0
JUNE 2018
EDEAN PROOF OF CONCEPT 
Final Evaluation Report
Evaluating a peer learning and
community theatre intervention to 
increase fertility awareness and family 
planning use 
Georgetown University’s Institute for Reproductive Health and Save the Children
1
© 2018, Institute for Reproductive Health, Georgetown University
Recommended citation: 
Spindler, Esther and Lillian Ojanduru. 2018. EDEAN Proof of Concept Final Evaluation Report:
Evaluating a peer learning and community theatre intervention to increase fertility awareness
and family planning use. Fertility Awareness for Community Transformation (FACT) Project. 
Washington, D.C.: Institute for Reproductive Health, Georgetown University.
This report was authored by Esther Spindler and Lillian Ojanduru on behalf of the Fertility 
Awareness for Community Transformation (FACT) Project. This report and the FACT Project are 
made possible by the generous support of the American people through the United States 
Agency for International Development (USAID) under Cooperative Agreement No. AID-OAA￾A-13-00083. The contents are the responsibility of Georgetown University’s Institute for 
Reproductive Health and Save the Children and do not necessarily reflect the views of 
Georgetown University, USAID, the United States Government.
The authors would like to acknowledge the time and efforts of the community leaders,
Community Development Officers (CDOs) and EDEAN Peer Group Moderators from Moroto 
and Napak Districts of Karamoja who carried out the implementation of the EDEAN proof of 
concept. We also recognize the many consultants and organizations responsible for data 
collection and analysis, including Sharon Tsui, Bernadit Anyait, Lokoru Mariano, Marissa Ware, 
Hanley Futz and Dr. Nazarius Mbona Tumwesigye, Cissie Namanda and Dr. John Lule from 
Makerere University School of Public Health (MUSPH). The authors gratefully acknowledge 
Shannon Pryor, Dickens Ojamuge, Hamis Basalirwa, Christopher Kugonza and Flora Aber from 
Save the Children, and Sam Okello, Jennifer Langoya, Emily Arach, and Thomas Odong from 
IRH for supporting field implementation and monitoring of activities. We thank Victoria 
Jennings, Rebecka Lundgren, Dominick Shattuck and Lauren VanEnck for shaping and 
guiding the study, Danielle McCadden for her program leadership and management; Lauren 
DuComb, Darya Dokshina and Tori Stevens for their programmatic and coordination support; 
Sammie Hill for her communications support; Christina Riley for quantitative analyses advice,
and; last but certainly not least, Jessica Velcoff for spearheading the initial formative research 
and concept design phases over four years ago.
FACT Project
Institute for Reproductive Health | Georgetown University
1825 Connecticut Avenue, N.W., Suite 699
Washington, DC 20009 USA
irhinfo@georgetown.edu
www.irh.org/projects/FACT_Project
2
TABLE OF CONTENTS
LIST OF FIGURES AND TABLES ....................................................................................................... 3
ACRONYMS ................................................................................................................................... 5
EXECUTIVE SUMMARY................................................................................................................... 6
INTRODUCTION............................................................................................................................ 11
DESIGNING ‘EDEAN’................................................................................................................... 14
EDEAN PROOF OF CONCEPT .................................................................................................... 18
PROOF OF CONCEPT EVALUATION.......................................................................................... 23
FEASIBILITY RESULTS...................................................................................................................... 35
ACCEPTABILITY RESULTS.............................................................................................................. 57
EFFECTIVENESS RESULTS .............................................................................................................. 57
CONCLUSION AND RECOMMENDATIONS .............................................................................. 81
REFERENCES................................................................................................................................. 85
ANNEX .......................................................................................................................................... 87
3
LIST OF FIGURES
Figure 1. Map of Karamoja Sub Region ……………………………………………….……….…….. 13
Figure 2. EDEAN Intervention Development Timeline (2014 - 2017) …………………................. 14
Figure 3. Karimojong men pre-test initial designs of fertility awareness topics ……………….. 17
Figure 4. EDEAN Intervention Sites ……………………………………………...………….................. 18
Figure 5. EDEAN Activity Components ……………………………………….…………….…..…..... 20
Figure 6. Peer Group Members attending meetings, rehearsals and performances, ………. 35
Figure 7. Percentage of total audience members in age cohorts, all performances ……… 36
Figure 8. Average number of audience members at performances, by module …………... 36
Figure 9. Percentage of Moderators following guidelines ………………………………...…....... 37
Figure 10. Moderators ability to facilitate EDEAN activities …………………...………………….. 39
Figure 11. Audience understanding of content delivered, for performances only ……........ 40
Figure 12. Percent Time Spent by Activity, disaggregated by type of activity ………………. 42
Figure 13. Proportion of time spent on all activities, disaggregated by Month …………….... 43
Figure 14. Number of FP clients from April 2015 to March 2017, by age cohorts …………….. 44
Figure 15. Intervention women and men with high fertility awareness ………………...………. 60
Figure 16. Control women and men with high fertility awareness ………………………………. 60
Figure 17. Intervention women and men who intend to use modern FP …………….………... 64
Figure 18. Control women and men who intend to use modern FP ……………………………. 64
Figure 19. Exposure to EDEAN performances, by intervention men and women ……………. 74
Figure 20. Frequency of EDEAN performance attendance, among exposed participants 74
LIST OF TABLES
Table 1. EDEAN Training Phases ……………………………………………………………………....... 20
Table 2. Sample Schedule of Activities ………………………………………………….……………. 21
4
Table 3. Timeline of Study Activities ………………………………………………………………........ 24
Table 4. Feasibility Methods …………………………………………………………………………...... 25
Table 5. Acceptability Methods …………………………………………………...…….…………...... 26
Table 6. Effectiveness Methods ………………………………………………………..……………..... 27
Table 7. Fertility Awareness Measures and Survey Questions …………………………………….. 29
Table 8. Approximate number of days spent on implementation activities, by month ….… 41
Table 9. Average time preparing and facilitating EDEAN activities, by Peer Moderators ... 43
Table 10. Summary of Q&A with service provider / administrator …………...………………..... 45 
Table 11. Socio-demographic characteristics of participants, at baseline …………….…...... 57
Table 12. Exposure at endline, among participants in intervention communities ………….... 59
Table 13. Fertility awareness at baseline to endline, among intervention and control …….. 60
Table 14. Women and men who answered fertility awareness questions correctly ………… 61
Table 15. Women and men who answered “true” when asked about menstruation myths 62
Table 16. FP outcomes baseline to endline, among intervention and control ….................... 63
Table 17. Self-efficacy to access and use FP, among intervention and control ..................... 65
Table 18. Attitudes about FP, among intervention and control ………………………..……..... 66
Table 19. Social norms around FP, among intervention and control …………………....…….. 67
Table 20. Couple communication about FP, among intervention and control …………….. 68
Table 21. Community communication about FP, among intervention and control………… 69
Table 22. Intervention effect on high fertility awareness, using difference-in-difference ..… 71
Table 23. Intervention effect on intent to use modern FP, using difference-in-difference … 72
Table 24. Intervention effect on current modern FP use, using difference-in-difference ….. 73
Table 25. Logistic regression results for fertility awareness and FP, by EDEAN exposure ….... 75
Table 26. Logistic regression results for secondary outcomes, by EDEAN exposure …........... 76
Table 27. Odds of current modern FP use, by correct knowledge of fertile window …........ 77
Table 28. Odds of intent to use modern FP, by fertility awareness individual measures ….... 78
5
ACRONYMS
CDO Community Development Officer
CMC Centre Management Committee
ECCD Early Childhood Care and Development
EDEAN Emorikinos Daadang Etogogongo Alatanakin Ngidwe
EFA Exploratory Factor Analysis
FACT Fertility Awareness for Community Transformation
FAM Fertility Awareness-Based Methods 
FGD Focus Group Discussion
FP Family Planning
ICRW International Center for Research on Women
IDI In-Depth Interviews 
IRH Institute for Reproductive Health, Georgetown University
IPV Intimate Partner Violence
KII Key Informant Interview
LAM Lactational Amenorrhea Method
LOE Level of Effort
mCPR Modern Contraceptive Prevalence Rate
M&E Monitoring & Evaluation
ODK Open Data Kit
OR Odds Ratio
PMC Population Media Center
POC Proof of Concept
SDM Standard Days Method
ToT Training of Trainers
TFR Total Fertility Rate
USAID United States Agency for International Development
VHT Village Health Team
6
EXECUTIVE SUMMARY
INTRODUCTION
This report shares EDEAN’s story, from intervention design to proof of concept testing and final 
evaluation findings in increasing fertility awareness and improve FP use in Karamoja region.
EDEAN is grounded from formative research, which showed that misconceptions about 
contraceptive side effects and pregnancy risk are key barriers to family planning (FP) use in 
Uganda’s Karamoja region, where the modern contraceptive prevalence rate is 6.5% (UBOS 
and ICF 2016). Formative research suggests that the transition from a nomadic to sedentary 
lifestyle – in which men ceased migrating to cattle kraals for extended periods of time –
altered couple dynamics on the use of post-partum abstinence as a traditional child spacing 
practice (IRH 2014). Karamoja’s cultural transition suggests that unconventional, yet culturally￾sensitive activities are necessary to reach couples with fertility and FP information.
ABOUT EDEAN
To address this gap, EDEAN was piloted under the Fertility Awareness for Community 
Transformation (FACT) Project as a peer learning and community theatre intervention to 
diffuse fertility awareness and FP information among hard-to-reach Karimojong communities. 
With the goal of increasing demand for FP, EDEAN is an eight month community-based 
intervention, covering four modules: 1. Couple Communication; 2. Menstruation; 3. Fertility, 
and; 4. FP. In each community, 10 male and 10 female youths between 18-25 years old learn 
about the above four topics through a series of gender synchronized same sex and mixed 
sex Peer Group Meetings. These Peer Group Meetings were facilitated by a trained male and 
female Peer Moderator pair from their local community. After finishing each peer group 
meeting module topic, the 20 Peer Group Members rehearse, and then enact a set of three 
performances to the community at large. During the Peer Group Members’ theatre 
performances, the Peer Moderator pair ask reflection questions to the audience, catalyzing 
larger discussion and reflection on existing norms, beliefs and information on fertility 
awareness and FP within their community. More information about the EDEAN solution can 
be viewed through a micro-documentary found here.
EVALUATION DESIGN AND METHODS
To understand potential for scale, EDEAN was pilot tested through a proof of concept phase
in late 2016 – early 2017 in six agro-pastoral communities of Moroto and Napak Districts in the 
Karamoja region of Uganda. A quasi-experimental study using mixed-methods was 
conducted to assess the delivery and potential scalability of EDEAN, as well as the 
effectiveness of the approach in increasing fertility awareness knowledge and improving 
attitudes and behaviors toward FP use. Methods of verification included baseline and endline 
household surveys with independent representative samples of intervention and control 
7
households (n=400 intervention and n=200 control); FGDs with 32 Peer Group Members and 
48 community members; interviews with 9 key informants; attendance and systematic 
observational data of 80 peer group meetings, rehearsals and theatre performances, and; 
service statistics from 14 local health facilities. 
Primary outcomes included self-reports of current modern FP use; intent to use modern FP, 
and; fertility awareness scores. Secondary outcomes included knowledge, attitudes and 
behaviors related to couple communication, menstruation, fertility and FP. Analyses were 
conducted through two estimation approaches: 1. Pre to post community-level analysis; and 
2; Post-intervention exposure level analysis. Community level data were analyzed to 
understand intervention effect at larger scale within the community (regardless of 
performance exposure), while exposure level data were analyzed to better understand 
associations between those who attended the performances and potential improvements in
fertility awareness and FP outcomes. For both analyses, bivariate descriptive comparisons 
were conducted, followed by multivariate logistic regression showing adjusted odds ratios. 
Content analyses were conducted on qualitative data from FGDs with community members 
and peer group members, and KIIs with service providers and community leaders.
KEY FINDINGS
Key findings are presented below vis-à-vis three evaluation result components to better 
understand the model’s potential scalability, including: 1. Intervention feasibility, 2. 
Intervention effectiveness in increasing fertility awareness and FP, and; 3. Acceptability of 
the model among participating stakeholders and communities.
Feasibility Delivering EDEAN Activities
Participation and Reach. Approximately 78% of the Peer Group 
Members attended all of the EDEAN Peer Group Meetings, rehearsals 
and performances. EDEAN was well received in the communities with 
large engaged audiences. Process data show wide reach, with a total 
of 6,669 points of contact made in the six communities from June to 
December 2016 (average 89 audience members at each 
performance). 
Fidelity. Manual and messaging fidelity were closely assessed given 
Karamoja’s resource scarce and low-literate environment (74% of the 
intervention household sample had no formal education). Systematic observations show Peer 
Moderators were able to follow the guidelines and manuals almost all the time (80%), ensuring
fertility awareness and FP message fidelity. Other findings suggest the combination of scripted 
and “create your own drama” storylines ensured fertility awareness messaging fidelity, while 
allowing Peer Group Members to tap into their own cultural and societal experiences to 
deconstruct fertility awareness and FP beliefs and norms in their communities. 
Female Peer Moderator with 
EDEAN Manual (Credit: Bridge 
Films)
8
Addressing gender and power dynamics. Finally, given the rigid cultural gender roles and 
structures in Karimojong society, ensuring equitable participation and discussion among men 
and women could be challenging. However, Peer Group Members also shared during FGDs 
that the gender synchronized meetings, facilitated more participatory and equitable 
discussion during the same-sex, and later mixed-sex meetings. 
Acceptability and Diffusion in Communities
Community acceptability. Qualitative results show high community acceptability of the 
EDEAN model across community leaders, service providers, and community members. The 
high levels can be attributed to EDEAN’s approach of using the community resource persons 
as peer moderators and peer group members. Community Development Officers (CDOs), 
who are respected in the communities, also played a vital role in the EDEAN implementation. 
Interviewed community leaders and members consistently reaffirmed that EDEAN was 
compatible with their culture and traditions. Engaging elders in intervention activities was key 
in encouraging young couples to attend the performances and in dispelling myths and “fear” 
around fertility awareness and FP topics.
Health Service Linkages within EDEAN activities. Providers perceived the EDEAN activities as 
a demand generation activity that could complement and improve their day to day FP work, 
generating increased numbers of FP clients. The qualitative findings also showed that health 
workers were interested in the EDEAN activities, and in some cases, attended some of the 
performances. The findings suggest that referral mechanisms to health systems need to be 
strengthened. Community based referral systems are notoriously challenging. EDEAN used a 
paper-based referral through FP invitation cards. Future programs might explore use of
alternative and/or mobile referrals which have shown greater success (Lebetkin 2015). 
Diffusion of FA/FP information among community members. Those who attended the EDEAN 
performances shared fertility awareness and FP information with their friends, neighbors, 
among others and brought them to attend subsequent performances. The performance 
locations – conducted at ECCD centers open to all in the community – also facilitated access 
and diffusion of information. Karamoja’s mobility patterns facilitated the speed and feasibility 
of information to be diffused and shared among community members.
Effectiveness Increasing Fertility Awareness and FP
Community Participant Profiles. Approximately 74% of respondents in the intervention 
community had no education compared to 81% in the control community, consistent with 
Karamoja’s low education levels. On average, respondents had about 3 children and had 
been married for eight years across both intervention and control groups. Both intervention 
and control groups reported low use of ever modern FP use in their lifetime, about 21% and 
18% of intervention and control respondents, respectively. The majority of participants in both 
9
intervention and control communities reported using traditional method of FP, and more 
specifically periodic abstinence (55% intervention vs. 62% control).
Community-level results on fertility awareness and FP. The proportion of respondents with high 
fertility awareness scores increased in intervention communities from 15% to 75% from baseline 
to endline, respectively. Respondents living in an intervention community were more than 
seven time more likely to have a high FA score, relative to those in control communities (aOR 
7.3; p<.0001). The effect appears stronger when controlling for demographic variables (aOR 
9.3; p<.0001). In intervention communities, intent to use a modern FP method increased 
among surveyed women (from 25% to 29%; p=0.3427) and men (8% to 17%; p=0.009) over the 
six month period. Consistent with these data, respondents in intervention communities were 
approximately three times more likely to intend to use a modern FP method (aOR 3.1; 
p=0.026). There were no community-level increases in current modern FP use among both 
intervention and control groups. At endline, 6% of intervention women and 14% of 
intervention men reported current modern FP use. 
Exposure-level results on fertility awareness and FP. At endline, 79% of surveyed respondents 
in intervention communities had heard of the EDEAN performances, while 65% reported 
viewing at least one performance. A higher proportion of women in intervention communities 
viewed the performances (79%), relative to men (51%). Post-intervention exposure analysis 
comparing those who viewed at least one performance (in intervention community) vs. no 
performance (in intervention and control communities) show that viewers (n=260) were more 
than twice as likely to have higher fertility awareness scores (aOR 2.5; p=<.0001), than non￾viewers (n=340); six times more likely to seek information about FP from a health worker (aOR 
6.5; p<.0001), and; significantly less likely to report that modern FP methods cause side effects 
(aOR 0.2; p<.0001), compared to non-viewers. Viewers were also more six times more likely to 
report intent to use modern FP (aOR 6.3; p=<.0001). Women’s reports of current modern FP 
use were similar and insignificant among the two viewer groups, with the majority reporting 
continued use of periodic abstinence. 
Associations between fertility awareness and modern FP use. Using the pooled endline 
sample, chi-square tests of associations found some significant associations between intent 
to use modern FP, and correct fertility awareness knowledge about the following topics: 1. 
Healthy vaginal secretions are a sign of fertility (OR 3.2; p<.0001), 2. Vaginal secretions come 
during the middle of the menstrual cycle (OR 2.1; p=0.0017), and; 3. Men are fertile every day 
(OR 4.8; p<.0001). Only one fertility awareness measure was significantly associated with 
current modern FP use: respondents who correctly identified their pregnancy risk during the 
fertile window were twice more likely to use a current modern FP method relative to those 
who identified the fertile window incorrectly (OR 2.0; p=0.0491). Overall, the relative low 
prevalence of current modern FP use in the sample (about 10%), limited the ability to find 
robust associations between fertility awareness and current modern FP use. Future research 
should be conducted on populations with more prevalent modern FP use to better 
10
understand associations between fertility awareness and FP, and other associated secondary 
factors. 
LESSONS LEARNED
Overall, proof of concept findings suggest that: 1. Performances were well attended and 
accepted by community members; 2. Moderators and Peer Group Members were able to 
deliver the EDEAN performances with fidelity despite low literacy levels, and; 3. Demand and 
intent to use modern FP methods exist, but continued use of prolonged abstinence persists 
due to the combination of cultural abstinence practices and limited FP services availability in 
the region. The findings suggest that EDEAN’s peer learning and community theatre 
approach was effective in increasing fertility awareness, creating a more supportive 
environment for FP and increasing intent to use modern FP. Qualitative findings reaffirm that 
viewers’ motivation to use a method of FP was embedded in their desire to have healthy 
families, reduce malnutrition and child mortality, all common in the resource-scarce region 
of Karamoja. However, the gap between participants’ high reports of intent to use a modern 
FP method, relative to their low current FP use, suggest that SBC demand generation activities 
focused on fertility awareness and FP should be paired with strong service delivery platforms
and linkages to increase FP uptake. 
As a SBC group engagement approach, EDEAN is an acceptable and feasible model to 
implement by low literate volunteer community resource persons to increase demand for FP 
information and services. Future research should test the EDEAN peer learning and 
community theatre model, paired with strong service delivery, to generalize the findings to 
other populations and communities. 
11
INTRODUCTION
Background
In 2013, Georgetown University's Institute for Reproductive Health (IRH) was awarded funding 
from the United States Agency for International Development (USAID) to implement the 
project, “ fertility awareness for Community Transformation (FACT)” (Cooperative Agreement 
No. OAA-A-13-00083). FACT is a five-year United States Agency for International Development 
(USAID)-funded project implemented by the Institute for Reproductive Health, Georgetown 
University (IRH) in partnership with the International Center for Research on Women (ICRW), 
Population Media Center (PMC), and Save the Children. USAID, IRH, and project partners 
envisioned the FACT Project as addressing fertility awareness and FP outside the traditional 
health system. 
The project aims to foster an environment where women and men can take actions to 
protect their reproductive health throughout the life-course. FACT is a research, intervention, 
and technical assistance project that is developing and testing unique interventions in Nepal, 
Rwanda and Uganda. These interventions are investigating two primary hypothesis:
 Increased fertility awareness improves family planning (FP) use
 Expanded access to fertility awareness -based methods (FAM) increases uptake of FP 
and reduces unintended pregnancies
As part of the first hypothesis, the goal of the Emorikinos Daadang Etogogogoto Alatanakin
Ngidwe (EDEAN) intervention - meaning “Let’s Come Together and Strengthen Child 
Spacing” in Nga’Karimojong - is to increase fertility awareness and improve FP use through 
existing community groups among hard to reach populations outside of the health system. 
Under this intervention, trained community peer group members diffused fertility awareness 
and FP messages through forum theatre performances to peer networks and the community 
at large in the Karamoja region of Uganda. 
About this Report
This report shares EDEAN’s story from formative research to proof of concept design through
final evaluation results. In the proof of concept testing phase, we were interested in 
understanding the feasibility, acceptability and effectiveness of EDEAN - as a peer learning 
and community theatre group intervention - in increasing fertility awareness and FP use. As 
such, this report presents not only information about the project solution cycle and
implementation processes of EDEAN, but the results of three different study components
related to “feasibility”, “acceptability” and “effectiveness” found as three separate results 
sections further below. The lessons learned and discussion sections of this report bring the 
different evaluation findings together, highlighting overarching implications and 
opportunities for further research. 
12
What is Fertility Awareness?
A systematic review of global literature conducted by IRH found that lack of fertility 
awareness and weak knowledge of pregnancy risk can be key barriers to FP access and use
(IRH 2013). Specifically, some evidence shows that inaccurate assessment of pregnancy risk 
reduces FP use at critical points in the life cycle (Kauyate 2010; Makinwa-Abedusaye 1992; 
Kaye, Suellentrop, & Sloup 2009). As well, in some settings, FP use is stigmatized or people 
believe it is not within human power to control fertility (Sedgh et al. 2007). Others may believe 
pregnancy requires frequent sex, or that having had sex without becoming pregnant 
indicates they may be infertile. Other 
researchers have found that negative 
perceptions of FP, including lack of self￾efficacy to control fertility, results in failure to 
discuss and seek FP services (Kauyate 2010; 
Makinwa-Abedusaye 1992; Kaye, Suellentrop, 
& Sloup 2009).
Based on the literature review conducted by 
IRH, FACT conceptualized and defined fertility 
awareness as actionable information about 
fertility throughout the life course and the 
ability to apply this knowledge to one’s own 
circumstances and needs. The ability to apply 
this information to one’s life requires individual 
knowledge, personal experience and skills, as 
well as an environment within the family and 
community that enables people to undertake 
appropriate actions.
Family Planning in Uganda
To reduce global unmet need for FP, the goal of the Family Planning 2020 (FP2020) initiative 
is to enable 120 million additional women and girls to use contraceptives by 2020. In Uganda, 
the total fertility rate (TFR) and unmet need for FP remain high, despite recent improvements. 
As such, Uganda’s FP2020 goal is to reduce unmet need to 10% by 2022 (FP2020 2017). Most 
recent data suggest improving FP trends at the national level. As of 2016, 28% of currently 
married women had an unmet need for FP, dropping substantially from 34% unmet need in 
2011 (Uganda Bureau of Statistics [UBS] and ICF 2018; 2012). The country’s modern 
contraceptive prevalence rate (mCPR) increased from 26% to 35% between 2011 and 2016, 
with about 1/5 of currently married women using injectables (19%; UBOS and ICF 2018; 2012). 
Knowledge of FP methods is almost universal in Uganda. When asked if they had heard of FP
methods, the majority of Ugandan women and men were aware of at least one method
(99%), and on average had heard of at least nine methods (UBOS and ICF 2018). Male 
 FERTILITY AWARENESS DEFINED
Fertility Awareness is actionable 
information about fertility throughout the 
life course and the ability to apply this 
knowledge to one’s own circumstances and 
needs. It includes basic information about 
the menstrual cycle, when and how 
pregnancy occurs, the likelihood of 
pregnancy from unprotected intercourse at 
different times during the cycle and at 
different life stages, and the role of male 
fertility. Fertility awareness also can 
include information on how specific FP 
methods work, how they affect fertility, 
and how to use them; and it can create the
basis for understanding communication 
about and correctly using FP.
13
condoms are the most known method (97% of women; 98% of men), followed by injectables
(96% and 89%) and pills (94% and 89%). Emergency contraception is the least known of 
modern methods (38% and 47%), followed closely by SDM (42% and 43%) and LAM (64% and 
48%; UBOS and ICF 2018).
The Karimojong in Uganda
The Karimojong are a cluster of communities settled in the 
semi-arid region of Northeast of Uganda. Known as a 
nomadic agro-pastoralist society, the Karimojong of Uganda 
are currently undergoing rapid social changes, shifting away 
from cattle-centered livelihoods to settled agriculture 
economies. The Karimojong have long faced challenges to 
maintaining their way of life including external political and 
economic pressure to settle down and ‘modernize.” The 
sparse literature on Karamoja’s transition shows that a
government disarmament campaign of the region in the 
2000’s, boosted security and stability in the region but also 
altered Karamoja’s social fabric (Stites and Akabwai 2009). FACT formative research 
conducted in 2014 suggests that this transition to a more sedentary lifestyle – in which men 
ceased to migrate to mobile cattle kraals for extended periods of time – altered couple 
dynamics around fertility and spacing. Specifically, couples’ ability to negotiate post-partum 
and periodic abstinence as a natural child spacing practice has become more difficult with 
the societal transition (IRH 2014).
To date, the Karamoja sub-region has some of the lowest 
national human and economic development indicators in 
Uganda. While the rest of Uganda progressed significantly 
over the last five years, FP-related indicators in Karamoja have 
stagnated. Karamoja continues to have the highest total 
fertility rate (TFR), jumping from 6.4 to 7.9 between 2011 to 
2016 (UBOS and ICF 2018; 2012), and the lowest rates of 
modern contraceptive use among currently married women, 
stagnating at 7% between 2011 and 2016, compared to the
current national average of 35% (UBOS and ICF 2018; 2012).
Karamoja also has the lowest demand for FP nationally (27%), 
compared to other regional ranges of 60% to 72% (UBOS and 
ICF 2018). Use of traditional methods is higher in Karamoja, 
relative to other regions. Women in Karamoja have the longest median duration of 
postpartum abstinence use, at about 11 months, compared to the 3 month national duration 
median (UBOS and ICF 2018). Following traditional family norms, Karimojong women desire 
large families (mean = 7.2 children; UBOS and ICF 2018) and close to 24% of women have 
begun childbearing between the ages of 15 to 19 years of age (UBOS and ICF 2018). Family 
Figure1| Karamoja sub-region of 
Uganda.
FERTILITY AND FP IN 
KARAMOJA
 Women on average 
have close to 8 children
(TFR: 7.9)
 Less than 1 in 10 
women use a modern 
FP method (mCPR: 7%)
 Close to 3 in 10 women 
have a demand for FP 
(Demand for FP: 27%)
14
size is also influenced by polygamy, which is fairly common (27% of men report multiple wives; 
UBOS and ICF 2018). 
Interestingly, the combination of traditional post-partum abstinence and low modern FP use 
result in a lower unmet need for FP than the national rate (20% of currently married Karimojong 
women compared to the 28% national average; UBOS and ICF 2018). Finally, evidence 
suggests that most Karimojong women and men do not seem to be receiving FP information 
at key points in the health system. About 45% of women not using a FP method discussed FP 
with a field worker or at a health facility in the past year (UBOS and ICF 2018). Opportunities
are also missed when women are present in health centers, as only 18% of postpartum 
mothers received FP counseling prior to discharge (UBOS and ICF 2012). These data suggest 
gaps in FP information and dissemination, providing an opportunity to spread FP messages
and demand among Karimojong women and men among harder accessed communities.
DESIGNING ‘EDEAN’
The EDEAN intervention was guided by the Solution Development Cycle - an iterative process 
for the discovery, design, and development of solutions - using formative research, 
participatory design, and intervention testing to develop a scalable solution (defined here as 
intervention). This section gives a brief snapshot of EDEAN’s intervention development from 
formative research to proof of concept testing. 
Figure 2| EDEAN Intervention Development Timeline (2014 - 2017).
Platform Assessment (April 2014)
Early Childhood Care and Development (ECCD) centers in Karamoja were identified as an 
ideal platform to integrate a potential community-based intervention on fertility awareness 
and FP. ECCDs in Karamoja aim to increase access to quality early childhood care and 
education, by providing 3-4 hours of daily caregiving activities to young children aged 0 to 8 
years old, in addition to monthly parenting sessions. Centers are managed by Center 
Management Committees (CMCs), comprised of 10 members, typically including teachers, 
15
caregivers and community leaders. These are members elected, respected and trusted by 
the communities. The ECCD management committees are provided with capacity building 
and training on parenting skills and health topics, and members conduct events in the 
community to share what they learned with other parents. This community platform was
chosen because Save the Children – as the FACT implementing partner - has been 
implementing community level activities through the ECCD centers in the Karamoja region.
In addition, the centers were well respected and trusted by the community. Finally, given that 
ECCD centers target young children, the platform was seen as a potential space to reach 
the target age group of 18 to 35 years old among parents attending the ECCD centers.
Community Development Officers (CDOs) were also identified as potential trainers and 
support resources of potential intervention activities. CDOs represent the Community Based 
Services Department, which has a mandate to mobilize communities to participate in the 
development process. CDOs are often respected and trusted by the community, making 
them key stakeholders for introducing topics covering fertility and FP to communities.
Formative Research (June - Aug 2014)
Following the platform assessment, formative research was conducted over a two month 
period in the Nadunget and Ngoleriet sub-counties of Karamoja to inform the intervention 
design. A full formative research report, highlighting the results from 20 focus group discussions 
and 12 in-depth interviews with ECCD members, ECCD caregivers and CMC members, 
community leaders, and FP providers, can be found here. A brief summary is provided below. 
Briefly, the FACT formative research results showed that despite the introduction of new health 
innovations and services, knowledge related to fertility and FP is still low among the 
Karimojong. With regard to fertility, participants, and in particular men, had little knowledge 
and many misconceptions about menstruation, fertile periods, vaginal secretions, and 
ultimately the likelihood of pregnancy. Participants were accepting of FP, linking child￾spacing to the health of children as important Karimojong values. This acceptability was often 
cited with the common traditional practice of extended post-partum abstinence, facilitated 
by men leaving to herd cattle at mobile cattle kraals for long periods of time. 
At the same time, formative research participants were increasingly familiar with modern 
methods of FP, including Implanon ® implants and Depo-Provera ®, despite very low use. 
Many women found these methods desirable, but most knew little about how the methods 
work and many held mistaken beliefs about FP. Further, they voiced strong concerns about 
side effects. Side effect-free FP methods were perceived as more acceptable, and 
participants had some exposure to Moon Beads1 and Lactational Amenorrhea Method 
(LAM). FP was said to be viewed as a woman’s issue, and low knowledge about modern FP 
among men was also said to be a barrier to couple communication. Men were identified as 
the primary decision-makers, and women were said to potentially face violence when 
 
1 Moonbeads is the marketed name for CycleBeads in Uganda, which was introduced in country through Population 
Service International (PSI).
16
bringing up the matter. Word of mouth and songs were identified as traditional 
communication channels to pass on information among communities.
Concept Design: From ‘Solution’ to ‘EDEAN’ (Oct 2014 – Jan 2015)
Findings from the formative research were used to design a locally driven solution through a 
series of meetings and workshops. A concept development workshop was conducted with 
local stakeholders, during which participatory design approaches including human centered 
design were used to identify potential key components of the solution. Suggestions were 
generated for feasible mediums for delivery of information and messages, resulting in a basic 
concept for the solution. During formative research dissemination events in four communities 
of Moroto and Napak, local community members discussed and voted on a local name for 
the “mystery solution.” The name “Emorikinos Daadang Etogogogitoth Alatanakithi Ngidwe
(EDEAN),” meaning “Let’s Come Together and Strengthen Child Spacing” in the 
Nga’Karimojong language. The name “EDEAN” was selected by Karimojong community 
members to emphasize the importance of bringing men and women together to discuss
fertility awareness and FP topics. During these series of meetings and based on formative 
research results, EDEAN was designed and developed to engage Karimojong women and 
men through community forum theatre to increase their knowledge about fertility awareness 
and FP and motivate their utilization of local FP services and methods.
Entry Meetings and Stakeholders Buy-In (January 2015)
To introduce the EDEAN intervention and develop buy-in at intervention sites, a series of 
introductory activities were conducted in the communities. In January 2015, Save the 
Children and IRH conducted district launch workshops with district officials in Moroto and 
Napak districts. The workshop included identification of roles of various stakeholders, including 
expected support from the district; composition and roles of a local advisory committee; and 
how the project would link to other FP programs in the districts. 
IRH and Save the Children held an inaugural meeting of a local Technical Advisory Group in 
Moroto in January 2015. This Karamoja Advisory Team (KAT) was formed as an advisory body
to provide guidance during the course of the intervention design, implementation and 
monitoring. Committee members were selected from stakeholders in the sexual and 
reproductive health (SRH) field, and included local district and sub-county officials, Christian 
and Muslim faith-based organizations, FP service providers, UN agencies, and representatives 
from the target community. KAT members had quarterly meetings in which they initially
reviewed the EDEAN materials for cultural appropriateness, conducted visits to the field, 
discussed process and monitoring data, addressed implementation issues and guided 
intervention processes forward. 
17
Material Development and Pre-testing (2015)
EDEAN materials were 
developed through
consultations with community 
members and local stakeholders. 
Visuals and key messages 
developed from the design 
activities were pretested within 
the communities before final 
materials for the intervention 
were developed. Messages 
pretested included key fertility 
awareness topics and FP 
messages such as: couple 
communication, menstruation, 
fertility, fertile window and FP. 
The purpose of this pre-testing 
phase was to design with key messages and visuals that were culturally appropriate and 
relevant to local communities. As seen in the picture to the above, key fertility awareness and 
FP messages were presented in the form of visuals or posters locally drawn by the 
communities. The final EDEAN materials included a Peer Moderator manual with meeting 
lessons plans and theatre performance storylines, based on social and behavior change 
(SBC) theory and community theatre practices. The manual curricula covered four modules 
including: 1. Couple communication; 2. Menstruation; 3. Fertility, and; 4. Family planning. Peer 
moderator manuals were accompanied by locally-adapted visual aids, including large 
posters, activity cards and reminder cards for the four topics, and game cards.
The SBC-designed theatre storylines were pre-tested by a Uganda-based theatre group, 
RAFIKI, following three days of orientation on the EDEAN intervention and key messages. 
RAFIKI reviewed the outlines of activities and short scripts to be conducted, held rehearsals 
and performed in three Karimojong communities. This pre-testing phase informed several key 
design components. First, the pretesting revealed that the performances were well-attended, 
and that the topics in the storylines were acceptable to the community. The decision was 
also made to include moderation after each scene of the drama as opposed to waiting until 
the end of the skit to hold moderation. This would allow for reflection on the different topics 
included in each scene facilitating ongoing discussion and reflection throughout the 
performance. Finally both scripted story line and “create your own drama” prompts were 
included in the final moderator manual to allow contextualized adaptation of key fertility 
awareness and FP messages.
Figure 3 | Karimojong men pre-test initial concepts and designs on fertility 
awareness topics |Photo Credit: Dickens Ojamuge, Save the Children 
International
18
EDEAN PROOF OF CONCEPT
Grounded in the formative research, design and pre￾testing phases described above, EDEAN was tested 
through a six month proof of concept phase from June
to Dec 2016 in six ECCD catchment areas of Moroto 
and Napak districts of Karamoja. Intervention 
components are fully described below. 
Intervention Summary
With the goal of increasing demand for FP, EDEAN was tested as an eight month community￾based intervention, covering four modules: 1. Couple Communication; 2. Menstruation; 3. 
Fertility, and; 4. FP. In each ECCD catchment area, 10 male and 10 female youths aged 18-
25 years old learned about the above four topics through a series of gender synchronized 
same sex and mixed sex Peer Group Meetings. These Peer Group Meetings were facilitated 
by a trained male and female Peer Moderator pair selected by local community members 
during entry meetings. After finishing each peer group meeting module topic, the 20 Peer 
Group Members rehearse, and then enacted a set of three performances to the community 
at large. During the Peer Group Members’ theatre performances, the Peer Moderator pair 
ask reflection questions to the audience, catalyzing larger discussion and reflection on 
existing norms, beliefs and information on fertility awareness and FP within their community. 
Trainers, Peer Moderators and Peer Group Members
Trainers: The Save the Children and IRH team trained CDOs, in addition to Assistant CDOs, as 
EDEAN trainers and supervisors to guide and support implementation of EDEAN activities. 
CDOs were chosen as trainers since they are the focal point persons for all community-based 
services at sub county level, hereby ensuring intervention acceptability and sustainability. 
After conducting trainings with Peer Moderators, the CDOs routinely provided supportive 
supervision and monitoring of the peer moderators and their group members to ensure that 
they adhered to the protocols and provide quality implementation. CDOs engaged the peer 
moderators in monthly reflection meetings to share project progress, any challenges faced 
and possible solutions for improvement. Generally, the CDOs were at the forefront of all the 
project activities in their respective sub counties.
Peer Group Members: The EDEAN peer group members were selected by CMC members and 
parents of children attending the ECCD center, and other community stakeholders. Selection 
criteria included being male or female, between the ages of 18 to 25 years old and being in 
marital union with a partner. In each ECCD catchment area, a total of 10 young females and 
10 young males were selected to participate as volunteer peer group members. Peer Group 
members were then trained by the peer moderators to conduct peer meetings, rehearsals 
and performance on fertility awareness topics for their communities. 
Figure 4| EDEAN Intervention Sites
19
Peer Moderators: Peer Moderators were selected by CMC members and EDEAN peer group 
members through initial community meetings. Peer Moderators, aged 18-30 years, were
selected for their potential to lead and facilitate peer meetings, rehearsals and performances 
in their communities. Profiles of the six moderator pairs can be found in Annex 2. In each ECCD 
catchment area, 1 female and 1 male facilitator pair were selected. Selection criteria were
previously determined by the project team and local communities and spearheaded by 
CDOs as follows: 
• He/she should be literate ( known to read and write in local language)
• Must be a resident of village and must not be a Village Health Team (VHT) member
• Be brave, not shy, and be able to explain issues clearly
• Be friendly and approachable at all times
• Be willing to work voluntarily for the community
• Be married and readily available when wanted
• Same age group with the people he/she will be leading in the peer meetings
• Have good morals and relate well with the community
• Be experienced with community mobilization skills
• Be a simple person, confident, humble and able to talk in the community, in addition 
to being respected by community members
EDEAN Solution Components
EDEAN consists of two core components: Peer Group Meetings and Theatre Performances. 
Additionally, EDEAN created linkages to the health system to facilitate access for those 
interested in using an FP method and worked with key influential stakeholders in the 
community to generate support for fertility awareness and FP at the community level. The 
intervention began with a series of introductory community meetings, followed by trainings of 
trainers (ToT) and Peer Moderators. A process flow of EDEAN activities from Peer Group 
Meetings to monitoring and reporting is found below: 
“
One day Save the Children called for a community meeting and requested the community 
members to select two peer moderators and 20 peer group members, who could 
spearhead the EDEAN project in my community. The community members set attributes 
that the members should possess to qualify. When elections were conducted I won the 
position of the female peer moderator. I was then trained by Save the Children and later 
started leading my peers in meetings and theatre performances.’’ – Female Peer 
Moderator, Moroto District
20
Figure 5| EDEAN Activity Components
Training of Trainers, and Training of Peer Moderators
Following their engagement in community entry activities in Nadunget and 
Ngoleriate sub-counties, CDOs were invited to participate in a ToT delivered 
by the project team and a Karamoja-based theatre organization (Ngikukus
Uganda Ltd). Following the EDEAN moderator guide, the CDOs subsequently 
trained the Peer Group Moderators on the four EDEAN content modules, 
group facilitation and drama skills, including developing, leading rehearsing
and acting out the developed dramas. Given Karamoja’s low literacy environment, the 
trainings were staggered across two training phases:
Table 1| EDEAN Training Phases
Phase Timing Couple Communication Module
Training 1 Following 
implementation 
launch
• ToT with CDOs conducted over three days covering the first 
two module topics of couple communication and 
menstruation
• The following week, CDOs replicated the training they 
received by training Peer Moderators through a five day 
training on the same two modules
Training 2 Halfway through 
implementation 
(Three months after 
Training 1)
• ToT with CDOs conducted over three days covering the last 
two topics of fertility and FP 
• The following week, CDOs replicated the training with Peer 
Moderators through a five day training on the same two 
modules
21
Peer Group Meetings 
During the Peer Group Meetings, Peer Moderators led 10 female and 10 male
Peer Group Members through a series of eight meetings over the course of 
29 weeks during which they learn core fertility awareness content related to 
couple Communication, menstruation, fertility, and FP. The peer group 
meetings were gender-synchronized, facilitating both single-sex and mixed￾sex meetings on each topic area. The date, time and venue of the peer 
meetings were decided by all group members. 
Rehearsals and Theatre Performances
Upon completing the Peer Group Meetings for a given module, Peer 
Moderators led Peer Group Members in rehearsing and then performing a 
set of 3 performances on the topic module to diffuse fertility awareness and 
FP information among the community. For each topic module, the first two 
performances were “scripted” storylines, following a standardized storyline 
from the EDEAN moderator manual. For the third “Create your own drama” 
performance, Peer Group Members developed their own storyline from a messaging prompt 
in the moderator manual, and as a group, decided on the story trajectory and ending of the 
performance. Audience members were mobilized by ECCD CMC members and Peer 
Moderators to attend the performances at the ECCD centers on the same day of
performance. A sample schedule of activities for the couple communication module is 
below, and a full schedule of activities can be found in Annex 2.
Table 2| Sample Schedule of Activities, Couple Communication Module
Week Number Couple Communication Module
Week 1 Single Sex Peer Group Meeting 
Week 2 Mixed Sex Peer Group Meeting
Week 3 Storyline 1 Theatre Rehearsal and Performance
Week 4 Storyline 2 Theatre Rehearsal and Performance
Week 5 Create Your Own Drama! Theatre Rehearsal and Performance
Linkages to Health System
To build linkages to and support for EDEAN, VHTs and health workers 
attended periodic EDEAN orientation and reflection meetings. These 
meetings provided information about the EDEAN activities. VHTs and health 
workers were invited to attend theatre performances and peer group 
meetings. Health center FP providers were encouraged to provide FP to 
those interested in a method, and oriented on the use of FP invitation cards. 
As a community-based platform, EDEAN created linkages with FP service providers so that 
those interested in using an FP method knew where to go to obtain one or get more 
information about FP methods. During performances, Peer Moderators gave out FP invitation 
22
cards to interested community members wanting more information about FP methods at the 
local health facility. 
Engaging Community Leaders
Stakeholders who were considered influential in the community such as 
community leaders and elders, among others, were invited to attend 
periodic sensitization and reflection meetings about their roles in supporting 
EDEAN. Given the strong cultural beliefs and myths around FP in Karamoja, 
engaging key opinion and cultural leaders was vital in dispelling myths and 
encouraging young community members to attend the performances. The periodic
reflection meetings with community leaders provided a space for community leaders to 
express and then discuss/ resolve any sensitive topics or issues (for example discussing 
secretions during performances). 
Supportive Supervision and Monitoring
CDOs and the FACT team provided ongoing support and coaching to the Peer Moderators 
through weekly check-in calls, field visits, and monthly in-person meetings. Supervision and 
mentoring visits were made during meetings and rehearsals, and to a lesser extent 
performances. The CDOs used coaching guides as a tool to record their observations and
interactions. This was also an opportunity for facilitators to ask questions and bring up 
challenges or issues with the CDO.
Monitoring of EDEAN activities was conducted through a collaborative process among Peer 
Moderators, CDOs and FACT project staff. CDOs and Peer Moderators were trained on the 
monitoring forms and process as part of the aforementioned EDEAN training workshops. They 
practiced completing the forms both using a hypothetical scenario during the classroom 
portion, and as part of a practical activity that was incorporated into the training. Peer 
Moderators completed attendance registries at the beginning of each Peer Group Meeting, 
Rehearsal and Performance. CDOs collected forms from Peer Moderators during monthly 
supervisory visits. CDOs reviewed the monitoring forms with facilitators before collecting them, 
to identify and correct errors in completion or missing information. 
A reflection meeting with CDOs and Peer Moderators was held at the end of every one of 
the four topic modules. This provided a space for CDOs and Peer Moderators to share 
successes and challenges, discuss issues they were facing, and collectively generate 
solutions. During this meeting, CDOs brought all of the forms that they had collected from 
Peer Moderators and gave them to the FACT program coordinator. Following the reflection 
meetings, a FACT EDEAN monitoring and evaluation (M&E) group meeting led by IRH and
Save the Children was held to discuss process data and improvement areas. The data were
useful in identifying issues in implementation, and the discussions promoted collective 
problem solving.
23
PROOF OF CONCEPT EVALUATION
Evaluation Purpose and Research 
The purpose of the proof of concept evaluation was to describe the delivery and 
effectiveness of the peer learning and community theatre approach in increasing fertility 
awareness and FP use in the Karamoja region of Uganda. In addition, the evaluation aimed 
to understand the acceptability and potential scalability of the EDEAN intervention to other 
parishes in the Karamoja Region of Uganda. Specifically, the study objectives and research 
questions of this study were to:
1) Describe the delivery and feasibility to potentially scale the EDEAN community theatre 
intervention at the community level 
2) Describe the acceptability of the EDEAN peer learning and community theatre 
approach among participating communities 
3) Assess the effectiveness of the EDEAN peer learning and community theatre 
approach in increasing fertility awareness and FP use in participating communities
Evaluation Design
EDEAN was tested through a proof of concept evaluation in late 2016 – early 2017 in six agro￾pastoral communities of Moroto and Napak Districts in the Karamoja region of Uganda. A 
quasi-experimental study using mixed-methods was conducted to assess the delivery and 
potential scalability of EDEAN, as well as the effectiveness of the approach in increasing
fertility awareness knowledge and improving attitudes and behaviors toward FP use. Ethical 
clearance for this study was obtained from the Georgetown University Institutional Review 
Board, The AIDS Support Organization (TASO) review Board, and the Uganda National 
Council of Science and Technology (UNCST). Local authorities in Karamoja were also 
informed and approved of study activities. 
The evaluation collected data related to 1. Feasibility; 2. Effectiveness, and 3. Acceptability 
of EDEAN’s proof of concept model. Methods included cross-sectional baseline and endline 
household surveys with representative samples of intervention and control households (n=400 
intervention and n=200 control); FGDs with 32 Peer Group Members and 48 community 
members; interviews with 9 key informants; attendance and attendance and systematic 
observational data of peer group meetings, rehearsals and theatre performances, and; 
service statistics from 14 local health facilities. 
24
Table 3| Timeline of Study Activities
ACTIVITIES
2016 2017
May 
June
Jul
Aug
Sept
Oct
Nov
Dec
Jan
Feb
Marc
hApril
May
IMPLEMENTATION ACTIVITIES
 Training Phase 1 and implementation of Couple 
Communication and Menstruation Modules
 Training Phase 2 and implementation of Fertility 
and FP Modules
EVALUATION ACTIVITIES
 Baseline Household Surveys
 Endline Household Surveys
 FGDs with Peer Group Members
 FGDs with community members
 KIIs with leaders and providers
 Service statistics from health facilities
 Systematic observations of EDEAN activities
 Community dissemination of preliminary results 
Sites were systematically selected in collaboration with Save the Children and district officials 
from the Community Development Office. Sites included seven ECCD catchment areas in 
the sub-counties of Ngoleriet, Nadunget and Rupa. Of these seven catchment areas, six were 
intervention sites linked to Peer Group Members implementing the EDEAN performances at 
the ECCD centers (Ngloleriet and Nadunget). The other ECCD catchment area in Rupa 
served as a control site. Catchment villages were systematically selected based on their 
proximity to the intervention ECCD center and distance to health services. The control ECCD 
catchment area in Rupa was selected with input from Save the Children based on its 
dominant ethnic group (Karimojong), participation in the ECCD program, distance from the 
other intervention sites (situated in a non-intervention sub-county), and proximity to other 
health services. 
Data collection methods and sources, sampling and analyses procedures are explained in 
more detail below by 1. Feasibility; 2. Acceptability, and; 3. Effectiveness evaluation methods.
1. FEASIBILITY ASSESSMENT METHODS 
To understand the delivery and potential scalability of EDEAN, monitoring data were 
collected as part of regular Save the Children monitoring and supervision activities across all 
six intervention communities. As well, two external local research consultants were hired to 
observe and document systematically identified peer group meetings, rehearsals and 
performances. A level of effort analysis was also conducted to understand time spent on 
monthly activities by Save the Children and IRH staff. Referrals to FP services were also 
25
assessed through data collection of service statistics and on-site interviews. Feasibility data 
were collected and analyzed as follows:
Table 4| Feasibility Methods 
Attendance forms recorded de-identified participant demographics at each peer group 
meeting, rehearsal and performance. In addition FP invitation cards given out during 
performances were recorded on attendance records. All data were then entered into an 
excel database by a trained data coordinator. Frequencies of attendance numbers and 
demographics were tabulated quarterly, and subsequently presented at the FACT EDEAN 
M&E group meetings with IRH and Save the Children.
Systematic Observations of activities were conducted by two trained research assistants in 
each of the intervention communities. During each observation, the trained research 
assistant would record observation criteria as per: 1. Message delivery (e.g. fidelity and 
correct messaging); 2. Message receipt (e.g. participation and discussion), and; 3. Facilitation 
skills. Observers also recoded the process and flow of activities, including type of props used, 
environment, and interaction of male and female peer group members and/or audience 
members. Observations prioritized message delivery and fidelity to better understand the 
potential scalability of EDEAN in disseminating accurate fertility awareness and FP messaging. 
During select observations, the research assistants conducted moderator de-briefs to better 
understand Peer Moderators’ experiences and time spent preparing for activities. At the end 
of select performances, observers also conducted short semi-structured interviews with male 
and female audience members to assess receipt and perceptions of the performances. At
intervention end, eight intervention debrief interviews were conducted with four Peer 
Moderator pairs to better understand their experiences implementing EDEAN. Quantitative 
FEASIBILITY. Describe the delivery and feasibility to potentially scale the EDEAN community theatre 
intervention at the community level. Methods of verification:
Attendance records  72 peer group meetings, 72 rehearsals, and 72 
performances
Observations and debrief
interviews
 69 observations of 21 peer group meetings; 25 
rehearsals; 23 performances
 25 short semi-structured interviews with audience members 
after the performances
 19 debriefs with Peer Moderator pairs after peer group 
meetings, rehearsals and performances
 8 post-intervention debrief interviews conducted 
separately with 4 Peer Moderator Pairs
Level of effort analysis  Monthly level of effort data collected from 5 EDEAN 
program staff in Uganda
Service Statistics  14 Facilities; Moroto Hospital, HCIII, HCII, HCI, drug shops, 
Marie Stopes mobile clinic
26
and qualitative data were entered and analyzed separately in an excel matrix, segmented 
by topic module and type of activity (peer group meetings, rehearsals and performances). 
Level of effort analysis was conducted to estimate the time needed to potentially implement 
or replicate EDEAN at a larger scale. Data were collected and sent monthly by five Save the 
Children and IRH program staff using a structured time-sheet, broken down by type of EDEAN 
activities spent each month (e.g. coordination, trainings, monitoring visits, etc.) LOE input 
were analyzed and tabulated according to the following project activities: 1. Administration 
and coordination; 2. Training of CDOs and/or moderators; 3. Supervisor support of peer group 
meetings, rehearsals and performances; 4. Support and coaching of supervisors / trainers 
(CDOs); 5. Monitoring and Evaluation, and; 6. Other activities (e.g. research). 
Service Statistics. Anonymous, un-linked primary service statistics were collected from 14 local 
health facilities in the intervention and control catchment areas before, during and after the 
period of the intervention. These data captured the number and types of FP methods 
provided, women and couples counseled on FP, and referrals for FP related services. This was 
done in collaboration with local health officials and with the approval of the Ugandan MOH. 
Data were compiled into an excel database by health facility staff, and then analyzed and 
visually displayed over time trends vis-a-vis different measures (e.g. new FP clients; types of FP 
methods sought).
2. ACCEPTABILITY ASSESSMENT METHODS 
To understand acceptability of the EDEAN activities, qualitative data was collected from Peer 
Group Members, community leaders and service providers and community members to 
determine acceptability, applicability and perceptions of EDEAN activities. 
FGDs with Peer Group Members. A total of four sex-disaggregated FGDs with 30 male and 
female Peer Group Members were conducted to explore their experiences with the EDEAN 
intervention, and understand how they applied fertility awareness and FP information in their 
own lives. Peer Group Members were systematically selected by Save the Children and 
consented to participate with other peer group members from different communities of the 
same sex. 
FGDs with Community Members. FGDs respondents were selected from the sample of endline 
household survey respondents (described further below). After completing the endline 
Table 5|Acceptability Methods
ACCEPTABILITY. Describe the acceptability and applicability of the EDEAN peer learning and 
community theatre approach among participating communities. Methods of verification:
Focus group 
discussions
 4 FGDs with 30 Peer Group Members (16 female / 14 male)
 6 FGDS with 48 community members who attended performances
(24 female / 24 male) 
Key informant 
interviews 
 3 Health providers; 3 VHTs; 3 Leaders and Elders
27
household surveys, respondents aged 15-35 years old were asked if they were interested in 
participating in a FGD with other members of their community and age groups. The FGDs 
were then segmented into same sex groups with community members aged 15 to 25 years 
old; and again with participants aged 26 to 35 year olds, who attended the EDEAN 
community theatre performances. Groups were segmented by sex and age to ensure a safe 
space for members to share and exchange information among their peers, and to better 
understand different experiences and perceptions based on the two different age groupings. 
The FGDs explored knowledge, application and diffusion of fertility awareness and FP 
information at the community level as a result of the community performances. 
KIIs with Community Leaders and Health Service Providers. Approximately three health 
service providers, three VHTs and three community leaders were systematically selected to 
participate in KIIs in intervention areas (total n=9). The KIIs collected information related to the 
respondent’s awareness and level of knowledge of fertility awareness and FP, in addition to 
perceptions of EDEAN activities.
Analysis. All qualitative interview and FGD events were recorded, analyzed and translated 
from Nga’karimojong to English, synthesized and coded for processing and analysis. Data 
analysis software (Dedoose) was used to organize and prepare data to identify themes and 
key information. Content analysis was performed, with the following specific themes of 
interest: acceptability of discussing fertility awareness through community theatre 
performances, whether and how community members applied fertility awareness 
information to their lives; overall perceptions of the intervention, and; perceived advantages 
and disadvantages of this approach in dissemination and debunking myths around fertility 
awareness and FP.
3. EFFECTIVENESS ASSESSMENT METHODS 
To assess intervention effectiveness on fertility awareness and FP outcomes, surveys were 
administered at the household level among the systematically assigned intervention and 
control communities at baseline, and again six months later, at endline (see Table 6).
Table 6|Effectiveness Methods
Household Survey
Sampling. Two independent samples of households in the intervention and control ECCD 
center catchment area villages were selected to participate at baseline and again at end 
EFFECTIVENESS. Assess whether the EDEAN peer learning and community theatre approach can lead 
to an increase in fertility awareness and FP use in participating communities. Methods of verification:
Household surveys in intervention 
and control communities 
 Baseline and endline interviews with 400 intervention and 
200 control households (independent, representative samples)
28
line to assess community-level changes in fertility awareness and FP, and other secondary 
outcomes. An ECCD center catchment area was defined by three to four surrounding 
villages served by the ECCD center programs. For each of the ECCD catchment area, a list 
of two to four villages with its approximate number of households were generated to serve 
as enumeration areas. Each selected village was segmented according to the target 
number of households needed to achieve the necessary number of interviews. Households 
were then systematically selected to participate dependent on the sample size needed for 
that particular village. Based on existing population estimates, every other household in each 
village were visited to participate in the survey.
The final sample was comprised of two community-level cross-sectional samples of 600 
respondents: 400 within intervention communities (200 women and 200 men) distributed 
proportionately across ECCD catchment areas where the EDEAN intervention was 
implemented; and 200 control participants (100 women and 100 men) in one control ECCD
catchment area (see table in Annex 3). Estimating an intervention effect of 10% FP uptake 
(from 7.8% existing FP use in Karamoja to 17.8% expected use), a random sample of 172 
women will permit 95% confidence that the true proportion who took up an FP method lies 
between 17.1% and 18.6%.
Recruitment of Participants. After households were systematically selected during the 
sampling process, interviewers went to each household in designated villages. If an eligible 
study participant resided at the household, interviewers would describe the study to them, 
and ask for their consent to participate. At least one return visit was made if the potential 
participant was not at home. Only one member per household was be eligible to be 
interviewed. Couples were not interviewed. Eligibility criteria to participate in the study 
included: 
 Being of reproductive age (15-49 years old for woman, any age for men)1
 Identify as either female or male 
 Reside in one of the intervention or control study catchment areas
 Agree to participate in study
Measures. Baseline and endline interview guides included questions about background 
characteristics, knowledge, attitudes and behaviors related to the four intervention modules: 
1. Couple communication; 2. Menstruation; 3. Fertility, and; 4. FP. The survey instrument also 
determined individual exposure to specific intervention components (sessions and 
performance attendance) in order to assess levels of community exposure to the intervention. 
Primary outcomes included increased fertility awareness, current use of a modern FP method 
and intent to use a modern FP method. Secondary outcomes included knowledge, attitudes 
and behaviors related to couple communication, fertility and FP.
 
1 Women and men 15-18 years old were consented and included to participate in the study if they were parents or 
married and therefore considered emancipated adults per Ugandan law
29
Indicators were constructed to measure the aforementioned primary and secondary 
outcomes of interest, as highlighted below. A full explanation of the scale construction can 
be found in Annex 3.
Fertility Awareness
A composite score for correct knowledge of fertility awareness was constructed using 14 
measures based on literature reviews earlier on in the FACT project. We defined fertility 
awareness as encompassing knowledge about the menstrual cycle, when pregnancy can 
occur, pregnancy risk and male and female fertility components. Each of the 14 measures 
were asked as knowledge-based questions, and subsequently recoded as correct or 
incorrect answers. The full list of fertility awareness measures and questions can be seen in 
Table 7 below. 
Table 7|Fertility Awareness Measures and Survey Questions
Answer values for each of the 14 fertility awareness questions included a correct answer 
response option, and an “other” response option (e,g. What is the sign that a girl is able to 
become pregnant? 1. The first menstruation; 2. Other [specify]). Answer options were never
read out loud to the respondent. For each fertility awareness question, answers were then
recoded as correct (1) or incorrect (0) answer values. Textual data (e.g., response given as 
“other”) were reviewed and recoded into either correct or incorrect. “Don’t know” or “no 
response” answer options were recoded as incorrect. Subsequently, a fertility awareness 
knowledge index score was created by summing all the correct responses for the 14 
knowledge response items. Lastly, the median of the knowledge score was identified and a 
COMPONENTS SURVEY QUESTIONS 
MENSTRUAL CYCLE 1. What is the menstrual period?
2. What is the menstrual cycle?
3. When does the menstrual cycle begin?
4. When does the menstrual cycle end?
5. How long do most menstrual cycles last? 
PREGNANCY 
RISK
6. When during her menstrual cycle is a woman most likely to get 
pregnant if you have unprotected sex? 
7. From one period to the next, are there certain days when a woman is 
more likely to become pregnant?
8. Let’s pretend a woman named [___] had sex yesterday during her 
fertile day. She is not using a method of family planning or doing 
anything to avoid pregnancy. Can [___] become pregnant?
MALE AND FEMALE 
FERTILITY 
9. What is the sign that a girl is now able to become pregnant?
10. What is the sign that a boy is able to get a girl pregnant?
11. Other than sexual arousal, what are healthy vaginal secretions in a 
woman a sign of?
12. When during the menstrual cycle do vaginal secretions come?
13. When are men fertile?
14. What determines whether a baby will be a boy or girl at birth?
30
dichotomous variable categorized the knowledge index score into those who scored above 
the median knowledge (high fertility awareness) and those who had less than or equal to the 
median level knowledge (low fertility awareness).
Self-efficacy to access and use FP 
A FP self-efficacy scale was constructed based on exploratory factor analysis (EFA)
composed of the following four measures (Cronbach alpha = 0.68). Items were answered 
and coded as a 5-point agree to disagree likert scale for each measure:
 I am confident I could use a method correctly all the time to delay or avoid pregnancy
 I am confident I could use a method correctly all the time to delay or avoid pregnancy 
even if my partner disagrees
 I have the information I need to make a decision about whether to use FP if I wanted 
to delay or avoid pregnancy
 I know where to obtain a method of FP to delay or avoid pregnancy
Scale items were recoded from a 5-point Likert to a 3-point Likert of agree, neutral, and 
disagree. A mean scale score was obtained by identifying the mean of the 4 scale items. 
Finally, a dichotomous variable was created to reflect respondents who scored less than or 
equal to the mean (low self-efficacy), and those who scored above (high self-efficacy).
Attitudes about FP
A FP attitudes scale was constructed based on exploratory factor analysis (EFA) composed 
of the following four measures (Cronbach alpha = 0.67). Items were answered and coded as 
a 5-point agree to disagree Likert scale for each measure:
 Using a FP method is safe for a woman’s health
 It is acceptable for a woman to bring up the topic of FP with her husband
 It is acceptable for a man to bring up the topic of FP with his wife
 FP methods are a way to continue the Karamoja culture and way of life
Scale items were recoded from a 5-point Likert to a 3-point Likert of agree, neutral, and 
disagree. A mean scale score was obtained by identifying the mean of the 4 scale items. 
Finally, a dichotomous variable was created to reflect respondents who scored less than or 
equal to the mean (less positive FP attitudes), and those who scored above (more positive FP 
attitudes).
Social norms about FP
Similar to the two above scales, a FP social norms scale was constructed based on 
exploratory factor analysis (EFA) composed of the below three measures (Cronbach alpha = 
0.68). Items were answered and coded as a 5-point agree to disagree Likert scale for each 
measure:
31
 Most couples in my community who want to prevent pregnancy use family planning
(descriptive norm) 
 Couples in my community who want to prevent pregnancy should use family planning
(injunctive norm) 
 I should use a family planning method if I want to prevent pregnancy (injunctive norm).
Scale items were recoded from a 5-point Likert to a 3-point Likert of agree, neutral, and 
disagree. A mean scale score was obtained by identifying the mean of the 4 scale items. 
Finally, a dichotomous variable was created to reflect respondents who scored less than or 
equal to the mean (less supportive FP norms) and those who scored above (more supportive 
FP norms).
Partner communication about FP 
A partner communication about FP score was constructed from two binary measures (Yes, 
No): 
 Do you feel that your opinion is taken into account when talking about family planning
with your partner?
 In the last three months, have you discussed using FP to delay or avoid pregnancy with 
your partner?
Observations were restricted to respondents with a current partner. A composite score was 
generated from 0 to 2 (0 = no partner communication, answered no to both; 1= moderate 
partner communication, answered yes to one question; 2 = high partner communication,
answered yes to both questions). 
Community communication about FP 
A community communication about FP score was constructed from three binary measures 
(Yes, No): 
 In the last 3 months, have you talked about fertile days during the menstrual cycle with 
anyone? 
 In the last 3 months, have you corrected someone when/if you heard them saying 
something untrue about menstruation?
 In the last 3 months, have you corrected someone when/if you heard them saying 
something untrue about family planning?
A composite score was generated from 0 to 3 (0 = no community communication, answered 
no to both; 1= low community communication, answered yes to one question; 2 = moderate 
community communication, answered yes to two questions; 3 = high community 
communication, answered yes to all three questions). 
32
Data Analysis. All primary and secondary indicators were analyzed using frequency tables 
and simple bivariate analysis, along with analysis of differences between key indicators in the 
control and intervention areas. Analyses were conducted through two approaches: 1.
Baseline to endline community-level analysis; and 2; Post-intervention exposure level analysis. 
For the baseline to endline community-level analysis, an intention to treat (ITT) approach was 
used which defined all respondents in intervention communities as “exposed” to intervention 
regardless of whether or not they attended a performance. While this approach aimed to 
understand community-level results, rather than individual, this approach generally results in 
a more conservative estimate of the treatment effect because it does not remove non￾compliers from the analysis. A difference-in-difference approach was used to estimate 
intervention causality on primary and secondary outcomes, controlling for time and 
intervention assignment. A difference-in-difference approach is considered the gold 
standard for estimating the causality of an intervention since it not only compares an 
intervention group to a control group, but it also controls for any time-variant factors from 
baseline to endline (that is any differences in outcomes over time due to non-intervention 
factors such as seasonality, migration, other NGO FP programs, etc.) To address the limitation 
of a smaller sample size at control sites (n=200), sample weights were added in the regression 
analyses to ensure equivalence between control (n=200) and intervention sample size
(n=400).
The ITT community-level analysis considered all community members in intervention 
communities exposed to the intervention regardless of whether or not they attended a 
performance. As a result, the second-stage exposure level analysis aimed to better 
understand individual-level associations between exposure to a performance and likelihood 
of experiencing a certain outcome. While more directly targeted, these data were only used 
at post-intervention (since no one was exposed at baseline). As a result, associations between 
EDEAN exposure and primary / secondary outcomes, do not take into account any temporal 
pre to post effects as a result of the intervention, and is a less robust estimation method to 
attribute causality. For both analyses, bivariate descriptive comparisons were conducted, 
followed by multivariate logistic regression showing adjusted odds ratios (aORs). 
Limitations. As this was a proof of concept study, we did not aim to conduct a large-scale
study in Karamoja, and as a result, sampled a small number of communities. The rather small 
sample size limited the robustness of the results. The study team opted to collect data from 
community-based independent samples at baseline and endline due to the high mobility 
and migration patterns of the region’s population. Since this was not a cohort study, the ability 
to establish individual-level impact – rather than community level – was limited. However, 
qualitative data were used wherever possible to assess self-reported changes as a result of 
the EDEAN intervention. The post-exposure level analysis aimed to pinpoint exposure and 
outcome-level associations, stratified by age and sex groups. However, the original study 
design was not purposely powered for stratified sampling, resulting in some cases, in wide 95% 
33
confidence interval ranges of sex and age stratified exposure results.1 The baseline survey was 
conducted approximately one month after the launch of EDEAN activities. Approximately 2-
3 peer group meetings on couple communication had been conducted (2-3 weeks) before 
baseline but this was not expected to have has a significant effect on fertility awareness and 
FP outcomes. 
Lastly, spillover exposure in control area was not collected and assessed at end line. Yet, 
process-level, observational data suggested that some EDEAN participants and audience 
members moved back and forth between the intervention and control communities. This 
made pre and post comparisons controlling for contamination difficult. As a result, we used 
post-intervention level exposure analysis as described above to better understand specific 
exposure associations between attending performances and the likelihood of experience a 
specific outcome. As well, where possible we tried to validate the quantitative results with the 
qualitative data described below.
 
1 95% confidence intervals were used for this study, defined as the likely range of an outcome’s odds ratio estimate 
that would be observed 95 out of 100 times if the study were repeated
34
 
EDEAN FEASIBILITY
RESULTS
35
FEASIBILITY RESULTS 
Feasibility results below are presented by 1. Peer Group Members profiles and participation; 
2. Reach of EDEAN activities; 3. Delivery and Reception of fertility awareness / FP messaging; 
4. Level of Effort in supporting and delivering EDEAN activities, and; 5. Linkages to FP services. 
1. PEER GROUP MEMBERS PROFILE AND PARTICIPATION
Profiles of Peer Group Members. Of the 120 Peer Group Members in the six communities, all 
(were within the 18-25 year old criteria, with an average age of 23 years old (23 yrs female;
22 yrs male). About 89% of peer group members were married, and on average, peer group 
members had about 1.8 children (Min=0, Max=6). The majority of peer group members had
no formal education (70%), 17% had attended an Alternative Basic Education in Karamoja 
(ABEK) educational centers, 3% had primary education, and 10% had some level of 
secondary education.
Peer Group Members’ Attendance of EDEAN activities. There was overall high attendance of 
EDEAN activities by peer group members across all four activities. Attendance rates were 
similar across male and female peer group members. As seen in Figure 6, attendance 
dropped off however by the FP topic in November and December. Observational data 
suggest that the attendance drop-off was likely due to temporary migration to search for 
water and pasture for their cattle during the months of November and December, in addition 
to festival season, when most men tend to move to other places to participate in marriage 
and funeral rite festivals. When disaggregated by type of activity and sex, the attendance 
show similar attendance patterns among male and female peer group moderators, including 
a slight drop in attendance during rehearsals among both groups, relative to peer group 
meetings and performances. This drop-off was reaffirmed by systematic observations which 
documented the challenges that peer moderators had in coordinating performances with 
peer group members who had not attended the rehearsals, especially toward the end of the 
intervention when compliance in activity attendance became more challenging.
Figure 6|Proportion of Peer Group Members attending meetings, rehearsals and performances, by topic (n=120)
(Source: Attendance forms)
74%
83% 85%
70%
0%
20%
40%
60%
80%
100%
June - July 2016 July - Aug. 2016 Sept. - Oct. 2016 Nov. - Dec. 2016 COUPLE COMMUNICATION MENSTRUATION FERTILITY FAMILY PLANNING
36
2. REACH OF COMMUNITY PERFORMANCES
Audience member attendance data were collected at each performance. Audience 
member numbers are defined as points of contacts, rather than unique individuals, since 
individuals often returned to more than one performance (and numbers may double-count 
individuals returning again). Across the 72 performances delivered in six months, 6,699 points 
of contact were made with community 
members at performances. Both women 
(53%) and men (47%) attended the 
performances. Majority of the audience 
members were married (76%). Of the 6,699 
contacts made, 40% were in EDEAN’s target 
age group of 15-25 years, and 30% between 
26-35 years. There were also significant number 
of elders 36+ years who attended the 
performances (27%), while 3% were very young 
adolescents (10-14 years old). 
 
As seen in Figure 8, an average of 89 audience 
members attended each performance. 
Attendance lowered during the months of 
November and December similar to trends in 
peer group members’ participation above, 
due to seasonal migration and festival trends.
Figure 8| Average number of audience members at performances, by module
 (Source: Attendance data)
The systematic observations suggest that the rehearsals were also key in increasing reach and 
mobilization for performances (it was not uncommon for community members to also attend 
the rehearsals). As well, reach varied with ECCD center side location, depending on the type 
96 100
88
70
June - July 2016 July - August 2016 Sept. - October
2016
Nov. - Dec. 2016
COUPLE 
COMMUNICATION
MENSTRUATION
FERTILITY
FAMILY 
PLANNING
10-14 yrs
3%
15-
19 yrs
12%
20-25 yrs
28%
26-35 yrs
30%
36+ yrs
27%
Figure 7| Percentage of total audience members in 
age cohorts, all performances (n=6,135)*
(Source: Attendance data; *Note: age not recorded for 
every respondent, number therefore lower here than the 
total points of contacts recorded)
37
of mobilization strategy used. Typically, where the ECCD CMC presence was strong, 
performance attendance was high. 
3. FERTILITY AWARENESS AND FP MESSAGE DELIVERY AND RECEPTION 
Systematic observations of peer group meetings, 
rehearsals and performances were conducted to 
better understand delivery, participation, and 
receipt of EDEAN activities. A full observation 
results table can be found in Annex 4.
Message Delivery, Fidelity and Facilitation
Given Karamoja’s low literacy context, research 
assistants scored Moderator Pairs’ ability to follow 
the Moderator manual’s sequenced activity plans 
and guidelines. Figure 9 shows fidelity of manual guidelines, as observed during peer group 
meetings, rehearsals and performances. Overall, during peer meetings, 85% of observed Peer 
Moderators were able to follow the manual all the time, while 15% followed some of the time. 
During rehearsals, 82% followed the manual all the time, 13% followed some of the time while 
5% did not follow. For performances, the percentage of those who followed the guidelines all 
the time dropped to 74%.
 Figure 9|Percentage of Moderators following guidelines (n=69)
 (Source: Systematic Observations)
Ensuring fertility awareness and FP message fidelity. In general, almost all of the Peer 
Moderators and Peer Group Members maintained the fidelity of the fertility awareness and 
FP messaging when delivering the EDEAN activities. Very few observations documented 
delivery of incorrect messaging. Examples of incorrect messaging included “the egg is 
fertilized in the uterus” during a fertility module performance, and “a condom can only burst 
85% 82% 74%
15% 13% 22%
5% 4%
0%
20%
40%
60%
80%
100%
Peer Group
Meetings
Rehearsals Performances
3. Did not follow guidelines
2. Some of the time
1. Almost all of the time
“Peer Group Members were seated 
under a tree by the ECCD, there were 
also older people and younger 
children playing nearby the EDEAN 
rehearsal area. From a distance there 
were small kraals of goats and sheep 
making noise…”
– Observation of EDEAN 
Rehearsal
38
when there is no space left at the tip,” during a peer group meeting and rehearsal on FP, 
respectively. Incorrect messaging were more likely to occur during the FP module, partly due 
to the large number of questions peer moderators received from participants, and their 
challenges in answering technical, detailed questions about how methods work (see Annex
4 for examples of questions and answers). In some cases, peer moderators – and in some 
cases with the help of other peer group members - were able to correct incorrect messaging 
by peer group members during rehearsals: 
“When the one Peer Group Member who was acting [in the rehearsal] as a 
VHT said that the injectable is for 3 years, members started murmuring looking 
at each other, even those who had stepped back came back concerned
about this [Peer Group Member’s statement]. One of them did not wait for the
Peer Group Member to complete, she said please correct it right now before 
you make mistakes tomorrow it is three months, not years.” 
- Observation of rehearsal on FP topic
As seen in Annex 4, messaging were adapted to the local context. In some cases, peer group 
members – and in particular female ones - substituted vocabulary taught in the EDEAN 
manual due to fear of community backlash in discussing sexual reproductive vocabulary
considered taboo in the community. 
“The moderator also asked the males to name the different parts of the organs 
someone at the back named the male reproductive organ as “achia” the others 
burst into laughter this took about sometime. One of the males also said no this is 
wrong it is a penis call the spade by its name”- Observation of Male Peer Group 
Meeting on Fertility topic
Peer moderators, with the support of the Trainers/Community Development Officers (CDO)
often played a crucial role during rehearsals in ensuring that correct language were used 
by performance (as seen in last quote). 
Adaptation of fertility awareness messaging to create own drama and songs–Whereas Peer 
Moderators were able to follow the manual well, peer group members valued being able to 
“create their own drama,” as part of the third storyline under each module. Some groups 
created and adapted songs around their performances, and in some cases, some groups 
emphasized and linked the benefits of child spacing to improving child nutrition. One Peer 
Group Member explained:
“The dramas that we created or the dramas that we thought on our own, were 
not those ones in the manual that they used for teaching us. We created 
dramas that affected us but not calling other people’s names.” Female Peer 
Group Member, 18-25 yrs, Nadunget Sub-County
Moderator’s ability to facilitate EDEAN activities. Peer moderators’ ability to facilitate EDEAN
activities improved overtime. During activities, observers scored Peer Moderators’ facilitation 
39
skills as: “very good,” “fair,” and “poor.” Overall the 73% of observed Peer Moderators were 
rated as very good in facilitating performances. The proportion of Peer Moderators who 
received “very good” facilitation skills rating were high for the couple communication
module (83%), menstruation (86%) and FP (80%) but was weaker for fertility (43%). The lower 
performance rating during the fertility sessions may be due in part due to the content of the 
module, covering taboo topics such as male and female reproductive organs. During
observed performances, audience members comprised of elders, relatives, in-laws, 
commented that topics were not culturally acceptable to discuss openly in the community. 
Figure 10 shows Peer Moderators’ ability to facilitate activities during Peer Group Meetings, 
rehearsals and performances. The results show that facilitators tended to perform better 
during performances, when fertility awareness and FP information were diffused to the 
community at large. 
Figure 10| Peer Moderators ability to facilitate EDEAN activities, by meetings, rehearsals and performances 
(n=69)
 
 (Source: Systematic Observations)
‘You have chosen a sheep!’ Navigating gender dynamics during EDEAN activities. FACT 
formative research conducted in 2014 showed rigid age and gendered social hierarchies in 
Karimojong society. The systematic observations, coupled with qualitative data, showed that 
the use of gender synchronized single-sex meetings, followed by mixed-sex meetings for peer 
group members worked well to ease any gender power dynamics and sensitivities between 
the male and female Peer Group Members: 
“When we first started, we would fear mentioning some words, girls were 
fearing to look at boys, and boys were fearing to look at girls. The boy would 
say, that is my sister, why are they talking about such words? But when we 
started mixing, we learned that they were just educating us.” - Female Peer 
Group Member, 18-25 yrs, Nadunget Sub-County
Observations captured some challenges with gender and power dynamics among peer 
group members especially during rehearsals. For instance, several observations frequently 
recorded male peer group members volunteering female peer group members during role￾playing activities, and in some cases would tease and or laugh at female peer group 
Very 
good, 
54%
Fair, 
41%
Poor, 
5%
Very 
good, 
57%
Fair, 
31%
Poor, 
16%
Very 
good, 
73%
Fair, 
15%
Poor, 
12%
Peer Group Meetings Rehearsals Performances
40
members. As one peer group member stated during an observation: ‘you have chosen a 
sheep!’ when reacting to a female peer group member’s shyness during a rehearsal.
Message Receipt and Comprehension
Overall, fertility awareness and FP message content delivered by the peer moderators were 
well understood by both peer group members during peer group meetings and rehearsals
(78% and 72% respectively); and by audience members during performances (77%). Figure 
11 shows understanding of messages by audience members delivered by Peer Moderators 
disaggregated by the four topic modules for performances only. Understanding of messages 
were very high for the menstruation and fertility modules. However, only 67% and 60% of the 
observed performances on couple communication and FP, respectively, were marked to be 
“very well” understood by audience members.
Figure 11| Audience understanding of content delivered, for performances only (n = 23)
 (Source: Systematic Observations)
Audience Members’ reaction to the performances. Select interviews with male and female
audience members after performances showed that they appreciated the performances, 
stating that the performances created positive change especially among men in the 
community. Systematic observation revealed that the audience members were interested in 
the performances and considered the topics as relevant and reflective of couple dynamics 
within their community: 
“These performances are good because look at the way that woman was able 
to talk to a very difficult man, most of our men here in Karamoja are like him. I 
really like it when she eventually managed to talk to him.” - Female audience 
member, 30 years old, attended Couple Communication Performance
67%
100%
80%
60%
33%
20%
40%
0% 20% 40% 60% 80% 100%
Couple Communication
Performances
 Menstruation
Performances
 Fertility
Performances
Family Planning
Performances
Very well Fair Poor
41
“I saw the women near him started to whisper at each other and then smile;
‘This one has really been given a lot of guts, look at the way she is talking without 
fear. How I wish we [had] so many like her’ – Observation of performance on 
couple communication
While elders dominated some of the audience conversations during performances, they also 
had positive reactions to the performances and expressed desires for “change” among 
young couples in their communities, many of which were now living different, more sedentary 
lives than elders themselves had lived. Observations captured dynamics and conversations 
between older and younger audience members, in particular related the ‘old’ methods of 
FP such as abstinence, and the new FP methods being introduced during performances:
“When the nurse [in the performance] talked about condoms and other 
methods of family planning available at the clinic, there were murmurs from the 
female audience members. One actually said that abstinence was better for 
them while they were growing up.” – Observation of performance on FP
Similar to peer group member dynamics during rehearsals, gender and age dynamics were 
also observed during performances. Typical performances observations recorded female 
and male audience members in two distinct groups, seated separately under the community 
tree. During moderation questions, observations recorded that young females would shy 
away when asked questions in the presence of elders. Moderation questions were frequently
answered by elders during performances since in Karimojong society, it is more culturally 
acceptable to “respect elders” when they talk.
4. LEVEL OF EFFORT NEEDED TO IMPLEMENT EDEAN
We collected monthly time input sheets from five program support staff, in addition to level 
of effort estimates from Peer Moderators during moderator debriefs, to better understand 
what it would take to implement and replicate EDEAN activities in other contexts. 
Time Spent by Program Staff
Time input sheets were submitted in hours, but converted into days for easier interpretation. 
Table 8 shows the approximate number of days spent by staff during each month of 
intervention implementation, across a range of categories. The results show that most time 
were spent during the month of August 2016, when initial intervention activities were being 
implemented on the ground. 
42
Table 8| Approximate number of days spent on each implementation activities, by month (n=5 staff)
CATEGORIES Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 TOTAL
1. Administration & 
Coordination 7.7 4.2 1.8 2.3 3.2 3.8 22.8
2. Training of CDOs and Peer 
Moderators 3.0 8.6 14.9 0.0 0.0 0.0 26.6
3. Support of Peer Group 
Meetings & Performances 2.4 0.9 0.3 1.8 0.0 0.8 6.1
4. CDO support / coaching 0.7 0.5 0.5 0.2 0.6 0.0 2.4
5. Moderator support / 
coaching 0.2 1.4 1.7 0.5 1.3 0.0 5.0
6. Community Engagement 
meetings 0.0 4.5 0.0 4.1 3.0 8.0 19.6
7. Monitoring & Supervision 2.6 20.1 4.9 5.4 13.0 3.9 49.8
8. Other activities 1.0 3.4 2.6 0.0 6.4 3.3 16.7
TOTALS IN DAYS 17.6 43.6 26.6 14.3 27.5 19.6 149.1
(Source: LOE input sheets)
Figure 12 shows the percent time spent on all activities over the course of the eight month 
intervention period from the five program staff. Program staff included: 1. Save the Children
Senior Program Officer; 2. Save the Children M&E Manager; 3. IRH Program Manager; 4. IRH 
Program Assistant, and; 5. IRH M&E Advisor. The data suggest that most of the time was spent 
on monitoring and supervision of activities on the ground, followed by the preparing for, and 
facilitating the trainings of CDOs and Peer Moderators. 
Figure 13 shows percentage of time 
spent on all activities by month. The 
month by month trends show that 
monitoring and supervision activities 
remained consistent throughout 
implementation activities, and that 
additional time was spent in the first 
three months of intervention on 
administration and coordination, in 
addition to training of CDOs and Peer 
Moderators. More time was spend in 
the latter three months on community 
engagement meetings with leaders 
and providers and on “other” activities 
(e.g. mobilization for EDEAN end of 
project celebration, and field visits to 
the field with EDEAN advisory meeting). 
Administration & 
Coordination
15%
Training of 
CDOs and Peer 
Moderators
18%
Support of Peer 
Group Meetings 
& Performances
4%
CDO support / 
coaching
2%
Coaching of Peer 
Moderators 
3%
Community engagement 
activitites
13%
Monitoring & 
Supervision 
33%
Other activities 
11%
Figure 12| Percent Time Spent by Activity over six months, 
disaggregated by type of activity (n=5 staff).
43
Figure 13| Proportion of time spent on all activities, disaggregated by Month (n=5)
 
(Source: LOE input sheets)
Time Spent by Peer Moderators
On average, Peer Moderators reported spending a little over 1.5 hours facilitating peer 
meetings and rehearsals and a little over 1 hour moderating performances (see Annex 4 for 
breakdown of average hours spent by type of activity and module). As seen in Table 9, Peer 
moderators reported spending more time and energy preparing for, and in facilitating 
rehearsals. During debrief interviews, Peer Moderators shared that the increased time spend 
during rehearsals was due to time spent mobilizing Peer Group members, and starting to 
develop characters and teaching Peer Group Members the new storyline and characters. 
Peer Moderators also shared that performances tended to be easier, relative to rehearsals. 
Overall, Peer moderators felt well prepared for the sessions, and shared that CDOs aided 
them in understanding the material, mentored them in the handling of peer group members, 
and helped mobilize participants and community acceptance of the project.
Table 9| Average time preparing and facilitating for EDEAN activities, reported by Peer Moderators (n=69)
(Source: Systematic Observations; Debrief Interviews with Peer Moderators)
Peer Moderators were also asked about benefits and challenges in working as a pair with 
another partner. In general, Peer Moderators appreciated their partners’ commitment to 
their shared role and collaboration to lead activities together. Some peer moderators 
indicated that their role in EDEAN improved their leadership skills, while also citing challenges 
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16
Administration & Coordination Training of CDOs and/or moderators
Peer group meetings, rehearsals and performance support CDO support / coaching
Moderator support / coaching Advocacy & community meetings
M&E Other activities
Month
% of Time Spent on Activities
Preparation time Facilitation time
Peer Group meeting 1 hour 1.6 hours
Rehearsal 2 hours 1.6 hours
Performance 0.6 hour (35 minutes) 1.2 hours
44
such as absenteeism, lack of incentives, members attending sessions while drunk, and fears
of discussing taboo topics around fertility awareness and FP. Some male peer moderators 
shared the perception of an imbalanced workload (male Peer Moderator taking on more of 
the workload), while females felt the work was shared evenly. As well, male Peer Moderator 
felt that peer group members respected them more than their female counterparts, feeling 
the need to take on the lead during facilitation of EDEAN activities. This was supported by 
short semi-structured interview data in which female Peer Moderators indicated spending less 
time preparing for the EDEAN activities. 
5. LINKAGES TO FP SERVICES
FP service statistics were collected from 14 health facilities to understand trends in nearby 
services use over time. The collected FP service statistics from the health facilities in Napak 
and Moroto districts, from primary health centers, to Moroto regional hospital show similar 
trends in FP clients by age cohorts from July 2016 to March 2017, although trends increased 
slightly after intervention start (see Figure 14).
 Figure 14| Number of FP clients from April 2015 to March 2017, by age cohorts
During service statistics collection, health service administrators were also asked a set of 
questions about their familiarity with EDEAN, receipt of FP invitation cards, and any reported 
stock outs. While some interviewed health administrators confirmed that clients had come in 
after being referred from EDEAN activities, very few reported seeing the FP invitation cards. 
Only health centers based in Moroto district, rather than Napak district, reported receiving FP 
invitation cards. Moreover, service providers in Napak also reported frequent stock-outs of FP 
methods, such as pills, IUDs, and Implanon (see Table 10). Interestingly, of the 27 reported FP 
invitation cards received in Moroto district, eight were received in the Rupa health center, 
0
100
200
300
400
500
600
Apr - Jun
2015
Jul - Sep Oct - Dec Jan - Mar
2016
Apr - Jun Jul - Sept Oct - Dec Jan - Mar
2017
15-19 20-25 26-35 >35
EDEAN Intervention
45
located in the control catchment area, suggesting potential diffusion and spillover of EDEAN 
activities into control communities.
LESSONS LEARNED ABOUT FEASIBILITY 
Peer moderators and peer group members can deliver EDEAN activities and messages
correctly. Despite early project concerns about low literacy levels in Karamoja, Peer 
Moderators were able to follow the guidelines in the manual and use job aids correctly for 
the EDEAN activities. Peer Moderators prepared in advance and learned FA/FP messages 
during single and mixed sex meetings before rehearsing and performing to the entire 
community. The moderator’s ability to follow the manual as a guide to the activities led 
delivery of correct messages. The PMs and PGMs were quick to adapt the messages shared 
into songs and created their “own drama”. The ability of the PMs to facilitate and direct 
dramas and of peer group members to quickly grasp the FA/FP messages enabled them to 
translate these messages in to songs and dramas. This was a good way to pass information 
to and engage low literate communities. 
Early-on, systematic engagement of community leaders is key in ensuring feasibility of 
activities. For EDEAN, engaging community leaders and CDOs early on in the project were 
crucial in introducing new taboo topics while also being respectful of community leaders’ 
authority and decision-making process on the introduction of new non-normative information 
within their communities. Community engagement – both in terms of leaders, CDOs and 
service providers – were also key from a technical support standpoint. For instance, Peer 
Moderators were not always able to answer all questions during performances, especially 
Table 10| Summary of Q&A with service provider / administrator during data collection
DISTRICT Familiarity of FP 
providers with EDEAN
Received 
community 
members from 
EDEAN activities?
FP Invitation 
cards 
received?
Stock outs in the last 3 
months?
NAPAK
(7 Facilities)
All were well aware of 
EDEAN, and each clinic 
had attended the 
reflection meetings, and 
were actively involved by 
CDOs 
Slightly more than 
half of the providers 
received clients who 
mentioned EDEAN. 
Providers referenced 
peer moderators as 
the link to the health 
system. 
No invitation 
cards 
received.
All clinics reported stock outs. 
The majority reported being 
out of pills (4), implants (5), 
IUDs (4). One clinic did not 
have any methods. Another 
clinic only offered non￾hormonal methods due to 
stock out.
MOROTO 
(7 Facilities)
Almost all of the FP 
providers, had heard of 
EDEAN, most had 
attended sensitization 
meetings and been in 
contact with CDOs and 
technical staff.
All but one facility 
received referrals, 
with one mentioning a 
peer moderator 
referral, while two 
others cited the 
performances. 
Cards were 
received at 
Nadunget 
Health Center 
(19) and Rupa 
(8).
The majority of clinics did not 
report any stock out (4). One 
clinic only offered non￾hormonal methods. Another 
reported being out of the 
injectable, implant, pills, IUDs, 
and condoms.
46
technical questions related to fertility and physiology (e.g. ovulation) and FP methods. In 
other cases, peer group members sometimes feared using vocabulary and images in the 
menstruation, fertility and FP modules. The presence and support of supervisors (CDOs) were 
key in answering technical questions, in addition to handling problems related to 
absenteeism and incentives. Lastly, as Karamoja is a resource-scarce environment, the 
absence of incentives were also a key challenge in the initial phase of intervention 
implementation. Involving community leaders and elders were instrumental in mitigating 
material expectations and ensuring continual participation of participants.
High demand exists for fertility awareness and FP information by peer group members and 
audience members. The receptiveness and desire to learn new information among the 
Karimojongs about EDEAN drew large crowds to the performances. EDEAN was well-received 
as new but relevant information that could enable the continuation of their long tradition of 
spacing children through postpartum and periodic abstinence. Given Karamoja’s tradition 
of songs and dramas, the theatre approach was attractive to both women and men, both 
of whom returned to multiple performances to learn about fertility awareness and FP topics. 
However, attendances of performances varied by type of mobilization. Where CMCs were 
strong in an ECCD centre, mobilization was strong and generated high attendance. 
Presence of elders is key, but challenges exist in balancing the participation of youth and 
elders during performances. Performances attracted large crowds, and targeting 18-25 year 
olds, the EDEAN target group, exclusively was not possible. Given that some of the ECCD 
centers don’t have permanent structures and performances were conducted out in the 
open community, narrow targeting of the intended 18 to 25 year old age group was difficult. 
The presence of different community members had both pros and cons: while elders tended 
to dominate conversation during moderation questions, their presence was also crucial in 
encouraging younger community members to learn and reflect on the new fertility 
awareness and FP information seriously.
Addressing gender and power dynamics. Finally, given the rigid cultural gender roles and 
structures in Karimojong society, ensuring equitable participation and discussion among men 
and women could be challenging. Not surprisingly, male and female Peer Group Members 
faced difficulties in openly discussing taboo topics and sex-related vocabulary without “fear.” 
Peer Group Members also shared during FGDs that the gender synchronized meetings 
facilitated more participatory and equitable discussion during the same-sex, and later mixed￾sex meetings. However, additional attention should be paid to these gendered dynamics 
during training, and routine monitoring and supervision.
47
 
EDEAN 
ACCEPTABILITY
RESULTS 
48
ACCEPTABILITY RESULTS
Community acceptability of EDEAN’s peer learning and community theatre model was also 
assessed given the proof of concept phase’s emphasis on potential for scalability. 
Acceptability results are presented below, highlighting: 1. Community members’ reaction to 
EDEAN; 2. Community leaders’ and service providers’ reaction to the EDEAN activities; 3. 
Diffusing and applying fertility awareness and FP information.
1. COMMUNITY MEMBER REACTIONS TO EDEAN
Qualitative FGDs conducted with community members and peer group members highlight 
that the EDEAN activities were generally well attended and welcomed by the communities. 
The majority of FGD participants expressed the desire for activities to be sustained, in order 
for communities to continue learning about the benefits of child spacing. In particular, 
community members cited the importance of receiving information on child spacing, 
because of the “loss of knowledge” around the cultural practice of postpartum abstinence 
as a child spacing mechanism.
“EDEAN was real positive in our community, the community was not negative or 
even arguing. They said, thank you for bringing this program because before
family planning was hard.” – Female Peer Group Member FGD Nadunget, 15 –
25 years old
Given Karamoja’s scarce resource environment, child malnutrition and mortality was a 
common concern raised by FGD participants. Given this context, FGD participants commonly 
associated the performances topics on fertility awareness and FP with information they could 
use to prevent child malnutrition and mortality. Accordingly, observations also captured that 
in some cases, peer group members included messaging and songs on child malnutrition in 
the “create your own drama” rehearsals and performances. The importance of FP as a way 
to prevent child morbidity and mortality was highlighted by both male and female FGD 
participants, and peer group members and audience members alike:
“People became happy with the project and said the government has 
brought good information that is teaching us well, it has brought knowledge. 
The information they are teaching us is to space children. In the past when 
children were not spaced, children were not surviving, others would die.” -
Male Peer Group Member, FGD Nadunget, 15 – 25 years old 
“I think I will first talk to my wife about spacing our children so that we can get 
food to feed them. Then I will also talk to my elder brother because he has many 
children.” - Male audience member, 22 years old
Overall, the consistent positive reactions shared during FGDs across different age groups, 
gender, and communities points to a large demand for sexual and reproductive health 
49
information in Karamoja, a hard to reach communities with very limited access to education 
and services. 
2. COMMUNITY LEADER AND PROVIDER PERCEPTIONS OF EDEAN
Interviewed community leaders and service providers were receptive to the EDEAN activities. 
Similar to the community member FGDs, leaders and providers urged the continuation of 
EDEAN activities, in addition to the expansion of activities to different regions of Karamoja.
Community leaders including religious leaders, local councils (LCs), cultural leaders, ECCD 
CMC members and opinion leaders - were engaged during community entry meetings and 
routine implementation activities to encourage community acceptability for intervention 
activities. During endline interviews, community leaders reported encouraging young 
couples to attend the performances and take-up the issue of child spacing seriously. Some 
credited the EDEAN messages as bringing them out of “darkness,” while others 
acknowledged that the EDEAN performances were well aligned with their culture and 
tradition of child spacing.
“The plays of EDEAN are showing good works. I have not yet seen anything so 
far that might be against our tradition. Like the injectable, condoms, they are 
things that work between you and your mind and your home. EDEAN has not 
come to kill people but instead to clarify the myths about family planning.”
– Male Religious Leader, Nadunget Sub-county
Health workers. Both service-based providers and community-based VHTs were engaged 
through periodic orientation meetings, during which community performances and the use 
of FP invitation cards were explained. Endline interviews suggest that providers and VHTs 
appreciated and valued the EDEAN performances and activities. In some cases, VHTs 
reported attending some of the performances. When present, observations also showed that 
VHTs were also able to provide technical support, answering questions raised by audience 
members that peer moderators were unable to answer. During interviews, service providers 
shared that EDEAN activities had eased their work by facilitating outreach activities and 
bringing in clients to facilities in need of FP information and services. Interviewed service 
providers recognized the value of the EDEAN messaging as a compliment to their FP work, a 
difficult topic to discuss among communities of Karamoja:
“Those youth are lively and they really pass the message according to the way 
they really feel. I am always happy when I hear them performing and talking 
about family planning, which is not an easy topic when you talk about it to 
the community, especially to the men. When you see young ladies and young 
men talking… At least the message has been understood and disseminated.”
- Female health worker, Nadunget Sub-county
Consistent with the service statistics data collected from local health services, some 
interviewed health providers expressed concerns and confusion about the invitation cards. 
Specifically, health workers did not always understand the use of the FP invitation cards. Given 
50
that the cards were in local language, health workers – whom are often not from Karamoja -
could not understand the content of the FP invitation cards. 
3. DIFFUSING AND APPLYING INFORMATION ABOUT FERTILITY AWARENESS 
AND FP 
During concept design, performances were identified as a potential vehicle to diffuse and 
spread information about fertility awareness and FP throughout Karimojong communities. 
Coupled with the aforementioned survey data results, qualitative data suggest that diffusion 
occurred beyond performances, and between peer group members and other community 
members in other non-performance settings. 
EDEAN messages were shared beyond performances. During FGDs, peer group members 
cited sharing and discussing the information from performances with others, but in particular
neighbors, siblings and friends, encouraging them to share the information with their partners. 
According to peer group members, the most common type of information shared among 
community members included the messages around couple communication, menstruation,
FP and child spacing. Consistent with Karamoja’s gendered hierarchies, information was 
typically shared among gendered lines: whereas women shared information with their fellow 
female peers (sisters, female friends), men shared more with their fellow males (brothers, male 
friends). The type of information shared also varied among men and women. For instance, 
male peer group members shared more information about condoms, and men’s role in 
determining the sex of a baby, while female peer group members shared information about
menstruation and FP.
Elders encouraged the diffusion of the EDEAN messages without fear. Consistent with 
observational data and interviews with community leaders, male peer group members 
reported that elderly men often encouraged them to spread the messages without “fear.” 
His finding reinforces the importance of the sensitization and reflection meetings as a key 
EDEAN component to engage influential reference groups such as community elders. 
“Those very days, elders started spreading messages and telling boys that they 
should start spreading messages to people without fear. The elders are helping 
so much. Some elders begun enjoying the performances […] Next time during 
the subsequent shows, the boy comes along with his friends. For example, he 
comes along with four, then next time those ones also keep on coming along 
with others.” - Male community member, FGD Ngoleriet, 15-25 years old 
Applying Couple Communication Messages. The first EDEAN module focused on couple
communication messaging, based on formative research showing community demand for 
couple communication skills to reduce couple conflicts and disagreements. Qualitative FGD 
data suggest that the couple communication messages were easily adopted by Karimojong 
couples attending the performances. During FGDs, participants commonly defined couple 
communication as “couples respecting each other” and “couples listening to each other.” 
The majority of FGD participants reported increased discussions with their partners about child 
51
spacing, their children’s health and shared caregiving responsibilities (e.g. taking them to 
school and/or providing food). Improvements in couple communication were shared by both 
peer group members and community members attending the performances. As seen in the 
second quote, while male community members also cited improvements, they also 
commonly pointed to their female partner as the initiator and instigator of arguments. 
“What changed in my life? Couple conversation. You need to share so much
information with your husband. Like if you are over producing children, you will 
go and say: ‘my husband it is like this: we are supposed to do family planning, 
we have to let children reach like 2 years upwards then we can start to have
another child, so that our home also looks healthy.’ ” - Female Peer Group 
Member, FGD Ngoleriet, 15-25 years old
“This teaching has been good for me, especially the one about listening to 
each other. It is about togetherness. That is why when she decides to quarrel in 
the near future I will just remind her that we made an agreement to listen to 
each other. That knowledge was good for me”. – Male Community Member, 
FGD Ngoleriet, 26 to 35 years old
In almost all the qualitative FGDs, couple communication was frequently referenced as a skill 
that has helped couples solve sources of conflict. The prevalence of violence cited in the 
qualitative data are consistent with anthropological research conducted in Karamoja, 
pointing to increased use of violence as Karimojong families adjusted to a sedentary lifestyle 
and consequent loss of their cattle-centered livelihood (Stites and Marshak 2016). Within the 
endline qualitative data, some EDEAN participants cited improved couple communication 
as having helped improve their relationships and reduced violence in their family.
“So many people no longer fight, they now do couple conversation in their 
homes and they like now sharing information because, somebody will 
remember what he/she learned from out, saying, this is the very thing they 
teach us, to remove fighting and violence at home.” - Female Peer Group 
Member, FGD Nadunget, 15-25 years old
The improvements in couple communication was not isolated to young community members. 
Interviewed community leaders, elders and CMCs also reported using and benefitting from
couple communication information with their own families and personal lives.
Applying Menstruation and Fertility Messages. FACT’s formative research (2014) identified a 
number of menstruation myths present in Karimojong, including that when menstruating, 
women cannot cook, cannot milk a cow, nor cannot go to the pumpkin garden, among 
other myths. Qualitative data suggest that the menstruation performances helped debunk 
some of these myths among peer group members and community members of both sexes. 
Interestingly, a few male community members cited the importance of “respecting” your wife 
during her menstruation, translated as not having sex with your partner when she is 
menstruating.
52
During the fertility module, peer group members and community members learned about 
the fertile and non/fertile days during the menstrual cycle. Qualitative data suggest that this 
information was shared among couples, even when only one partner attended the 
performance. Qualitative data suggest that information passed on was not only conveyed 
in the taught fertility awareness methodology. For instance, in many cases secretions were 
shared among networks as “whitish” substances that come a few days after the menstrual 
period: 
“We talked about this information first of all with my wife. She told me that she 
has discovered the days when she is on her menstrual period [cycle]. When the 
blood shades off then something that is whitish like water starts to come. Then 
something thick like semen starts to come then that is when you are able to 
conceive if we slept together. That is the information I heard from my wife” –
Male Community Member, FGD Ngoleriet, 15 to 25 years old
Qualitative results suggest that community members valued information related to menstrual 
cycle and timing of the fertile window. In particular, audience members viewed the 
information and knowledge they gained from the menstruation and fertility topics as 
information that could aid them in preventing pregnancies: 
“I used that one [method] of the menstrual cycle and found that, there are 
some days when a child cannot get in your womb even when you sleep with
a man. I use the method of the menstrual cycle when I know, these days, I am 
going to sleep with my husband and these days I cannot sleep with my 
husband” – Female Community Member, FGD Nadunget, 15 to 25 years old
“If men get to know the menstruation cycle of their wives, it can help them to 
use this as a method of family planning and also avoid unwanted pregnancies.”
- Audience member reaction after performance on menstruation
While this is encouraging in that it shows interest in, and general understanding of this key 
topic, it is also important to recognize that EDEAN performances did not attempt to convey 
specific information that is essential for using, for example, the TwoDay Method or SDM. 
Consideration should be given to either providing that information (preferable via trained 
community or facility providers) or clarifying with performance audiences that this is general 
information, and not to be used as a FP method. 
Applying FP Messages. Family planning was the last module introduced in the EDEAN set of 
intervention activities. During qualitative FGDs and interviews, community members and peer 
group members acknowledged their increased awareness of modern FP methods and how 
they work. This was particularly salient among men, who expressed increased interest in
learning about new modern methods of FP like the pills, injectables, implants and permanent 
methods (vasectomy).
53
“At least now we know that there are some other methods of family planning. 
Unlike those days, there was only the method of the kraal [post-partum 
abstinence] that we knew” – Male Community Member, FGD Ngoleriet, 15 to 
25 years old
While most community members and peer group members reported applying the learned FP 
information by discussing child spacing, a few participants cited either using a method of FP 
or knowing of someone who had taken up a method. Consistent with the survey results which 
showed that men increased their intent to use modern FP, and in particular condoms, one 
male community member shared how he approached his wife about using condoms: 
“I used the knowledge and decided to use condoms. Maybe when we are 
going to sleep or when we are sleeping, I get my condom and I dress in it, then 
I inform my wife that instead of us spoiling our family let us share this very thing 
such that we can space children”. – Male Community Member, FGD Ngoleriet, 
15 to 25 years old
“There is a woman called [….] she went to the performances and she was 
carrying her child and was still young. She did go and then got an injection 
even now as we talk she is having the injection.” – Female Community Member, 
FGD Nadunget, 15 to 25 years old
During FGDs, some community members shared that they continued to space their children 
through the use of post-partum and periodic abstinence, while others did not specify if and/or
what methods they were using. Some male FGD participants shared that since the start of 
EDEAN, their wives had not yet conceived. Specifically, a few male peer group members 
reported discussing with their wives the use of LAM as a way of spacing birth.
‘I stood there and listened to everything they discussed, I listened. Then I 
discovered that this information is vital. Like for me since my wife is still having 
one child I should leave that one to grow, to grow till when I have discovered 
that he has grown then I can sleep with the mother. – Male Community 
Member, FGD Ngoleriet, 15 to 25 years old
Overall, the qualitative results suggest that communities are demanding fertility awareness 
and FP information and have intent to use modern FP methods, but in many cases are 
continuing the use of post-partum and periodic abstinence as a method of child spacing. 
Interestingly, the qualitative results also highlight that couple communication appears to be 
a vital and integral part of applying the EDEAN messages into their own lives. Lastly, the results 
suggest that the messaging of the menstrual cycle game helped women and men learn 
about their bodies, and that there was interest in using this information to prevent 
pregnancies. 
54
LESSONS LEARNED ABOUT ACCEPTABILITY 
EDEAN’s peer learning and community theatre model was highly accepted in Karamoja. 
EDEAN’s high acceptability can be attributed to EDEAN’s approach of using the community 
resource persons as peer moderators and peer group members. CDOs also played vital role 
in the EDEAN implementation and are respected in the communities. Interviewed community
leaders and members consistently reaffirmed that EDEAN was compatible with their culture 
and traditions. Engaging elders in intervention activities was key in encouraging young 
couples to attend the performances and in dispelling myths and “fear” around fertility 
awareness and FP topics.
Potential to integrate fertility awareness and FP messaging with nutrition and maternal and 
child health (MCH) activities. The EDEAN messages were relevant and applicable due to the 
current scarce resource environment in Karamoja. Hunger in Karamoja is a driving factor 
behind interest in FP and desire to space children. Qualitative respondents repeatedly raised 
the harsh realities of their day to day life, including poverty, child malnutrition, and child 
mortality as key factors for wanting to learn more about FP methods. As private investments 
increase to the region and foreign assistance transitions away from humanitarian assistance, 
programs should explore ways to integrate the EDEAN fertility awareness and FP messages 
into existing nutrition and MCH programming.
Health providers welcomed the EDEAN activities and have potential to be more involved. 
Providers perceived the EDEAN activities as a demand generation activity that could 
complement and improve their day to day FP work, generating increased numbers of FP 
clients. However, the findings also suggest that the FP invitation cards may not have been an 
effective linkage and referral system. However, the findings also suggest that referral 
mechanisms to health systems need to be strengthened. Community based referral systems 
are notoriously challenging. EDEAN used a paper-based referral through FP invitation cards. 
Future programs might explore use of a paper-based plus or mobile referral which have 
shown greater success (Lebetkin 2015). 
Wide diffusion of FA/FP information with others in the community. Karamoja’s mobility patterns 
facilitated the speed and feasibility of information to be diffused and shared among 
community members. Those who attended the EDEAN performances shared FA/FP 
information with their friends, neighbors, brothers and sisters among others and brought them 
to attend in subsequent performances. The performance locations – conducted at ECCD 
centers open to all in the community – also facilitated access and diffusion of information. 
The diffusion of information shared by qualitative participants also supports our assumption 
that messaging may have spilled over to control communities (as shown in survey results). 
Participants learned and applied information about the menstrual cycle game, and in some 
cases to prevent pregnancy. The results show that some community members shared that 
they used the knowledge of the menstrual cycle game as a “method” to space their children. 
55
While better than not using any method at all, caution should be undertaken when teaching 
fertility awareness and in particular the menstrual cycle games in context with low literacy, 
low modern FP use and limited services availability. In such contexts, emphasis on cycle 
regularity and eligibility should be emphasized and special considerations should be given 
toward gender power dynamics and couple agreement on how the information will be used 
(for example, in planning vs. preventing a pregnancy). One potential avenue could be 
including the participation of a health provider during intervention activities such as the 
menstrual cycle game to clarify menstrual cycle eligibility, and emphasize the presence of FP 
methods and services at nearby health facilities. 
Couple communication was valued as skill to reduce intimate partner conflict and violence. 
While IPV was not directly addressed in the EDEAN curricula, the qualitative data suggest 
clear improvements in couple communication between couples who attended the 
performances. Both men and women shared that the skills they learned around couple 
communication from the performances helped resolve conflicts related to income 
generation, household chores and child spacing decisions. As well, men’s increased 
awareness and interest in FP and child spacing juxtapose other studies showing that Ugandan 
men in particular have very limited accurate knowledge and access to information about FP 
methods (Thummalachetty, et al. 2017). Despite the couple-level improvements, the 
qualitative data –while triangulated with the systematic observational data – suggest that 
gender disparities still remain and that more could be integrated within the EDEAN model to 
incorporate more gender deconstructing content and reflection. 
56
 
EDEAN 
EFFECTIVENESS
RESULTS
57
EFFECTIVENESS RESULTS 
Effectiveness results are presented below by 1. Community-wide results baseline to endline;
2. Exposure level results at endline, and; 3. Fertility awareness and FP outcome association 
results. Community level data were analyzed to understand intervention effect at wider scale 
within the community (regardless of performance exposure), while exposure level data were 
analyzed to describe more in-depth associations between those who viewed the 
performances and potential improvements in fertility awareness and FP outcomes. Lastly, 
associations between high fertility awareness scores and, 1. Current modern FP use, and; 2. 
Intent to use modern FP are presented to assess whether a high fertility awareness score is 
associated with improved FP outcomes.
1. COMMUNITY-WIDE RESULTS
Demographic Characteristics
Table 11 shows survey respondent demographics at baseline. At baseline, survey respondents
in intervention and control groups had similar age distributions (on average 30 years old). All 
survey respondents identified themselves as Karimojong (data not shown). Consistent with 
Karamoja’s low education levels, there were no significant differences in educational 
background between intervention and control respondents: 74% of respondents in the 
intervention community had no education compared to 81% in the control community at 
baseline. About 92% of intervention respondents were living with a partner compared to 98% 
of control respondents. On average, respondents had about 3 children and had been 
married for eight years across both groups. Both intervention and control groups reported low 
lifetime use of modern FP, about 21% and 18% of intervention and control respondents, 
respectively. The majority of participants in both intervention and control communities 
reported using traditional method of FP, and more specifically periodic abstinence (55% 
intervention vs. 62% control). 
58
Table 11| Socio-demographic characteristics of participants, at baseline (n=600)
(Source: Baseline Survey)
Demographic tables disaggregated by sex, and by total sample at endline can be found in 
Annex 5). Significantly different demographic variables between intervention and control 
groups were controlled for in the adjusted regression models presented further below. 
Community Exposure to EDEAN Performances
At endline, respondents in intervention communities were asked if they heard about, and/or 
attended the EDEAN theatre performances. Table 12 shows levels of EDEAN exposure for all 
intervention participants, and disaggregated by male and female samples. Approximately 
DEMOGRAPHICS Intervention Control
n=400 % n=200 % P-value
Sex 
Male 
Female
199
201
49.8
51.3
100
100
50.0
50.0
N/A
Age
Mean (Std)
Min, Max
29.8 (8.4)
15, 49
30.1 (7.4)
17, 49
0.1354
Highest level of school attended 
None
ABEK
Some primary education 
Completed primary 
Some secondary 
Completed secondary 
297
33
45
10
13
2
74.3
8.3
11.3
2.5
3.3
0.5
161
7
23
5
2
2
80.5
3.5
11.5
2.5
1.0
1.0
0.1354
Have job or work 
Yes
No 
223
177
55.8
44.3
139
61
69.5
30.5
0.0012
Type of work
Farming/ agriculture
Formal paid work
Informal paid work
98
9
116
44.0
4.0
52.0
27
10
102
19.4
7.2
73.4
<.0001
Currently living with partner 
Yes
No
367
33
91.8
8.3
196
4
98.0
2.0
0.0027
Time married or living together
Mean (Std)
Min, Max 
8.13 (6.55)
0.1, 30
7.97 (6.79)
0.5, 32.0
0.7804
Husband has another wife
1 wife
2 + wives
171
196
46.6
53.4
111
85
56.6
43.4
0.0232
Number of living children 
Mean (Std)
Min, Max
3.3 (2.7)
0, 18
3.1 (2.1)
0, 12
0.3547
Ever use modern FP
Yes
No 
84
316
21.0
79.0
35
165
17.5
82.5
0.3379
Type of FP method ever used
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
CycleBeads
TwoDay Method
LAMW
Withdrawal
Abstinence 
 Other
14
0
2
11
3
0
0
0
1
1
60
10
219
20
3.5
0.0
0.5
2.8
0.8
0.0
0.0
0.0
0.3
0.3
15.0
2.5
54.8
5.0
7
0
0
1
1
0
0
0
1
0
29
13
124
1
3.5
0.0
0.0
0.5
0.5
0.0
0.0
0.0
0.5
0.0
14.5
6.5
62.0
0.5
59
79% of surveyed intervention participants had heard of the EDEAN performances, while 65% 
reported attending at least 1 performance. About 16% of the intervention sample identified 
themselves as EDEAN peer group members (39 female study participants and 25 male study 
participants). Peer moderators and peer group members were a primary source of 
information for hearing about the EDEAN intervention, especially for women. At endline, 69% 
and 57% of the community members indicated that they heard of EDEAN intervention from 
Peer Moderators (PMs) and Peer Group Members (PGMs) respectively.
Table 12| Exposure at endline, among participants in intervention communities, by total, male and female (n=400)
(Source: Endline Survey)
Baseline to Endline results
Baseline to endline descriptive results are presented below for outcomes related to: 1. Fertility 
awareness, 2. FP use; 3. Self-efficacy to use and access FP; 4. FP attitudes; 5. Social norms 
around FP; 6. Partner communication about FP, and; 7. Community communication about 
fertility awareness and FP.
Fertility Awareness
Table 13 shows changes over time from baseline to endline in key fertility awareness and FP
outcomes in both intervention and control communities. A fertility awareness score was 
considered high if a score was above the median score of 8 (e.g. answering questions about 
fertile window; signs of onset fertility in boys and girls). Both intervention and control 
ENDLINE
TOTAL SAMPLE FEMALE MALE
n=400 % n=202 % n=198 %
Have you heard of the EDEAN 
community theatre performances?
Yes
No
315
85
78.8
21.2
189
13
93.6
6.4
126
72
63.6
36.4
Have you attended any EDEAN theatre 
performances? 
Yes
No
260
140
65.0
35.0
160
42
79.2
20.8
100
98
50.5
49.5
Are you an EDEAN peer group member?
Yes
No
64
336
16.0
84.0
39
163
19.3
80.7
25
173
12.6
87.4
How did you hear about the community 
performances? (multiple response)
ECCD Center
ECCD CMCs
Save the Children
CDO
EDEAN Peer Group Moderator
EDEAN Peer Group Member
Community leader
My partner
Family member 
Friend
Other 
91
51
92
40
179
149
18
7
1
24
2
22.8
12.8
23.0
10.0
44.8
37.3
4.5
1.8
0.3
6.0
0.5
28
27
26
5
122
99
16
4
1
22
1
13.9
13.4
12.9
2.5
60.4
49.0
7.9
2.0
0.5
10.9
0.5
63
24
66
35
57
50
2
3
0
2
0
31.8
12.1
33.3
17.7
28.8
25.3
1.0
1.5
0.0
1.0
0.0
How many performances did you attend?
One to three
Four to six
Seven to nine
Ten to twelve
136
45
26
52
34.0
11.3
6.5
13.0
86
26
15
33
42.6
12.9
7.4
16.3
50
19
11
19
25.3
9.6
5.6
9.6
60
communities had higher fertility awareness scores at endline, suggesting possible spillover 
effects in control communities. A higher percentage of participants in the intervention group 
increased fertility awareness scores (from 15% to 75%). 
Table 13| Fertility Awareness at baseline to endline, among intervention and control groups (n=600)
PRIMARY OUTCOMES INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=400 % n=400 % P-Value n=200 % n=200 % P-Value
Fertility Awareness Score
High (1 >8)
Low (0 ≤8)
60
340
15.0
85.0
298
102
74.5
25.5
<.0001 60
140
30.0
70.0
105
95
52.5
47.5
<.0001
(Source: Baseline and Endline Surveys)
Figures 15 and 16 show high fertility awareness score proportions from baseline to endline 
segmented by male and female survey respondents in intervention (left) and control 
communities (right). The changes over time suggest that both women and men in 
intervention communities increased their fertility awareness knowledge over time, but men 
experienced the highest knowledge jumps: at baseline, 14% of men had high fertility 
awareness scores, compared to 81% of men at endline. Women and men in control 
communities also experienced improvements in fertility awareness scores among, albeit by 
smaller margins. 
Figure 15| Intervention women and men with 
high fertility awareness scores (>8), at pre and 
post (n=400) 
 
Figure 16| Control women and men with 
high fertility awareness scores (>8), at pre and 
post (n=200) 
(Source: Baseline and Endline Survey; Note: p-values: * <.05; ** <.01; ***<.001)
Table 14 shows the proportion of women and men by intervention arm who answered the 14 
Fertility Awareness questions correctly at baseline and endline. The proportion of women and 
men who answered questions correctly about the menstrual cycle more than doubled from 
baseline to endline. Women and men in control areas also experienced improvements in 
menstrual cycle knowledge, although these marginal gains were smaller from baseline to 
16% 14%
68%***
81%***
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
WOMEN MEN
33%
27%
56%**
49%**
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
WOMEN MEN
BASELINE
(n=201)
ENDLINE
(n=202)
BASELINE
(n=199)
ENDLINE
(n=198)
BASELINE
(n=100)
ENDLINE
(n=100)
BASELINE
(n=100)
ENDLINE
(n=100)
61
endline. Knowledge of pregnancy risk was relatively high for intervention women (93.5% at 
baseline), and no improvements were observed over time. In contrast, the proportion of men 
who knew of fertile days during the menstrual cycle jumped from 65% to 90%, while a 
decrease in knowledge was found among control men (84% to 52%). Knowledge of the fertile 
window – or when pregnancy risk can occur – was much lower among respondents, although 
knowledge improved over time among intervention respondents. About 22% of intervention 
women were able to identify the correct fertile window at baseline, increasing to 42% at 
endline; while only 7% of intervention men were able to answer the same question correctly, 
more than doubling to 18% at endline. Improvements in fertile window knowledge were not 
found among control men, but not control women. 
Table 14| Proportion of women and men who answered Fertility Awareness questions correctly, by intervention 
arm and time point (n=600)
FERTILITY AWARENESS 
MEASURES 
WOMEN MEN
INTERVENTION CONTROL INTERVENTION CONTROL
Baseline %
(n=201)
Endline %
(n=202)
Baseline %
(n=100)
Endline %
(n=100)
Baseline %
(n=199)
Endline %
(n=198)
Baseline %
(n=100)
Endline %
(n=100)
MENSTRUAL CYCLE
Menstrual period is when 
woman has monthly bleeding 93.0 99.0 100.0 100.0 92.0 99.0 96.0 100.0
Menstrual cycle is all the days 
before the start of one 
menstrual period to the day 
before the next one starts
38.3 78.2 56.0 78.0 44.8 86.4 67.0 92.0
Menstrual cycle begins on the 
first day of bleeding
37.1 87.6 45.0 56.0 35.7 83.8 54.0 88.0
Menstrual cycle ends on the 
day before the start of the 
next menstrual period 
26.4 76.7 48.0 72.0 23.6 81.3 42.0 62.0
Regular menstrual cycle lasts 
between 26 to 32 days 36.3 85.2 34.0 44.0 36.2 81.3 29.0 49.0
PREGNANCY RISK
There are certain days in a 
menstrual cycle when women 
are more likely to become 
pregnant
93.5 91.6 83.0 88.0 64.8 89.9 84.0 52.0
A woman is most likely to get 
pregnant if you have 
unprotected sex halfway 
between two menstrual cycles
21.9 42.1 2.0 2.0 6.5 17.7 5.0 34.0
You can become pregnant if 
had unprotected sex during 
fertile day 
75.6 80.7 74.0 68.0 85.9 89.4 88.0 98.0
MALE AND FEMALE FERTILITY
First bleeding is sign girl is 
able to become pregnant 93.0 97.5 96.0 93.0 84.9 94.4 90.0 71.0
First ejaculation is sign boy is 
able to get a girl pregnant 45.3 67.3 70.0 42.0 81.9 95.0 84.0 87.0
Other than sexual arousal, 
healthy vaginal secretions are 
sign of fertility
50.3 85.2 62.0 98.0 72.9 81.3 58.0 94.0
Vaginal secretions come 
during the middle of the 
menstrual cycle
11.0 38.1 58.0 83.0 10.1 21.2 6.0 46.0
62
FERTILITY AWARENESS 
MEASURES 
WOMEN MEN
INTERVENTION CONTROL INTERVENTION CONTROL
Baseline %
(n=201)
Endline %
(n=202)
Baseline %
(n=100)
Endline %
(n=100)
Baseline %
(n=199)
Endline %
(n=198)
Baseline %
(n=100)
Endline %
(n=100)
Men are fertile every day 58.2 83.2 80.0 99.0 89.5 87.4 96.0 100.0
Baby’s sex is determined by 
father’s sperm 8.0 24.8 0.0 0.0 3.0 31.3 9.0 0.0
(Source: Baseline and Endline Surveys)
Based on menstruation myths identified in formative research, changes in myths and 
misconceptions around menstruation were measured at baseline and endline among survey 
respondents. As highlighted in Table 15, the proportion of intervention women and men who 
reported believing menstruation myths (asked as “true” or “false” answer options), dropped 
from baseline to endline. These include beliefs that the pumpkin garden will dry if a woman 
gardens during her period (intervention women: 80% to 51%; intervention men: 68% to 54% 
from baseline to endline); and that a woman is unclean or “polluted” during her menstrual 
period (intervention women: 70% to 54%; intervention men: 81% to 74% from baseline to 
endline). 
Table 15| Proportion of women and men who answered “true” when asked about menstruation myths, by 
intervention arm and time point (n=600)
MENSTRUATION 
MEASURES 
WOMEN MEN
INTERVENTION CONTROL INTERVENTION CONTROL
Baseline %
(n=201)
Endline %
(n=202)
Baseline %
(n=100)
Endline %
(n=100)
Baseline %
(n=199)
Endline %
(n=198)
Baseline %
(n=100)
Endline %
(n=100)
If a woman touches a cow 
during her period, the cow 
will go blind
50.0 35.6 61.0 68.0 57.8 50.5 67.0 72.0
A woman is unclean or 
“polluted” during her 
menstrual period
69.5 53.5 88.0 79.0 80.9 74.2 98.0 93.0
If a woman goes to the 
pumpkin garden when she has 
her menstrual period, the 
plants will dry
80.1 51.0 97.0 100.0 67.8 53.5 82.0 71.0
(Source: Baseline and Endline Surveys)
In addition, intervention women and men were more likely to believe that a woman is able 
to milk a cow during her menstrual period (intervention women: 26% to 50%; intervention men: 
33% to 44% from baseline to endline); and that a woman can clean the nose of a child during 
her period (intervention women: 16% to 43%; intervention men: 51% to 72% from baseline to 
endline.1 No improvements over time were found among control female or male 
respondents. 
Family Planning Use 
Trends over time show increases in intent to use modern FP, but limited improvements in 
current modern FP use. As seen in Table 16, intervention respondents’ reported intent to use 
a modern FP method increased significantly from baseline to endline, increasing from 17% to 
 
1 Data not shown in table
63
23% over time (p=0.0219). Specifically, intervention respondents reported increased intent to 
use pills (0% to 5% baseline to endline), injectable (2% to 7%) and condoms (3% to 5%). No 
significant changes were reported in the control group in intent to use a modern FP method.
There were no significant improvements over time in current use of modern FP, decreasingly 
slightly over time for both intervention and control groups. Intervention respondents’ reported 
current modern FP use decreased from baseline to endline, 14% to 12% (p=0.2868), although 
this change was not significant. Trends in reported abstinence use decreased slightly among 
intervention respondents, from 39% to 33% baseline to endline, respectively.
Table 16| FP outcomes baseline to endline, among intervention and control groups (n=600)
(Source: Baseline and Endline Surveys)
Table 16 above that the use of modern FP use decreased slightly from 14% to 12% among the 
intervention sample (p=0.2868). Sex-disaggregated data suggest that current use of modern 
FP reported by intervention women decreased from baseline to endline from 16% to 6%, 
respectively, and this decrease was mainly attributed specifically to decreased use of LAM 
over time (14% to 4%). Current use of modern FP by intervention men remained unchanged 
at 14% at both baseline and endline. 
FAMILY PLANNING USE
MEASURES
INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=400 % n=400 % P-Value n=200 % n=200 % P-Value
Intends to use modern FP in 
next 3 months 66 16.5 92 23.0 0.0219 14 7.0 8 4.0 0.1882
Type of Method Intend to Use 
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
CycleBeads
TwoDay Method
LAM
Withdrawal 
Abstinence 
 Other
13
0
0
7
14
0
0
0
0
0
32
1
149
9
3.3
0.0
0.0
1.8
3.5
0.0
0.0
0.0
0.0
0.0
8.0
0.3
37.3
2.3
20
0
19
26
16
1
0
1
1
0
8
0
168
4
5.0
0.0
4.8
6.5
4.0
0.3
0.0
0.3
0.3
0.0
2.0
0.0
42.0
1.0
4
0
1
2
1
0
0
0
0
0
6
2
97
0
2.0
0.0
0.5
1.0
0.5
0.0
0.0
0.0
0.0
0.0
3.0
1.0
48.5
0.0
1
0
0
2
0
0
0
0
0
0
5
0
103
0
0.5
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
2.5
0.0
51.5
0.0
Currently using modern FP 57 14.3 47 11.8 0.2868 23 11.5 17 8.5 0.3173
Type of FP Method Using
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
CycleBeads
TwoDay Method
LAM
Withdrawal 
Abstinence 
 Other
9
0
1
5
1
0
0
0
1
0
45
2
157
4
2.2
0
0.3
1.3
0.3
0.0
0.0
0.0
0.3
0.0
11.3
0.5
39.3
1.0
14
0
4
3
1
0
0
0
1
0
28
0
131
0
3.5
0.0
1.0
0.8
0.3
0.0
0.0
0.0
0.3
0.0
7.0
0.0
32.8
0.0
5
0
0
0
0
0
0
1
0
0
20
10
93
1
2.5
0.0
0.0
0.0
0.0
0.0
0.0
0.5
0.0
0.0
10.0
5.0
46.5
1.0
3
0
0
0
0
0
0
0
0
0
14
0
96
0
1.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
7.0
0.0
48.0
0.0
64
Figures 17 and 18, show trends in intention to use modern FP use by male and female 
respondents. Both women and men in intervention communities reported increased intent to 
use a modern FP method at endline. Approximately 25% to 29% of intervention women 
reported an intent to use modern FP from baseline to endline, respectively (p=0.3427). Among 
intervention men, 8% reported an intent to use a modern FP method at baseline, doubling 
significantly to 17% at endline (p=0.009). Intent to use modern FP was generally lower among 
control respondents, decreasing insignificantly from baseline to endline. 
Figure 17| Intervention women and men who 
intend to use modern FP, at baseline and endline 
(n=400)
Figure 18| Control women and men who intend 
to use modern FP, at baseline and endline 
(n=200)
(Source: Baseline and Endline Surveys; Note: p-values: * <.05; ** <.01; ***<.001)
Self-efficacy to access and use FP
Table 17 shows the proportion of intervention and control respondents with high self efficacy 
to use and access FP (defined by mean score > 3.8) at baseline and endline, followed by the 
individual measures used to construct the self-efficacy score. Overall, both study arms 
reported high FP self-efficacy at baseline, increasing significantly from baseline to endline 
and in particular among the control group (intervention 62% to 69%; p=0.027; control 73% to 
94%; p<0.0001). The individual measures show a high proportion of intervention and control 
respondents who agreed that they felt confident they could use a method correctly to delay 
or avoid pregnancy (intervention 88% to 90%; control 94% to 98%), and that they knew where 
to obtain a method of FP to delay or avoid pregnancy (intervention 80% to 86%; control 86% 
to 99%). 
25%
8%
29%
17%**
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
WOMEN MEN
8% 6% 6% 2%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
WOMEN MEN
BASELINE
(n=201)
ENDLINE
(n=202)
BASELINE
(n=199)
ENDLINE
(n=198)
BASELINE
(n=100)
ENDLINE
(n=100)
ENDLINE
(n=100)
BASELINE
(n=100)
65
Table 17| Self-efficacy to access and use FP at baseline to endline, among intervention and control groups
(n=600)
(Source: Baseline and Endline Surveys)
Sex-disaggregated data suggest that the high baseline ceiling effect and improvements 
were primarily due to the reports of male respondents. Among intervention men, 66% had 
high self-efficacy scores, compared to 82% at endline (p=0.0003), compared to 79% at 
baseline to 100% at endline for control men (p<0.0001).
Attitudes about FP
The proportion of intervention and control respondents with more positive attitudes about FP
(defined by mean score > 2.9) increased significantly for both intervention and control 
respondents over time (intervention 70% to 92%; p<0.0001; control 86% to 95%; p=0.001).
Among intervention respondents, 78% reported agreeing that FP methods are a way to 
continue the Karimojong way of life at baseline, increasing to 97% at endline. The majority of 
intervention and control respondents agreed that using FP methods is safe for a woman’s 
health at both baseline and endline (intervention 93% to 97%; control 96% to 100%), while a 
larger proportion of intervention respondents agreed that it is acceptable that a women 
bring up the topic of FP with her husband (88% to 96% baseline to endline). Trends in FP 
attitudes from baseline to endline were similar between female and male respondents in 
respective intervention and control groups (data not shown). 
INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=400 % n=400 % P-Value n=200 % n=200 % P-Value
SELF-EFFICACY TO ACCESS AND USE FP SCORE
% with high self-efficacy
( mean > 3.8) 242 62.2 272 69.0 0.027 143 72.6 188 94.0 <.0001
SELF-EFFICACY MEASURES
% who agree with following statements:
I am confident I could use a 
method correctly all the time 
to delay or avoid pregnancy
350 87.7 359 89.8 0.316 187 93.5 197 98.5 0.011
I am confident I could use a 
method correctly all the time 
to delay or avoid pregnancy 
even if my partner disagrees
287 71.9 298 74.5 0.380 158 79.0 188 94.0 <.0001
I have the information I need 
to make a decision about 
whether to use FP if I wanted 
to delay or avoid pregnancy
324 81.2 342 85.5 0.088 180 90.0 197 98.5 <.0001
I know where to obtain a 
method of FP to delay or 
avoid pregnancy
322 80.7 344 86.0 0.037 172 86.0 198 99.0 <.0001
66
Table 18| Attitudes about FP at baseline to endline, among intervention and control groups (n=600)
(Source: Baseline and Endline Surveys)
Social Norms and Expectations about FP use
Table 19 shows respondent’s perceptions of social norms and expectations of FP use among 
couples. The proportion of respondents who reported more supportive social expectations 
around FP use among couples (defined by mean score > 2.7) increased for both intervention 
and control groups, although control respondents were more likely to report supportive FP 
norms at both time points (intervention 62% to 81%; p<0.0001; control 73% to 99%; p<0.0001).
At baseline, 74% of intervention respondents agreed with the descriptive social expectation
that most couples in their community who want to prevent a pregnancy use a method of FP, 
increasing to 92% of respondents at endline (control 84% to 99%), while 80% to 95% of 
intervention respondents agreed that couples in their community who want to prevent 
pregnancy should use a method of FP from baseline to endline (control 82% to 100%). 
INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=400 % n=400 % P-Value n=200 % n=200 % P-Value
FP ATTITUDES SCORE
% with more positive FP 
attitudes
( mean > 2.9) 
276 70.2 365 92.0 <.0001 168 85.7 189 94.5 0.001
FP ATTITUDES MEASURES
% who agree with following statements:
Using a FP method is safe for 
a woman’s health 369 92.5 388 97.0 0.003 191 95.5 199 99.5 0.010
It is acceptable for a woman 
to bring up the topic of FP 
with her husband
349 87.5 385 96.3 <.0001 181 90.5 193 96.5 0.015
It is acceptable for a man to 
bring up the topic of FP with 
his wife
347 87.0 389 97.3 <.0001 185 92.5 198 99.0 0.001
FP methods are a way to 
continue the Karamoja culture 
and way of life
310 77.7 386 96.5 <.0001 190 95.0 196 98.0 0.103
67
Table 19| Social norms around FP at baseline to endline, among intervention and control groups (n=600)
(Source: Baseline and Endline Surveys)
Additional analysis shows that an increasing proportion of intervention men reported more 
perceived supportive norms around FP over time, increasing from 53% to 76% from baseline 
to endline, respectively. However intervention women’s reports remained unchanged over 
time. 
Couple Communication about FP 
As seen in Table 20, respondents in intervention communities reported increased partner 
communication about FP over time, relative to respondents in control communities. 
Approximately 58% of intervention respondents reported high communication with their 
partner about FP (both discussing FP in the last three months and feeling that their opinion is 
valued when discussing FP), increasing to 73% of intervention respondents at endline 
(p<0.0001). No significant improvements were found among the control group (74% baseline 
and endline; p=0.4103). More specifically, 60% to 74% of intervention respondents reported 
talking about FP with their partner in the last three months from baseline to endline 
respectively (control 75% to 78%). Within the intervention group, improvements in couple 
communication from baseline to endline were similar between female and male respondents 
(data not shown). 
INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=400 % n=400 % P-Value n=200 % n=200 % P-Value
FP SOCIAL NORMS
% who report more supportive 
FP norms ( mean > 2.7) 237 61.7 321 81.3 <.0001 143 72.6 197 98.5 <.0001
FP SOCIAL NORMS MEASURES
% who agree with following statements:
Most couples in my 
community who want to 
prevent pregnancy use 
family planning 
297 74.4 368 92.0 <.0001 168 84.0 198 99.0 <.0001
Couples in my community who 
want to prevent pregnancy 
should use family planning 
320 80.2 381 95.3 <.0001 164 82.0 199 99.5 <.0001
I should use a family 
planning method if I want to 
prevent pregnancy 
339 85.0 370 92.5 0.001 189 94.5 199 99.5 <.0001
68
Table 20| Couple Communication about FP at baseline to endline, among intervention and control groups (n=600)
(Source: Baseline and Endline Surveys; Note: restricted to respondents living with a partner)
Community Communication about Fertility Awareness and FP 
Table 21 shows that both intervention and control respondents reported slightly higher 
community communicating about fertility awareness and FP from baseline to endline. About 
14% of intervention respondents reported high communication with other in their community 
about fertility awareness and FP (talking about fertile days during the menstrual cycle with 
someone, correcting someone when they heard them saying something untrue about 
menstruation, and; when saying something untrue about FP methods) increasing to 25% of 
intervention respondents at endline (p=0.0004). Increases in high communication with others 
in the community were also found in the control group (from 8% to 25%; p<0.0001). While a 
larger proportion of respondents in intervention communities reported talking about fertile 
days and correcting something untrue about menstruation from baseline to endline, the 
proportion of intervention respondents correcting something untrue about FP methods 
stayed relatively the same at both time points (34%). 
INTERVENTION CONTROL
Baseline Endline Baseline Endline
n=367 % n=344 % P-Value n=196 % n=189 % P-Value
COUPLE COMMUNICATION ABOUT FP
High communication
Low communication
No communication
206
131
21
57.5
36.6
5.9
243
88
2
73.0
26.4
0.6
<.0001
140
46
3
74.1
24.3
1.6
140
42
7
74.1
22.2
3.7
0.4103
COUPLE COMMUNICATION MEASURES
% of respondents who report
Feeling that their opinion is 
taken into account when 
talking about FP with their 
partner
327 89.3 326 94.8 0.006 185 94.4 174 92.1 0.364
Discussing FP to delay or 
avoid pregnancy with their 
partner in last 3 months
220 60.1 253 73.6 <.0001 146 74.5 148 78.3 0.378
69
Table 21| Community Communication about Fertility Awareness and FP at baseline to endline, among 
intervention and control groups (n=600)
(Source: Baseline and Endline Surveys)
Difference-in-Difference Results: Intervention effect across time
Following the descriptive comparisons, bivariate and multivariate logistic regression models 
were conducted – using difference-in-difference estimation – to assess potential effect of 
living in an intervention community and experiencing improved fertility awareness and FP 
outcomes. The models assumed an intention to treat (ITT) approach, defining all respondents 
in intervention communities as “exposed” to intervention regardless of whether or not they 
attended a performance. The difference-in-difference estimation was used to estimate 
causality of the intervention since it not only compares an intervention group to a control 
group, but also controls for any time-variant factors from baseline to endline (that is 
controlling for any differences in outcomes over time due to non-intervention factors such as 
seasonality, migration, other NGO FP programs, etc.) While this approach aimed to 
understand community-wide effect, rather than individual effect, this approach generally 
results in a more conservative estimate of the treatment effect because it does not remove 
non-exposed community members from the analysis.
Multivariate logistic regression models were used reporting AORs, and controlled for 
significantly different at baseline and endline were controlled for in the model, including:
INTERVENTION CONTROL
Baseline Endline Endline Baseline
n=400 % n=400 % P-value % n=200 % n=200 P-value
COMMUNITY COMMUNICATION ABOUT FERTILITY AWARENESS FP
High communication
Moderate communication
Low communication
No communication
55
36
97
200
14.2
9.3
25.0
51.6
99
30
68
198
25.1
7.6
17.2
50.1
0.0004 16
12
24
141
8.3
6.2
12.4
73.1
49
0
9
139
24.9
0.0
4.6
70.6
<.0001
COMMUNITY COMMUNICATION MEASURES
% of respondents who report that in the last three months
Talked about fertile days 
during the menstrual cycle 
with someone
111 27.8 163 40.8 <.0001 37 18.5 52 26.0 0.071
Corrected someone when
heard them saying something 
untrue about menstruation
92 23.1 125 31.3 0.009 24 12.0 49 24.5 0.001
Corrected someone when
heard them saying something 
untrue about FP methods
134 33.6 137 34.3 0.823 37 18.5 56 28.0 0.043
70
 Gender
 Age
 Education
 Employment
 Living with a marital partner 
 Number of children
Tables 22 - 24 show the results of the adjusted regression models on fertility awareness and FP
outcomes using the difference-in-difference estimation, controlling for time trends from 
baseline to endline between intervention and control groups. The results show the effect of 
living in an intervention community – regardless of whether a survey respondent attended a 
performance - on the likelihood of experiencing a high fertility awareness score, reporting 
current use of modern FP and intent to use modern FP. In each table, three separate effect 
variables are presented: 1. DID effect, or the likelihood of experiencing an outcome if living 
in an intervention community, controlling for time trends; 2. Time effect, or the likelihood of 
experiencing an outcome at endline, relative to baseline, and; 3. Intervention effect, or the 
likelihood of experiencing an outcome if intervention community, pooling both baseline and 
endline time points. Each of the three tables show results of three separate regression models: 
Model 1. Unadjusted and not controlling for other demographics or predictor variables; 
Model 2. Adjusted for significantly different demographic variables between intervention and 
control groups; Model 3. Adjusted for both demographic and secondary predictors (e.g. FP 
attitudes, FP self-efficacy, etc.)
Table 22 shows that respondents living in an intervention community were more than seven 
time more likely to have a high FA score, relative to those in control communities (aOR 7.3; 
p<.0001). The effect appears stronger when controlling for demographic variables (aOR 9.3; 
p<.0001), in addition to secondary outcome indicators (aOR 13.8; p<.0001). This is consistent 
with descriptive results showing that the proportion of respondents with high fertility awareness 
increased in both intervention and control communities, but significantly more so among 
intervention communities (15% to 75% baseline to endline in intervention; 30% to 53% in 
control). Predictors of high fertility awareness score included being older (aOR 1.3; p<.0001), 
and having higher levels of education (aOR 1.3; p<.0001), and having high communication 
with others in the community about fertility awareness and FP topics (aOR 1.7; p<.0001). 
Respondents reporting more supportive FP norms and expectations in their community were
less likely to have a high FA score (aOR 0.63; p=0.037) Further regression modeling exploring 
heterogeneous effects among male and female respondents separately, show that both 
women and men had high fertility awareness scores when controlling for demographic 
variables, although the likelihood of having a high fertility awareness score were higher 
among intervention men (male aOR 19.0; p<.0001; female aOR 9.8; p<.0001).
71
Table 22| Intervention effect on high fertility awareness, using difference-in-difference
(Source: Baseline and Endline Surveys; Note: Living with a partner variable omitted from Model 3 due to multicollinearity 
with couple communication about FP variable)
As seen in Table 23 respondents in intervention communities were approximately three times 
more likely to intend to use a modern FP method, and this effect was significant (aOR 3.11; 
p=0.026). This effect is consistent with descriptive data showing a proportional increase in 
intent to use modern FP from 17% to 23% baseline to endline among intervention respondents.
The results show a number of sociodemographic predictors of modern FP use intentions, 
including gender (aOR 2.5; p<.0001), age (aOR 0.65; p<.0001), education (aOR 1.5; p<.0001) 
and having formal employment (aOR 2.8; p<.0001). Further regression models explored 
gendered differences in modern FP use intent (data not shown here) and showed that
likelihood of reporting an intent to use a modern FP method was statistically significant and 
strong among men in intervention community (aOR 8.5; p= 0.032), but not women (aOR 2.0; 
p=0.332). The results suggest that having high self-efficacy to use FP and community 
Fertility Awareness 
Unadjusted model 1
Fertility Awareness 
Adjusted demographics 
model 2
Fertility Awareness 
Adjusted demographics + 
predictors model 3
OR 
(95% CI) 
P-value aOR 
(95% CI)
P-value aOR 
(95% CI)
P-value
Effect variables
DID (Time * Intervention)
7.33
(4.21, 12.77)
<.0001 9.28
(5.21, 16.52) <.0001 13.77
(7.17, 26.42)
<.0001
Time
2.64
(0.21, 0.51)
<.0001 2.56
(1.67, 0.52) <.0001 2.39
(1.49, 3.84)
<.0001
Intervention
0.33
(1.75, 3.99) <.0001 0.34
(0.22, 3.92) <.0001 0.25
(0.15, 0.41)
<.0001
Sociodemographic variables
Gender
1.00
(0.74, 1.36) 0.994
1.32
(0.92, 1.90) 0.131
Age 
1.25
(1.10, 1.42) <.0001
1.26
(1.10, 1.46) 0.001
Education
1.32
(1.14, 1.52) <.0001
1.16
(0.99, 1.37) 0.075
Employment 
1.08
(0.79, 1.47) 0.636
0.92
(0.64, 1.33) 0.674
Has child/ren
0.96
(0.89, 1.05) 0.387
0.94
(0.86, 1.04) 0.225
Living with a partner
0.44
(0.27, 0.73) 0.002 -- --
Predictor variables
Current modern FP use
1.23
(0.76, 1.98) 0.392
FP self-efficacy
1.08
(0.72, 1.62) 0.700
FP attitudes
1.55
(0.89, 2.69) 0.118
FP norms
0.63
(0.41, 0.97) 0.037
Couple communication about 
FP 
0.86
(0.62, 1.19) 0.354
Community communication
about FA and FP 
1.70
(1.46, 1.98) <.0001
Number observations n=1,200 n=1,200 n=1,016
72
communication about fertility awareness and FP are also significantly associated with
intending to use modern FP (aOR 3.8; p<.0001; aOR 1.3; p=0.006).
Table 23| Intervention effect on intent to use modern FP, using difference-in-difference 
(Source: Baseline and Endline Surveys; Note: Living with a partner variable omitted from Model 3 due to multicollinearity 
with couple communication about FP variable)
The regression results in Table 24 show that respondents in intervention communities were no 
more likely to use modern FP than those in control communities (aOR 1.1; p=0.797). These 
results are consistent with the descriptive results showing that use of modern FP stalled at 
about 14% to 12% among the intervention sample over time. Both age and education were 
significantly associated with current modern FP use: those younger were more likely to be 
using a modern of FP (aOR 0.84; 0.046), in addition to those with higher levels of education 
(aOR 1.2; p=0.037). Interestingly, having more positive attitudes about FP were inversely 
associated with current use of modern FP (aOR 0.4; p=0.004). 
Intent to use modern FP 
Unadjusted model 1
Intent to use modern FP 
Adjusted demographics 
model 2
Intent to use modern FP 
Adjusted demographics + 
predictors model 3
OR 
(95% CI) 
P-value aOR 
(95% CI)
P-value aOR 
(95% CI)
P-value
Effect variables
DID (Time * Intervention)
2.72
(1.04, 7.10) 0.041
3.11
(1.14, 8.46) 0.026
3.67
(1.24, 10.91) 0.019
Time
0.55
(0.23, 1.35) 0.194
0.42
(0.17, 1.05) 0.064
0.31
(0.12, 0.85) 0.022
Intervention
2.63
(1.44, 4.82) 0.002
2.01
(1.06, 3.78) 0.032
1.95
(0.98, 3.89) 0.057
Sociodemographic variables
Gender
2.51
(1.64, 3.86) <.0001
4.11
(2.46, 6.88) <.0001
Age 
0.65
(0.54, 0.78) <.0001
0.67
(0.54, 0.82) <.0001
Education
1.47
(1.25, 1.72) <.0001
1.42
(1.17, 1.73) <.0001
Employment 
2.82
(1.86, 4.30) <.0001
2.53
(1.55, 4.11) <.0001
Has children
1.09
(0.96, 1.23) 0.193
1.06
(0.92, 1.22) 0.429
Living with a partner
2.23
(1.35, 3.68) 0.002 -- --
Predictor variables
High fertility awareness 
0.89
(0.54, 1.48) 0.656
FP self-efficacy
3.84
(2.25, 6.55) <.0001
FP attitudes
0.55
(0.30, 1.02) 0.056
FP norms
1.29
(0.74, 2.25) 0.367
Couple communication about 
FP 
1.13
(0.74, 1.72) 0.568
Community communication
about FA and FP 
1.29
(1.08, 1.54) 0.006
Number observations n=1,199 n=1,199 n=1,016
73
Table 24| Intervention effect on current modern FP use between intervention and control groups and across time, 
using difference-in-difference estimation
(Source: Baseline and Endline Surveys; Note: Living with a partner variable omitted from Model 3 due to multicollinearity 
with couple communication about FP variable)
2. EXPOSURE LEVEL RESULTS AT ENDLINE
Following the community-level analyses, exposure level analyses were conducted on the 
endline sample only to better understand individual-level associations between exposure to a 
performance and likelihood of experiencing higher fertility awareness and positive FP 
outcomes. This secondary analysis was conducted since the above ITT community-level 
analysis considered all community members in intervention communities exposed to the 
intervention regardless of whether or not they attended a performance. The exposure level 
analysis is a more targeted analysis approach, but only uses endline data since no one was 
exposed at baseline. As a result, associations between EDEAN exposure and outcomes did
Current modern FP use 
Unadjusted model 1
Current modern FP use 
Adjusted demographics 
model 2
Current modern FP use 
Adjusted demographics + 
predictors model 3
OR 
(95% CI) 
P-value aOR 
(95% CI)
P-value aOR 
(95% CI)
P-value
Effect variables
DID (Time * Intervention)
1.12
(0.51, 2.44) 0.780
1.11
(0.50, 2.45) 0.797
1.28
(0.54, 3.01) 0.579
Time
0.71
(0.37, 1.38) 0.319
0.72
(0.37, 1.40) 0.330
0.61
(0.31, 1.23) 0.167
Intervention
1.28
(0.76, 2.15) 0.345
1.23
(0.72, 2.07) 0.448
1.09
(0.61, 1.92) 0.779
Sociodemographic variables
Gender
1.00
(0.67, 1.48) 0.982
1.14
(0.74, 1.76) 0.547
Age 
0.84
(0.71, 1.00) 0.046
0.87
(0.73, 1.04) 0.133
Education
1.19
(1.01, 1.39) 0.037
1.14
(0.95, 1.37) 0.153
Employment 
0.90
(0.60, 1.35) 0.616
0.70
(0.44, 1.09) 0.117
Has children
1.11
(1.00, 1.23) 0.055
1.10
(0.99, 1.23) 0.090
Living with a partner
0.88
(0.46, 1.69) 0.697 -- --
Predictor variables
High fertility awareness 
1.25
(0.78, 2.00) 0.350
FP self-efficacy
1.59
(0.97, 2.60) 0.067
FP attitudes
0.44
(0.25, 0.77) 0.004
FP norms
1.50
(0.90, 2.51) 0.124
Couple communication about 
FP 
0.97
(0.66, 1.43) 0.871
Community communication
about FA and FP 
1.13
(0.95, 1.34) 0.163
Number observations n=1,199 n=1,199 n=1,016
74
not take into account any temporal baseline to endline effects outside of the intervention, 
and is a less robust estimation method to attribute causality. For this section, those exposed to 
at least one performance were considered “viewers,” and non-exposed as “non-viewers.”
Community Exposure to EDEAN Performances 
As seen in Figure 19, almost all women (94%) and two-thirds of men (64%) had heard about 
the EDEAN performances in intervention communities and more female respondents in the 
intervention communities viewed the performances than men (79% and 51%, respectively). 
Figure 19| Exposure to EDEAN performances at endline, by intervention men and women (n=400) 
The number of performances attended was similar for men and women. As reflected in Figure 
20, half of the participants viewed two to three performances (52%), while 1/5 went 10 to 12
times (20%). 
Figure 20| Frequency of EDEAN performance attendance among exposed participants (n=260)
 (Source: Endline Survey)
Association between viewing performances and key FA/FP outcomes
Table 25 shows aOR results from logistic regression models comparing viewers versus non￾viewers (viewers defined as having attended at least one performance) for the total endline
94%
64%
79%
51%
0% 20% 40% 60% 80% 100%
Women
(n=202)
Men
(n=198)
Attended an EDEAN performance Heard of EDEAN performances
52%
17%
10%
20%
0% 10% 20% 30% 40% 50% 60%
1-3 times
4-6 times
7-9 times
10-12 times
75
sample; female sample (female viewers vs. female non-viewers at endline); male sample 
(male viewers vs. male non-viewers at endline); 15-25 year old sample (15 – 25 year old viewers
vs. 15-25 year old non-viewers at endline), and so forth with higher age categories. 
The results show that individuals who viewed the theatre performances were more than twice 
as likely to have higher fertility awareness (aOR=2.52; p=<.0001), compared to non-viewers. 
Relative to non-viewers, performance viewers were more than five times more likely to report 
intent to use modern FP (aOR=5.68; p=<.0001), and twice more likely to report current modern 
FP use (aOR=2.02; p=0.007), the latter of which was mainly attributed to exposed men’s 
reported use of condoms. Women’s reports of current modern FP use were similar and 
insignificant among female viewers and non-viewers. At endline, approximately 21% of 
women viewers compared to 35% of non-viewers reported use of abstinence as method of 
FP. Qualitative findings suggest that while some women and men knew of peers going to 
health services to seek more information about FP after performances, others reported that 
peers continued use of post-partum abstinence practices (sleeping in different houses until 
the child is about eight months). 
Associations between viewing at least one performance and key fertility awareness and FP 
outcomes were stronger for younger individuals: viewers between the ages of 15 to 25 years
were 4 times more likely to report current modern FP use (aOR 4.05; p=.028), and 10 times more 
likely to report intent to use modern FP (aOR 10.08; p=<.0001), than 15-25 year old non-viewers.
Table 25| Adjusted logistic regression results for fertility awareness and FP, by sex and age
(Source: Endline Surveys; p-values: * <.05; ** <.01; ***<.001; model adjusted for age, employment, education, living 
with partner, having children)
Logistic regression models were also conducted on the same key fertility awareness and FP
outcomes, but restricted to individuals in the sample who desire to prevent pregnancy (see 
Annex 5). Desire to prevent pregnancy was assessed by asking respondents “do you want to 
get pregnant in the next 12 months?” Any respondent answering “no” was considered to have 
a desire to prevent a pregnancy. In total, 74.7% of the endline sample reported a desire to 
prevent a pregnancy (n=448); these proportions were similar when segmenting male and 
female respondents. The results show that among those who desire to prevent a pregnancy, 
viewers were more than twice as likely to have a high fertility awareness score, relative to non￾viewers (aOR 2.66; p=0.0002). Male viewers were 14 times more likely to report intent to use a 
Total aOR
(n=600)
Female aOR
(n=302)
Male aOR
(n=298)
15-25 yrs aOR
(n=231)
26 – 35 yrs aOR
(n=236)
36+ yrs aOR
(n=133)
High fertility 
awareness 
(score > 8 )
2.52
(1.66, 3.82)
p<.0001
2.23
(1.29, 3.86)
p=0.0042
2.48
(1.20, 5.09)
p=.0135
2.31
(1.18, 4.49)
p=0.0141
2.28
(1.14, 4.56)
p=0.0203
3.31
(1.06, 10.33)
p=0.0392
Current modern FP 
use
2.02
(1.12, 3.66)
p=0.0066
1.19
(0.44, 3.18)
p=0.7318
2.84
(1.25, 6.44)
p=0.0127
4.05
(1.17, 14.09)
p=0.0278
1.62
(0.66, 3.97)
p=0.2879
1.35
(0.28, 6.59)
p=0.7074
Intent modern FP use 
in next 3 months
5.68
(3.04, 10.61)
p<.0001
6.58
(2.87, 15.08)
p<.0001
8.06
(2.51, 25.91)
p=0.0005
10.08
(3.93, 25.86)
p<.0001
2.92
(1.08, 7.86)
p=0.0342
8.20
(0.54, 124.56)
p=0.1297
76
modern FP method relative to male non-viewers (aOR 14.38; p=0.0215), while 15-25 year old 
viewers were 22 times more likely to report an intent to use modern FP, relative to non-viewers
of the same age range (aOR 22.16; p=0.0005). 
Table 26 below shows associations between viewing at least one performance and other 
outcomes, such as self-efficacy to use FP and health seeking behaviors, among others. The 
results show that individuals who viewed the theatre performances close to six times more likely 
to seek information about FP from a health worker (aOR 5.79; p<.0001), and; significantly less 
likely to report that modern FP methods cause side effects when asked what they had heard 
about modern FP methods (aOR 0.16; p=<.0001), compared to non-viewers. The segmented 
sex and age analysis suggest that performances had differential impacts on different viewers: 
associations were significantly stronger for those 15-25 years old, relative to older age groups. 
Across gender, female viewers were more likely to report higher partner communication about 
FP and community communication about menstruation and fertility, relative to unexposed
(aOR 8.07; p<.0001). 
Table 26| Adjusted logistic regression analysis results for secondary outcomes, by age and sex
(Source: Endline Surveys; p-values: * <.05; ** <.01; ***<.001; model adjusted for age, employment, education, living with 
partner, having children)
 
1 aOR not shown since no unexposed men (0) responded this answer affirmatively / negatively
Total aOR
(n=600)
Female aOR
(n=302)
Male aOR
(n=298)
15-25 yrs aOR
(n=231)
26 – 35 yrs aOR
(n=236)
36+ yrs aOR
(n=133)
High FP self-efficacy 0.81
(0.52, 1.28)
p=0.3643
0.46
(0.26, 0.81)
p=0.0071
6.49
(1.93, 21.85)
p=0.0025
1.72
(0.78, 3.80)
p=0.1789
0.75
(0.36, 1.58)
p=0.4533
0.36
(0.11, 1.17)
p=0.0889
Sought info about FP 
from health worker in 
last 3 months
6.04
(3.47, 10.51)
p<0.0001
3.26
(1.58, 6.73)
P=0.0014
11.68
(4.69, 29.11)
p<0.0001
5.97
(2.36, 15.09)
p=0.0002
6.38
(2.604, 15.63)
p<.0001
5.55
(1.38, 22.37)
p=0.0159
FP methods help 
prevent pregnancy 
4.99
(2.60, 9.56)
p<.0001
19.09
(6.51, 56.03)
P<.0001
0.60
(0.20, 1.81)
p=0.3639
10.07
(2.70, 37.62)
p=0.0006
2.97
(1.20, 7.30)
p=0.0176
6.02
(0.96, 37.80)
p=0.0554
FP methods cause side 
effects 
0.25
(0.14, 0.44)
p<.0001
0.10
(0.05, 0.20)
<.0001
-------
1
0.54
(0.22, 1.33)
p=0.1820
0.12
(0.04, 0.37)
p=0.0002
0.12
(0.02, 0.69)
p=0.0182
High FP attitudes 1.18
(0.58, 2.37)
p=0.6514
1.14
(0.50, 2.63)
p=0.7570
2.19
(0.49, 9.78)
p=0.3058
1.68
(0.63, 4.49)
p=0.3001
0.70
(0.20, 2.46)
p=0.5774
1.93
(0.11, 34.91)
p=0.6554
High support for FP 
social norms score
1.17
(0.66, 2.08)
p=0.5823
0.51
(0.21, 1.22)
p=0.1278
2.48
(0.96, 6.45)
p=0.0621
0.90
(0.35, 2.28)
p=0.8209
2.18
(0.83, 5.68)
p=0.1125
0.59
(0.12, 3.03)
p=0.5308
High FP communication 
with partner
2.33
(1.44, 3.74)
p<.0001
2.56
(1.41, 4.68)
p=0.0021
1.97
(0.91, 4.25)
p=0.0853
2.94
(1.37, 6.33)
p=0.0037
3.98
(1.58, 10.04)
p=0.0034
1.43
(0.51, 4.00)
p=0.4981
High communication on
fertility awareness and 
FP with community
4.15
(2.79, 6.18)
p<.0001
8.07
(4.08, 15.98)
p<.0001
2.57
(1.48, 4.47)
p=0.0008
5.07
(2.69, 9.54)
p<.0001
3.89
(2.00, 7.58)
p<.0001
3.56
(1.23, 10.29)
p<.0189
77
3. ASSOCIATION BETWEEN HIGH FERTILITY AWARENESS AND FP USE
The relationship between high fertility awareness and modern FP use within the endline sample 
was explored, using contingency tables with chi square tests. Any statistically significant 
associations were followed by OR calculation to determine the odds of experiencing an FP 
outcome based on a high fertility awareness score. This process was repeated for the 
composite high fertility awareness score and the 14 individual fertility awareness measures (for 
each question answered correctly).
There were no significant associations between having an overall high fertility awareness score 
and current modern FP use nor intent to use modern FP, respectively. However, a few 
significant associations were observed for individual fertility awareness measures and current 
modern FP use (1 measure), and intent to use modern FP (4 measures). 
As seen in Table 27, respondents who correctly identified the fertile window were twice more
likely to use a current modern FP method relative to those who identified the fertile window 
incorrectly (OR 2.02; p=0.0491). 
Table 27 | Odds of current modern FP use, by correct knowledge of fertile window (n=600)
Current FP Use
Odds Ratio (95% CI) P-value
A woman is most likely to get 
pregnant if she has unprotected 
sex halfway between two 
menstrual cycles
2.02 (1.00, 4.08) 0.0491
(Source: Endline Surveys; p-values)
Table 28 shows significant associations between individual measures and intent to use a 
modern FP method in the next three months. Answering the following three measures correctly 
were all statistically significantly associated with an increased likelihood of intent to use 
modern FP: 1. healthy vaginal secretions are a sign of fertility (OR 3.23; p<.0001), 2. Vaginal 
secretions come during the middle of the menstrual cycle (OR 2.14; p=0.0017), and; 3. Men 
are fertile every day (OR 4.82; p<.0001). Interestingly, respondents who knew that most 
menstrual cycles last between 26 to 32 days were half as likely to intend to use modern FP, 
relative to those who did not answer the question correctly (OR 0.49; p=0.0078).
Table 28 | Odds of intent to use modern FP, by fertility awareness individual measures (n=600)
Intent to use FP
Odds Ratio (95% CI) P-value
Other than sexual arousal healthy 
vaginal secretions are a sign of fertility
3.23
(1.89, 5.52) <.0001
Vaginal secretions come during the 
middle of the menstrual cycle
2.14 
(1.33, 3.45) 0.0017
Men are fertile every day
4.82
(2.73, 8.49) <.0001
Most menstrual cycles last between 26 
to 32 days
0.49 
(0.28, 0.83) 0.0078
 (Source: Endline Surveys)
78
LESSONS LEARNED ABOUT EFFECTIVENESS
EDEAN led to higher fertility awareness among performance viewers and the community at 
large. The pilot results strongly suggest that surveyed respondents in intervention communities 
had increased fertility awareness – that is knowledge about male and female fertility, 
pregnancy risk, and fertile window - relative to respondents in the control community. The 
community-level effect was strong among both male and female respondents, even when 
including non-compliers in the intervention sample (those who did not attend the 
performances) into the modeling analyses. Beyond the community level effect, having high 
fertility awareness was also significantly and positively associated with performance viewing.
These findings suggest that the peer learning and community theatre can not only increase 
fertility awareness among those who viewed the performances, but also among others at the 
community-at-large, regardless of performance viewing. The findings highlight that 
community communication about fertility awareness and FP was significantly associated with 
having a high fertility awareness score, suggesting that communication and diffusion of the 
performance messages may have enabled the community-wide improvement in fertility 
awareness. 
Intent to use modern FP can be generated through a peer learning and community theatre 
model in Karamoja. Increases in intent to use modern FP were reported from baseline to 
endline among participants in intervention communities, relative to those in control 
communities. The community-level results were further corroborated by the post-exposure 
level results which showed that participants who viewed the performances, especially women
and youth between 15-25 years old, were more likely to intend to use modern FP than those 
who did not view the performances. These results suggest an increased demand for modern 
FP methods as a result of the performances, circling back to the region’s need to find effective 
communication channels to increase demand for FP services and methods. As well, the 
findings suggest significant positive associations between education and employment and 
current modern FP use. Programs and investments focused on increasing education and 
employment opportunities, while also integrating fertility awareness and FP information, in 
Karamoja could also potentially improve overall FP outcomes in the region. 
In settings with low use of FP services like Karamoja, demand generation activities like 
community theatre need to be paired with strong service delivery platforms to enable 
behavior change from intent toward use of modern FP methods. The gap between 
participants’ high reports of intent to use a modern FP method, relative to their low current FP 
use, suggest that SBC demand generation activities focused on fertility awareness and FP may 
need to be paired with strong service delivery platforms. The community level findings show 
mixed results on current use of modern FP. While intervention women had increased odds of 
using a current modern method of FP at endline, current modern FP use was still low at 
approximately 6%. The majority of female and male respondents continued to use periodic 
abstinence as a main method of modern FP use. The limited effect on current modern FP use 
79
is not surprising given that as a pilot intervention, the length of implementation was limited to 
6 months. 
EDEAN performances resonated with women and men in hard to reach communities. At 
endline, close to 7 out of 10 women in intervention communities had attended the 
performances, compared to 5 out of 10 men. The vast majority of women and men had also 
heard of the performances (79% and 64% respectively). The results suggest that that women 
who attended several performances (four or more) had higher fertility awareness scores and 
higher intent to use modern FP, compared to those who attended few performances (one to 
three). The finding that male viewers benefitted equally, regardless of their performance 
dosage, suggest that a low intervention dosage (one to three performances) may be enough 
to engage men in intervention activities and improve their fertility awareness and related FP 
outcomes. 
Several fertility awareness indicators were associated with FP outcomes. Four of the 14 
individual fertility awareness knowledge-based questions had a positive association with intent 
to use modern FP. These included measures related to male fertility (men are fertile every day)
and female fertility (vaginal secretions are a sign of fertility; secretions come during the middle 
of the menstrual cycle). One measure related to pregnancy risk (a woman is most likely to get 
pregnant if she is fertile “for several days’ halfway between two periods she has unprotected 
sex halfway between two menstrual cycles) was associated with higher modern FP use. The 
refinement of the exact body of fertility awareness questions is still under development. 
However based on the EDEAN data, no significant associations were found between having 
a high fertility awareness score (defined as answering more than 8 out of 14 fertility awareness 
questions correctly) and current modern FP use, nor intent to use modern FP.
Future research is needed among other populations to confirm and ascertain a fertility 
awareness and FP association. In the case of Karamoja, the FACT project purposely intended 
to work among a very hard to reach group with very low modern FP use. While lack of 
knowledge of fertility and pregnancy risk may be a key barrier to using modern FP in Karamoja, 
not surprisingly qualitative and process data suggest that structural access and lack of facility￾based services, among other strong socio-gender normative factors may also be contributing 
to the low modern FP use. For instance, at the time of the endline data collection, Marie Stopes 
clinics - often the sole source of FP methods in Karimojong communities - had closed down in 
the region. Further research may be needed among populations with higher levels of current 
modern FP use to ascertain associations between fertility awareness and FP use, and the intra￾relationships of FA, and other secondary outcomes, such as social norms, self-efficacy and 
partner communication about FP. 
Further research is needed to assess community-level impact of the peer learning and 
community theatre model. Significant improvements fertility awareness and FP-related 
outcomes were experienced among control communities, suggesting potential spillover 
effects and messaging diffusion among neighboring communities. This finding is not surprising 
80
given the migratory movements of Karimojong community members. While the suggested 
diffusion of EDEAN messaging is promising, the spillover effects limited the potential to isolate 
a community-level intervention impact, using difference-in-difference analysis, on fertility 
awareness and FP outcomes. In response to these study limitations, the study attempted to 
generate exposure-level associations between performance exposure and fertility awareness 
and FP outcomes, and found significant strong associations across gender and age groups. If 
this proof of concept peer learning and community theatre model is brought to larger scale, 
alternative research design may be needed to isolate a community-level intervention impact.
81
 
CONCLUSIONSo What?
82
CONCLUSION AND RECOMMENDATIONS
The EDEAN proof of concept evaluation results suggest that hard to reach Karimojong 
communities are, in fact, ready and willing to use modern FP methods to improve the health 
and well-being of their families, if investments in health and FP services can be increased in 
the region. The quantitative and qualitative evaluation results are promising, with potential for 
future adaptation and scale. If coupled with strong service delivery platforms and community 
engagement activities, the peer learning and community theatre model has potential to 
achieve improvements in FP use and creating an enabling environment for FP among hard to 
reach populations. Key take away points and recommendations are highlighted below. 
KEY TAKE AWAYS
 Feasibility. Trained, low literate peer moderators are able to learn and deliver correct FA/FP 
messages, follow their moderator manual guidelines and ensured overall messaging
fidelity. Messages about fertility awareness and FP were equally well understood by 
audience members attending the performances. The feasibility of the EDEAN peer learning 
and community theatre approach suggest that the model could be scaled up beyond a 
proof of concept phase, although considerations for scale include volunteer retention and 
necessary imaging and text adaptations for other contexts.
 Acceptability. EDEAN is acceptable in the communities and by the leaders including 
service providers. Engaging influential reference groups - and in the case of Karamoja, 
specifically community elders – was key in generating demand and acceptability of the 
performances and reflection of taboo and sensitive topics. Messages about fertility 
awareness and FP were widely diffused throughout the communities, and messages about 
couple communication appeared to be the most relevant and valuable skills shared by 
interviewed community members. As well, a key component of fertility awareness is the 
ability to apply this information to one’s own life. The findings suggest that community 
members were interested in applying fertility awareness and FP messages to their own lives. 
Their interest needs to be supported with effective counseling and increased access to 
modern FP methods.
 Effectiveness. The peer learning and community theatre approach, combined with 
community engagement activities, is effective in increasing FA, and increasing intent to 
use FP. However, the gap between participants’ high reports of intent to use a modern FP 
method, relative to their low current FP use, suggest that SBC demand generation activities 
focused on fertility awareness and FP may need to be paired with strong service delivery 
platforms and linkages to increase FP uptake. Future scale-up and/or adaptation efforts 
should ensure that the peer learning and community theatre model is paired with strong 
service delivery linkages and platforms. 
83
RECOMMENDATIONS
The EDEAN model has potential to be adapted and implemented at larger scale, if combined 
with a strong service delivery component. Recommendations for both future programming 
and research are below.
Recommendations for Programming at Scale
 Stronger linkages with health facilities are needed to increase modern FP use. As a 
community group engagement approach intended to function outside the health system, 
EDEAN did not directly provide FP services, rather FP invitation cards were used to link those 
interested in FP methods to health facilities. For future adaptations, health providers and 
community health workers (CHWs) could be engaged more actively, by attending 
performances and providing FP information at those performances.
 Ensure appropriate community engagement is a part of the peer learning and community 
theatre model. Both CMCs and community leaders were key in ensuring the acceptability 
and feasibility of the project, and future adaptations or replications will need to ensure that 
key community engagement activities are part of the intervention model. 
 Ensure that monthly reflection meetings are integrated into the model for Peer Moderators 
to strengthen content, facilitation and theatre skills especially on sensitive and more 
technical topics like fertility and FP, while also keeping in mind training burden. Ensure that 
CDOs – or identified trainers - supporting the peer moderators especially when Peer 
Moderators may be underperforming.
 Considerable attention should be paid to gendered power dynamics during training, and 
routine monitoring and supervision. Given the firm gender and social roles and structures 
in Karimojong society, observations showed that ensuring equitable participation and 
discussion about taboo topics among men and women could be challenging. Future 
activities should continue to be conducted in a gender-synchronized approach in order 
to ensure participatory and equitable discussion during same- and mixed-sex meetings. 
 Address and discuss issues of incentives and other expectations that can affect the project 
attendance and participation at the beginning of the project. Explicit discussions are 
warranted to emphasize that there are no additional incentives for Peer Moderators and 
Peer Group Members.
 The EDEAN activities and audiences may need further segmentation. Other strategic 
approaches should be explored to increase the participation of adolescents in the 
community theatre performances. As well, the EDEAN activities should be synchronized 
according to the local context. In particular the FP module may need to be taught earlier 
on in the intervention activities sequence.
 When teaching the menstrual cycle game, ensure that enough focus is given to sharing 
and understanding information about menstrual cycle regularity and the benefits and 
limitations of using the menstrual cycle information as a way to prevent pregnancy. As well, 
the presence of a health provider during intervention activities such as the menstrual cycle 
84
game could help emphasize the presence of FP methods and services at nearby health 
facilities. 
Recommendations for Future Research:
 The proof of concept evaluation did not collect quantitative data related to IPV, yet the 
qualitative data suggest that reductions in couple conflict as a result of couple 
communication may have occurred. Further quantitative research could be conducted 
to better understand linkages between couple communication - as incorporated in the 
peer learning and community theatre model – and reduction of family violence especially 
in post-conflict contexts like Karamoja. 
 Increased attention is being given to better understanding intervention dosage and 
segmented impacts of SBC interventions. The evaluation results confirmed that the peer 
learning and community theatre model had differential associations on male versus 
female, and young versus older exposed participants. However, as a less-resource 
intensive proof of concept evaluation, the sample design was not powered enough to 
identify and attribute different intervention impacts across different population segments. 
Further research should test and confirm the associations found in this evaluation, including 
for instance, the differences in dosage comparisons between men (no differences 
between going to a few versus many performances) and women (women attending more 
performances were more likely to have improved outcomes). 
 Finally, we conducted fertility awareness and FP tests of associations to test the FACT 
hypothesis that increased fertility awareness leads to increased FP use. Our findings found 
a few associations between intent to use modern FP and limited number of fertility 
awareness methods. However, the relative low prevalence of modern FP use in the region 
(about 7%), limited the ability to find robust associations. Future research should be 
conducted on populations with more prevalent modern FP use to better understand 
associations between fertility awareness and FP, and other associated secondary factors. 
The FACT Project begun work in the Karamoja region in 2014 in recognition of the potential 
challenges ahead in working with one of the most excluded and marginalized communities
of Uganda. Four years later - and with the efforts of the communities and implementation
stakeholders along the way - the proof of concept findings today suggest that Karimojong are 
interested and would like to learn more about modern FP methods, and that the theatre 
model is an appropriate vehicle to increase the demand, in combination with strong FP service 
delivery and platforms. 
85
REFERENCES
Daily Monitor (June 20, 2012). Waiting for Karamoja to develop: Of Uganda’s uneven 
development. Retrieved from 
http://www.monitor.co.ug/SpecialReports/ugandaat50/Waiting+for+Karamoja 
+to+develop/-/1370466/1430994/-/3h1ccoz/-/index.html (accessed December 2014)
Family Planning 2020 (July 11, 2017). Family Planning 2020 Commitment: Government of 
Uganda. Retrieved from: http://ec2-54-210-230-186.compute-1.amazonaws.com/wp￾content/uploads/2017/08/Govt.-of-Uganda-FP2020-Commitment-2017-Update-SO￾CL.pdf
Haub, C., & Gribble, J. (2011). The World at 7 Billion. Population Bulletin,66 (2), 1-12.
Institute for Reproductive Health. (2013). Fertility Awareness across the Life Course: A 
comprehensive literature review. FAM Project. Washington, DC. Institute for Reproductive 
Health, Georgetown University.
Kaye, K, Suellentrop, K, and Sloup, C (2009). The Fog Zone: How Misperceptions, Magical 
Thinking, and Ambivalence Put Young Adults at Risk for Unplanned Pregnancy. 
Washington, D.C.: The National Campaign to Prevent Teen and Unplanned Pregnancy.
Kouyate, R (2010). LAM and the Transition Barrier Analysis, Sylhet, Bangladesh. USAID/ACCESS 
Report.
Lebetkin, E. 2015. Situation Analysis of Community-Based Referrals for Family Planning: A 
Review of the Evidence and Recommendations for Future Research and Programs. 
Arlington, VA: Advancing Partners & Communities.
Makinwa-Adedusoye, P (1992). Sexual behavior, reproductive knowledge and contraceptive 
use among young urban Nigerians. International Family Planning 
Perspectives: 18(2):66-70.
Sedgh G, et al. (2007). Women with an unmet need for contraception in developing countries, 
and their reasons for not using a method. New York: Guttmacher Institute.
Stites, E., & Marshak, A. (2016). Who are the Lonetia? Findings from southern Karamoja, 
Uganda. The Journal of Modern African Studies, 54(2), 237-262. 
doi:10.1017/S0022278X16000021
Stites E. & Akabwai, D. (2009). Changing Roles, Shifting Risks: Livelihood Impacts of 
Disarmament in Karamoja, Uganda. Feinstein International Center: Medford, MA. 
86
Thummalachetty, N., Mathur, S., Mullinax, M., DeCosta, K., Nakyanjo, N., Lutalo, T., Brahmbhatt, 
H., and Santelli, J. S. (2017). Contraceptive knowledge, perceptions, and concerns 
among men in Uganda. BMC Public Health, 17, 792. http://doi.org/10.1186/s12889-017-
4815-5
Uganda Bureau of Statistics (UBOS) and ICF. 2018. Uganda Demographic and Health Survey
2016. Kampala, Uganda and Rockville Maryland, USA: UBOS and ICF.
UBOS and ICF. 2012. Uganda Demographic and Health Survey 2011. Kampala, Uganda: UBOS 
and Calverton, Maryland: ICF International Inc.
87
ANNEX 1. FACT EDEAN RESOURCES
1. EDEAN Website and Resources
2. EDEAN Solution Brief
3. EDEAN Micro Documentary Video
4. EDEAN Formative Research Report
5. EDEAN Blog “In EDEAN, Youth Leaders Create a Space to Reflect on Fertility and Family 
Planning”
6. EDEAN Blog “Through EDEAN, Communities “Come Together to Strengthen Child 
Spacing”
7. EDEAN Evaluation Tools
88
ANNEX 2. INTERVENTION TABLES
Table 29 | Profiles of Peer Moderators
SEX AGE EDUCATION MARITAL 
STATUS
OCCUPATION RELATIONSHIP WITH 
ECCD
PREVIOUS COMMUNITY 
EXPERIENCE
Peer Moderator Pair 1 
Male 25 Secondary Married Farmer None Coordinating group events, 
facilitating activities in the 
community and mobilize for group 
events
Female 20 None Married None None None
Peer Moderator Pair 2 
Male 25 Primary Single Farmer None Mobilize for group events
Female 19 Primary Married Moderator Supervising members Supervising or mentoring 
members, mobilize for group 
events
Peer Moderator Pair 3 
Male 22 Completed 
high school
Married Business 
man
Translating speeches 
to visitors, mobilizing 
community for the 
ECCD activities, built 
the pit latrine for the 
center
Coordinating group events, 
supervising or mentoring members, 
mobilizing for group events, 
facilitating activities in the 
community
Female 30 Primary Married Lead 
mother
None Mobilizing for group events, 
preparing reports.
Peer Moderator Pair 4 
Male 28 Secondary Married Business 
man
Participated in the 
building of the ECCD 
center and built the 
stands for the children 
to play
Coordinating group events, 
supervising or mentoring members 
and mobilizing for group events
Female 28 Secondary Married None None None
Peer Moderator Pair 5 
Male 24 Secondary Married Farmer Participated in the 
construction of the 
ECCD center
Coordinating group events, 
supervising or mentoring members, 
mobilizing for group events and 
facilitating activities in the 
community
Female 22 Primary Married None None Coordinating group events
Peer Moderator Pair 6 
Male 23 Secondary Single Teacher None Mobilizing for group events and 
facilitating activities in the 
community
Female 20 Secondary Married Moderator Mobilizing community 
members for meetings 
and conduct outreach 
in communities.
Supervising or mentoring members 
and mobilizing for the group 
events
89
Table 30 | EDEAN Schedule of Activities
WEEK # ACTIVITY 
Week 1 Training of Trainers (ToT) #1 (3 Days)
Topics: Couple Communication and Menstruation 
Week 2 Training of Peer Moderators #1 (5 Days)
Topics: Couple Communication and Menstruation 
1. COUPLE COMMUNICATION MODULE 
Week 3 Couple Communication Single Sex Peer Group Meeting 
Week 4 Couple Communication Mixed Sex Peer Group Meeting
Week 5 Couple Communication Storyline 1 Theatre Rehearsal and 
Performance
Week 6 Couple Communication Storyline 2 Theatre Rehearsal and 
Performance
Week 7 Couple Communication Create Your Own Drama! Theatre 
Rehearsal and Performance
Week 8 Reflection Meetings 
2. MENSTRUATION MODULE 
Week 9 Menstruation Single Sex Peer Group Meeting
Week 10 Menstruation Mixed Sex Peer Group Meeting
Week 11 Menstruation Storyline 1 Theatre Rehearsal and Performance
Week 12 Menstruation Storyline 2 Theatre Rehearsal and Performance
Week 13 Menstruation Create Your Own Drama! Theatre Rehearsal and 
Performance
Week 14 Reflection Meetings 
Week 15 TOT #2 (3 Days)
Week 16 Training of Peer Moderators #2 (5 Days)
3. FERTILITY MODULE 
Week 17 Fertility Single Sex Peer Group Meeting 
Week 18 Fertility Mixed Sex Peer Group Meeting
Week 19 Fertility Storyline 1 Theatre Rehearsal and Performance
Week 20 Fertility Storyline 2 Theatre Rehearsal and Performance
Week 21 Fertility Create Your Own Drama! Theatre Rehearsal and 
Performance
Week 22 Reflection Meetings
4. FAMILY PLANNING MODULE 
Week 23 Family Planning Single Sex Peer Group Meeting 
Week 24 Family Planning Mixed Sex Peer Group Meeting
Week 25 Family Planning Storyline 1 Theatre Rehearsal and Performance
Week 26 Family Planning Storyline 2 Theatre Rehearsal and Performance
Week 27 Family Planning Storyline 3 Theatre Rehearsal and Performance
Week 28 Family Planning Create Your Own Drama! Theatre Rehearsal and 
Performance
Week 29 Final Reflection Meeting and Community Celebration
90
ANNEX 3. EVALUATION METHODS TABLES
Table 31 | Household survey sample
ECCD Center Catchment Area
FEMALE MALE TOTAL
Intervention 
Nadunget Sub-County, Moroto District
1. Nangorit ECCD (2 villages)
2. Nachele ECCD (2 villages) 100 100 200
Ngoleriet Sub-County, Napak District
3. Nagule angolol ECCD (2 villages) 
4. Longariama ECCD (3 villages)
5. Narongor ECCD (2 villages) 100 100 200
TOTAL (I) 200 200 400
Control 
Rupa Sub-county, Moroto District
 6. Rupa ECCD (4 villages) 100 100 200
TOTAL (I + C) 300 300 600
91
Table 32 | Construction of Outcome Indicators
Indicator Measures incorporated (include questions and question numbers) Narrative Summary of Indicator Construction 
FA score - q201. First bleeding: sign girl is able to become pregnant";
- q202. First ejaculation: sign boy is able to get girl pregnant"; 
- q203. Menstrual period is when woman has monthly bleeding";
- q204. Menstrual cycle is all the days before the start of one 
menstrual period to the day before the next one starts"; 
- q205. Menstrual cycle begins on the first day of bleeding"; 
- q206. Menstrual cycle ends on the day before the start of the 
next menstrual period"; 
- q207. Menstrual cycle lasts between 26 to 32 days"; 
- q208. There are certain days in a period women are more likely 
to become pregnant"; 
- q209. A woman is most likely to get pregnant if you have 
unprotected sex halfway between two menstrual cycles"; 
- q210. You can become pregnant if had unprotected sex during 
fertile day"; 
- q211. Other than sexual arousal, healthy vaginal secretions are 
sign of fertility"; 
- q212. Vaginal secretions come during the middle of the 
menstrual cycle"; 
- q214. Men are fertile every day"; 
- q215. Baby’s sex is determined by father’s sperm"
 First, fertility awareness knowledge items 
(q201-q217) were coded numerically. 
Textual data (e.g., specify other) were 
reviewed and recoded into numeric data, 
according to the categories provided in the 
question. 
 Second, fertility awareness knowledge items 
were coded correct or incorrect. Correct 
responses were assigned a numeric value of 
1. Incorrect responses were assigned a 
numeric value of 0. 
 Third, a knowledge index score was 
created by summing all the correct 
responses for q201-q212, q214-q215. 
 Fourth, the median of the knowledge score 
was identified and a dichotomous variable 
categorized the knowledge index score into 
those who had less than or equal to the 
median level knowledge, and those who 
scored above the median knowledge. 
Current use 
of modern FP
- Q412 Have you or your partner ever used a family planning 
method?
- Q416 Are you or your partner currently using a family planning 
method?
- Q417 What method of family planning are you or your partner 
using? (mark all that apply)
 First, ensured all responses to current FP 
method (q417) were coded numerically. 
Textual data (e.g., specify other) were 
reviewed and recoded into numeric data, 
according to the categories provided in the 
question.
 Second, current use of modern FP was 
constructed using variables q412, q416, 
and 417: 
 Respondents who reported currently using 
any of the following a modern FP method in 
q417 were coded as currently using a FP 
method. 
 Respondents who reported currently using a 
non-modern FP method in q417, not 
currently using FP method in q416), and 
who have never used a FP method in q412 
were coded as not currently using a FP 
method.
Intent to use 
modern FP
- Q418 Do you plan on [using] or [continue using] a family 
planning method in the next three months?
- Q419 Which method of family planning do you intend to use?
 First, ensured all responses to current FP 
method (q419) were coded numerically. 
Textual data (e.g., specify other) were 
reviewed and recoded into numeric data, 
according to the categories provided in the 
question.
 Second, intent to use of modern FP in the 
next 3 months was constructed using 
variables q418 and q419: 
 Respondents who reported intending to use 
any of the modern FP methods in q419 in 
the next three months were coded as having 
an intent to use a modern FP method. 
92
 Respondents who reported no, undecided, 
don’t know, or no response to intending to 
use a FP method (q418), or who selected a 
non-modern FP method in q419 were coded 
as not having an intent to use a modern FP 
method in the next three months. 
Self-efficacy 
to use and 
access FP
- Q421a I am confident I could use a method correctly all the 
time to delay or avoid pregnancy";
- Q421b I am confident I could use a method correctly all the 
time to delay or avoid pregnancy even if my partner 
disagrees"; 
- Q421f I have the information I need to make a decision about 
whether to use FP if I wanted to delay or avoid pregnancy";
- Q421g I know where to obtain a method of FP to delay or 
avoid pregnancy."
 The final self-efficacy items were identified 
using exploratory factor analysis (Cronbach 
alpha = 0.684608). Scale items were 
recoded from a 5-point Likert to a 3-point 
Likert of agree, neutral, and disagree. A 
mean scale score was obtained by finding 
the mean of the 4 scale items. Finally, a 
dichotomous variable was created to 
reflect respondents who scored less than or 
equal to the mean, and those who scored
above. 
Attitudes 
about FP
- Q308a Using a FP method is safe for a woman’s health 
- Q308c It is acceptable for a woman to bring up the topic of FP 
with her husband 
- Q308d It is acceptable for a man to bring up the topic of FP 
with his wife 
- Q308f FP methods are a way to continue the Karamoja culture 
and way of life 
 The final FP attitude items were identified 
using exploratory factor analysis (Cronbach 
alpha = 0.673875). Scale items were 
recoded from a 5-point Likert to a 3-point 
Likert of agree, neutral, and disagree. A 
mean scale score was obtained by finding 
the mean of the 4 scale items. Finally, a 
dichotomous variable was created to 
reflect respondents who scored less than or 
equal to the mean, and those who scored 
above. 
Social norms 
on FP use 
among 
young 
couples
- Q308g Most couples in my community who want to plan or 
prevent pregnancy use family planning
- Q308h Couples in my community who want to plan or prevent 
pregnancy should use family planning 
- Q308i I should use a family planning method if I want to plan or 
prevent pregnancy 
 The final social norms around FP items were 
identified using exploratory factor analysis 
(Cronbach alpha = 0.681678). Scale items 
were recoded from a 5-point Likert to a 3-
point Likert of agree, neutral, and 
disagree. A mean scale score was obtained 
by finding the mean of the 3 scale items. 
Finally, a dichotomous variable was 
created to reflect respondents who scored 
less than or equal to the mean, and those 
who scored above.
Partner 
communicati
on about FP
- Q110 are you currently living together with a person who is 
your husband/wife or boyfriend/girlfriend?
- Q303 Do you feel that your opinion is taken into account when 
talking about planning pregnancy with your partner?
- Q305 In the last three months, have you discussed using FP to 
delay or avoid pregnancy with your partner?
 This variable was created for respondents 
who had a current partner (q110=1). A 
composite score up to 2 was created using 
2 variables: q303 and q305. 
Community 
communicati
on about FP
- Q503 In the last 3 months, have you talked about fertile days 
during the menstrual cycle with anyone? 
- Q504 In the last 3 months, have you corrected someone when/if 
you heard them saying something untrue about menstruation? 
- Q505 In the last 3 months, have you corrected someone when/if 
you heard them saying something untrue about family planning? 
 A composite score up to 3 points was 
created using 4 variables: q503, q504 and 
q505.
93
ANNEX 4. FEASIBILITY RESULTS
Table 33 | Observation scores, Message delivery and participation
 Couple communication Menstruation Fertility Family planning 
Peer 
meetings 
(4)
Rehears
als (5)
Perform
ances
(6)
Peer 
meeting 
(6)
Rehea
rsals 
(6)
Performa
nces (7)
Peer 
meeting(6
)
Rehea
rsals 
(6)
Performa
nce (5)
Peer 
meeting 
(5)
Rehea
rsals 
(8)
Performa
nce (5)
Moderators ability to follow the manual activity plan 
1.Almost all 
of the time
75% 60% 67% 67% 83% 83% 100% 83% 67% 100% 100% 80%
2. Some of 
the time
25% 20% 17% 33% 17% 17% 0% 17% 33% 0% 0% 20%
 3. Did not 
follow 
guideline. 
0% 20% 17% 0% 0% 0% 0% 0% 0% 0% 0% 0%
Moderators ability to facilitate (write and explain the response)
 Very good 25% 60% 83% 83% 29% 86% 67% 83% 47% 40% 56% 80%
 Fair 75% 20% 17% 17% 43% 14% 33% 17% 29% 40% 44% 0%
 Poor 0% 20% 0% 0% 43% 0% 0% 0% 28% 20% 0% 20%
Peer group members understood the content and message shared.
1.Yes very well 50% 60% 67% 83% 100% 100% 100% 67% 80% 80% 63% 60%
2.Partially 
understood
50% 20% 33% 17% 0% 0% 0% 33% 20% 20% 37% 40%
3. Not well 
understood
0% 20% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
PGMs responded and discussed questions asked by PMs
1.Yes a lot 0% 75% 50% 83% 50% 71% 67% 17% 60% 60% 13% 40%
2.Yes, 
sometimes
100% 25% 50% 17% 50% 29% 33% 83% 40% 40% 87% 60%
3. No, Not at 
all
0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%
94
Table 34 | Examples of messages delivered and discussion questions, by topic module
Topic Module Examples of messages delivered Examples of questions and answers
Couple 
Communication
 “Child spacing or family planning does not mean stop 
producing but produce as many but spaced”
 “Couples should learn to listen when the other is talking or 
passing a message and learn to ask questions if they do not 
understand what the partner is trying to say”
 “There is need for couples to discuss and decide together 
whether to use family planning or not but the decision should 
be agreeable to both.”
 “Wives and husbands should sit down and talk if there is a 
problem at home”
 “Good couple communication makes a family stronger”
 “With family planning and child spacing, children will not fall 
sick, be malnourished, they will be able to go to school” 
Question: Now if you ask a woman who does not space her children to do so in this 
Karamojong, will her?
Answer: That is the mere reason we are here. To learn this information about family 
planning so that we become change agents to people like that in this community. Look 
when you talk to this person and she gets to sit down to think and they will wonder 
where they missed out, it will make them rethink the information they were given and 
act on it
Menstruation  “Menstruation is natural and comes from God” 
 “Menstruation is a sign that a girl is able to become a mother”
 “Myths about menstruation are lies”
 “Pests spoil vegetables not bleeding women”
 “Menstruation does not harm plants and animals” 
 “In the middle of the cycle occur secretions where a female is 
most likely to conceive”
 “Secretions are healthy”
Question: Why is it that some women even give birth when they have never seen their 
menstruation?
Answer: it is God who created them like and some women do experience 
menstruation even when they do not see the blood but you can tell by looking at their 
hands.
Fertility  “When a man begins wet dreams it means he can make a woman 
pregnant”
 Sperms can survive in the uterus for up to 5 days 
 “It is the man’s sperm that determines a child’s sex and but neither 
the husband or wife can dictate on the child they want to have”
 “Once a woman starts menstruating, it is a sign that she can get 
pregnant as long as secretions occur”
 Secretions are a sign of fertility
 “It is always good to consult the VHT and visit the health center to 
know more about family planning”
Question: Why is the egg fertilized at the corner of the fallopian tubes? 
Answer: That is how God created a man that it should be like that but in case there is 
anything else you need an answer for, come and I’ll let you know about what I know. 
Family Planning  Family planning methods are a way couples can space their 
pregnancies
Question: How about the condoms when we need them can they be given for free or 
we need to have some money?
95
 There are three LAM Conditions:
o A child should be below 6 months
o The mother should breast feed exclusively without giving the 
child any food
o The mother should have not received back her periods
 Pills, injectable, IUDs, implants and tubal ligation may make a 
woman experience some side effects in the beginning such as 
heavy bleeding, irregular periods or cramps but he added 
that these do not last.
 “Healthy timing is waiting till one turns 18 to get pregnant and 
healthy spacing is leaving a pace of 2 years between births” 
Answer: Condoms are available at the hospitals or health centers or even with the 
VHTs for free of charge”.
Question: How much time do IUDs take for it to stop functioning?
Answer: It take strictly one year and then there was this peer group member with a 
manual she decided to read out from the manual, then the moderator apologized and 
then gave in the correction according to the one the peer group member read.
Question: A peer group member asked the peer moderator whether vasectomy is like 
castration. He actually asked if the testicles are physically crushed.
Answer: The P.M answered him that the testicles are not crushed but rather the sperm 
duct cut. The peer again asked if the process for vasectomy would be reversed such 
that a man can have children again. The Peer Moderator and some peers told him 
that the method is permanent."
96
Table 35 | Average time facilitating EDEAN activities during systematic observations, disaggregated by module and 
type of activity (n=69)
(Source: Systematic Observations)
Couple 
Communication
Menstruation Fertility Family 
Planning
Single-sex 
female 
meetings
0.6 hour
(34 mns)
1.5 hours 1.5 
hours
2.1 hours
Single-sex 
male 
meetings
N/A 1.0 hour 1.0 
hour
1.7 hours
Mixed-sex 
meetings
1.9 hours 1.5 hours 1.9 
hours
1.9 hours
Rehearsals 1.8 hours 1.5 hours 1.5 
hours
1.7 hours
Performances 2.0 hours 0.8 hour 
(46 mns)
1 hour 0.8 hour 
(45 mns)
97
ANNEX 5. EFFECTIVENESS RESULTS
Table 36 | Socio-demographic characteristics of participants, at baseline (n=600)
(Source: Endline Survey)
DEMOGRAPHICS 
ENDLINE
Intervention Control
n=4
00
% n=2
00
% P-value
Sex 
Male 
Female
198
202
49.5
50.5
100
100
50.0
50.0
N/A
Age
Mean (Std)
Min, Max
28.2 (7.6)
16, 49
32.6 (7.9)
19, 49
<.0001***
Highest level of school 
None
ABEK
Some primary 
Completed primary 
Some secondary 
Completed secondary 
More than secondary 
298
35
43
10
9
3
2
74.5
8.8
10.8
2.5
2.3
0.8
0.5
154
10
25
7
2
2
0
77.0
5.0
12.5
3.5
1.0
1.0
0.0
0.6646
Have job or work 
Yes
No 
211
189
52.8
47.3
90
110
45.0
55.0
0.0735
Type of work
Farming/ agriculture
Formal paid work
Informal paid work
121
8
82
57.4
3.8
38.9
13
5
72
14.4
5.6
80.0
<.0001***
Currently living with
partner 
Yes
No
344
56
86.0
14.0
189
11
94.5
5.5
0.0018*
Time married or living 
together
Mean (Std)
Min, Max 
8.1 (7.08)
0.1, 32.0
9.8 (7.57)
1.0, 34.0
0.0100*
Husband has another wife
1 wife
2+ wives
150
194
43.6
56.4
98
91
51.9
48.2
0.0678
Number of living children 
Mean (Std)
Min, Max
3.1 (2.5)
0, 14
3.8 (2.4)
0, 11
0.4166
Ever use modern family 
planning 
Yes
No 
70
330
17.5
82.5
27
173
13.5
86.5
0.2096
Type of FP methods ever 
used (multiple response)
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
CycleBeads
TwoDay Method
LAM
Withdrawal 
Abstinence 
Other
16
0
7
6
8
0
0
0
2
0
39
1
193
6
7.1
0.0
3.1
2.7
3.5
0.0
0.0
0.0
0.9
0.0
17.3
0.4
85.4
2.7
3
0
0
2
0
0
0
0
0
0
22
0
168
0
1.8
0.0
0.0
1.2
0.0
0.0
0.0
0.0
0.0
0.0
13.0
0.0
99.4
0.0
98
Table 37 | Socio-demographic characteristics of women, baseline and endline (n=302)
(Source: Baseline and Endline Surveys)
DEMOGRAPHICS BASELINE ENDLINE 
Total Intervention (I) Control (C) I vs. C Total Intervention (I) Control (C) I vs. C
n=301 % n=201 % n=100 % P value n=302 % n=202 % n=100 % P value
Age
Mean (Std)
Min, Max
31.0 (8.2)
18.0, 49.0
30.8 (8.3)
18.0, 49.0
31.2 (7.8)
18.0, 49.0
0.5117 29.6 (7.8)
17.0, 49.0
28.1 (7.1)
17.0, 49.0
32.8 (8.3)
19.0, 49.0
0.0004
Highest level of school 
attended 
None
ABEK
Some primary 
Completed primary 
Some secondary 
Completed secondary 
238
29
30
3
1
0
79.1
9.6
10.0
1.0
0.3
0.0
153
25
19
3
1
0
76.1
12.4
9.5
1.5
0.5
0.0
85
4
11
0
0
0
85.0
4.0
11.0
0.0
0.0
0.0
0.1828 246
20
33
3
0
0
81.5
6.6
10.9
1.0
0.0
0.0
167
14
19
2
0
0
82.7
6.9
9.4
1.0
0.0
0.0
79
6
14
1
0
0
79.0
6.0
14.0
1.0
0.0
0.0
0.6845
Have job or work 
Yes
No 
132
169
43.9
56.2
82
119
40.8
59.2
50
50
50.0
50.0
0.1296 85
217
28.2
71.8
47
155
23.3
76.7
38
62
38.0
62.0
0.0074
Type of work
Farming/ agriculture
Formal paid work
Informal paid work
12
2
118
9.1
1.5
89.4
4
2
76
4.9
2.4
92.7
8
0
42
16.0
0.0
84.0
0.0593 7
4
74
8.2
4.7
87.1
1
2
44
2.1
4.3
93.6
6
2
30
15.8
5.3
79.0
0.0250
Living with partner 
Yes
No
282
19
93.7
6.3
183
18
91.0
9.0
99
1
99.0
1.0
0.0052 268
34
88.7
11.3
177
25
87.6
12.4
91
9
91.0
9.0
0.3823
Time married 
Mean (Std)
 Min, Max
9.7 (7.1)
0.1, 32.0
9.5 (7.0)
0.1, 30.0
10.0 (7.5)
0.6, 32.0
0.5684 7.7 (6.8)
0.1, 32.0
7.5 (7.3)
0.1, 32.0
7.9 (5.8)
1.0, 26.0
0.0064
Husband has another wife
1 wife
2+ wives
186
96
66.0 
34.0
114
69
62.3
37.7
72
27
72.7
27.3
0.0776 168
100
62.7
37.3
111
66
62.7
37.3
57
34
62.6
37.4
0.9905
Number living children 
Mean (Std)
Min, Max
3.2 (1.8)
0, 9
3.4 (1.8)
0, 9
2.9 (1.6)
0, 8
0.0427 3.3 (1.8)
0, 9
3.2 (1.8)
0, 9
3.5 (1.9)
0, 8
0.1509
Ever use modern FP
Yes
No 
Don’t know
160
138
2
53.3
46.0
0.7
111
87
2
55.5
43.5
1.0
49
51
0
49.0
51.0
0.0
156
146
0
51.7
48.3
0.0
84
118
0
41.6
58.4
0.0
72
28
0
72.0
28.0
0.0
Type of FP methods ever 
used (multiple response)
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
CycleBeads
TwoDay Method
LAM
Withdrawal 
Abstinence 
Other
0
0
2
5
2
0
0
0
1
1
64
10
125
16
0.0
0.0
1.3
3.1
1.3
0.0
0.0
0.0
0.6
0.6
40.0
6.3
78.1
10.0
0
0
2
4
1
0
0
0
1
1
49
10
81
16
0.0
0.0
1.8
3.6
0.9
0.0
0.0
0.0
0.9
0.9
44.1
9.0
73.0
14.4
0
0
0
1
1
0
0
0
0
0
15
0
44
0
0.0
0.0
0.0
2.0
2.0
0.0
0.0
0.0
0.0
0.0
30.6
0.0
89.8
0.0
0
0
6
6
6
0
0
0
1
0
32
1
137
6
0.0
0.0
3.9
3.9
3.9
0.0
0.0
0.0
0.6
0.0
20.5
0.6
87.8
3.9
0
0
6
4
6
0
0
0
1
0
11
1
66
6
0.0
0.0
7.1
4.8
7.1
0.0
0.0
0.0
1.2
0.0
13.1
1.2
78.6
7.1
0
0
0
2
0
0
0
0
0
0
21
0
71
0
0.0
0.0
0.0
2.8
0.0
0.0
0.0
0.0
0.0
0.0
29.2
0.0
98.6
0.0
99
Table 38 | Socio-demographic characteristics of men, baseline and endline (n=299)
DEMOGRAPHICS BASELINE ENDLINE 
Total Intervention (I) Control (C) I vs. C Total Intervention (I) Control (C) I vs. C
N=299 % n=19
9
% n=100 % P value N=298 % n=198 % n=100 % P value 
Age
Mean (Std)
Min, Max
29.3 (8.0)
15.0, 49.0
28.8 (8.4)
15.0, 49.0
30.3 (7.0)
17.0, 49.0
0.0202 29.8 (8.1)
16.0, 49.0
30.3 (7.0)
17.0, 49.0
29.8 (8.1)
16.0, 49.0
0.0002
Highest level of school 
attended 
None
ABEK
Some primary education 
Completed primary 
Some secondary 
Completed secondary 
 More than secondary 
200
11
38
12
14
4
0
73.6
3.7
12.7
4.0
4.7
1.3
0.0
144
8
26
7
12
2
0
72.4
4.0
13.1
3.5
6.0
1.0
0.0
76
3
12
5
2
2
0
76.0
3.0
12.0
5.0
2.0
2.0
0.0
0.6184 206
25
35
14
11
5
2
69.1
8.4
11.7
4.7
3.7
1.7
0.7
131
21
24
8
9
3
2
66.2
10.2
12.1
4.0
4.6
1.5
1.0
75
4
11
6
2
2
0
75.0
4.0
11.0
6.0
2.0
2.0
0.0
0.2938
Have job or work 
Yes
No 
230
69
76.9
23.1
141
58
70.9
29.2
89
11
89.0
11.0
0.0004 216
82
72.5
27.5
164
34
82.8
17.2
52
48
52.0
48.0
<.0001
Type of work
Farming/ agriculture
Formal paid work
Informal paid work
113
17
100
49.1
7.4
43.5
94
7
40
66.7
5.0
28.4
19
10
60
21.4
11.2
67.4
<.0001 127
9
80
58.8
4.8
37.0
120
6
38
73.2
3.7
23.2
7
3
42
13.7
5.8
80.8
<.0001
Currently living with partner 
Yes
No
281
18
94.0
6.0
184
15
92.5
7.5
97
3
97.0
3.0
0.1202 265
33
88.9
11.1
167
31
84.3
15.7
98
2
98.0
2.0
0.0004
Time married together
Mean (Std)
 Min, Max
6.4 (5.7)
0.2, 29.0
6.8 (5.8)
0.2, 29.0
5.9 (5.2)
0.5, 26.0
0.9111 7.7 (6.8)
0.1, 32.0
7.5 (7.3)
0.1, 32.0
7.9 (5.8)
1.0, 26.0
0.6268
Husband has another wife
1 wife
2+ wives
96
185
34.2
65.8
57
127
31.0
69.0
39
58
40.2
59.8
0.1210 80
185
30.2
69.8
39
128
23.4
76.7
41
57
41.8
58.2
0.0016
Number of living children 
Mean (Std)
Min, Max
3.3 (3.1)
0, 18
3.3 (3.3)
0, 18
3.3 (2.6)
0, 12
0.9111 3.3 (3.0)
0, 14
2.9 (3.0)
0, 14
4.1 (2.7)
0, 11
0.0019
Ever use modern FP
Yes
No 
Don’t know
No response
283
58
3
0
79.6
19.4
1.0
0.0
152
46
1
0
76.4
23.1
0.5
0.0
86
12
2
0
86.0
12.0
2.0
0.0
239
50
8
1
80.2
16.8
2.7
0.3
142
49
7
0
71.7
24.8
3.5
0.0
97
1
1
1
97.0
1.0
1.0
1.0
Type of FP methods ever 
used (multiple response)
Male condom 
Female condom
Pills
Injectable
Implant
IUD
Vasectomy
Tubal ligation 
MoonBeads/ CycleBeads
TwoDay Method
LAM
Withdrawal 
Abstinence 
Other
21
0
0
7
2
0
0
0
1
0
25
13
218
5
8.8
0.0
0.0
2.9
0.8
0.0
0.0
0.0
0.4
0.0
10.5
5.5
91.6
2.1
14
0
0
7
2
0
0
0
0
0
11
0
138
4
9.2
0.0
0.0
4.6
1.3
0.0
0.0
0.0
0.0
0.0
7.2
0.0
90.8
2.6
7
0
0
0
0
0
0
0
1
0
14
13
80
1
8.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.2
0.0
16.3
15.1
93.0
1.2
19
0
1
2
2
0
0
0
1
0
29
0
224
0
8.0
0.0
0.4
0.8
0.8
0.0
0.0
0.0
0.4
0.0
12.1
0.0
93.7
0.0
16
0
1
2
2
0
0
0
1
0
28
0
127
0
11.3
0.0
0.7
1.4
1.4
0.0
0.0
0.0
0.7
0.0
19.7
0.0
89.4
0.0
3
0
0
0
0
0
0
0
0
0
1
0
97
0
3.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1.0
0.0
97.0
0.0
(Source: Baseline and Endline Surveys)
100
Table 39 | Adjusted logistic regression analysis results for fertility awareness and FP outcomes, by desire to prevent 
pregnancy
(Source: Endline Surveys; model adjusted for age, employment, education, living with partner, having children)
Table 40 |Adjusted logistic regression analysis results for secondary outcomes, by desire to prevent pregnancy
Source: Endline Surveys; model adjusted for age, employment, education, living with partner, having children)
 
1 aOR not shown since no control men (0) responded this answer affirmatively / negatively
Total aOR
(n=448)
Female aOR
(n=225)
Male aOR
(n=223)
15 - 25 yrs aOR
(n=150)
26 - 35 yrs aOR
(n=194)
36+ yrs aOR
(n=104)
High fertility 
awareness Score
( fertility awareness 
> 8 )
2.66
(1.60, 4.41)
p=0.0002
2.20
(1.09, 4.48)
p=0.0289
3.23
(1.35, 7.75)
p=0.0085
1.58
(0.61, 4.10)
p=0.3448
2.04
(0.98, 4.25)
p=0.0579
6.56
(1.57, 27.43)
p=0.0100
Current modern FP 
use
1.47
(0.76, 2.81)
p=0.2509
0.72
(0.22, 2.28)
p=0.5710
2.65
(1.07, 6.56)
p=0.0347
4.09
(0.86, 19.57)
p=0.0774
1.43
(0.57, 3.62)
p=0.4498
0.56
(0.09, 3.53)
p=0.4501
Intent modern FP use 
in next 3 months
9.65
(3.70, 25.20)
p<0.0001
9.50
(2.90, 31.17)
p=0.0002
14.38
(1.48,139.47)
p=0.0215
22.16
(3.85, 127.44)
p=0.0005
5.02
(1.45, 17.40)
p=0.0109
9.55
(0.44, 206.20)
p=0.1499
Total
(n=448)
Female
(n=225)
Male
(n=223)
15 - 25 yrs
(n=150)
26 - 35 yrs
(n=194)
36+ yrs
(n=104)
High FP self-efficacy 0.72
(0.41, 1.26)
p=0.2502
0.16
(0.07, 0.38)
p<0.0001
18.25
(3.63,91.88)
p=0.0004
2.11
(0.73, 6.09)
p=0.1696
0.37
(0.15, 0.90)
p=0.0276
0.49
(0.12, 1.97)
p=0.3143
Sought info about FP 
from health worker in 
last 3 months
5.45
(2.68, 11.10)
p<0.0001
3.23
(1.21, 8.66)
p=0.0197
9.47
(2.79, 32.15)
p=0.0003
11.73
(2.35, 58.61)
p=0.0027
5.89
(1.91, 18.18)
p=0.0020
1.62
(0.30, 8.92)
p=0.5780
FP methods help 
prevent pregnancy (q. 
what do you know 
about FP methods?)
4.82
(2.26, 10.32)
p<0.0001
14.43
(4.63, 44.96)
p<.0001
0.83
(0.24, 2.88)
p=0.7634
33.97
(3.44, 335.65)
p=0.0026
2.99
(1.17, 7.65)
p=0.0224
2.83
(0.42, 19.09)
p=0.2858
FP methods cause side 
effects (q. what do you 
know about FP 
methods?)
0.16
(0.08, 0.32)
p<0.0001
0.06
(0.02, 0.15)
p<.0001
--------
1
0.52
(0.16, 1.64)
p=0.2616
0.05
(0.01, 0.20)
p<0.0001
0.07
(0.01, 0.89)
p=0.0401
High FP attitudes 0.92
(0.37, 2.29)
p=0.8531
0.47
(0.16, 1.42)
p=0.1811
4.07
(0.28, 59.54)
p=0.3054
1.14
(0.29, 4.47)
p=0.8534
0.62
(0.13, 3.06)
p=0.5564
1.21
(0.08, 18.50)
p=0.8910
High support for FP 
social norms score
1.43
(0.74, 2.75)
p=0.2856
0.35
(0.11, 1.11)
p=0.0748
3.32
(1.16, 9.48)
p=0.0253
0.79
(0.26, 2.38)
p=0.6761
2.01
(0.71, 5.72)
p=0.1881
3.28
(0.25, 43.34)
p=0.3664
High FP communication 
with partner
0.32
(0.18, 0.56)
p<0.0001
0.17
(0.07, 0.41)
p<0.0001
0.64
(0.27, 1.48)
p=0.2938
0.19
(0.06, 0.55)
p=0.0024
0.31
(0.12, 0.78)
p=0.0125
0.62
(0.18, 2.11)
p=0.4454
High communication 
with fertility awareness 
and FP with community
0.50
(0.32, 0.78)
p=0.0026
0.16
(0.07, 0.37)
p<.0001
1.08
(0.57, 2.03)
p=0.8186
0.34
(0.15, 0.78)
p=0.0109
0.67
(0.33, 1.36)
p=0.2631
1.16
(0.37, 3.65)
p=0.7952