Declining HIV prevalence and risk behaviours in Zambia: evidence from surveillance and population-based surveys

OBJECTIVE: To examine trends in HIV prevalence and behaviours in Zambia during the 1990s. METHODS: The core Zambian system for epidemiological surveillance and research has two major components: (i) HIV sentinel surveillance at selected antenatal clinics (ANC) in all provinces; and (ii) population-based HIV surveys in selected sentinel populations (1996 and 1999). The former was refined in 1994 to improve the monitoring of prevalence trends, whereas the latter was designed to validate ANC-based data, to study change in prevalence and behaviour concomitantly and to assess demographic impacts. RESULTS: The ANC-based data showed a dominant trend of significant declines in HIV prevalence in the 15--19 years age-group, and for urban sites also in age-group 20--24 years and overall when rates were adjusted for over-representation of women with low education. In the general population prevalence declined significantly in urban women aged 15--29 years whereas it showed a tendency to decline among rural women aged 15-24 years. Prominent decline in prevalence was associated with higher education, stable or rising prevalence with low education. There was evidence in urban populations of increased condom use, decline in multiple sexual partners and, among younger women, delayed age at first birth. CONCLUSIONS: The results suggested a dominant declining trend in HIV prevalence that corresponds to declines in incidence since the early 1990s attributable to behavioural changes. Efforts to sustain the ongoing process of change in the well-educated segments of the population should not be undervalued, but the modest change in behaviour identified among the most deprived groups represents the major preventive challenge.