The proportion of tuberculosis cases in Tanzania attributable to human immunodeficiency virus

Abstract

Background: Routine data obtained from the National Tuberculosis and Leprosy Programme (NTLP) of Tanzania have shown a constant increase in the notified number of tuberculosis (TB) cases since 1982. Possible causes include an improved reporting system, improvement in health services after the introduction of short course chemotherapy (SCC), and human immunodeficiency virus (HIV) infection. This paper examines to what extent the increased TB case detection rate can be attributed to HIV infection, by calculating the population attributable risk for various years.

Method: The prevalence of HIV infection was obtained from data of the National AIDS Control Programme and the relative risk of HIV for developing TB from a case-control study and the literature.

Results: Between 1985 and 1989 the increase was the highest among women aged 15-24 years and men aged 25-34 years; age groups in which HIV prevalence is highest. In the case-control study HIV prevalence among blood donors was 9.4% and among smear-positive pulmonary TB patients 51.6%, giving an odds ratio (OR) of 8.1 (95% confidence interval (CI): 4.4-16.3). For all TB cases the OR was 11.8. In a population with an HIV prevalence of 10%, about 40% of the smear-positive TB patients are attributable to HIV. The excess of TB cases in the entire country between 1982 and 1989 can be attributable to HIV infection. This has implications for TB control and socioeconomic consequences in the country.

PIP: The increase seen in the incidence of tuberculosis (TB) in many developing countries in the early 1980s was at first through to be the result of better case detection, but it soon became clear that HIV infections were influencing this increase. To determine the extent that HIV infection has increased TB case detection rates in Tanzania, data were analyzed from the National TB and Leprosy Programme, the National AIDS Control Programme, and a case-control study conducted for three months in 1990. Cases were all 128 newly registered cases of TB in three districts. Controls were 1558 blood donors in these districts. HIV prevalence among the cases was 51.6%, with no differences in sex, residence, or type of TB. HIV prevalence was highest among 25-34 year olds. HIV prevalence in controls was 9.4%, with no variation by age or sex. The odds ratio for association between HIV infection and new smear-positive TB, stratified by age, was 8.1. The age-stratified offs ratio for HIV infection and any type of TB was 11.8. The population attributable risk for 1990 was in the order of 30%, which means that, without HIV, the increase in TB cases seen after 1985 would not have occurred. It is essential to improve TB programs to minimize the looming increase in the annual risk of infection. Also, HIV control programs will have a large effect on TB control programs, and collaboration between the two should be encouraged.