Research

Patterns of sexual violence in Eastern Democratic Republic of Congo: reports from survivors presenting to Panzi Hospital in 2006

Susan A Bartels^1,2 , Jennifer A Scott³ , Denis Mukwege⁴ , Robert I Lipton¹ , Michael J VanRooyen^2,5,6 and Jennifer Leaning^2,6

¹  Department of Emergency Medicine, Beth Israel Deaconess Medical Center, Boston, USA

²  Harvard Humanitarian Initiative, Cambridge, USA

³  Department of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Boston, USA

⁴  Hôpital de Panzi, Bukavu, South Kivu, Democratic Republic of Congo

⁵  Department of Emergency Medicine, Brigham and Women's Hospital, Boston, USA

⁶  Harvard School of Public Health, Boston, USA

author email corresponding author email

Conflict and Health 2010, 4:9doi:10.1186/1752-1505-4-9

Published:	5 May 2010

Abstract

Background

Despite the signing of international peace agreements, a deadly war continues in the Democratic Republic of Congo (DRC) and sexual violence is a prominent modus operandi of many military groups operating in the region.

Methods

Retrospective cohort study of women who presented to Panzi Hospital in 2006 requesting post-sexual violence care. Data was extracted and analyzed to describe the patterns of sexual violence.

Results

A total of 1,021 medical records were reviewed. A majority of attacks occurred in individual homes (56.5%), with the fields (18.4%) and the forest (14.3%) also being frequent locations of attack. In total, 58.9% of all attacks occurred at night. Of the four primary types of sexual violence, gang rape predominated (59.3%) and rape Not Otherwise Specified (NOS) was also common (21.5%). Sexual slavery was described by 4.9% of the survivors and a combination of gang rape and sexual slavery was described by 11.7%. The mean number of assailants per attack was 2.5 with a range of one to > 15. There were several demographic predictors for sexual slavery. Controlling for age, education level and occupation, a marital status of "single" increased the risk of sexual slavery (OR = 2.97, 95% CI = 1.12-7.85). Similarly, after controlling for other variables, age was a significant predictor of sexual slavery with older women being at a slightly reduced risk (OR = 0.96, 95% CI = 0.92-0.99). Women who experienced sexual slavery were 37 times more likely to have a resultant pregnancy in comparison to those who reported other types of sexual violence (OR = 37.50, 95% CI = 14.57-99.33).

Conclusions

Among sexual violence survivors presenting to Panzi Hospital in 2006, the majority of attacks occurred in women's own homes, often at night. This represents a pattern of violence that differs from other conflict settings and has important implications regarding protection strategies. Sexual violence in South Kivu was also marked with a predominance of gang rape, thus increasing the risk of serious injury as well as the likelihood of an individual woman contracting a sexually transmitted infection (STI). Sexual slavery was noted to be more common among young, single women and was found to have a high rate of resultant pregnancy.