Theses Doctoral

The Effects of Conflict on Fertility Desires and Behavior in Rwanda

McGinn, Therese J.

Rwanda experienced genocide from April to July 1994 during which over 800,000 people were murdered. Among the far-reaching changes that followed this event among individuals and in society overall, the Rwandan Demographic and Health Surveys (DHS) showed that contraceptive prevalence declined from 13% in 1992 to 4% in 2000 among married women of reproductive age.

This dissertation has two hypotheses concerning Rwandan women's fertility preferences and behavior following the genocide. It is hypothesized that, first, high levels of conflict reduced women's desire for a child or for additional children and second, that women who experienced relatively high levels of conflict were more likely to act on their wish to not have a child or another child by using modern contraceptives than were women who experienced relatively low levels of conflict.

The study's logistic regression dependent (outcome) variables were desire for a or another child and the use of modern contraceptives; the source for these data was the 2000 DHS. Three groups of independent variables were included: socio-demographic variables, also from the 2000 DHS, included age, number of living children, education level, urban/rural residence and socio-economic status; availability of family planning services, assessed using women's perception of distance as a barrier to obtaining health care for themselves, from the 2000 DHS, and quality of health services, assessed with data from the 2001 Service Provision Assessment; and experience of conflict, measured as the percentage of the 1994 commune populations that resided in refugee camps in 1995. Communes were considered `high migration' if 10 percent or more of their populations migrated to camps and `low migration' if less than 10 percent of their populations migrated to camps. Women who lived in high migration communes were considered to have relatively high experience of conflict and those who lived in low migration communes were consider dot have relatively low experience of conflict.

Analysis showed that residents of high migration communes were significantly less likely to want a or another child as compared to residents of low migration communes (OR = .74); it appeared that the social environment of high migration had a dampening effect on desire for children. The analysis also showed that residents of high migration communes were significantly less likely to use a modern contraceptive method than were those of low migration communes (OR = .57), even though they were less likely to want a or another child and even when family planning services were reasonably available.

The reasons for these results are unclear, and many factors may contribute. The generalized trauma experienced by the population may have had a numbing effect, in which taking action in any domain was difficult. Women may have felt pressured by society to have children as the society emerged from war, despite their own preferences. The population may also have distrusted government health facilities - the only source of services for most - in light of the interactions with officials during and after the genocide. However, another set of reasons specific to women and women's health may also have influenced the findings. There is a pervasive social stigma around reproductive health; these services have generally lagged behind other primary health care components. Moreover, rape was used as a weapon of war in the genocide; these experiences may have reduced women's willingness to seek reproductive health services specifically. Finally, the Rwandan genocide and its preparation were decidedly misogynistic; this pervasive dehumanization may have made it particularly difficult for women to seek care for their sexual and reproductive health needs and desires. This complex personal, social, physical and political context may explain why Rwandan women who may not have wanted a child or additional children nonetheless did not consistently act on their desires in the years following the 1994 genocide.

The dissertation includes a series of essays providing the author's personal perspective on working in Rwanda in the 1980s and 1990s and being present in the country at the start of the genocide in April 1994.

Geographic Areas


  • thumnail for 295882_pdf_290117_9BED99E0-0160-11E4-9B4A-944DEF8616FA.pdf 295882_pdf_290117_9BED99E0-0160-11E4-9B4A-944DEF8616FA.pdf application/pdf 885 KB Download File

More About This Work

Academic Units
Sociomedical Sciences
Thesis Advisors
Parker, Richard G.
Dr.P.H., Mailman School of Public Health, Columbia University
Published Here
June 27, 2017