Relationship of tobacco smoking to cause-specific mortality: contemporary estimates from Australia
Grace Joshy, Kay S, Katherine Anne Thurber, Sam Egger, Marianne Weber, Peter Sarich, Jennifer Welsh, Rosemary Korda, Amelia Yazidjoglou, Mai Nguyen, Ellie Paige, Michelle Gourley, Karen Canfell, Emily Banks
Abstract
BACKGROUND: Tobacco industry activities and reduced smoking prevalence can foster under-appreciation of risks and under-investment in tobacco control. Reliable evidence on contemporary smoking impacts, including cause-specific mortality and attributable deaths, remains critical. METHODS: Prospective study of 178,169 cancer- and cardiovascular-disease-free individuals aged ≥ 45 years joining the 45 and Up Study in 2005-2009, with linked questionnaire, hospitalisation, cancer registry and death data to November 2017. Cause-specific mortality hazard ratios (HR) by smoking status, intensity and recency were estimated, adjusted for potential confounding factors. Population attributable fractions were estimated. RESULTS: There were 13,608 deaths during 9.3 years median follow-up (1.68 M person-years); at baseline, 7.9% of participants currently and 33.6% formerly smoked. Mortality was elevated with current versus never smoking for virtually all causes, including chronic lung disease (HR = 36.32, 95%CI = 26.18-50.40), lung cancer (17.85, 14.38-22.17) and oro-pharyngeal cancers (7.86, 4.11-15.02); lower respiratory infection, peripheral vascular disease, oesophageal cancer, liver cancer and cancer of unknown primary (risk 3-5 times as high); and coronary heart disease, cerebrovascular disease and cancers of urinary tract, pancreas, kidney, stomach and prostate (risk at least two-fold); former versus never-smoking demonstrated similar patterns with attenuated risks. Mortality increased with smoking intensity, remaining appreciable for 1-14 cigarettes/day (e.g. lung cancer HR = 13.00, 95%CI = 9.50-17.80). Excess smoking-related mortality was largely avoided with cessation aged < 45 years. In 2019, 24,285 deaths (one-in-every-six deaths, 15.3%), among Australians aged ≥ 45 years, were attributable to tobacco smoking. CONCLUSIONS: Smoking continues to cause a substantial proportion of deaths in low-prevalence settings, including Australia, highlighting the importance of accelerated tobacco control.