HIV prevalence and incidence in Uganda on the way up

Michael Carter
Published: 26 August 2008

HIV prevalence and incidence in rural Uganda appear to be increasing, researchers report in a study published in the August 20th edition of AIDS. Coinciding with these increases, the investigators observed changes in HIV sexual risk behaviours in certain groups. The study was conducted in villages in rural Uganda and the trends it revealed mirror other evidence from Uganda pointing to increases in HIV prevalence and incidence.

Uganda has long been regarded as an HIV success story. The Ugandan government initiated a robust response to the epidemic which was praised as a model response to the HIV epidemic bringing about a substantial fall in HIV prevalence.

But in recent years data from antenatal clinics and voluntary testing and counselling facilities have suggested a levelling off of HIV prevalence in the country.

Investigators therefore looked at long-term trends in HIV prevalence and incidence (new infections) in an ongoing study involving individuals living in villages in rural south-west Uganda. The investigators also examined trends in sexual behaviour to see if any changes in risk behaviour could be linked with changes in HIV incidence.

The researchers analysed data collected between 1989 and 2005. From 1989 to 1999 individuals in 15 villages were included in the study, with on average 3722 adults a year having an HIV test. A further ten villages were added to the study in 1999, increasing the average annual number of HIV tests to 6422.

HIV prevalence declined from a peak of 8.5% in 1990/91 to a low of 6.2% in 1999/2000 before increasing again to 7.7% in 2004/05. The fall in HIV prevalence observed during the 1990s was statistically significant (p < 0.01).

Estimates of HIV incidence were based upon 14,449 individuals who had a first negative HIV test before October 2007 and two or more further HIV tests. Like HIV prevalence, there was a fall in new HIV infections for much of the 1990s, with incidence falling from 7.5 per 1000 person years in 1990 to 4.1 per 1000 person years in 1998. After 1999, however, there was a slight increase in incidence until 2004 when it reached 4.1 per 1000 person years. But in 2005 there was a sudden drop in incidence to 2.5 per 1000 person years, the lowest figure ever recorded. The investigators think, however, that this drop in 2005 is likely to have been a rogue finding as preliminary data for 2006 once again suggested an increase in new HIV infections.

Patterns in HIV incidence according to age emerged. Generally incidence increased until individuals reached their late 20s, before declining. But in the period between 2000 and 2004 the investigators noted a second peak of incidence in men aged 40-44 (9.5 per 1000 person years, compared to 7.3 per 1000 person years in 1995-99). The investigators also observed a second slight peak in incidence for women aged 30-34.

Next the investigators looked at reported sexual behaviour to see if this could explain the changes in prevalence and incidence they had seen.

The proportion of young people reporting one or more casual partner increased between 1997 and 2004 before dipping in 2005. It was only amongst 35-44 year olds that the increase in casual partners was statistically significant (p = 0.003). This coincided with the overall stabilisation of HIV incidence in the general population, and with the increase in incidence seen in older men.

The investigators also observed a reduction in the number of individuals in their 20s reporting condom use with casual partners. Amongst 20-24 year olds, this dropped from 74% in 1997 to 51% in 2005. However, condom use amongst older age groups appeared to be increasing, with the proportion of those aged 45 or over saying they used condoms with casual partners increasing from 10% in 1997 to 24% in 2005.

Although the proportion of pregnancies amongst never-married women aged 15-19 fell rapidly from 1990-98, this trend stopped after 1998.

Starting in 1997, the investigators asked individuals aged 16-19 the age at which they first had sex. There was a steady increase until 2005. The investigators also asked 15-18 year olds if they had ever had sex. There was a decline in the proportion of young men answering yes between 1993 and 1997, but an increase between 1998 and 2000. The investigators note that this increase coincided with the period when HIV incidence in the general cohort was increasing.

“HIV prevalence in Uganda is no longer falling and is beginning to rise at least in some parts of the population”, write the investigators, a finding consistent with other reports from Uganda. “Factors influencing the recent trends of the epidemic are not yet clear, but there are indications that the observed changes in trend may be partly explained by increased sexual risk behaviour.”

The investigators conclude, “Uganda has been highly successful in controlling the HIV epidemic in…the past. To solidify this success, the ongoing efforts in HIV prevention need to be re-strengthened, using all strategies known to reduce HIV transmission.”

Reference

Shafer L.A. et al. HIV prevalence and incidence no longer falling in southwest Uganda: evidence from a rural population cohort 1989-2005. AIDS 22: 1641-1649, 2008.