Abstract

In Botswana, tuberculosis is the single most important cause of patient disease and death. According to WHO, Botswana has the highest tuberculosis notification rate in the world. The Botswana Tuberculosis Control Programme has been underway for 12 years without a satisfactory study of the trend of disease. This project will support a tuberculin survey of the next primary school intake of students to determine: the tuberculin positivity rate; the BCG scar rate; the proportion of tuberculin positives due to tuberculosis; and calculate the annual risk of tuberculosis infection. The project will provide data to assess the Tuberculosis Control Programme and a statistical basis for future routine surveys.

Post-Project Summary

The study was carried out in 1989. The universe under study consisted of all first-grade children in all primary schools in Botswana , for a total of 44 833. Using stratified, multistage cluster technique, the researchers selected a sample of 3 692 schoolchildren - one class each (approximately 30 students) from 121 schools in three different areas (strata), one of average, low and high tuberculosis prevalence. Tuberculin tests were carried out by two experienced Mantoux testers, the presence or absence of a BCG scar was noted and the children were re-vaccinated with BCG. The Mantoux reaction was measured 72 hours after the test and the researchers examined the distribution of the reactions to determine the modes of the different reactions, i.e. small, intermediate and strong. A second tuberculin test was administered 3 415 of the children 8-12 weeks later and the results compared.

It was assumed that all non-specific sensitivity would produce reactions of intermediate size, and that the right hand tail of the distribution would represent infected persons. It was also assumed that the infected cases would exhibit about the same reaction both times, whereas the sensitivity of the non-specific reactors would be compounded by the BCG vaccination, giving a stronger reaction the first time around.

The results showed a very small number of infected individuals (<16 mm reaction): only four in all, two with and two without a BCG scar. This level of infection (0.1%) was much lower than that observed in 1956 (4.45%). It also contrasted favorably with the high rate of sputum positive cases recorded in the general population in 1987: 140 per 100 000 population. Of these cases, 48% were found in the age group 40 and over, which represents only 18% of the population. This suggests that the high incidence of sputum positive cases reflects the high rate of infection that characterized the childhood of this cohort. The decline of tuberculosis in the younger age group, which represents 35% of the population, was attributed to improvements in the standard of living, the introduction of short-course chemotherapy in the mid-1980s and BCG vaccination. While the trend is encouraging, concern remains that rapid spread of HIV infection may slow down or reverse the decline in tuberculosis.

Recipient Institution(s)