War related sexual violence and it's medical and psychological consequences as seen in Kitgum, Northern Uganda: A cross-sectional study

Northern Uganda has suffered chronic warfare for the last 20 years, with a conflict waged between the central government army (the Uganda Peoples Defence Forces; UPDF) and a vicious rebel group, the Lord's Resistance Army. This conflict has led to the mass traumatisation of the population including the abuse of human rights, and the forced displacement of over 2 million people (80% of the population in the region) into internally displaced persons camps. Since 2006, the Sudan sponsored peace talks between the Uganda government and the Lord's Resistance Army have brought relative peace to the region with people gradually returning to their homes; as a result both government and the international community have begun focusing on the rehabilitation and resettlement of the war survivors. However often such rehabilitation and resettlement programs fail to address the mental health and reproductive health needs of war survivors; when these needs are addressed it is done in a very superficial manner.

This paper will examine the specific long term health consequences of war related sexual violence among women from two internally displaced persons camps in Kitgum district, in war affected Northern Uganda, with the aim of highlighting the burden of these problems.

Study respondents were recruited in 2005 from 2 out of 21 internally displaced persons camps (IDPs) in Kitgum district (approximate population, 300,000) when war was still ongoing and the security situation in the region extremely precarious. The camps which were selected, Mucwini (population 22,438) and Padibe (population 35,006) were some of the biggest and worst affected by the war. This study was part of a medical intervention to offer emergency reproductive health care, psychological support and surgical services to women affected by the war [8]. The intervention was undertaken by Isis-WICCE (an international women's organization) working in collaboration with professional associations of gynaecologists, psychiatrists and surgeons of Uganda.

The medical intervention involved an initial screening stage where trained non-medical community interviewers resident in the two camps administered a structured screening questionnaire to identify respondents who had suffered war trauma and had psychological, gynaecological or surgical complications. Apart from documenting these problems through the screening interview, respondents who were found to require further medical evaluation and treatment were referred to the camp based health workers. Respondents who were found to require further evaluation and treatment including surgery were referred to the visiting team of specialist who were based at Kitgum district hospital. The study employed a purposive cross-sectional study design where respondents who wanted to access the services of the medical intervention were consecutively recruited and subjected to the screening interview.

A structured questionnaire was used to screen the war survivors for war trauma and it's psychological and medical consequences. It included items on the following; a) Socio-demographics - change of district, sex, age, religion, marital status, highest educational attainment, and employment status, b) War trauma- derived from the commonly reported forms of war trauma in Uganda [9] grouped under i) loss of loved ones (5 items)- including death as a result of war of the spouse, children, parents or other close relatives, ii) war related sexual violence (11 items)- including single episode rape, gang rape, homosexual rape, sexual comforting and abduction with sex, iii) physical trauma (10 items)-including whether the respondent suffered beatings or being kicked, gunshot injury, burn injuries, being forced to carry heavy loads for long distances and severe tying of the hands behind the back (locally called Kandoya), iv) psychological trauma (6 items)- including whether the respondent witnessed someone being killed, ever being forced to sleep in the bush or swamps for days, weeks and whether ever abducted, c) Perpetrators of war trauma (9 items)- possible perpetrators of war trauma included the Lord's Resistance Army, government army (UPDF), police, and prisons officers, d) Psychological consequences - including psychological dependence on alcohol (alcoholism) assessed using the C.A.G.E [10], iii) psychological distress (assessed using the WHO Self Report Questionnaire-20 [11], 1994), iv) suicide attempt (whether had attempted suicide in a life-time and in the last 12 month), v) whether had homicidal ideas. e) Gynaecological problems[12, 13] (13 items)- including leaking urine (vaginal fistula), leaking faeces (rectal fistula), vaginal and perineal tears, sexual dysfunction and abnormal vaginal discharge, f) Surgical problems[12, 13] (9 items)- including chronic backache, broken limbs, disfigurement as a result of burns and swellings on limbs, g) HIV serological status, and h) personal risk assessment of contracting HIV over next few years-assessed as low, medium or high.

Questionnaires were reviewed every evening for errors and inconsistencies which were then brought to the attention of the interviewers. Data was entered using SPSS version 10.0. and analysed using STATA. War related sexual violence was the dependent variable. As part of data analysis, frequencies were generated, and unadjusted analyses undertaken using Chi square and Odds ratios as the test statistics. Factors found to be significantly associated with war related sexual violence in unadjusted analyses were entered into multiple logistic regression model to determine their independent contribution to war related sexual violence. For this multivariable analysis, a logistic regression model was fitted using a stepwise backward elimination procedure with a liberal significance level of 0.15 so that all potentially important variables would be included in the model. The model was then verified by forward selection.

Scientific and ethical clearance for the medical intervention and research was sought by Isis-WICCE from both the Ministry of Health of Uganda and the Uganda National Council of Science and Technology. Informed consent was obtained from the respondents after adequate explanation of the study procedures and both the immediate and anticipated benefits of the screening procedure/study. This manuscript was reviewed and cleared by the Uganda National Health Research Organisation (UNHRO).

813 respondents were screened during this study of whom 573 (70.5%) were women or girls. The gender ratios displayed in this study probably reflect the manner in which the medical intervention was publicized in the camps- as the intervention mainly addressed women's health concerns. This analysis focused on the women and girls, with most of them (90.2%) displaced from within their mother district of Kitgum. The majority were over the age of 24 years (88.7%), mainly Christian (99.3%), married or ever have been married (82.3%), of no or little formal education (less than 8 years of formal education; 94.9%) and mainly employed as peasant farmers (87.6%).

Over a quarter (28.6%) of the women and 6.7% of the men reported having suffered at least one form of war related sexual violence (Table 1). The most frequently reported forms of war related sexual violence among women included defilement (14.5%), abduction with sex (7.9%), attempted rape (4.9%), forced marriage (4.4%) and heterosexual rape-single episode (4.2%). Among the men the most frequently reported forms of war related sexual violence was homosexual rape (4.2%). Other commonly reported forms of war trauma experiences reported included, death of at least one loved one (67.7%), having suffered at least one physical trauma (52.0%), and having suffered at least one psychological trauma (86.7%).

The main perpetrators of war trauma in general among women were the rebel Lord's Resistance Army (71.4%) and the government soldiers (23.4%), a similar pattern was also reflected for war related sexual violence where the main perpetrators were the rebel Lord's Resistance Army (80.5%) and the government soldiers (23.2%; Table 2),

Table 3, a comparison of the sexual violations committed by the rebel Lord's Resistance Army (LRA) and the government soldiers reveals that the main method of sexual trauma in both groups were defilement (LRA- 55.3%; government soldiers- 42.1%), abduction with sex (LRA- 27.3%; government soldiers- 36.8%) and attempted rape (LRA- 15.2%; government soldiers- 18.4%). There were however important differences between the two groups of perpetrators with government soldiers proportionally using more the method of sexual comforting (26.3%) than the LRA (10.6%) and the LRA proportionally using more the method of heterosexual rape- single episode (12.9%) than government soldiers (7.9%).

About three quarters (72.4%) of the respondents had 'at least one gynaecological complaint' with the most commonly reported complaints being chronic lower abdominal pain (52.2%), abnormal vaginal bleeding (27.4%), infertility (26.6%), genital sores (25.3%), and swellings in the abdomen (22.5%).

The respondents also had high levels of both psychological and surgical morbidity including alcoholism (18.7%), significant psychological distress (scores ≥6 on the WHO SRQ-20; 69.4%), attempted suicide life-time (15.8%), having at least one surgical complaint (75.6%) and known HIV positive status (4.8%) (although the majority of respondents (82.3%) did not know their HIV status.

Among women, the socio-demographic factors significantly associated with war related sexual violence were, age ≤ 44 years where 35.2% suffered war related sexual violence as compared to 19.3% in the ≥ 45 years age group (OR, 2.28; 95%CI, 1.52-3.40; p = < 0.001), and being Catholic where 32.3% suffered war related sexual violence as compared to 23.5% in the Protestants (OR, 1.55; 95%CI, 1.06-2.28; p = 0.04) (Table 4).

Among women, the medical and psychosocial factors significantly associated with war related sexual violence were, suffered war related physical trauma (OR, 1.55; 95%CI, 1.07-2.24; p = 0.02), and having 'at least one gynaecological complaint' (OR, 2.86; 95%CI, 1.75-4.67; p = < 0.001) (Table 5).

The specific gynaecological complaints associated with war related sexual violence included infertility (OR, 2.16; 95%CI, 1.44-3.24; p = < 0.001), chronic lower abdominal pain (OR, 1.98; 95%CI, 1.35-2.91; p = < 0.001), abnormal vaginal bleeding (OR, 1.86; 95%CI, 1.24-2.79; p = 0.002) and sexual dysfunction (OR, 2.09; 95%CI, 1.27-3.45; p = 0.003) (Table 6).

The logistic regression model computed showed that 2 domains of factors were independently associated with war related sexual violence, that is the predisposing factors-age of less than or equal to 44 years (AOR, 1.96; 95%CI, 1.23-3.13; p = 0.004) and being Catholic (AOR, 1.65; 95%CI, 1.07-2.54; p = 0.03) and the medical consequences of war related sexual violence-suffering from a gynaecological problem (AOR, 2.36; 95%CI, 1.39-4.00; p = 0.002) (Table 7).