Incidence of lung cancer among patients with tuberculosis: a nationwide cohort study in Lithuania.

Abstract

Although studies have found a relationship between tuberculosis (TB) and lung cancer, results for the long-term risk and the role of confounding factors remain inconclusive. To examine the risk of lung cancer in a Lithuanian cohort of 21 986 TB patients. During follow-up (1998-2012), 477 patients developed lung cancer. Standardised incidence ratios (SIRs) and 95% confidence intervals (95%CIs) were calculated to compare the incidence of lung cancer among cohort participants with the general population. Cox regression was used to estimate the effect of risk factors on lung cancer risk among TB patients. Lung cancer SIRs were 3.55 (95%CI 3.24-3.89) among all cohort participants, 4.48 (95%CI 4.04-4.96) among smokers and 1.93 (95%CI 1.56-2.36) among non-smokers. SIRs declined substantially >3 years after TB diagnosis in non-smokers, although they remained significantly elevated for ⩾10 years among smokers. In the multivariable analysis, smoking, male sex and older age substantially increased the risk of lung cancer. A statistically significant association with education, employment, site of TB and alcohol consumption did not persist in the lag-time analysis. Most of the observed excess risk of lung cancer in the TB cohort possibly reflects the confounding effects of reverse causation, closer medical surveillance and shared risk factors, particularly smoking.