Higher Smoking Prevalence Among People with HIV in Almost All World Regions
March 24, 2021

Higher Smoking Prevalence Among People with HIV in Almost All World Regions

People living with HIV are approximately two-thirds more likely to be current tobacco smokers than HIV-negative individuals, according to the findings of a global meta-analysis published in AIDS. Elevated smoking prevalence was observed in both men and women with HIV and in all world regions except the Western Pacific.

“This systematic review and meta-analysis is the first to synthesize all studies that directly compare worldwide smoking prevalence in people with HIV and people without HIV. The review comprised a comprehensive, tailored search strategy, standardized data extraction, quality appraisal by multiple reviewers and a priori defined subgroup analyses,” write the authors. “We show that people living with HIV are more likely to be current tobacco smokers than HIV-seronegative individuals, a finding maintained among men and women and consistent in four of five World Health Organization regions.”

The findings underline the importance of integrating smoking cessation into routine HIV care.

Research has consistently shown that people living with HIV are more likely to be current or former tobacco smokers than people in the general population. The high prevalence of smoking is a major reason why, despite major advances in treatment and care, the overall life expectancy of people with HIV, even when in the context of successful antiretroviral therapy, still lags that of their HIV-negative peers.

However, most of the research comparing rates of tobacco smoking between people with and without HIV has been conducted in Europe and North America. This means that prevalence rates are unclear in other world regions. Moreover, whether smoking rates differ according to gender also requires clarification.

Dr. Peter Johnston and colleagues at the Universities of Sheffield and Nottingham sought to address these limitations in current knowledge by conducting a meta-analysis of published research. The inclusion criteria were strict. Only observational studies comparing current smoking rates according to HIV status were eligible. Research including people with HIV but focusing on smoking-related disease was excluded. Smoking methods included cigarettes, pipes, cigars and hookah. The research did not include e-cigarettes, chewing tobacco and snuff.

A total of 37 eligible studies were identified. They were conducted between 1990 and 2018 and included a little over 111,000 people with HIV and just under 11 million people without HIV. Of these, 38,336 people with HIV (18,241 male, 18,095 female) and 709,358 HIV-negative individuals (298,334 male, 411,024 female) were current smokers.

By World Health Organization region, nine of the studies were from Europe (Denmark, France, Spain and the UK), 17 from the Americas (Brazil, Canada, Haiti and the US), seven from Africa (Rwanda, South Africa, Tanzania, Uganda and Zimbabwe), two from the western Pacific (Australia and China) and just one from south-east Asia (India). An additional global study provided data on 28 low- and middle-income countries.

Pooling the results of all 37 studies showed that people living with HIV were approximately two-thirds more likely to be current tobacco smokers compared to people without HIV (OR = 1.64; 95% CI, 1.45-1.85).

HIV-positive men were 68% more likely to smoke than HIV-negative men (OR = 1.68; 95% CI, 1.44-1.95) and women with HIV had twice the odds of smoking compared to women without HIV (OR = 2.16; 95% CI, 1.77-2.63).
People with HIV were more likely to be current tobacco smokers than people without HIV in Africa (OR = 1.21; 95% CI, 1.06-1.37), the Americas (OR = 2.16; 95% CI, 1.83-2.55), Europe (OR = 2.32; 95% CI, 1.59-3.38) and south-east Asia (OR = 1.62; 95% CI, 1.38-1.91). The only WHO region where the difference in smoking odds failed to reach statistical significance was the western Pacific (OR = 1.80; 95% CI, 0.82-3.93).

Studies rated by the investigators as having a higher risk of bias were somewhat more likely to show that smoking was more common among people with HIV (OR = 2.06; 95% CI, 1.57-2.71) than studies with a lower risk of bias (OR = 1.54; 95% 1.35-1.76), but HIV-positive individuals were still more likely to be current smokers than people without HIV.

The study was not designed to explain why people with HIV were more likely to be current smokers. Nevertheless, the investigators offer a number of possible reasons based on the findings of earlier research. These include: a low quit rate; a high prevalence of depression; use of recreational drugs; social exclusion and lower educational attainment.

“Our results highlight the need for robust strategies to help worldwide populations of people living with HIV to quit smoking,” comment the researchers. “The development of effective strategies for smoking cessation is a clear priority, especially in low resource settings.”

Authored by Michael Carter


Johnston PI et al. Worldwide relative smoking prevalence among people living with and without HIV: a systematic review and meta-analysis. AIDS, published ahead of print, January 2021.
doi: 10.1097/QAD.0000000000002815