UC San Diego

Background: Cigarette smoking is thought to be a major driver of the global tuberculosis epidemic. However, limited evidence exists to support a causal relationship between smoking and Mycobacterium tuberculosis (Mtb) infection. Cigarette smoking and Mtb infection are highly prevalent among persons who inject drugs, and smoking cessation interventions might help reduce tuberculosis risk in this population. Objectives: This dissertation research sought to expand the current knowledge of the relationship between cigarette smoking and Mtb infection and to describe patterns of cigarette smoking and quit attempts among persons who inject drugs. Methods: Data from longitudinal studies of persons who inject drugs in Tijuana, Mexico, were used to conduct the following studies: 1) a longitudinal analysis of the association between self-reported number of cigarettes smoked per day and incident Mtb infection; 2) a cross-sectional study to describe the association between salivary cotinine levels and Mtb infection; and 3) a cross-sectional study to describe cigarette smoking patterns and quit attempts that lasted [greater than] 1 day during the past six months. QuantiFERON-TB Gold In-Tube (QFT) was used to determine Mtb infection. Results: For Study 1, Poisson regression analysis of 129 participants who were QFT negative at baseline found no evidence of an association between self- reported number of cigarettes smoked per day and QFT conversion after controlling for age, gender, education, and alcohol. For 208 participants in Study 2, using multivariable logbinomial regression models adjusted for sex and education, cotinine levels were not associated with QFT positivity when included as smoking categories, but were independently associated with QFT positivity when included as an ordinal variable (prevalence ratio = 1.09 per + 1 cotinine level; 95% confidence interval = 1.01 - 1.17). Study 3 found that 124 (20.6%) of the 601 current cigarette smokers had made a recent quit attempt. Conclusion: A dose-response relationship was observed between smoking intensity and Mtb infection when cigarette smoke exposure was measured using salivary cotinine but not when using self-report. One out of five participants reported recently making attempts to quit smoking. Cigarette smoking cessation interventions should be implemented for persons who inject drugs and other populations at high risk for tuberculosis.