Women who smoke like men die like men who smoke: findings from two Australian cohort studies Konrad Jamrozik, 1 Deirdre McLaughlin, 2 Kieran McCaul, 3 Osvaldo P Almeida, 3,4,5 Kore Yiee Wong, 2 Dimitrios Vagenas, 6 Annette Dobson 2 ABSTRACT Background There is controversy about whether men and women with similar smoking histories have similar incidence and mortality rates from smoking related diseases. Objective To compare mortality rates from all causes of death and various smoking related causes for men and women smokers categorised by numbers of cigarettes smoked and for ex-smokers by time since quitting. Methods This was a 10-year follow-up study with deaths identified from the National Death Index. The setting was two cohort studies in Australia established in 1996. Participants were: men (n¼12 154) and women (n¼11 707) aged (mean (SD)) 72.1 (4.4) and 72.5 (1.5) years, respectively, when recruited. The main outcome measure was HRs for men and women separately and RRs calculated from combined analyses using proportional hazards models (for deaths from all causes) and competing risks proportional hazards models (for specific causes). Results HRs for deaths from all causes for men (n¼3549 deaths) and women (n¼2665 deaths) among smokers increased with amount smoked and for ex-smokers decreased with time since quitting. Similar effects were found for various groups of smoking-related conditions with the dose-response effects largest for lung cancer and chronic obstructive pulmonary disease. The ratios of HRs for women relative to men were near unity and the 95% CIs included unity for almost all comparisons. Conclusions The data provide strong evidence that men and women with similar patterns of smoking experience similar rates of death due to smoking. INTRODUCTION Smoking has been linked to numerous respiratory, neoplastic and cardiovascular diseases such as chronic obstructive pulmonary disease, lung cancer, asthma, coronary heart disease and stroke. There are dose-response relationships between these diseases and cigarette consumption. For example, the risk of lung cancer increases with increasing number of cigarettes smoked daily as well as with increased duration of smoking. 1e3 There is conflicting evidence on gender differences in the magnitude of risk associated with smoking. It has been argued that risk for women may be different from men due to biological factors such as smaller lung capacity and body size or hormonal effects, or behavioural factors such as smoking different types of cigarettes, less inhalation and differences in health service use. 4e6 Such differences might affect disease incidence, case death or both. 7 For lung cancer, case control studies suggested women have higher incidence rates than men 8 but prospective studies have shown no differences. 9 10 Similarly, there are reports of higher incidence rates of cardiovascular disease associated with smoking among women than men 5 11 and of similar mortality. 12 One possible way of clarifying if gender mediates the effects of smoking on health outcomes is to determine if the mortality hazard is different for men and women. For example, if women were more susceptible to the carcinogenic effects of smoking than men, one would expect to find a higher mortality hazard attributable to lung cancer among women for the same amount smoked. We designed the present study to deter- mine if the mortality hazard from various smoking- related causes by smoking pattern (ie, amount of cigarettes smoked per day, and time since quit smoking) is different for women compared with men. METHODS Study design This paper uses smoking data obtained from two population-based longitudinal studies that began in 1996. These data were linked to national mortality data up to 10 years later. The paper conforms to the STROBE (‘STrengthening the Reporting of OBser- vational studies in Epidemiology ’) guidelines for observational studies. Setting and participants Data were derived from the 1921e1926 birth cohort of the Australian Longitudinal Study on Women’s Health (ALSWH) and the Health in Men Study (HIMS). The detailed methods for both studies have been described elsewhere previously. 13 14 Briefly, the ALSWH is a survey of the health and well-being of three cohorts of women who were aged 18e23 years (1973e1978 birth cohort), 45e50 years (1946e1951 birth cohort) and 70e75 years (1921e1926 birth cohort) when recruited in 1996. The project uses mailed questionnaires to collect self-report data on health and related variables every 3 years. Women were selected from the Australian national health insurance database (Medicare), which includes all citizens and perma- nent residents. Stratified random sampling was used with intentional oversampling of women from rural and remote areas. In the 1921e1926 birth cohort, 39 000 women were initially invited to participate; of these 1100 were not contactable and 2366 were ineligible. Of the remaining women (35 534), 12 614 responded. The ethics committees See Editorial, p 251 1 School of Population Health and Clinical Practice, University of Adelaide, Adelaide, South Australia, Australia 2 The University of Queensland, School of Population Health, Brisbane, Queensland, Australia 3 Western Australian Centre for Health & Ageing, Centre for Medical Research, University of Western Australia, Crawley, Western Australia, Australia 4 School of Psychiatry & Clinical Neurosciences, University of Western Australia, Crawley, Western Australia, Australia 5 Department of Psychiatry, Royal Perth Hospital, Perth, Western Australia, Australia 6 Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Queensland, Australia Correspondence to Annette Dobson, The University of Queensland, School of Population Health, Herston Road, Herston, Queensland 4006, Australia; a.dobson@sph.uq.edu.au Received 25 July 2010 Accepted 3 March 2011 Published Online First 11 April 2011 258 Tobacco Control 2011;20:258e265. doi:10.1136/tc.2010.039172 Research paper group.bmj.com on March 6, 2013 - Published by tobaccocontrol.bmj.com Downloaded from