Smoking and Tuberculosis Risk: High Attributable Risk in a Low-Incidence Environment
Justin T Denholm1,2, James M Trauer1,4 and Emma S McBryde1,3,4
1Victorian Tuberculosis Program, Doherty Institute for Infection and Immunity, Melbourne, Victoria; 2Department of Microbiology and Immunology, University of Melbourne, Parkville, Victoria; 3Victorian Infectious Diseases Service, Royal Melbourne Hospital, Parkville, Victoria; 4Burnet Institute, Prahran, Victoria
Abstract: Objectives: To evaluate the potential impact of smoking cessation strategies on TB incidence in Victoria, Australia.
Methods: A previously validated model of TB in migrants was adapted to concentrate on the attributable TB risk arising from smoking. This model included realistic estimates of population smoking rates, and incorporation of risk factors for TB disease such as smoking, HIV and diabetes. A baseline scenario was compared with an alternative scenario in which a hypothetical public health strategy eliminating smoking was introduced.
Outcome measures: The primary outcome was the number of TB cases in the Australian state of Victoria in the year 2050. Secondary outcomes considered were total case burden during the period considered and the population attributable fraction of TB cases related to smoking.
Results: Under the baseline scenario, a median of 318 cases of TB occurred in the target year of 2050, corresponding to an estimated incidence of 3.5 cases/100,000 population. The alternative scenario, where no smoking occurred, found 272 cases (2.95 cases/100,000 population) occurring in 2050; a 14.5% reduction in TB cases. Over the entire period considered, 1650 fewer cases of TB occurred in the alternative scenario, suggesting a population attributable fraction of 11.2% of TB cases from smoking in the Australian context.
Conclusions: Smoking cessation strategies may have disproportionate benefits in TB incidence reduction. Such benefits should be considered in additional to other well-recognised population health effects from smoking reduction.
Keywords: Tuberculosis, smoking, public health, mathematical modeling.