HIV, HSV-2 and sexual behaviour
The long-term population based study gave us the opportunity to investigate the HIV epidemic and its impact in Karonga District from its first introduction through to the current era of widespread antiretroviral therapy (ART), using conventional and molecular epidemiology.
HIV was first introduced in the early 1980s. We have samples from 1982 which include the earliest known subtype C sequences in the world. Even by the early 1980s there were several distinct introductions, and we have documented a shift from a mixture of subtypes (C, A, D) in the early 1980s to an epidemic dominated by subtype C by the late 1980s.
Using data from whole population surveys in parts of the district, controls from case-control studies of tuberculosis and leprosy, and antenatal clinic surveillance, we have shown that HIV prevalence increased from <1% in the early 1980 to 2% in the late 1980s to peak at about 13% in the late 1990s. The highest prevalence was found in the urban and periurban areas, and was initially more common among the most educated and those of higher socio-economic status. Within the demographic surveillance area (where HIV prevalence is around the average for the district as a whole) HIV prevalence is 9% in women and 7% in men. Methodological work has assessed the biases in using antenatal clinic data (for both HIV and HSV-2 estimates) and the biases in general population surveys when increasing numbers of people already know their status.
The demographic surveillance area allows us to measure the direct and indirect impact of HIV and of ART, on mortality, morbidity (including tuberculosis), fertility, and other outcomes. We showed the devastating effect of HIV on adult and child mortality and were the first to show improved survival at the whole population level following ART roll-out. Associations with and impacts on tuberculosis, non communicable diseases and fertility are covered in other sections.
Sexual behaviour surveys were conducted in the demographic surveillance area. These have shown little change in the age at first sex or marriage in women over time, and a slight decrease in men. There was a strong association between age at menarche and age at first sex, and women with earlier menarche were more likely to have HSV-2 infections; there was no association with HIV. Looking historically, we found that HSV2 was high in this community before the increase in HIV. Methodological work involving triangulation of measures showed under-reporting of sexual activity by young men as well as by young women, and challenged UNAIDS recommendations on how to measure concurrency.
Current studies (2012-2016)
Current studies on HIV itself (rather than its link with other diseases) focus largely on ART uptake and the impact of increasingly decentralised care and of the introduction of “Option B+” (the policy of giving ART for life to all HIV-positive pregnant women, whatever their stage) on retention in care and survival. In the context of a population based study of non-communicable disease in both the Karonga Demographic Surveillance site and our new urban study site in Area 25 of Lilongwe, we have conducted repeat HIV serosurveillance and associated reports of ART clinic attendance. In Karonga we collect linked ART clinic data from consenting attenders.