Section: Research Letters
Child rape violates human rights and causes immediate and long-term health problems for the child. In the 1998 South Africa Demographic and Health Survey, we assessed frequency of rape in a nationally representative study of 11 735 women aged 15-49 years. 153 (1.6%, 95% CI 1.2-1.9%) of these women had been raped (forced or persuaded to have sex against their will) before the age of 15 years. Our results show that younger women were significantly more likely to report rape than older women (p<0.0001). The largest group of perpetrators (33%) were school teachers. Our findings suggest that child rape is becoming more common, and lend support to qualitative research of sexual harassment of female students in schools in Africa.
Correspondence to: Dr Rachel Jewkes (e-mail: rjewkes@mrc.ac.za)
Sexual abuse of girls is a problem worldwide, and a growing concern in sub-Saharan Africa. Such abuse constitutes a profound violation of human rights, and has been associated with long-term mental and physical health consequences.[1] Coerced sex in dating relationships, including forced sexual initiation, is not rare: results of a study[2] in Cape Town, South Africa, showed that 72% of pregnant teenagers and 60% of teenagers who had never been pregnant had had coerced sex, but little research has been done on sexual abuse in earlier childhood.
In 1998, we did a cross-sectional study, the South Africa Demographic and Health Survey, of a nationally representative sample of 11 735 women aged 15-49 years. We selected women by multistage random methods that were stratified by province and urban or non-urban regions. The sampling frame in each province consisted of enumeration areas that had been demarcated for the 1996 census. Weighting factors have been applied to the data to make it representative of the country. Actual numbers are not provided since these values are not meaningful in a non-selfweighting sample. The study complied with WHO guidelines for research on violence against women, and every participant provided written informed consent. The Medical Research Council's ethics committee approved the study.
We asked women if they had been "forced to have sexual intercourse against their will by being threatened, held down, or hurt in some way" or "persuaded to have sexual intercourse when they did not want it". Women who responded that they had, were asked if these events had happened before the age of 15 years, in which case we concluded that they had been raped. The response rate for the survey was 90%.
153 (1.6%; 95% CI 1.2-1.9) of the women interviewed reported rape before the age of 15 years, Two (1.3%) of whom had been raped by a gang of three and ten men. The figure shows the proportion of women in each 5-year age-group who reported child rape. 85% of the rapes took place in children aged 10-14 years, and 15% between 5 and 9 years. School teachers were the most common child rapists, and were responsible for 33% (95% CI 23-42) of rapes. Relatives were also a major group (21%), as were strangers or recent acquaintances (21%), and boyfriends (10%).
Proportion of women in each age group reporting rape before age 15 years
Bars indicate 95% CIs.
Educational status and type of residence were not associated with childhood rape, but ethnic origin (based on the apartheid-defined population categories), province, and age-cohort were. After adjustment for province and age, more white women reported rape than African women (table). Women in the Northern Cape, Free State, and North West, reported rape significantly less often than those in the Western Cape (after adjustment for ethnic origin and age, table). Women were significantly less likely to report rape with increasing age (table). Age-by-province and age-by-race interactions were not significant.
Few nationally representative studies have been done on child rape. In the USA, 1.9% of women have reported rape or attempted rape before age 12,[3] a finding close to that of our study. The negative age trend suggests that child rape is becoming more common in South Africa, although reporting bias cannot be ruled out. Our findings, however, lend support to perceptions of increases in the number of child rapes reported in the media, seen in health facilities, and reported to the police. In 1998, the rate of rape and attempted rape of girls aged 0-17 years reported to the South Africa police services was 47.1 per 100 000 people, but we could not investigate for a trend because earlier data was not disaggregated by age.
The role of school teachers in child rape has been reported in many other African countries.[4] Through such abuse, not only is the girl's body violated, but also her right to education. Not only is risk of HIV-1 transmission increased through the sexual act, but in many countries, child rape has increased the likelihood of unsafe sexual practices during later years, including having multiple partners, participation in sex work, and increase in risk of rape in adulthood.[5] Intergenerational sex also fuels the HIV-1 epidemic by providing foci of infection within every emerging age group, leading to transmission of the virus to peers once children reach the stage of consensual sexual activity. Many girls are forced to leave school because of pregnancies fathered by teachers and beca use of harassment by teachers.[4] A girl's ability to reach her economic and social potential is thus reduced, and likelihood of subsequent dependency on sex for payment rises. They and their children are also placed at much greater risk of many other health problems.
Our findings confirm that rape of girls, especially in school, is a substantial public-health problem in South Africa. Research is needed to understand its broader social context, nature, and magnitude, and to develop interventions for primary prevention and prevention of its long-term health consequences. Effective action to address rape and sexual harassment of girls in schools is needed. South Africa has only recently made sexual relations, consensual or otherwise, between staff and students a serious offence that results in dismissal of staff. Many countries do not have such legislation. Enforcement still presents a substantial challenge.
Contributors
R Jewkes designed the questions on child sexual abuse, ensured adherence to ethics, and wrote the report. J Levin was the study statistician, and analysed the data. D Bradshaw was the main investigator, and N Mbananga was the project manager. All researchers contributed to all decisions on planning, design, and project implementation, and writing of the report.
Conflict of interest statement: None declared.
Acknowledgments
The 1998 South Africa Demographic and Health Survey was initiated and mostly funded by the National Department of Health. The survey was coordinated by the Medical Research Council, with technical assistance from MACRO International. Fieldwork was done by the Centre for Health Systems Research and Development of the Free State University, and USAID provided additional funding. The Department of Health of each province provided provincial co-ordinators who assisted in facilitating the survey, Statistics South Africa provided the sampling frame and sample details, and the Human Sciences Research Council monitored the quality of fieldwork. The study sponsors had no role in study design, data collection, data analysis, or data interpretation or in the writing of the report.
We thank the many field workers, supervisors, editors, and provincial managers for their contribution, and the people who shared intimate details of their lives in the interviews.
Multiple logistic regression model
Legend for Chart: A - Characteristic B - Odds ratio (95% CI) C - p A B C Province Western Cape 1.00 <0.0001 Eastern Cape 0.51 (0.22-1.17) Northern Cape 0.34 (0.14-0.82) Free State 0.25 (0.08-0.83) Kwazulu Natal 0.87 (0.35-2.15) North West 0.27 (0.08-0.91) Gauteng 1.45 (0.63-3.34) Mpumalanga 1.63 (0.67-3.94) Northern 0.48 (0.16-1.45) Ethnic origin Black African 1.00 0.0098 Coloured 2.03 (0.89-4.61) White 2.57 (1.41-4.69) Indian 0.56 (0.18-1.75) Age group (linear trend) 0.74 (0.66-0.82) <0.0001
1 Mccauley J, Kern DE, Kolodner K, et al. Clinical characteristics of women with a history of childhood abuse: unhealed wounds. JAMA 1997; 277: 1362-68.
2 Jewkes R, Vundule C, Maforah F, Jordaan E. Relationship dynamics and adolescent pregnancy in South Africa. Soc Sci Med 2001; 52: 733-44.
3 Tjaden P, Thoennes N. Full report of the prevalence, incidence and consequences of violence against women. National Institute of Justice, Washington: 2000.
4 Omaar R, de Waal A. Crimes without punishment: sexual harassment and violence against female students and universities in Africa. Discussion Paper Number 4, African Rights, July 1994.
5 Garcia-Moreno C, Watts C. Violence against women: its importance for HIV/AIDS. AIDS 2000; 14 (suppl): S253-65.
~~~~~~~~
By Rachel Jewkes, MBBS, Gender and Health Group, Medical Research Council, Private Bag X385, Pretoria 0001, South Africa; Jonathan Levin, PhD, Gender and Health Group, Medical Research Council, Private Bag X385, Pretoria 0001, South Africa; Nolwazi Mbananga, MD, Gender and Health Group, Medical Research Council, Private Bag X385, Pretoria 0001, South Africa and Debbie Bradshaw, PhD, Gender and Health Group, Medical Research Council, Private Bag X385, Pretoria 0001, South Africa