       Document 2535
 DOCN  M94A2535
 TI    HIV incidence in a cohort of Burundian male workers.
 DT    9412
 AU    Tobi S; Buzingo T; Sokal D; Kamuragiye A; Rubayiza E; Alexandre Y;
       Family Health International, Burundi.
 SO    Int Conf AIDS. 1994 Aug 7-12;10(1):282 (abstract no. PC0048). Unique
       Identifier : AIDSLINE ICA10/94370040
 AB    OBJECTIVE: To study longitudinally the epidemiology of new HIV-1
       infection in a cohort of workers in the capital city, to provide
       information for policy makers about the scope and direction of the AIDS
       epidemic in Burundi. METHODS: Between 1990 and 1993, 2,430 workers from
       five companies in Bujumbura were enrolled into the cohort. New
       volunteers were encouraged to participate each year. Men who were HIV-1
       negative at the time of recruitment were retested yearly to determine
       the annual incidence rate of new infection. A KAP survey was
       administered at the time of testing to provide information about risk
       factors for infection. HIV testing was by Elisa with confirmation by
       Western Blot. RESULTS: More than 50% of the workers volunteered for the
       study. Among those who volunteered, Sixty-one percent of seronegative
       men were followed in 1991, 76% in 1992, and 77% in 1993. Prevalence
       among new volunteers was 14.3% in 1990, 13.8% in 1991, 13.0% in 1992,
       and 10.6% in 1993 (chi 2 test for trend = 2.19, p = .14). The 1990/1991
       incidence rate was 2.1/100 person-years. In 1991/1992 it was 2.1/100
       person-years, and in 1992/1993 it was 2.5/100 person-years. Risk factors
       for seroconversion included occurrence of urethritis (IRR = 2.8, 95% CI
       = 1.2-6.4), reported occurrence of genital ulcer (IRR = 4.2, 95% CI
       1.9-9.5), and having more than one sexual partner (IRR = 1.93, 95% CI =
       1.1-3.3) during the risk period (i.e., since last seronegative test).
       CONCLUSIONS: To the extent that newly recruited workers represent the
       general populations of workers, there is an apparent leveling off of HIV
       prevalence. Despite this observed trend, the incidence rate of HIV-1
       continues to be high. Although observed prevalence may result form
       participation bias, this finding reinforces the need to exercise caution
       in interpretation of prevalence data.
 DE    Burundi/EPIDEMIOLOGY  Cohort Studies  Human  HIV
       Infections/*EPIDEMIOLOGY  *HIV-1  Knowledge, Attitudes, Practice  Male
       Prevalence  Risk Factors  MEETING ABSTRACT

       SOURCE: National Library of Medicine.  NOTICE: This material may be
       protected by Copyright Law (Title 17, U.S.Code).

