The balance of self-reported heterosexual activity in KAP surveys and the AIDS epidemic in Africa

The AIDS epidemic in Africa depends on heterosexual behavior, about which little is known. National Knowledge, Attitudes and Practices (KAP) surveys are a potentially important data source but are compromised because women report less-often, much less-sexual activity than men. This imbalance is investigated using questions on the type of activity (with regular or casual partners and involving compensation) and on the duration of abstinence (time since last act), as well as other socioeconomic variables. Evidence from a hazard analysis and other information suggest that the imbalance may originate in the failure of respondents to answer accurately rather than the systematic undersampling of particular respondents, such as prostitutes. Lf so, the imbalance does not reflect an epidemic driven by a small group of high-activity women-the core group model-but rather a more diffuse sexual network. The contrasting implications of these hypotheses for the dynamics of the epidemic have been stressed by theoretical epidemiologists. The imbalance seems to originate with respondents who have particular socioeconomic characteristics. Furthermore, the absolute level of activity inferred from the hazard analysis is substantially higher than what is directly reported with possible importance for the evolution of the epidemic. These findings have implications for the usefulness of KAP surveys in general and how they can be refined in particular.