1. Impact of armed conflict on
maternal and reproductive health
in Sub-Saharan Africa
Primus Che Chi
PhD Disputation, Kreftsenteret Auditorium
March 9, 2016

2. Acknowledgements
Supervisors
• Professor Henrik Urdal
• Professor Johanne Sundby
Adjudication committee
• Senior Advisor Michael Tawanda
• Senior Research Fellow Helge Brunborg
• Associate Professor Benedikte Lindskog
Institutions: Faculty of Medicine, UiO and PRIO
Family, friends & colleagues
PhD Defense_Primus Che Chi 2

3. Overview
• Background
• Purpose of the study
• Study design and
procedures
• Key findings and
discussion
• Conclusions
• Recommendations
PhD Defense_Primus Che Chi 3

4. Background
Armed conflict:
• Public health problem.
• Important contributor to the social and
political determinants of health.
• Driver of poverty and health inequity.
• Leading to poor maternal and child outcomes.
PhD Defense_Primus Che Chi 4

5. Purpose of the study
Aim
To assess the impact
of armed conflict on
maternal and
reproductive health
(MRH) in sub-Saharan
Africa
• Case studies of
recent post-conflict
societies:
Burundi and
Northern Uganda
Objectives:
• Impact of armed conflict on
maternal mortality and fertility
levels
• Stakeholders’ perceptions of
the effects of armed conflict on
MRH services and outcomes
• Determinants of women’s
utilisation of MRH services
• Barriers to the effective
delivery of emergency
obstetric and neonatal care
(EmONC)
PhD Defense_Primus Che Chi 5

6. Study design & procedures
Multidisciplinary and multi-method
Quantitative: Cross-national time-series regression
analysis (IV: Armed conflict; DV: MMR & TFR).
Qualitative: semi-structured in-depth interviews
(IDIs) and focus group discussions (FGDs) with key
stakeholders (health providers, women of
reproductive age, policy-makers etc.) in Burundi &
Northern Uganda.
PhD Defense_Primus Che Chi 6

7. Key findings and discussion
PhD Defense_Primus Che Chi 7

8. Model 1 Model 2 Model 3 Model 4
Battle-related deaths (ln) 0.008
(0.007)
-0.008
(0.006)
-0.031***
(0.007)
-0.001
(0.007)
Neighboring conflict 0.009
(0.008)
0.006
(0.007)
0.004
(0.007)
-0.015
(0.008)
Battledeaths (ln)*
Poverty
0.035***
(0.006)
1970-74 (reference category)
1975-79 -0.208**
(0.070)
0.154*
(0.074)
0.134
(0.072)
0.045
(0.080)
1980-84 -0.345***
(0.070)
0.287**
(0.088)
0.231**
(0.087)
0.197*
(0.098)
1985-89 -0.556***
(0.083)
0.324**
(0.107)
0.270*
(0.105)
0.256*
(0.122)
1990-94 -0.829***
(0.092)
0.292*
(0.125)
0.213
(0.123)
0.243
(0.147)
Constant 16.51***
(1.018)
17.33***
(1.420)
15.33***
(1.185)
17.55***
(1.794)
N 1,124 1,018 1,018 668
R sq, overall 0.03 0.15 0.16 0.10
Armed conflict and total fertility rates, 1970-2005
PhD Defense_Primus Che Chi 8

9. Armed conflict and maternal mortality rates, 1990-2005
Model 5 Model 6 Model 7 Model 8
Battle-related deaths (ln) 0.012*
(0.005)
0.007
(0.005)
0.007
(0.006)
0.003
(0.006)
Neighboring conflict -0.014**
(0.005)
-0.013**
(0.005)
-0.013**
(0.005)
0.006
(0.007)
Battledeaths (ln)*
Poverty
-0.0003
(0.004)
1990-94 (reference category)
1995-99 -0.155***
(0.032)
-0.133***
(0.032)
-0.133***
(0.032)
0.169**
(0.059)
2000-04 -0.280***
(0.109)
-0.238***
(0.043)
-0.238***
(0.044)
0.076*
(0.035)
HIV prevalence 0.052***
(0.008)
Population (ln) 0.256
(0.132)
0.245
(0.142)
0.243
(0.143)
0.247
(0.245)
Constant 2.882*
(1.149)
5.060***
(1.341)
3.068*
(1.244)
5.832*
(2.286)
N 561 532 532 354
R sq, overall 0.05 0.40 0.40 0.54
PhD Defense_Primus Che Chi 9

10. Key findings: Qualitative studies
PhD Defense_Primus Che Chi 10
Country Study areas
Participants/ Informants
Total
Women LHPs NGOs
Burundi Bujumbura
Marie and
Ngozi
provinces
11
Interviews
& 2 FGDs
9
Interviews
& 1 FGD
11
Interviews
& 1 FGD
31
interviews
& 4 FGDs
Uganda Northern
Uganda
10
Interviews
& 2 FGD
12
Interviews
& 1 FGD
10
Interviews
& 1 FGD
32
interviews
& 4 FGDs
All countries
21
interviews
& 4 FGDs
21
interviews
& 2 FGDs
21
interviews
and 2
FGDs
63
interviews
& 8 FGDs

11. Key finding 2 – Effects of conflict
The perceived effects of the conflict on MRH
outcomes included:
• increased maternal and newborn morbidity and
mortality
• high prevalence of HIV/AIDS and SGBV
• increased levels of prostitution, teenage
pregnancy and clandestine abortion
• high fertility levels
Relocation to government recognized IDP camps
reportedly improved access to health services for
many women
PhD Defense_Primus Che Chi 11

12. Armed conflict and access and quality of MRH services
Event
• Armed conflict
Mechanisms
•Destruction of health
facilities
•Looting of medical
supplies at health
facilities
•Shutdown of health
facilities
•Fleeing of local health
providers from conflict
zone
•Limited movement to
operational health
facilities due to insecurity
•Irregular opening hours of
health facilities
•Disruption of medical
supplies to health facilities
•Displacement of
populations away from
health facilities
•Targeted killing of local
health providers (Burundi)
•Favouritism in the
provision of health
services on ethnic basis
(Burundi)
•Abduction of health
providers (N .Uganda)
Outcomes
•Poor access to
MRH services
•Limited access to
MRH services
•Poor quality MRH
services
PhD Defense_Primus Che Chi
11

13. Key finding 3 – Determinants: utilisation
PhD Defense_Primus Che Chi 13

14. Key finding 4 – Barriers: EmONC
Study settings
Themes Subthemes Burundi Northern
Uganda
Human resources-related challenges Acute shortage of trained personnel X X
Demoralised personnel and perceived lack of
recognition
X
Perceived poor living conditions and poor
remuneration for personnel
X X
High personnel turnover X X
Increasing workload and high burn-out X X
High levels of staff absenteeism in rural health
centres
X
Poor level of coordination among key EmONC
personnel resulting in delays to provide emergency
services
X
Systemic and institutional failures Poorly operational ambulance service for referrals X
Inefficient drug supply system X
Inefficient referral system X
Lack of essential installations, supplies and
medications
X X
Poor allocation of limited resources X
Poor harmonization and coordination of EmONC
training curriculum nationally
X
Weak/ incomprehensive training curriculum X
Poor data collection and monitoring system X X
Inequity in the distribution of EmONC facilities
between urban and rural areas
X X
PhD Defense_Primus Che Chi 14

15. Conclusions
• Armed conflicts have a substantial negative
impact on MRH
• In post-conflict settings, women’s utilization of
MRH services is affected by a complex set of
factors cutting across the individual, socio-
cultural, political and health system domains.
• The delivery of EmONC services post-conflict
health systems is hampered by a series of
human resources-related challenges, and
systemic and institutional failures.
PhD Defense_Primus Che Chi 15

16. Recommendations
• Prioritise the delivery of quality EmONC
services.
• Integration of refugees and/or IDPs health
services with those of local host communities.
• Bridge ethnic, religious or political inequalities
in the delivery of health services .
• Coordinate the location and construction of
IDP camps with local government.
PhD Defense_Primus Che Chi 16