6
6.8
10.7 14.1
7.2
11.8
15.1
8.9
13.7
18.2
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Jan-June2004 July-Dec 2004 Jan-June 2005 July-Dec 2005
Within one day Within 3 days
Within 1 week
1.1 2.5 3.3 1.4 2.7 3.3 0.7 1.4 1.9
13.7
22.4
28.9
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Within one day Within 3 days
Within 1 week
68
Figure 30. Postpartum home visits by ANMs, AWWs or Change Agents 
Care-seeking for Newborns 
Patterns of care-seeking for newborns increased during the program, but nonetheless only 
a minority of newborns received care from trained providers. Among all newborns, the 
proportion that received a check-up from a trained provider within one month of birth increased 
from 5% at baseline to 23% at endline (Figure 31). Levels of reported newborn complications2
declined in the intervention area from baseline (57%) to endline (42%). A similar, but smaller 
decline in level of reported illness were also seen in the comparison area, from 51% to 42%. 
Little is known about the reliability of retrospectively reported neonatal illness, so this change 
should be interpreted with caution. It may or may not reflect a real change in the prevalence of 
serious complications. Reported care-seeking showed some shifts in both areas (Figure 32, Table 
23). In the intervention area, an increased proportion of women reported taking sick newborns 
for care from a trained provider (allopathic, qualified doctors, lady health visitors or ANMs) at 
endline (31%) compared with baseline (20%). However, only 1 in 3 newborns received this type 
of care, and 27% received no care from outside the home. There was a slight decline in the rate 
of care-seeking from a trained provider in the comparison area. 
 
2
 Complications included fever, breastfeeding problems, breathing trouble, fast breathing, chest in-drawing, 
drowsiness, jaundice, diarrhea, persistent vomiting, felt cold, red or discharging eye, skin pustules, umbilical redness 
or discharge, unconsciousness, and convulsions. 
1.1 2.5 3.3 1.4 2.7 3.3 0.7 1.4 1.9
13.7
22.4
28.9
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline Endline Baseline Endline
Within one day Within 3 days
Within 1 week
69
Figure 31. Newborns that received a check-up from a trained provider within one month 
11.9 11.4
5.0
22.7
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Comparison Intervention
Baseline End of Project
Difference of differences: p<0.00
11.9 11.4
5.0
22.7
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Comparison Intervention
Baseline End of Project
11.9 11.4
5.0
22.7
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Comparison Intervention
Baseline End of Project
Difference of differences: p<0.00
Figure 32. Newborns with complications that received care from a trained provider 
0
10
20
30
40
50
60
70
80
90
100
 
Comparison Intervention
Baseline End of Project
Difference of differences: p<0.00
0
10
20
30
40
50
60
70
80
90
100
 
Comparison Intervention
Baseline End of Project
35.1
30.8
20.0
30.5
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
Baseline End of Project
Difference of differences: p<0.00
70
Table 24. Care-seeking for Sick Newborns 
Comparison Intervention 
Baseline Endline Change Baseline Endline Change
Sought Care From:* 
Qualified doctor, nurse, 
LHV or ANM 
35.1 30.8 -4.3 20.0 30.5 10.5 
Homeopathic or Ayurvedic 
doctor 
1.3 20.4 19.1 1.6 16.4 14.8 
Untrained doctor 17.3 19.4 2.1 38.1 20.5 -17.6 
Others 21.0 14.5 -6.5 19.0 11.3 -7.7 
No care sought outside of 
the home 
27.1 24.8 -2.2 22.6 27.3 4.7 
 
Total Numbers of 
Newborns with 
Complications 
3,142 2,858 5,087 3,747 
* Multiple responses possible. 
71
CHAPTER 8. EQUITY OF PROGRAM COVERAGE 
This chapter presents data on the equity of program coverage by geographical sectors and 
socio-cultural indicators. The program attempted to cover all geographic sectors in the 
intervention district and to improve equity of coverage regardless of socio-cultural factors by 
targeting coverage to mothers of lower economic status and caste. Indicators are presented to 
show distribution of program activities and practices at baseline and at the end of the project. 
Comparison of Geographical Equity 
Figures 31-38 show program coverage by geographical sector in the intervention district only. 
Though there is some variation among sectors, no sector stands out as having much higher or 
lower coverage for all of the indicators. The proportion of recently delivered women receiving 
visits by auxiliary nurse midwives increased significantly from baseline to endline and showed 
some variation across sectors with a high of 42% in Puredalai and a low of 27% in Banikodar 
(Figure 31). Aganwadi worker coverage increased even more dramatically from baseline to 
endline (Figure 32), and change agents achieved about 30% in all sectors (Figure 33). AWW 
coverage was highest in Banki (56%) and lowest in Dariyabad (33%), while CA coverage was 
highest in Harak (33%) and lowest in Banikodar (25%). Antenal care coverage increased in all 
sectors but was highest in Banki (51%, Figure 34) at endline while Dariyabad had the lowest 
coverage (31%). Most sectors saw slight improvements in skilled attendance at birth from 
baseline to endline, and there was less variation in coverage at endline (16%-28%, Figure 35). 
Clean cutting and tying practices and breastfeeding on the first day increased significantly in all 
sectors with relatively even distribution (Figures 36-37). The proportion of RDW receiving 
postpartum visits in the first three days went from around 1% in most sectors to between 20 and 
25% in all sectors (Figure 38). Therefore, though geographic variations were present, program 
coverage was relatively equal across sectors and improved within each sector over the course of 
the intervention. 
Figure 33. Auxiliary Nurse Midwife Visit Coverage by Sector, Intervention District 
Only 
14.5
26.5
16.0
38.1
15.4
30.2
14.2
34.5
11.6
32.5
15.6
42.4
16.7
40.3
8 9
33.7
4.1
28.8
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
14.5
26.5
16.0
38.1
15.4
30.2
14.2
34.5
11.6
32.5
15.6
42.4
16.7
40.3
8 9
33.7
4.1
28.8
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
72
Figure 34. Anganwadi Worker Visit Coverage by Sector, Intervention District Only 
5.0
35.6
12.3
56.0
5 2
33.3
5.8
41.6
4.5
50.8
2.6
49.1
7.9
48.4
3.3
37.6
3.6
48.6
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
5.0
35.6
12.3
56.0
5 2
33.3
5.8
41.6
4.5
50.8
2.6
49.1
7.9
48.4
3.3
37.6
3.6
48.6
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
Figure 35. Change Agent Visit Coverage by Sector, Intervention District Only 
1.1
24.8
0.1
29.5
1.5
27.2
0.1
29.5
0.0
31.2
0.0
33.4
0 3
28.1
0.2
26.5
0.0
30.7
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
1.1
24.8
0.1
29.5
1.5
27.2
0.1
29.5
0.0
31.2
0.0
33.4
0 3
28.1
0.2
26.5
0.0
30.7
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
73
Figure 36. Antenatal Care Coverage by Sector, Intervention District Only 
Figure 37. Skilled Birth Attendance at Home or Facility by Sector, Intervention District 
Only
20
27
23 23
19
23
17
24
15
21
9
16
23
28
13
18
15
21
10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
20
27
23 23
19
23
17
24
15
21
9
16
23
28
13
18
15
21
10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
20.1
36.4
30.7
51.1
18.4
30.9
20.7
36.9
15.1
34.0
17.6
39.7
17.4
40.2
16.8
32.3
11 3
34.1
0 
10 
20 
30 
40 
50 
60 
70 
80 
90 
100 
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
Proportion of RDW
74
Figure 38. Clean cutting and tying practices by Sector, Intervention District Only 
39.2
72.8
31.2
72.1
37.4
68.3
29.5
62.9
37.4
72.3
30.5
63.2
29.0
68.2
32.2
70.9
27.1
65.0
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
39.2
72.8
31.2
72.1
37.4
68.3
29.5
62.9
37.4
72.3
30.5
63.2
29.0
68.2
32.2
70.9
27.1
65.0
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
Figure 39.Breastfeeding Initiation on Day of Birth by Sector, Intervention District Only 
11.0
59.0
30.5
73.2
17.1
66.4
17.1
65.4
16.3
64 0
13.8
66.4
20.9
72.0
14.8
62.6
14.9
71.2
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
11.0
59.0
30.5
73.2
17.1
66.4
17.1
65.4
16.3
64 0
13.8
66.4
20.9
72.0
14.8
62.6
14.9
71.2
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
75
Figure 40. Received postpartum visit within three days by Sector, Intervention District 
Only 
1.5
21.7
1.5
23.0
2.5
23.3
1.0
23.2
0.8
21.2
1.0
20.8
3.2
25.3
0.9
19.6
0.6
23.0
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
1.5
21.7
1.5
23.0
2.5
23.3
1.0
23.2
0.8
21.2
1.0
20.8
3.2
25.3
0.9
19.6
0.6
23.0
0 10 20 30 40 50 60 70 80 90 100
Proportion of RDW
Banikodar Banki Dariyabad Fatehpur Harakh Puredalai Ramnagar Siddhaur Nindura
Baseline Endline
Comparison of Socio-cultural Equity 
Program coverage was examined using two socio-cultural indicators: standard of living 
index (SLI) and caste group (scheduled castes/tribes, other backward classes, and other classes). 
The SLI is a summary measure that was calculated by the same method used in NFHS-II, in 
which points are assigned based on the family’s house type, toilet facitliy, source of lighting, 
main cooking fuel, source of drinking wate and possession of certain durable goods such as car, 
tractor, bicycle, electric fan, television, sewing machine, etc. This measure, along with caste 
classifications which are designated by the Government of India, provide economic and socio￾cultural markers of the study population. Program indicators are presented here for baseline and 
endline to show equity of coverage by socio-economic groups. 
Program Coverage by Standard of Living Index (SLI) in the Intervention District 
Indicators of program coverage in the intervention district are presented by SLI 
categories: low, medium, or high. Coverage by SLI appears more equitable compared to 
geographical distribution. The proportion of RDW receiving visits from ANMs increased in all 
categories by about 20%, with an endline range of 32-40% (Figure 39). AWW visit coverage 
increased even more dramatically by about 40% in each SLI category, with a rage of 44-48% at 
end of program (Figure 40). Visits by CAs were also equitable at about 30% of RDW in all 
groups at endline (Figure 41). Figure 42 shows that antenatal care coverage had a significant 
76
difference between SLI categories at both baseline and endline (Low: 31%, High: 58%), but 
coverage in all groups increased. Skilled attendance at birth showed little to no improvement 
from baseline and also had unequal distribution by SLI group (Low: 16%, High: 42%, Figure 
43). Clean cutting and tying practices and initiation of breastfeeding on the first day showed 
marked increases and were equitable across groups, ranging only a few percentage points 
(Figures 44-45). Postpartm visitation within three days increase by about 20% and was almost 
exactly equal in all groups at endline (Figure 46). Home visitation and preventive behaviors 
improved over the course of the intervention and appear equitable among SLI groups, while 
care-seeking for antenatal check-ups and skilled attendance remained unequal by group. 
Figure 41. Auxiliary Nurse Midwife Visit Coverage by SLI 
10.1
32.1
13.1
34.3
17.8
39.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
77
Figure 42. Anganwadi Worker Visit Coverage by SLI 
Figure 43. Change Agent Visit Coverage by SLI 
29.0 28.7 30.4
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium
high
4.6
43.6
5.2
44.8
7.7
48.2
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
78
Figure 44. Coverage of one of more Antenatal Check-ups 
Figure 45.Trained Birth Attendance (Home or Facility) by SLI 
11.0
15.8 15 3
24.0
40.4 42.3
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
11.9
30.9
17 9
37.7 38.0
57.8
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
79
Figure 46. Clean cutting and tying practices by SLI 
Figure 47.Breastfeeding Initiation on Day of Birth by SLI 
Figure 48. Postpartum visit within three days by SLI 
13.6
65 2
16.8
66.7
28.1
73 3
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
31.1
67.6
32 3
67.7
36.2
70.5
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
80
1.0
22 5
1.4
22.4
2.7
22 5
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
low medium high
Baseline Endline
81
Comparison of Coverage by Scheduled Castes and Backward Classes vs. Other 
Groups 
In this section, recently delivered women are categorized by caste into three categories: 
scheduled castes/tribes, other backward classes, and other classes. Because there was no 
baseline data collected on caste, this section compares coverage in the intervention and 
comparison districts at endline. Coverage of antenatal home visits visits by ANMs was equitable 
in the intervention district (33-36%, Figure 47) and slightly higher than the comparison group, 
while coverage by AWWs was also equitable (40-45%, Figure 48) but significantly higher than 
the comparison district. The proportion of RDW visited by CAs in the intervention group ranged 
from 26-30% by caste (Figure 49). Again, antenatal check-ups and skilled birth attendance 
showed the most inequality among groups (Figures 50-51). Neither of these indicators was 
considerably better in the intervention group versus the comparison group, and in fact the 
comparison group had a higher proportion of skilled birth attendance for the “other classes” 
category. Practices of clean cutting and tying and initiation of breastfeeding the first three days 
were quite equal across castes (68-69%, 66-70%, respectively, Figures 52-53). And finally, 
coverage of postpartum home visits was about 20% higher than the comparison district and 
ranged from 18-24% in the different castes (Figure 54). Anganwadi worker and change agent 
coverage for antenatal and postpartum visits showed slightly better coverage for lower castes at 
endline. This could be due to the fact that anganwadi workers and change agents are more likely 
to be from lower castes and, therefore, are more comfortable visiting women of lower castes. 
While program coverage seems to have improved for lower socio-economic groups, access to 
care remains unequal, which could be reflective of many barriers such as knowledge, cost, 
distance, or cultural factors. 
Figure 49. Auxiliary Nurse Midwife Antenatal Visits by Caste 
13.1 15.7
22.5
33.0 34.4 35.9
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
82
Figure 50. Anganwadi Worker Antenatal Visits by Caste 
Figure 51. Change Agent Antenatal Visit Coverage by Caste 
9.3 9.5 12.2
45.0 45.5
40.1
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
29.8 29.3
25.6
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
scheduled castes/tribes other backward classes
other classes
83
Figure 52. Antenatal Check-up by Caste 
Figure 53.Trained Birth Attendance at Home or Facility by Caste 
22.6
28.6
50.7
33.6
37.2
49 5
0
10
20
30
40
50
60
70
80
90
100
Proportion of RD
W
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
15.5
21 6
42.9
17.3
23.9
34.8
0
10
20
30
40
50
60
70
80
90
100
Proportion of RD
W
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
84
Figure 54. Clean cutting and tying practices by Caste 
Figure 55.Breastfeeding Initiation on Day of Birth by Caste 
41.8 40 2
50.4
67.7 68.1 69.1
0
10
20
30
40
50
60
70
80
90
100
Proportion of RD
W
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
21.3 21.4
36.4
66.4 66 5
70.3
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
85
Figure 56. Received postpartum visit within three days by Caste 
2.4 2.5 4.0
24.0 22.1
18.3
0
10
20
30
40
50
60
70
80
90
100
Proportion of RDW
Comparison Intervention
scheduled castes/tribes other backward classes
other classes
86
CHAPTER 9. TRENDS IN NEONATAL MORTALITY 
Table 24 shows the mortality rates of the comparison and intervention areas at baseline and at the 
end of the project. Neonatal mortality rates were relatively constant in both the comparison and 
intervention areas. The neonatal mortality rate in the comparison area was 47.3 per 1,000 live 
births (confidence interval 42.3-52.7) at baseline and 49.8 per 1,000 live births at endline. The 
neonatal mortality rate in the intervention area was 50.5 per 1,000 live births (confidence interval 
44.6-55.5) at baseline and 47.8 per 1,000 live births (confidence interval 43.2-52.6) at the 
endline. No considerable differences were seen in early neonatal mortality, with baseline and 
endline rates at 21.3 per 1000 live births and 18.1 per 1000 live births, respectively, in the 
comparison area and 23.4 per 1000 live births and 22.6 per 1000 live births, respectively, in the 
intervention area. Stillbirth rates also remained relatively stable. 
Figure 55 shows the probability of neonatal death by study arm. Overall there is no difference in 
the risk of dying between the comparison and intervention arms. Figure 56 shows the probability 
of death in the intervention arm by baseline and endline. Again, an inconsequential difference is 
observed in the risk of dying between baseline and endline. 
87
Table 25. Mortality Rates 
Comparison Intervention 
Baseline Endline Baseline Endline 
Number Rate Number Rate Number Rate Number Rate 
Stillbirths 207 30.1 232 35.3 316 33.0 281 32.9 
 
Timing of Neonatal 
Deaths 
 
 1st day 78 11.7 109 17.2 123 13.3 103 12.5 
 1-6 days 142 21.3 115 18.1 217 23.4 187 22.6 
 7-27 days 95 14.3 92 14.5 128 13.8 105 12.7 
 
Total Neonatal 
Deaths 
315 47.3 316 49.8 468 50.5 395 47.8 
Total Live Births 6,654 6,342 9,268 8,267 
 
Total births 6,861 6,574 9,584 8,548 
Figure 57. Kaplan-Meier Estimates of Probability of Neonatal Death, Intervention 
district vs. Control district at Endline 0.00 0.01 0.02 0.03 0.04 0.05
0 10 20 30
Days after birth
Comparison Intervention
0.00 0.01 0.02 0.03 0.04 0.05
0 10 20 30
Days after birth
Comparison Intervention
88
Figure 58. Kaplan-Meier Estimates of Probability of Neonatal Death, in Intervention 
district, Baseline vs. Endline 
 
0.00
0.01
0.02
0.03
0.04
0.05
0 10 20 30
Days after birth
Baseline End 
0.00
0.01
0.02
0.03
0.04
0.05
0 10 20 30
Days after birth
Baseline Endline
89
CHAPTER 10. CONCLUSIONS AND RECOMMENDATIONS 
 
The Integrated Nutrition and Health Program II (INHP II) has been a continuation of CARE￾India’s longstanding collaboration with the Government of India. This cycle of the program 
included a new emphasis on promoting the health and survival of newborns, a group with 
specific health needs that have received less emphasis in traditional child survival programs. 
Several community-based efficacy trials have been conducted to examine various strategies for 
improving newborn survival with significant success. This is the first large-scale effectiveness 
trial of a package of neonatal health interventions implemented through the existing government 
and non-governmental health systems. The program inputs that were provided were achieved at 
scale, with a limited cost and time frame. Thus, this evaluation research yields useful insights for 
future INHP programs, Indian health system policies and the global network of program planners 
and researchers in maternal and newborn health. It has also led to important lessons learned for 
this type of evaluation research. Finally, this program was evaluated in one of the more 
economically and socially disadvantaged areas of India, rural Uttar Pradesh. Therefore, this 
research can help health planners to understand what can be achieved realistically in a limited 
resource setting within India using the existing government health services. 
Program accomplishments and lessons learned 
At the beginning of the project, community-based health service providers had little 
knowledge of newborn health, because this topic had seldom been emphasized in previous 
trainings. Knowledge of some basic aspects of maternal health care –such as the “five cleans” of 
delivery – was also limited at the outset. One of the major challenges of this program was to train 
a large number of community-based health workers in maternal and newborn health in a short 
period of time. The difficulty of accomplishing this task should not be understated. Within the 
newborn evaluation research area, workers were scattered over hundreds of villages. Most of the 
anganwadi workers have limited access to telephones and transportation outside their villages, 
and as honorary workers they receive modest compensation for their work. The auxiliary nurse 
midwives have multiple competing priorities in their job responsibilities. In addition to training 
the ANMs and AWWs, CARE recruited and trained a new cadre of community health volunteers 
known as change agents. The logistics involved in bringing these groups together for training 
was quite challenging. The completion of three rounds of two-day trainings of the large group of 
workers is a remarkable accomplishment of this program. 
 In addition to completing the trainings, this evaluation suggests that the trainings elevated 
workers’ knowledge of maternal and newborn health. Knowledge scores rose significantly in the 
intervention area while remaining similar in the comparison area. Nonetheless, some deficiencies 
remained in some workers’ knowledge. For example, none of the ANMs or CAs and only 2.5% 
of the AWW received more than 75% of the possible points on the knowledge score. Knowledge 
of maternal and newborn danger signs, in particular, was somewhat lower than expected for 
cadres of workers who were meant to effectively recognize and refer serious complications. 
These findings suggest that the information imparted during the trainings was not retained 
adequately by the community-based service providers. One explanation is that this information 
was not conveyed adequately in training. Another explanation is that the field-based support for 
90
these groups was insufficient to sustain their gains in knowledge. The challenge of maintaining 
the knowledge of community-based workers is large scale government programs is common. 
The next phase of the program should include clear guidelines to maintain the knowledge and 
skills of the community-based workers through the use of job aids and refresher trainings. 
These strategies should be field-tested and evaluated on a pilot level before being scaled up to 
the entire program. 
The baseline survey found that the ICDS and community-based component of the Ministry 
of Health programs, which are implemented mainly through the anganwadi workers and 
auxiliary nurse midwives, failed to reach most women and newborns, particularly economically 
vulnerable women, younger mothers and those at greatest risk for neonatal mortality. Coverage 
improved remarkably by the endline. One of critical lessons learned from this project is that it 
is possible to improve coverage of proven interventions at scale using sustainable strategies.
The increases in coverage were achieved mostly by improved management and training, without 
increasing the number of paid community-based workers. Of the three types of workers, the 
anganwadi workers were most successful in redirecting their work to improve program coverage. 
However, the anganwadi workers and auxiliary nurse midwives were far more successful at 
reaching women during pregnancy than they were at contacting women during delivery or during 
the early postpartum period. This is clearly a factor of the differences in the windows of 
opportunity, and it may also reflect cultural norms about receiving outside visitors during 
childbirth and postpartum. Although essential newborn care practices can and should be taught 
during pregnancy, other community-based trials suggest that it is essential to have contact with 
mothers and newborns in the days immediately after birth. This contact can facilitate the 
recognition and management of newborn complications. It can also enhance the adoption of 
supportive newborn care practices for small or weak newborns, such as kangaroo mother care. 
Skilled birth attendance and access to emergency obstetric care have long been the hallmarks of 
the safe motherhood movement and are critical to reducing maternal and newborn mortality. 
Therefore, the next phase of the program should find strategies to improve coverage with 
skilled attendants at delivery and in the early postpartum period and to increase the 
communities’ capacity for prompt referral of women with obstetric and postpartum 
complications and sick newborns. 
 The recruiting and training of change agents was a new innovation of this program. As 
mentioned previously, the accomplishment of this large-scale training was an important program 
achievement. Deploying the change agents was one strategy to improve program coverage 
overall, and particularly to reach out to women from economically or socialized marginalized 
groups. In terms of the first goal, change agents reached about 30% of pregnant women and 
about 20% of newborns less than three days old. Training of change agents alone, therefore, was 
insufficient to achieve high levels of ICDS program coverage at national scale. However, taken 
together with increased coverage of the ANMs and AWWs, providers reached 3.5 times greater a 
proportion of women with antenatal home visits, and the coverage of postpartum visits within a 
week of delivery rose from less than 1% to 29%. Examining the relative contribution of change 
agents to this increase in coverage, 3% of women were visited only by change agents during the 
antenatal period, and 15% were visited only by change agents during the first week postpartum. 
In other words, change agents made a relatively small contribution to total program coverage 
during the antenatal period and a much larger contribution during the neonatal period, in terms of 
91
their direct provision of home visits. However, they may have improved coverage indirectly by 
assisting the ANMs and AWWs in identifying women in need of services or by encouraging 
pregnant women to seek care. In terms of the second goal of improving program equity, change 
agents were no more likely to contact women of lower castes or lower standard of living 
households than they were to contact women from other groups. The question arises of whether 
the cost of recruiting, training and maintaining support for the change agents merits the return on 
the investment. These data should be used in conjunction with CARE’s internal data on the 
program inputs in order to assess this question more fully. 
The impetus behind the training and increased coverage were to improve the quality of 
counseling received by mothers and their families. Assessing the quality of counseling through a 
survey is challenging, because women were asked to recall conversations that took place one 
month to two years before the survey was conducted. The survey used both prompted and 
unprompted questions to elicit women’s recall of topics that were discussed in visits by auxiliary 
nurse midwives, anganwadi workers and change agents during pregnancy. Examining the 
unprompted responses, it seems that the improvements in quality of counseling were about 20-30 
percentage points at most for any given topic, and the mean number of topics discussed 
improved only slightly. However, looking at the prompted responses, women recalled far more 
of the topics being discussed, indicating greater improvements in quality. Therefore, the quality 
of counseling may have improved modestly during the course of the intervention, but the 
inconsistency in women’s responses call for some caution in interpreting this statement. 
The rationale behind this intervention is that health education will lead to improved 
knowledge which will then improve maternal and newborn care practices. The practices might be 
considered in two categories, preventive care practices and care-seeking. Many critical behaviors 
related to newborn health improved during the course of the intervention. Almost all of the 
preventive care practices, such as sterile cord care and early breastfeeding initiation, have 
improved significantly in the intervention area from baseline to endline, while little change was 
observed in the comparison area. These changes are useful and have important implications for 
newborn health. The baseline survey identified many practices that put newborns at increased 
risk of mortality. Some ethnographic studies suggest that these practices are firmly entrenched in 
local understanding of risks and benefits to the newborn and mother. This study shows these 
newborn care practices can be changed through community-based health education campaigns. 
This is a significant finding of this study. Some variations in findings exist when the endline 
survey is contrasted with the adequacy survey findings, as will be discussed further. Although 
the endline survey may overstate the behavior change somewhat, the evaluation research largely 
suggests that many of the essential newborn care practices have been adopted by a large 
proportion of households in the study area. However, women’s knowledge of maternal and 
newborn danger signs showed almost no improvement. This finding suggests a deficiency in the 
program, as prompt recognition of danger signs is one prerequisite to timely and appropriate 
care-seeking. 
The data related to care-seeking for maternal and newborn care practices is more difficult 
to interpret. Although antenatal care coverage improved, quality of care was poor and changed 
little from baseline to endline. As a result, it is unclear whether a greater proportion of women 
with pregnancy complications received adequate treatment. Improvements in skilled birth 
92
attendance were of little practical significance, and no evidence suggests that a greater proportion 
of women with delivery complications sought care over the course of the intervention. In terms 
of neonatal care-seeking, data suggest that the proportion of newborns with serious 
complications that were taken for care from a trained provider increased by about 20 percentage 
points. This improvement is promising, but less than half of newborns (45%) with serious 
complications received care from a trained provider. In addition, these reports are based on 
complications reported by mothers, which may or may not correspond to medical diagnoses. 
Care-seeking is affected by many factors outside of the scope of this intervention, such as 
availability and access to health care. However, the limited improvements in care-seeking may 
help interpret the mortality data, as will be discussed further. 
Antenatal intervention coverage was higher than postpartum coverage, but still 45% of 
women were not reached by the intervention. This level of coverage represents a significant 
improvement but the quality of ANC did not improve. It is a highly encouraging finding of this 
study that it is possible to expand government health and ICDS program coverage without 
increasing the number of workers. The increased coverage suggests that the improvements in 
management and convergence between the two government ministries took root at the village 
level. Nonetheless, coverage, particularly during the postpartum period, may have been too low 
to achieve the desired impact on neonatal mortality. 
Interpreting different findings within the study methods 
The interpretation of the improved program coverage and behavior change is complicated 
by discrepant findings between the periodic adequacy surveys and the endline survey. For some 
key program indicators, the periodic adequacy surveys yielded very different findings than the 
endline survey. It is implausible that this would be a result of sampling differences. It is also 
implausible that the differences reflect a change in the manner of administering the survey, since 
the same survey agency administered both surveys using the same instruments, the same group 
of researchers oversaw training for the data collection and the data quality assurance found the 
data to be of outstanding quality. Therefore, one aspect of interpreting the endline findings is to 
understand how this could have occurred. 
 One possible explanation would be that since the endline survey involved long recall, 
mothers’ recollection of their past behaviors was not always consistent. That is, women reported 
different behaviors at different times because their recollections were not always the same. 
However, this would not explain why the changes were measured in the intervention area but not 
in the comparison area. 
Another explanation is that respondents reported ideal behaviors rather than actual 
behaviors. Some respondents who gave birth earlier in the intervention may not have known 
about the desired behaviors at the time of pregnancy or delivery, such as the essential newborn 
care practices promoted through the intervention. They also may not have known about these 
behaviors at the time of the adequacy surveys, or they knew about these behaviors but reported 
their actual behaviors. Later in the intervention they may have been exposed to the intervention, 
and then during the endline survey incorrectly reported that they had adopted these behaviors. 
As a result, at the end of the project they reported these desired behaviors as if they had practiced 
93
them at the time of birth. This explanation seems plausible. Therefore, some data suggest that 
the endline survey may overstate the extent of the increases in program coverage and the 
improvement in key practices. Nonetheless, the findings suggest that, at minimum, knowledge 
of the essential newborn care has improved substantially in the study area. 
Lack of impact on neonatal mortality 
 The key outcome indicator of this study was the neonatal mortality rate (NMR). The 
study was designed to detect a 20% or greater decrease in NMR. These data suggest no 
statistically significant change in neonatal mortality following two years of intervention. To 
understand how so many aspects of the progam could have improved without showing a 
corresponding decrease in neonatal mortality, it is useful to review the common causes of 
neonatal death in this community and discuss the extent to which the intervention addressed 
these causes. Figure 57 presents the causes of neonatal deaths in both districts as ascertained by 
the baseline verbal autopsy study. Preterm birth was the leading cause of death, attributed to 28% 
of deaths. One proven strategy to reduce preterm birth that was a component of this program was 
extra care for newborns, such as extra feeding, extra attention to hygiene and added warmth 
through skin-to-skin contact and other measures (13). It is unclear where preterm or low birth 
weight newborns received extra care at the household level. It is difficult to distinguish preterm 
and low birth weight babies in this analysis, given that newborns are seldom weighed after birth 
and gestational age is usually not calculated. The second leading cause of death was serious 
infections, such as sepsis and pneumonia. Many of the preventive care strategies included in the 
intervention have been shown to be effective at reducing neonatal infections, such as clean cord 
care and immediate breastfeeding (13). However, the two strategies that could be expected to 
have the greatest impact on deaths due to infections – case management of pneumonia and 
emergency neonatal care – were not included in the intervention package. While recognition of 
maternal and newborn danger signs was part of the intervention, providers’ knowledge and 
RDWs’ recall of these signs was low, suggesting care-seeking and referral for serious 
complications was low. Family care (clean home delivery, hygienic cord care, thermal care and 
breastfeeding promotion), which was included in the intervention, could be expected to reduce 
infection deaths by 20-50% if coverage approached 80% or higher (16). Birth asphyxia is the 
third most common cause of death, and none of the intervention components explicitly targeted 
birth asphyxia. Deaths due to congenital abnormality could be reduced 1-10% through 
periconceptual folic acid supplementation; but only about 20% of women in the intervention 
district took an adequate amount of iron-folate tablets, and they may have taken the supplements 
too late to prevent birth defects. Tetanus deaths could be reduced by tetanus toxoid immunization 
and clean cord care, indicators that improved during the intervention. However tetanus was 
identified as a cause of only about 4% of deaths, a small contribution to the total burden. 
Similarly, diarrhea deaths could be reduced up to 60% through the the family care practices 
promoted through the intervention, but only 1% of deaths were due to diarrhea (16). Therefore, 
some proven interventions that might be expected to have the greatest impact on neonatal 
mortality in this setting were not included in this program, or were included but were not 
adequately implemented to impact mortality. 
 
94
Figure 59. Causes of Neonatal Deaths Established by Baseline Verbal Autopsy Study 
Tetanus
4%
Congenital 
Abnormality
6%
Preterm birth
28%
Birth Asphyxia 
and Injury
14%
Serious 
infection
24%
Diarrhea
1%
Not Identified
23%
 Two other factors are also important for interpreting the mortality data. The first is that 
the verbal autopsy methods found no cause for 23% of deaths. This is partially a reflection of the 
limitations of verbal autopsy methodology. However, it also demonstrates that the signs and 
symptoms of neonatal complications can be difficult to recognize and that newborns need to be 
assessed by trained workers. The second factor is that most deaths occurred very early in the 
neonatal period. For example, 26% of deaths in the intervention area occurred on the first day of 
life and 73% occurred within the first week. The incidence of death in the early neonatal period 
was higher than expected and warrants a great deal of emphasis on intrapartum care and care￾seeking during the first week of life, particularly the first three days. At endline, the proportion 
of women in the intervention area that gave birth with a skilled attendant was 22%, 16% of 
women reported taking their newborn for a check-up during the first three days after birth and 
6% of women were visited at home by an ANM, AWW or change agent during the first three 
days after birth. Furthermore, the data suggests that in the INHP II program, service providers’ 
ability to detect newborn danger signs was less than optimal. Mothers’ knowledge of newborn 
danger signs showed limited improvement during the intervention. Additionally, if 
complications were recognized, it may have been difficult to seek care because of limited access, 
high cost, unsatisfactory facilities, distance, or cultural norms surrounding newborn care, 
although this evaluation does not have facility assessment data to tease out these barriers. As a 
result, it is clear that many serious newborn illnesses went unrecognized or untreated. 
Recommendations: 
 Our assessment is that this program has laid the groundwork for the promotion of 
newborn health through India’s health and ICDS programs. Some of the lessons learned from 
this evaluation research should inform the next cycle of program development and 
implementation. 
One critical aspect of future program development will be to improve ongoing 
supervision and quality of counseling for service providers. It will be important to assure that the 
95
service providers have resources (such as cue cards or pamphlets) to refresh their own 
knowledge in the field. It is equally important to continue to improve the field-based supervision 
and ongoing refresher training that the workers receive. The value of supervision and continuous 
training in the field has been recognized by CARE, and steps were taken, particularly after the 
midterm review, to improve these activities. From an evaluation research perspective, it will be 
critical to determine how to best assess the quality of counseling at the household level. This 
evaluation was based entirely on maternal recall, an approach that has both strengths and 
weaknesses. Future evaluations could consider direct observation or exit interviews of mothers to 
assess quality of visits. Program planners should also seek ways to maintain knowledge at the 
household level. One strategy to do this is the individual monitoring systems, but this approach 
was later rejected in Uttar Pradesh as not being scalable. Health education that improves 
women’s knowledge of dangers signs for themselves and their newborns will be one critical step 
toward improving care-seeking behaviors. 
 A substantial amount of program resources were devoted to recruiting and training the 
volunteer change agents. However, their contribution to the increase program coverage was 
limited. Also, their performance on knowledge tests suggests that they may have had difficulty 
retaining the information coveyed during trainings. For future program planning, it may be more 
appropriate to ask the change agents to focus on community mobilization activities, instead of 
providing health education or recognition and referral of complication. At the same time, 
improved training and supervision for the existing cadres of government workers, AWWs and 
ANMs, would be beneficial. 
 The lack of skilled birth attendance in the study community was clearly a barrier to 
reducing neonatal mortality. The program sought to improve awareness of the need for skilled 
birth attendance in the presence of complications and increase preparations necessary for 
institutional deliveries, such as saving money and arranging transport. However, no changes 
were measured in the coverage of skilled birth attendance, either when examining all recently 
delivered women or women with a self-reported delivery complication. Future program 
planning should: (a) assess the access to skilled birth attendance and emergency obstetric care in 
the communities; (b) determine community barriers to accessing skilled care. 
Finally, the program should focus particularly on strategies to improve program coverage 
during the first three days of life. Routes for delivering case management of pneumonia and 
serious infections should be explored. Although this intervention is more complex than other 
interventions, it can be expected to have great impact in this setting, when combined with 
preventive and supportive neonatal care practices. 
After the intervention began, the Lancet neonatal series and other research became 
available to further the research community’s understanding of the most effective interventions 
to reduce neonatal mortality. Increasing efforts are being made to understand and reduce the 4 
million newborn deaths that occur globally. A recent update on the Lancet neonatal series noted, 
“Although we know what we should be aiming for – high coverage of several simple cost￾effective interventions – major questions remain about how to get there” (21). This study is 
significant to researchers’ understanding of the benefits and challenges of implementing a 
newborn health program within an existing government health and development system. 
96
97
REFERENCES 
1. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year? 
Lancet 2003;361(9376):2226-34. 
2. Lawn JE, Cousens S, Zupan J. 4 million neonatal deaths: when? Where? Why? Lancet 
2005;365(9462):891-900. 
3. Hyder A. Burden of disease for neonatal mortality in South Asia and Sub-Saharan Africa. 
Washington DC: Save the Children Federation-US; 2001. 
4. Saving Newborn Lives. State of the World's Newborns. Washington DC: Save the 
Children Federation-US; 2001. 
5. NFHS-II. National Family Health Survey-II 1989-1999. Mumbai: International Institute 
for Populations Sciences and ORC-Macro; 2001. 
6. Lawn J, Shibuya K, Stein C. No cry at birth: global estimates of intrapartum stillbirths 
and intrapartum-related neonatal deaths. Bull WHO 2005;83:409-417. 
7. Child Health Research Project. Special report, Reducing perinatal and neonatal mortality 
Oct. 1999. 
8. UNICEF. State of the World's Children. New York: UNICEF; 2001. 
9. Bang A. Pneumonia in neonates: can it be managed in the community? Arch Dis Child 
1993;68:550-556. 
10. Bang AT, Bang RA, Baitule SB, Reddy MH, Deshmukh MD. Effect of home-based 
neonatal care and management of sepsis on neonatal mortality: field trial in rural India. 
Lancet 1999;354(9194):1955-61. 
11. Bang AT, Bang RA, Reddy HM, Deshmukh MD. Methods and the baseline situation in 
the field trial of home-based neonatal care in Gadchiroli, India. J Perinatol 2005;25 Suppl 
1:S11-7. 
12. Bryce J, Boschi-Pinto C, Shibuya K, Black RE. WHO estimates of the causes of death in 
children. Lancet 2005;365(9465):1147-52. 
13. Darmstadt GL, Bhutta ZA, Cousens S, Adam T, Walker N, de Bernis L. Evidence-based, 
cost-effective interventions: how many newborn babies can we save? Lancet 
2005;365(9463):977-88. 
14. Knippenberg R, Lawn J, Darmstadt G, Begkoyian G, Fogstad H, Walelign N, et al. 
Systematic scaling up of neonatal care in countries. Lancet 2005;365:1087-98. 
15. USAID/India. A strategic approach for USAID's investments to improve child health and 
nutrition in India. New Dehli March 2001. 
16. Paul V. Newborn care in India - A promising beginning, but a long way to go. In: 
Newborn Care in South-East Asia Region - Current Status and Priorities. Report of the 
Regional Expert Group Meeting, Nov. 16-17, 1998. 
17. NFHS India. National Family Health Survey; 1998-99. 
18. Bang AT, Reddy HM, Deshmukh MD, Baitule SB, Bang RA. Neonatal and infant 
mortality in the ten years (1993 to 2003) of the Gadchiroli field trial: effect of home￾based neonatal care. J Perinatol 2005;25 Suppl 1:S92-107. 
19. Pratinidhi A, Shah U, Shrotri A, Bhodani N. Risk-approach strategy in neonatal care. 
Bull World Health Organ 1986;64:291-297. 
98
APPENDIX: BEHAVIOR CHANGE MATRIX 
Lifecycle period 1: Pregnancy: (i) First Trimester 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (Mother): 
Report probable pregnancy 
(cessation of menses) to CA / AWW / 
ANM 
Seek antenatal checkup, ask ANM 
for expected date of delivery 
Consume 1 additional meal / day 
Eat available variety of foods, 
including cereals, pulses, green leafy 
vegetables, other vegetables and 
fruits. Consume milk, milk products, 
meat, fish and eggs if available and 
feasible 
Procure and consume SN 
Take rest – lie down for 1-2 hours 
during the day 
Seek treatment for fever from ANM / 
PHC / Hospital 
Make birth plan (with family) 
Begin saving for delivery 
Decide on whether to deliver at home 
or hospital (own choice, 
ANM/doctor’s advice) 
Identify hospital for hospital delivery 
Identify trained birth attendant for 
home delivery (CA/AWW/ANM’s 
advice) 
Begin self-monitoring antenatal care 
Change Agent: 
Report probable pregnancy to the 
AWW and ANM at the earliest 
(usually, when a second month 
passes without menses) 
Advise early antenatal checkup at the 
NHD / hospital 
Advise monthly attendance at NHD 
Counsel on diet, rest 
Begin making birth plans with family: 
Counsel on saving for delivery 
Help family plan place of delivery, 
and identify hospital / TBA 
Learn about whether woman plans to 
deliver in this village or parental 
village 
Explain the need for IFA tablets and 
TT from fourth month onwards 
Refer to facility 
Initiate and encourage self￾monitoring 
TBA: 
Note expected date of delivery 
Help CA / AWW / ANM provide 
antenatal care to mother 
AWW: 
Register pregnancy 
Report pregnancy to ANM; get 
confirmation and expected date of 
delivery from ANM 
Provide SN / THR as applicable 
Visit family at least once, along with 
CA 
Learn from CA about the family and 
their plans for birth 
Counsel on diet, rest and birth plan 
(as for CA) 
Keep CBO informed once a month 
(facilitate monthly CBO meeting) 
ANM: 
Register pregnancy 
Learn from CA and AWW about the 
family and their plans for birth 
Check up woman on NHD – obstetric 
history, pallor, baseline blood 
pressure, weight 
Enquire whether pregnancy was 
desired and planned 
Note and enquire into failure of 
contraception 
Advise against second trimester 
Supervisors: 
Help diagnosis of pregnancy, and of 
gestational age in doubtful situations 
Enquire about families likely to be 
difficult 
Help find/provide support to poor 
families 
Ensure SN/THR being provided 
Note problems for discussion at block 
(related to registration and 
contraceptive failure) 
Act on decisions taken at block 
Review social map and quality of 
CBO monitoring process 
Observe NHD and identify supply 
and training needs 
CDPO: 
Track registered vs expected 
pregnancies and address possible 
left-outs 
Track proportion of women 
registering early 
Track proportion of pregnant women 
getting SN/THR 
Attend to problems brought up by 
supervisors or the CBOs/PRI 
Lifecycle period 1: Pregnancy: (i) First Trimester, continued
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Family (husband, in-laws, parents): 
Report probable pregnancy 
(cessation of menses) to CA / AWW / 
ANM 
Ensure that woman gets checked by 
the ANM or at the PHC/hospital, and 
decide on where she will deliver, 
identify hospital or trained TBA 
Note expected date of delivery from 
ANM 
Ensure that woman stops doing hard 
labor 
Ensure that woman gets rest for two 
hours every afternoon and a full 
night’s sleep 
Make birth plans as above 
Ensure immediate treatment of any 
illness 
CBOs: 
Mark new pregnancies on map 
Identify and discuss unintended 
pregnancies 
Identify and discuss women with 
problems 
Identify and discuss families likely to 
be problematic 
Identify and discuss areas of village 
with no reported pregnancies 
Support AWW/CA in their work 
termination of pregnancy / pre-natal 
sex determination as needed 
Estimate expected date of delivery 
and inform woman, CA, AWW 
Advise hospital delivery if indicated 
(previous stillbirth / caesarian 
section, major medical problems, 
very small stature) and feasible 
Counsel on diet, rest and birth plan, 
and for forthcoming TT, IFA, and 
chloroquine 
Refer woman with illness in first 
trimester to appropriate referral 
center; advise not to take any 
treatment till fourth month without 
doctor’s advice 
MO (PHC): 
Track registered vs expected 
pregnancies and address possible 
left-outs 
Track proportion of women 
registering early 
Keep special tag on areas not 
covered by ICDS 
Handle referrals 
Attend to problems brought up by 
supervisors or the CBOs/PRI 
Agenda for BLAC/BLRM: 
Clusters / sectors / villages reporting 
less than 30% early registration – 
reasons 
Clusters / sectors reporting 25% less 
pregnancies than previous quarter – 
reasons 
Contraceptive failures 
Coverage in non-ICDS areas 
Problems identified by CDPO / MO 
Supply issues 
Training needs 
Problems brought up by individual 
supervisors 
Unusual problems 
Lifecycle period 1: Pregnancy: (i) First Trimester, continued 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (Mother): 
Report pregnancy to CA / AWW / 
ANM if not yet reported 
Seek antenatal checkup at NHD, at 
least twice this trimester 
Seek two doses of TT injections, one 
month apart 
Consume IFA, 1 tablet / day, or 1 
tablet twice a day, as advised by 
ANM/AWW 
Consume mebendazole 1 tablet 
twice daily for 3 days (one course) 
Consume 1 additional meal / day 
Eat available variety of foods, 
including cereals, pulses, green leafy 
vegetables, other vegetables and 
fruits. Consume milk, milk products, 
meat, fish and eggs if available and 
feasible 
Procure and consume SN 
Consume chloroquine, 2 tablets once 
every week, as advised by 
ANM/AWW 
Seek treatment for fever from ANM / 
PHC / hospital 
Change Agent: 
Encourage antenatal check up, 2 
doses of TT and consumption of 
IFA,mebendazole, and chloroquine 
tablets as advised 
Refer any problems woman has with 
IFA, mebendazole or chloroquine to 
ANM 
Encourage diet and rest as before 
Make birth plan with family, including 
plans for birth in parental village if the 
woman is going away. Discuss 
possible danger signs: 
Any bleeding before childbirth or 
excessive bleeding at or after 
childbirth 
Convulsions at any time 
Sudden appearance of facial / hand / 
genital swelling 
Severe headache 
Prolonged labor, beyond 12 hours of 
mild pain, or beyond 2 hours of 
active, severe pains 
Onset of labor more than a month 
before the expected date of delivery 
AWW: 
Visit family at least once during this 
trimester, and more frequently if 
there are problems 
Track progress of the mother at 
NHDs and through the CA, with 
respect to diet, rest, IFA, 
mebendazole chloroquine and birth 
plans 
In case of women planning to deliver 
in parental village, ensure that she 
has received two doses of TT and 
the clean delivery kit, is aware of 
danger signs and has made 
contingency birth plans before 
leaving 
In case of late registrations, ensure 
that the woman is seen by the ANM 
at the first available opportunity 
Ensure SN/THR, and supply of IFA 
Keep CBO informed once a month 
Supervisors: 
As for first trimester, plus 
Enquire into causes of late 
registration 
Note problems with supplies and 
report to block 
Help handle problems of families not 
participating in NHDs / receiving 
services 
Check on supplies 
Enquire about and address problems 
in consuming IFA, mebendazole and 
chloroquine 
Review social map and quality of 
CBO monitoring process 
Lifecycle period 1: Pregnancy: (ii) Second Trimester
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Make birth plan (with family): 
Continue saving for delivery 
Decide on place of delivery (hospital 
or home) if not yet decided, and 
identify trained TBA if planning home 
delivery 
Identify hospital to rush to in case of 
danger signs 
Accumulate enough clean cloth for 
the delivery / baby 
Self-monitor progress of own ante￾natal care 
Family (husband, in-laws, parents): 
Ensure she gets checked up at least 
once, gets two doses of TT and takes 
IFA, mebendazole and chloroquine 
tablets as advised by the AWW/ANM 
Ensure that woman continues to get 
enough food and rest; help reduce 
her workload 
Report any problems to the 
AWW/ANM 
Make birth plans as above; 
especially, identify hospital, vehicle 
for transport and person to 
accompany woman in an emergency 
In case of late detections, ensure that 
the woman is seen by the ANM at the 
first available opportunity, and that 
birth plans begin immediately 
Identify women likely to arrive newly 
in late pregnancy and inform AWW 
Initiate and encourage self￾monitoring, if not yet done 
TBA: 
Maintain contact with family and 
encourage them to follow the advice 
of the CA, AWW and ANM 
CBOs: 
Identify and discuss women not 
seeking services or consuming 
tablets as advised 
Identify and discuss adverse 
outcomes 
Identify women not likely to have 
needed support in emergency 
situations, and discuss ways of 
community support 
Identify shortcomings in service 
provision and take up the matter 
through PRI if need be 
 ANM: 
Conduct at least two antenatal 
checkups: pallor, BP, weight, fundal 
height 
Advise woman and family about any 
specific actions to be taken, including 
advice for hospital delivery 
Confirm expected date of delivery 
and inform/remind woman of this 
Decide if woman needs therapeutic 
dose of IFA and advise CA, AWW 
and woman of this 
Ensure two doses of TT are given 
during this trimester 
Ensure supply of IFA, mebendazole 
and chloroquine to pregnant woman 
Handle referrals from AWW/CA 
In case of late registrations, ensure 
at least one antenatal checkup and 
two TT doses are completed at 
earliest possible, and that birth plans 
begin immediately 
Track drop-outs with help of CA and 
AWW 
Counsel on clean delivery kit to 
women likely to deliver in some other 
village 
CDPO: 
As for first trimester, plus – 
Address supply problems 
MO (PHC): 
As for first trimester, plus – 
Address supply problems 
Agenda for BLAC/BLRM: 
Clusters / sectors / villages reporting 
less than 50% completion of TT2 in 
second trimester – reasons 
Clusters / sectors reporting less than 
30% or more than 70% of women as 
anemic – reasons 
Coverage in non-ICDS areas 
Problems identified by CDPO / MO 
Supply issues 
Training needs 
Problems brought up by individual 
supervisors 
Unusual problems 
Lifecycle period 1: Pregnancy (iii) Third Trimester 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (Mother): 
At least two antenatal checkups 
during this trimester 
Diet as in second trimester, IFA and 
chloroquine as advised by ANM 
Increase period of rest to at least two 
hours during the day 
Seek treatment for fever from ANM / 
PHC / hospital 
Finalize birth plans with family: 
Learn about what can possibly go 
wrong at or around childbirth 
(bleeding, unusual swelling and 
convulsions, prolonged labor, 
premature labor), and make 
contingency plans with family 
Inform family members / CA / AWW 
immediately in case of such 
problems 
Change Agent: 
Maintain touch with woman and 
family 
Encourage antenatal checkup, diet, 
rest, IFA, chloroquine as advised by 
ANM/AWW 
In case of new arrivals in village, 
enquire about previous antenatal 
care, ensure earliest possible 
checkup by ANM or PHC, and 
discuss birth plans and contingencies 
with family at the earliest 
Counsel on danger signs and finalize 
birth plans with family and TBA: 
advise preparedness for immediate 
referral to hospital in case of: 
Any bleeding before childbirth or 
excessive bleeding at or after 
childbirth 
Convulsions at any time 
Sudden appearance of facial / hand / 
genital swelling, or severe headache 
AWW: 
Track progress of the mother at 
NHDs and through the CA, with 
respect to diet, rest, IFA, chloroquine 
and birth plans 
Visit family at least twice during this 
trimester and counsel on diet, rest, 
IFA and chloroquine 
Ensure birth preparedness is 
complete and satisfactory 
Address/refer problems related to 
IFA and chloroquine consumption 
In case of late arrivals, help expedite 
TT doses and birth preparedness 
Ensure SN/THR, and supply of IFA 
Keep CBO informed 
Supervisors: 
Conduct random home visits to 
assess quality of birth preparedness 
and help make improvements 
Review list of women advised 
hospital delivery and their birth plans 
with ANM 
Review and note all adverse events 
and report to block 
Review social map and quality of 
CBO monitoring process 
Seek poorest households and 
identify and address any specific 
problems there 
Lifecycle period 1: Pregnancy (iii) Third Trimester
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Continue saving, identify clean site of 
delivery that is well-ventilated and 
away from smoke, with help of 
CA/AWW/ANM, and get clothes 
ready for childbirth 
Procure and keep clean delivery kit 
handy (OR keep ready two clean 
blades, several pieces of clean 
thread, and soap for hand-washing) 
Continue self-monitoring 
Family (husband, in-laws, parents): 
As in second trimester, plus – 
Ensure woman does not have to 
indulge in hard labor 
Finalize birth plans: 
Understand danger signs (as above) 
Finalize arrangements for transport 
to hospital, cash and persons to 
accompany in case of danger signs 
Discuss contingency plan with TBA 
and CA 
Prepare for normal birth at home or 
at hospital, as advised by ANM 
(clean delivery kit, clean clothes, 
clean site) 
Prolonged labor, beyond 12 hours of 
mild pain, or beyond 2 hours of 
active, severe pains 
Onset of labor more than a month 
before the expected date of delivery 
Counsel on need to avoid bathing the 
newborn for a week, initiating BF 
within 1 hour after birth and avoiding 
prelacteal feeds, water, any other 
liquids or solids for the first 6 months 
(exclusive breast feeding) 
Support and encourage self￾monitoring 
TBA: 
Participate in birth planning with 
family, CA, AWW, ANM and 
cooperate with the plan 
Help family identify danger signs 
CBO: 
As above, plus 
Identify women due to deliver within 
the next month and review their birth 
plans, with special attention to 
women without resources: plan for 
them 
 ANM: 
Conduct at least two antenatal 
checkups and keep family, CA and 
AWW informed about any changes 
needed in birth plans 
Identify nonanemic women who have 
completed 100 days of IFA and 
advise them to stop taking IFA 
Ensure supply of IFA and 
chloroquine 
Handle referrals from AWW/CA 
In case of late registrations, ensure 
at least one antenatal checkup and 
two TT doses are completed at 
earliest possible, and that birth plans 
are completed satisfactorily 
Track drop-outs with help of CA and 
AWW 
Supply clean delivery kits to all 
women 
CDPO: 
As for second trimester, plus – 
Track adverse events reported vs 
expected 
MO (PHC): 
Track proportion of anemic women in 
the third trimester vs first trimester 
Track adverse events reported vs 
expected 
Agenda for BLAC/BLRM: 
Clusters / sectors / villages reporting 
higher proportion of anemic women 
in late pregnancy 
Coverage in non-ICDS areas 
Problems identified by CDPO / MO 
Supply issues 
Training needs 
Unusual problems 
Lifecycle period 2: From Delivery to 28 days (i) Delivery and the first two hours 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (Mother): 
Carry out birth plan 
Initiate BF within 1 hour 
Skin-to-skin contact for very 
small newborn, as advised by 
CA 
Recognize danger signs in self 
(bleeding, prolonged labor, 
premature labor, swelling on 
face/hand/genitals, severe 
headache) and newborn (not 
feeding well) and seek 
immediate care 
Family (husband, in-laws, 
parents): 
Carry out birth plan 
In case of home delivery, 
summon TBA and CA at onset 
of labor 
Ensure 5 cleans and use of 
clean delivery kit 
Avoid bath, immediately after 
tying and cutting cord, wipe 
newborn dry and wrap with 
multiple layers of dry, clean 
cloth 
Assist early BF, as soon as 
baby is wrapped up 
In case of danger signs in 
mother and newborn, seek 
immediate care 
Change Agent: 
Be present with the family at childbirth 
Help summon trained birth attendant 
Inform AWW of impending birth 
Ensure that preparations for childbirth are as 
planned 
Help ensuring 5 cleans, and use of clean 
delivery kit where available 
Ensure avoidance of bath and of provision of 
adequate warmth 
Support early and exclusive BF 
Assist in recognition of danger signs in mother 
and in activating contingency plans 
Assess feeding, activity and maturity of 
newborn, and determine need for referral 
Leave the home only after making sure that 
bleeding is normal and newborn has taken a 
good first breast feed 
TBA: 
Practice 5 cleans (use clean delivery kit OR 
clean site, hands, blade, thread, cloth) 
Avoid bath, immediately after tying and cutting 
cord, wipe newborn dry and wrap with multiple 
layers of dry, clean cloth 
Initiate BF immediately after wrapping and 
counsel on EBF; avoid pre-lacteal feeds 
Identify danger signs in mother and assist 
family in decision making for referral to 
predetermined facility 
Leave the home only after bleeding has 
stopped and breast feeding is well-established
CBO: 
Support family in taking correct decisions and 
acting upon them 
AWW: 
Try to attend childbirth whenever 
possible 
Weigh child immediately after birth 
(or within the first week); record 
the weight and report to ANM 
Assist in recognition of danger 
signs in mother and in activating 
contingency plans 
Assess birth weight, feeding, 
activity and maturity of newborn, 
and determine need for referral 
Record and report birth, death, 
and any adverse events 
ANM: 
Try to be present at every home 
delivery that happens during a visit 
to the village 
Meet family, AWW, TBA and CA at 
first opportunity after delivery and 
obtain details of the delivery; 
assess whether everything went 
according to plan; discuss 
deviations from plan with AWW, 
TBA and CA 
Observe and assess newborn and 
examine mother 
Support CA or AWW in providing 
care for small newborn 
Record and report birth, death and 
adverse events 
Supervisors: 
Enquire about each delivery in each 
village, and its outcome 
Probe all adverse events (stillbirths, 
deaths, complications, referrals, etc), 
identify where things went wrong and 
record observations 
CDPO: 
Track adverse events reported; 
compare with expected 
Compare reported with expected 
births, analyze by sector and enquire 
about discrepancies 
MO (PHC): 
Track adverse events reported; 
compare with expected 
Compare reported with expected 
births, analyze by cluster and enquire 
about discrepancies, with special 
attention to non-ICDS areas 
Agenda for BLAC/BLRM: 
Take note of all adverse events 
Discuss serious maternal 
complications and deaths 
Discuss preventable newborn deaths 
For each adverse event, identify 
system and behavior failures; identify 
positive deviants and possible 
solutions, and plan strategies for 
correction 
Lifecycle period 2: From Delivery to 28 days (ii) 2 hours post-delivery to 7 days post-delivery 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (mother): 
Practice exclusive breastfeeding 
Consume 1 additional meal/day 
Eat available variety of foods, including 
cereals, pulses, green leafy vegetables, other 
vegetables and fruits. Consume milk, milk 
products, meat, fish and eggs if available and 
feasible 
Consume SN/THR 
Consume IFA 1 tablet daily, or as advised by 
CA/AWW/ANM 
Seek assistance from CA/AWW for newborn 
who does not feed vigorously; provide 
expressed breast milk with clean cup and 
spoon as long as breast feeding is not 
adequately vigorous 
Provide more frequent feeding for a small 
newborn 
Continue to keep the newborn warm; take 
special care to ensure this constantly in a 
small newborn 
Minimize number of people who handle the 
newborn; ensure this particularly in the small 
newborn 
Seek care in hospital for danger signs in 
mother (bleeding, high fever, foul discharge) 
or newborn (poor feeding, poor activity) 
Seek BCG/OPV 0 (if hospital-delivered) 
Family (husband, parents, in-laws): 
Encourage EBF 
Provide 1 additional meal/day, help improve 
quality of diet, encourage consumption of SN 
Change Agent: 
Visit home 12 hours after 
birth and reassess mother 
and newborn 
Visit home once daily for this 
week if all is well 
Visit home more frequently 
on the first day and twice 
daily for the first week when 
the newborn needs special 
care 
Assess newborn for feeding 
and activity at each visit, and 
refer to predetermined facility 
if these are not satisfactory 
Recognize and address 
problems with breastfeeding 
Encourage diet, rest, IFA 
Recognize danger signs in 
mother and refer to 
predetermined facility 
TBA: 
Avoid bathing newborn till the 
end of the week 
Encourage EBF 
Encourage family to follow 
advice of CA/AWW/ANM 
CBO: 
Support family in taking 
correct decisions in care and 
referral of mother and 
AWW: 
Visit home on day of birth, day 3 and 
day 7, or daily if birth weight less 
than 2 kg, or with problems 
Weigh newborn at the earliest during 
the week, and record as birth weight 
Counsel on EBF and problem 
solving, refer if necessary 
Support mother in care of small 
newborn: help identify feeding that is 
less than adequately vigorous, help 
in expressing breast milk and using 
clean means to feed expressed 
breast milk 
Recognize danger signs in mother 
and newborn and take timely action 
for referral 
ANM: 
Meet family, AWW, TBA and CA at 
first opportunity after delivery and 
obtain details of the delivery; assess 
whether everything went according to 
plan; discuss deviations from plan 
with AWW, TBA and CA 
Observe and assess newborn and 
examine mother 
Support CA or AWW in providing 
care for small newborn 
Address breastfeeding problems 
Record and report birth, death and 
adverse events 
Begin counseling about plans for the 
Supervisors: 
Make visits to homes reporting recent 
births 
Enquire about each delivery in each 
village, and its outcome 
Probe all adverse events (stillbirths, 
deaths, complications, referrals, etc), 
identify where things went wrong and 
record observations 
Identify and address problems in 
AWW making immediate postpartum 
visit 
CDPO: 
Track adverse events reported vs 
expected 
Compare reported vs expected 
births, analyze by sector and enquire 
about discrepancies 
Address problems related to AWW 
finding time to provide newborn care 
MO (PHC): 
Track adverse events reported vs 
expected 
Compare reported vs expected 
births, analyze by cluster and enquire 
about discrepancies 
Track and analyze newborn deaths; 
make visits where possible to homes 
with newborns 
Agenda for BLAC/BLRM: 
Support mother in care of small newborn: 
help identify feeding that is less than 
adequately vigorous, help in expressing 
breast milk and using clean means to feed 
expressed breast milk 
Recognize danger signs and take timely 
action for referral 
newborn, and acting upon 
them 
Provide additional support to 
family to care for small 
newborn, and for referrals 
Mark birth / death / adverse 
event on map 
next child and for contraception Take note of all adverse events 
Discuss serious maternal 
complications and deaths 
Discuss preventable newborn deaths 
For each adverse event, identify 
system and behavior failures; identify 
positive deviants and possible 
solutions, and plan strategies for 
correction 
Identify and address training needs 
Lifecycle period 2: From Delivery to 28 days (ii) 8 days post-delivery to 28 days post-delivery 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (mother) 
Practice EBF 
Consume 1 additional meal/day 
Eat available variety of foods, including 
cereals, pulses, green leafy vegetables, 
other vegetables and fruits. Consume 
milk, milk products, meat, fish and eggs if 
available and feasible 
Consume SN/THR 
Continue to take rest and be with the 
newborn as much as possible; keep the 
small newborn close to body all the time, 
and check frequently to ensure that 
palms and soles are not cold 
Consume IFA 1 tablet daily, or as 
advised by CA/AWW/ANM 
Continue extra care of small newborn 
(warmth, feeding, minimal handling) 
Decide on birth spacing method to delay 
next birth, or appropriate method to limit 
conceptions; seek counseling from ANM 
Family (husband, parents, in-laws): 
Encourage EBF 
Provide 1 additional meal/day, help 
improve quality of diet, encourage 
consumption of SN (for mother) 
Continue extra care of small newborn 
Provide support so that the mother has 
time for rest and for care of the newborn 
Support mother in her decision to adopt 
spacing method; husband be equally 
responsible for decision and action 
Change agent: 
Visit home once a week if all is 
well and once a day in the 
second week after birth for 
newborn needing special care 
Encourage EBF, warmth, diet, 
rest, IFA 
Advise timely DPT1/OPV1 
Recognize danger signs in 
mother and newborn and refer 
to predetermined facility 
Identify the need for 
contraception and inform ANM 
Refer any problems to ANM 
TBA: 
Support advice given by 
CA/AWW/ANM 
Ensure exclusive breast 
feeding; avoid water or other 
liquids or solids 
CBO: 
Support family in taking correct 
decisions in care and referral of 
mother and newborn, and 
acting upon them 
Track families ensuring early 
breastfeeding 
Provide additional support to 
care for small newborn 
Mark birth / death / adverse 
event on map 
AWW: 
Visit home about 
once a week 
Counsel on EBF and 
any breast feeding 
problems 
Counsel on weight 
gain of child 
Report newborn 
deaths to ANM 
Identify need for 
contraception and 
refer to ANM 
Keep a tag on the 
mother and child via 
the CA 
Keep CBO informed 
Refer any problems 
to ANM 
ANM: 
Track health of 
newborn with help of 
CA/AWW 
Handle referrals 
Supervisors: 
Visit homes with adverse events and try to understand 
events; report to block 
Enquire about outcomes of mothers and newborn with 
danger signs / referrals 
Observe special care of small newborn, identify 
problems, report 
Ensure pregnancy outcome data have been 
accurately recorded and reported 
Identify villages with exemplary behavior changes 
CDPO: 
Compare reported vs expected neonatal deaths; verify 
a sample; compare different sectors and probe 
discrepancies 
Identify positive deviants and show recognition 
Insist on birth weights 
MO (PHC): 
Probe a sample of newborn deaths to understand 
cause of death 
Probe quality of care issues in special care of small 
newborn, identify and address training needs 
Identify TBA training needs; other training needs 
Handle referrals 
Agenda for BLAC/BLRM: 
Take note of all adverse events 
Discuss serious maternal complications and deaths 
Discuss preventable newborn deaths 
For each adverse event, identify system and behavior 
failures; identify positive deviants and possible 
solutions, and plan strategies for correction 
Identify and address training needs 
Lifecycle period 3: 1-5 Months (until end of sixth month) 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (mother): 
Practice EBF 
Seek BCG/DPT/OPV as advised 
by CA/AWW/ANM 
Practice EBF 
Consume 1 additional meal/day, 
Eat available variety of foods, 
including cereals, pulses, green 
leafy vegetables, other 
vegetables and fruits. Consume 
milk, milk products, meat, fish 
and eggs if available and 
feasible 
Consume SN 
Decide on birth spacing method 
if need to delay next birth; seek 
counseling from ANM 
Continue to monitor for child and 
self 
Family (husband, parents, in￾laws): 
Seek BCG/DPT/OPV as advised 
by CA/AWW/ANM 
Encourage EBF; avoid giving 
water or other liquids or solids to 
the child 
Provide 1 additional meal/day, 
help improve quality of diet, 
encourage consumption of SN 
(for mother) 
Support mother in her decision to 
adopt spacing method; husband 
be equally responsible for 
decision and action 
Change agent: 
Remind family of BCG/DPT/OPV 
vaccine days (on NHDs); counsel 
on immunization 
Counsel on EBF and help resolve 
breastfeeding problems 
Counsel on diet, rest and IFA 
Identify need for contraception 
and refer couple to ANM 
Identify mothers who return from 
parental village, expedite 
vaccine/contraceptive use 
Encourage to continue monitoring 
for child and self 
TBA: 
Support advise given by 
CA/AWW/ANM 
CBO: 
Track vaccine progress of each 
child on map; identify left-outs 
and drop-outs 
Identify and discuss families 
having problems with EBF 
Identify and discuss families 
having problems with using 
contraceptives 
Identify supply problems related 
to vaccines, contraceptives, etc., 
and take appropriate action 
through PRI 
AWW: 
Weigh child monthly at NHD 
Visit family once in second, 
third and fourth months to 
remind about vaccines, and in 
sixth month to begin advising 
about CF from seventh month 
Encourage EBF; discourage 
giving water 
Track vaccine / NHD left-outs 
and drop-outs with help of 
CAs, identify problems, 
discuss with ANM/supervisor 
Refer problems with 
contraceptive use to ANM 
Identify mothers who return 
from parental village, 
expedite 
vaccine/contraceptive use 
ANM: 
Encourage EBF 
Counsel on diet, rest and IFA 
Track NHD left-outs and 
drop-outs with help of 
CAs/AWW 
Provide FP services and 
counseling 
Handle referrals, including 
STIs 
Ensure cold chain 
maintenance for vaccines 
Supervisors: 
Use social map to assess left-out areas from all 
services 
Identify any geographical clusters of left-outs and 
drop-outs and probe reasons 
Observe NHDs and identify gaps in training or 
supplies 
Attend to specific problems brought up by 
AWW/ANM 
Identify issues related to cold chain and follow-up 
care of vaccine and of FP method users 
Note and report problems to block 
CDPO: 
Identify sectors / villages having good / poor 
coverage of vaccination (use DPT3/OPV3 within 6 
months age as indicator), probe reasons 
Track proportion of normal grade children 
MO (PHC): 
Identify clusters / villages having good / poor 
coverage of vaccine and FP services, probe reasons 
Handle referrals 
Address quality of care issues in vaccination and FP 
services, particularly cold-chain and adequacy of 
follow-up support for families with problems 
Tag complication rates 
Agenda for BLAC/BLRM: 
Review coverage situation cluster-wise and village￾wise if needed 
Identify and address gaps in cold chain and supplies 
Identify and address training needs 
Lifecycle period 4: 6-12 Months (i) 6-8 months (until end of 9th month) 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO (PHC) 
Woman (mother): 
Practice appropriate complementary 
feeding 
Feed small frequent meals 2-3 
times/day, of modified soft mashed 
(not watery) home foods e.g. rice, dal 
Introduce foods by feeding little at a 
time to gradually increasing the 
quantity, also increase consistency 
gradually 
Feed SN 
Feed child yourself, interact with 
child, feed from a separate bowl to 
assess how much the child has eaten 
Feed ill child frequently during illness 
and after child recovers. Breastfeed 
frequently 
Continue BF 
Switch to OCPs if appropriate and 
desired 
Continue to monitor for child and self 
Family (husband, in-laws, parents): 
Encourage and assist appropriate 
complementary feeding 2-3 times/d 
and feeding of SN 
Participate in feeding demonstrations 
Encourage continued BF 
Support mother in continued use of 
spacing method 
Change Agent: 
Counsel on appropriate CF 2-3 
times/d and continued BF 
Counsel on attending feeding 
demonstrations 
Closely follow quantity of food eaten 
by child 
Counsel on adequate weight gain 
Remind to complete measles and 
vitamin A at 9 months 
Inquire about mother’s comfort with 
spacing method; refer for further 
counseling if she has problems 
(Identify couple’s problems and refer) 
Encourage to continue monitoring for 
child and self 
CBO: 
Host feeding demonstrations 
Identify and discuss families unwilling 
to start CF 
Identify and discuss children not 
gaining adequate weight 
AWW: 
Counsel and record appropriate CF 
2-3 times/d, and continued BF 
Remind to complete measles and 
vitamin A at 9 months 
Weigh children monthly and counsel 
on adequate weight gain 
Identify and refer of sick and 
malnourished children 
Provide spacing method and 
continue counseling as required 
ANM: 
Treat sick children 
Counsel on adequate CF and 
continued BF 
Counsel mothers of children not 
gaining adequate weight 
Supervisors: 
Use social maps to identify any 
geographical clusters of left-outs and 
drop-outs and probe reasons 
Talk to sample of mothers about CF; 
identify current barriers to starting or 
sustaining CF and address them 
Observe NHDs and identify gaps in 
training or supplies 
Attend to specific problems brought 
up by AWW/ANM 
Identify issues related to cold chain 
and follow-up care of vaccine and of 
FP method users 
CDPO: 
Track proportion of normal grade 
children 
MO (PHC): 
Handle referrals 
Keep note of measles, pertussis and 
diarrhea cases in the community; act 
on detecting outbreaks 
Agenda for BLAC/BLRM: 
Identify and address gaps in supplies 
Identify and address training needs 
Identify villages / clusters with low / 
high proportions of normal grade 
children – identify causes of 
continued low proportions / positive 
deviants; strategize ways of making 
improvements 
Lifecycle period 5 & 6: 12 Months onwards 
Mother and the Family Change Agent, TBA, CBO AWW and ANM Supervisors, CDPO, MO 
(PHC) 
Woman (mother): 
Seek measles vaccination and vitamin A at 
9 months 
Practice appropriate CF 
Feed small frequent meals 4-5 times/d, of 
modified home foods e.g. rice, dal, 
chapatti, green leafy vegetables, other 
vegetables and fruits. Feed milk products, 
eggs, meat, fish if available and feasible 
Feed SN 
Assist the child in feeding, interact with 
child, feed from a separate bowl to assess 
how much the child has eaten 
Frequent feeding of ill child during illness 
and after child recovers. Breastfeed 
frequently 
Continue BF 
Give pediatric iron supplements 
Seek Vitamin A every 6 months until five 
doses are complete 
Continue use of appropriate FP method 
Continue to monitor for child and self 
Family (husband, in-laws, parents) 
Encourage appropriate complementary 
feeding 4-5 times/d and feeding of SN 
Encourage continued BF 
Encourage consumption of pediatric iron 
supplements 
Continue use of appropriate FP method 
Change Agent: 
Follow up 12 month olds and 
counsel on appropriate CF 4-5 
times/d and continued BF 
Help arrange for pediatric iron 
supplements 
Counsel on importance of 
pediatric iron supplements 
Counsel on adequate weight gain 
Counsel on continuation of FP 
method; refer problems to ANM 
Encourage to continue monitoring 
for child and self 
CBO: 
Identify children not being given 
adequate CF; encourage and 
support families in providing CF 
Identify and discuss children not 
gaining adequate weight 
AWW: 
Provide pediatric iron supplements 
Counsel on importance of pediatric 
iron supplements 
Counsel and record appropriate CF 4-
5 times/d and continued BF 
Counsel on adequate weight gain 
Counsel on completing any 
uncompleted vaccine schedules 
Remind to seek second dose of 
vitamin A at 18 months, and every 6 
months thereafter, and DPT/ booster 
ANM: 
Treat sick children 
Counsel on continued adequate CF 
and continued BF 
Provide DPT booster and six monthly 
doses of Vitamin A 
Counsel mothers of children not 
gaining adequate weight 
Counsel on continuing FP methods, 
and solve problems related to them 
Supervisors: 
Identify reasons for drop- outs 
from services 
Ensure accurate growth 
monitoring 
CDPO: 
Track proportion of normal grade 
children 
MO (PHC): 
Handle referrals 
Keep note of measles, pertussis 
and diarrhea cases in the 
community; act on detecting 
outbreaks 
Agenda for BLAC/BLRM: 
Closely monitor proportion of 
normal grade children and 
vaccine drop-outs