This article reports on evidence for behaviour change in response to AIDS among women in two rural areas of Manicaland Province, Zimbabwe. It examines self-reported data on two overlapping areas of behaviour: (1) actions taken to avoid HIV-1 infection; and (2) fertility practices. The latter were used to assess the validity of the former, given that self-reported behaviour data are notoriously problematic. It is concluded that while self-reported behaviour change is exaggerated, the true level of change has nonetheless been significant and includes delayed onset of sexual relations, increased use of condoms and, possibly, increased monogamy.

Reported actions taken to avoid HIV-1 infection and differentials in fertility practices were correlated with data on demographic, social and psychological factors. Differentials in fertility practices were associated with heightened risk perception—particularly when based on personal acquaintance with AIDS patients—but not with greater knowledge of HIV-1/AIDS. Results from the study suggest that effective behaviour change in Manicaland is facilitated by greater knowledge, experience and personal risk perception but obstructed by low female autonomy, marital status and economic status, and by male labour migration and alcohol consumption.

Gaps in knowledge included misconceptions about the distinction between HIV-1 and AIDS, the influence of STDs, perinatal transmission, and incorrect modes of transmission. Better knowledge was associated with education, religion, travel and media exposure. Personal risk perception was quite high (42%) and correlated with non-marriage, media exposure and contact with medical services. Few respondents knew close relatives with HIV/AIDS (4%) but nearly a quarter of those who felt in danger of infection said this was because friends and relatives were dying of AIDS. Many reported credible behavioural responses, some of which would only be effective given their partner's co-operation.

Intensified behaviour interventions are needed which should include peer-education initiatives targeting men and individuals without access to modern media. The epidemic may accelerate fertility decline in rural Zimbabwe through behavioural as well as biological change.