Decreasing HIV-1 seroprevalence in young adults in a rural Ugandan cohort

Abstract

Objective: To assess the trend in HIV-1 seroprevalence in an adult population in Uganda.

Design: An observational cohort study with four year follow up.

Setting: A cluster of 15 villages in rural Uganda.

Subjects: All residents of the 15 villages--about 10,000 people.

Main outcome measure: Prevalence of HIV-1 infection as assessed by enzyme immunoassay.

Results: During the five year period the overall standardised seroprevalence of HIV-1 showed little change; 8.2% in 1990, 7.6% in 1994. Among males aged 13-24 years the prevalence decreased from 3.4% to 1.0% (P for trend < 0.001); among females of the same age the corresponding values were 9.9% and 7.3%. The decrease was greatest in males aged 20-24 years and females aged 13-19 years.

Conclusion: This is the first report of a decline in HIV-1 prevalence among young adults in a general population in sub-Saharan Africa with high overall HIV-1 prevalence. It is too early to conclude that the epidemic in this population is in decline, but the results of this study should be reason for some cautious optimism and encourage the vigorous pursuit of AIDS control measures.

PIP: An observational cohort study with 4-year follow-up measured the prevalence of HIV-1 infection as assessed by enzyme immunoassay to evaluate the trend in HIV-1 seroprevalence in an adult population in Uganda. The area of study was a cluster of 15 neighboring villages containing about 1/3 of the population of a subcounty of Masaka district in south west Uganda. During 1990-4 the cohort was surveyed annually by using procedures similar to those at baseline. In the 15 study villages 4167 adults (13 years or older) were enrolled during the initial survey and had an unambiguous serostatus. Those remaining in the study area, leaving the study area, or joining from outside the area had 5.5%, 11.5%, and 16.3% progressively higher (standardized) rates of HIV-1 infection. After 4 years the total adult population was 5649; 88% of these had had blood samples taken at some time and had an unambiguous serostatus. During the 5-year period the overall age and sex standardized seroprevalence showed little change from 8.2% at round 1 to 7.6% at round 5. There was, however, a considerable decrease in prevalence in males 13-24 years old (from 3.45 to 1.0%; p 0.001) and a suggestion of a corresponding decrease in females (from 9.9% to 7.3%; p = 0.08). In the age group 13-19, seroprevalence in males was very low at both round 1 (0.2%) and round 5 (0.4%); in young adult females the rate fell from 4.5% (27/601) to 2.4% (13/531; p = 0.09). In men 20-24 years old rates declined from 11.8% (28/237) to 2.7% (5/187; p 0.001). During the period 1989 to mid-1994 the incidence of HIV-1 infection among adults in the cohort remained at about 7/1000 person years. There was a suggestion of decrease in incidence among males 13-24 years old from 6.4/1000 person years (7/1094) to 2.6/1000 (4/1530; p = 0.14); however, among females of the same age the corresponding rates were 8.2/1000 (9/1100) and 9.7/1000 (14/1442; p = 0.7). This is the first report of a decline in HIV-1 prevalence among young adults in a general population in Sub-Saharan Africa with high overall HIV-1 prevalence.