Variation in incidences of tuberculosis in subgroups of South African gold miners

Abstract

Objectives: To determine risk factors and to identify groups at high risk of tuberculosis (TB) among South African gold miners.

Methods: A retrospective cohort analysis was carried out with hospital and personnel databases on a random sample of the workforce of Freegold Mines, served by Ernest Oppenheimer Hospital, Welkom, South Africa. The outcome measure of interest was a reported first episode of TB while employed at Freegold. A person-years analysis was carried out to estimate incidence ratios between different categories of age group, period, cumulative service, mining occupation, and silicosis status. A separate analysis was carried out on a subgroup of men (non-random) whose HIV status was known.

Results: The main finding of the study was that TB was more strongly associated with age than expected with a rate ratio of reported incidence of 21 for the oldest age group compared with the youngest, after adjustment for period, cumulative service, occupation, and silicosis status. There was also a significant association between TB and occupations such as drilling with a rate ratio of 2.3 compared with low dust surface and maintenance workers, after adjustment for age, period, cumulative service, and silicosis. Analysis of the HIV tested subgroup showed that these results are unlikely to be the result of confounding due to HIV infection. Prevalence of HIV in this group has been rising sharply since 1991, but the increase was slowing towards the end of the study period.

Conclusion: A profile of mineworkers who are at high risk of TB can be defined by age, mining occupation, silicosis status, and HIV infection with predicted rates of 100/1000 person-years. TB screening programmes should take special cognisance of high risk groups of gold miners.

PIP: This study determines risk factors for tuberculosis (TB) and identifies groups among South African gold miners at high risk for contracting TB. A retrospective cohort analysis was conducted with hospital and personnel databases on a randomly selected sample of the labor force of Freegold Mines. A person-years analysis was carried out to examine incidence ratios between different age groups, period, cumulative service, mining occupation, and silicosis status, with a separate analysis on a subgroup of men with known HIV status. Findings reveal that TB was associated with age, with a rate ratio of 21 for the oldest age group compared with the youngest, after adjustment for period, cumulative service, occupation, and silicosis status. There was also a significant relationship between TB and occupation, such as drilling, with a rate ratio of 2.3, compared with workers of low dust surface maintenance, after adjustment for age, period, cumulative service, and silicosis. Analysis of the HIV-tested subjects indicated that these findings are unlikely to be the result of confounding due to HIV infection. Prevalence of HIV infection in this group has been increasing sharply since 1991, but the increase was slowing towards the end of the study period. In conclusion, age, mining occupation, silicosis status, and HIV infection with predicted rates of 100/1000 person-years can define a profile of miners who are at high risk of developing TB. TB screening programs should therefore take special cognizance of high-risk groups of gold miners.