Tobacco use among people living with HIV: analysis of data from Demographic and Health Surveys from 28 low-income and middle-income countries.

BACKGROUND: Tobacco use among people living with HIV results in excess morbidity and mortality. However, very little is known about the extent of tobacco use among people living with HIV in low-income and middle-income countries (LMICs). We assessed the prevalence of tobacco use among people living with HIV in LMICs.

METHODS: We used Demographic and Health Survey data collected between 2003 and 2014 from 28 LMICs where both tobacco use and HIV test data were made publicly available. We estimated the country-specific, regional, and overall prevalence of current tobacco use (smoked, smokeless, and any tobacco use) among 6729 HIV-positive men from 27 LMICs (aged 15-59 years) and 11 495 HIV-positive women from 28 LMICs (aged 15-49 years), and compared them with those in 193 763 HIV-negative men and 222 808 HIV-negative women, respectively. We estimated prevalence separately for males and females as a proportion, and the analysis accounted for sampling weights, clustering, and stratification in the sampling design. We computed pooled regional and overall prevalence estimates through meta-analysis with the application of a random-effects model. We computed country, regional, and overall relative prevalence ratios for tobacco smoking, smokeless tobacco use, and any tobacco use separately for males and females to study differences in prevalence rates between HIV-positive and HIV-negative individuals.

FINDINGS: The overall prevalence among HIV-positive men was 24·4% (95% CI 21·1-27·8) for tobacco smoking, 3·4% (1·8-5·6) for smokeless tobacco use, and 27·1% (22·8-31·7) for any tobacco use. We found a higher prevalence in HIV-positive men of any tobacco use (risk ratio [RR] 1·41 [95% CI 1·26-1·57]) and tobacco smoking (1·46 [1·30-1·65]) than in HIV-negative men (both p<0·0001). The difference in smokeless tobacco use prevalence between HIV-positive and HIV-negative men was not significant (1·26 [1·00-1·58]; p=0·050). The overall prevalence among HIV-positive women was 1·3% (95% CI 0·8-1·9) for tobacco smoking, 2·1% (1·1-3·4) for smokeless tobacco use, and 3·6% (95% CI 2·3-5·2) for any tobacco use. We found a higher prevalence in HIV-positive women of any tobacco use (RR 1·36 [95% CI 1·10-1·69]; p=0·0050), tobacco smoking (1·90 [1·38-2·62]; p<0·0001), and smokeless tobacco use (1·32 [1·03-1·69]; p=0·030) than in HIV-negative women.

INTERPRETATION: The high prevalence of tobacco use in people living with HIV in LMICs mandates targeted policy, practice, and research action to promote tobacco cessation and to improve the health outcomes in this population.

FUNDING: South African Medical Research Council and the UK Medical Research Council.