AEGiS-13IAC: A new method of estimating HIV prevalence in men using antenatal clinic sentinel surveillance.

13th International AIDS Conference

Durban, South Africa - July 9-July 14, 2000

A new method of estimating HIV prevalence in men using antenatal clinic sentinel surveillance.

Int Conf AIDS 2000 Jul 9-14; 13:(abstract no. ThOrC669)

Glynn JR, Buve A, Carael M, Musonda R, Kahindo M, Macauley I, Hawken M, Tembo F, Zekeng L
J.R. Glynn, London School of Hygiene and Tropical Medicine, Keppel St, London WC1E 7HT, United Kingdom, Tel.: +44 207 927 2423, Fax: +44 207 436 4230, E-mail: j.glynn@lshtm.ac.uk

BACKGROUND: Much of our knowledge of HIV prevalence relies on surveillance in women attending antenatal clinics (ANC). Much less is known about HIV prevalence in men. We have evaluated the accuracy of a new method of estimating male prevalence, by asking pregnant women the age of the father of the child.

METHODS: HIV prevalence in the general population was available from a random sample of 1500-2000 adults in each of Yaounde, Cameroon, Kisumu, Kenya, and Ndola, Zambia as part of the Multisite Study on the Heterogeneity of HIV Epidemics in African Cities. Enhanced sentinel surveillance was conducted in ANCs in each of the three cities. Male prevalence was estimated using the age of the father of the child, directly standardised to the age distribution of men seen in the population survey, using 5 year age groups.

RESULTS: About 1500 women were interviewed in the ANCs in Yaounde and Kisumu and 1000 in Ndola. All but 6% in Yaounde and less than 2% in the other sites gave an age for the father of the child. Few women reported fathers aged less than 20 years so the estimates for this group are unreliable. The HIV prevalence in the ANCs was 5.5% in Yaounde, 30.6% in Kisumu and 27.3% in Ndola. The estimated prevalence in men was 3.8% (95% CI 2.9-4.7) in Yaounde, 25.7% (23.3-28.1) in Kisumu and 24.4% (20.4-26.8) in Ndola. The measured prevalence in men in the population was 4.1% (2.9-5.6) in Yaounde, 19.8% (16.7-23.1) in Kisumu, and 23.2% (20.0-26.8) in Ndola

CONCLUSIONS: In all sites the estimate based on father's age was closer to the true male prevalence than the crude result from the surveillance, and in two sites it was very close to the true prevalence. This method is only likely to give a good estimate of male HIV prevalence if concordance of HIV status between partners is high or if the proportion of discordant partners is similar for HIV positive men and women. The method should be tested in other sites.

Keywords: AEGIS, Prevalence, HIV Infections, Sentinel Surveillance, HIV Seropositivity, Interviews, Cameroon, Zambia, Kenya, Human, Male, Female, Adult, Child, Pregnancy, epidemiology, utilization, methods
000709
ThOrC669