Alcohol and problem drinking as risk factors for tuberculosis

Abstract

[Background] Tuberculosis is a major public health concern for South Africa which has one of the highest recorded incidence rates in the world. Previous research [1998 South African Demographic and Health Survey (SADHS)] reported a crude association between alcohol use and tuberculosis. This study aimed to examine evidence for a relationship, and the size thereof, between alcohol consumption and previous tuberculosis in the 2003 SADHS as a means of informing tuberculosis prevention. [ Methods ] This study was a secondary analysis of cross sectional data collected as part of the 2003 SADHS. Tuberculosis lifetime risk was derived from respondent reports of past tuberculosis episodes based on being informed by a healthcare worker. Alcohol consumption, problem drinking as well as selected explanatory variables were generated from similar questions from the adult questionnaire of the SADHS. The CAGE questionnaire was used to measure symptoms of alcohol problems. Logistic regression was used to model the relationship between past tuberculosis and both alcohol consumption and CAGE. [ Results ] Current and previous alcohol consumption were found to be associated with an increase in odds of tuberculosis in both men and women, with odds ratios ranging iii from 1.1 (95% CI 0.9 - 2.5) to 2.8 (95% CI 1.4 - 5.7) after adjusting for potential confounding effects of socioeconomic factors, smoking, nutritional status and age. Having a CAGE score of either 1 to 2 or 3 to 4 was associated with a doubling [OR 2.2 (95% CI 1.0 - 4.8) and quadrupling [OR 4.4 (95% CI 1.4 - 13.4)] in the odds of tuberculosis respectively. [ Discussion ] and conclusion Behavioural and biological mechanisms of effect of alcohol on tuberculosis may explain the findings. Impairment of the immune system, both acute and long term, has been suggested as the mechanism of increased susceptibility to tuberculosis. On the other hand, high risk living conditions and behaviour associated with problem drinking provide potential for increased exposure and susceptibility to tuberculosis infection. The study was able to control for several potentially confounding socioeconomic predictor variables although not HIV infection. The results complement a body of research that has documented the adverse effects of alcohol consumption on health in general and tuberculosis specifically. The findings thus provide more evidence for public health practitioners to tackle the problem of tuberculosis via specific efforts to control alcohol use and abuse, in addition to other methods of tuberculosis control.