SEXUAL BEHAVIORS RELEVANT TO HIV TRANSMISSION IN A RURAL AFRICAN POPULATION - HOW MUCH CAN A KAP SURVEY TELL US

KAP surveys have been proposed as a means to gather quantitative infor mation on AIDS-related sexual behaviors, but the validity of survey re sults has not been tested. The validity of data gathered during a KAP survey in a rural district in Northern Uganda (N = 1486) was examined analyzing expected behavioral patterns, agreement of partner reports, and concordance of number of sexual contacts across gender. Patterns o f sexual behavior and age trends are as expected. More men (50%) than women (18.5%) reported premarital sex. The likelihood of sexual interc ourse before marriage increases with age at first marriage and with ed ucation. Women marry 5 years earlier than men, and the number of marri ages increases with age. Peak incidence of casual sex occurs before ag e 25. The male/female ratio of casual sex is 4, as compared to about 3 in other African surveys. Single men are 2.5 times more likely to eng age in casual sex than married males. Agreement of partner reports was examined for 392 couples selected by chance. 86% of the couples agree d on being polygamous or monogamous. On average men reported 1.3 (SD = 0.7) wives as compared to women reporting 1.5 (SD = 0.89) wives (P < 0.001). 16.8% of women declared more, and 2.8% less cowives than their husband (r = 0.65). Self-reports on frequency of sexual intercourse i n the past month were examined for 256 monogamous couples. Mean freque ncies differ (5.24 +/- 5.1 for men, 4.43 +/- 4.7 for women, P < 0.001) . 42.8% of couples are in agreement within +/-1 unit (r = 0.44). The t otal number of extra-marital and marital sex acts, as well as the tota l number of partners reported by each gender are similar. There is, ho wever, a striking gender difference in reporting of casual partners in the past year. Data were found to be accurate at the aggregate level. However, accuracy of reporting at the individual level was found to b e low. The gender difference in reporting of casual partners may be du e to female underreporting, to not having captured prostitutes or to a different perception of the meaning of casual partnership. All KAP su rveys should include a validity analysis, so as to provide a sense of the accuracy of the surveys and allow for comparison of the quality of different KAP surveys. There is an urgent need for a standardized app roach to validating the findings from AIDS-related KAP surveys. Some o f the indirect methods described here could be relevant for further us e.