Prevalence of HIV-1 infection in rural, semi-urban and urban villages in southwest Tanzania: estimates from a blood-donor study

AIDS. 1994 Jul;8(7):971-6. doi: 10.1097/00002030-199407000-00015.

Abstract

Objectives: To measure the prevalence of HIV-1 infection in different subgroups of blood donors and to identify groups at high risk of acquiring HIV-1.

Methods: Between March 1988 and April 1991 all blood donors at Ilembula Lutheran Hospital, Tanzania were asked about their age, marital status, home village and occupation, and tested for the presence of HIV antibodies using a first generation, whole virus lysate enzyme-linked immunosorbent assay (ELISA). Some negative (n = 265) and nearly all positive samples (439 out of 485) were subjected to confirmatory testing including recombinant and peptide-based ELISA and Western blot assays.

Results: A total of 3474 male and 1287 female blood donors were studied. The overall HIV-1 prevalences for men and women were 6.6 and 7.0%, respectively, with a higher prevalence in urban villages (13.6 and 15.0%, respectively), an intermediate prevalence in semi-urban villages (7.2 and 7.9%, respectively), and a lower prevalence in rural villages (3.7 and 3.0%, respectively). HIV-1 infection occurred mostly in men aged 20-44 and women aged 15-34 years. Urban donors, but not semi-urban and rural donors from the highlands, had a higher HIV-1 prevalence (21.4%) than the corresponding group from the lowlands (10.2%). Apart from area of residence, HIV-1 infection was found to be associated with occupation and marital status. There was an increase in HIV-1 prevalence, although not statistically significant, during the period studied. None of the blood donors were positive for HIV-2.

Conclusions: Male and female donors from urban and semi-urban villages, non-farmers from urban villages, and unmarried donors were identified as high-risk groups, which is consistent with more extensive risk behaviour in urban communities. In addition to using an HIV test with high sensitivity, the importance of pre-donation selection of blood donors is stressed.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • AIDS Serodiagnosis
  • Adolescent
  • Adult
  • Blood Donors / statistics & numerical data*
  • Female
  • HIV Seroprevalence*
  • HIV-1*
  • Humans
  • Male
  • Middle Aged
  • Pregnancy
  • Rural Population / statistics & numerical data
  • Sensitivity and Specificity
  • Suburban Population / statistics & numerical data
  • Tanzania / epidemiology
  • Urban Population / statistics & numerical data