An investigation of the excess of lung cancer in young Tasmanian women aged 25-44 years during the period 1983-92

This thesis reports an investigation into the causes of higher female-than-male rates of lung cancer among 25-44 year olds in Tasmania, Australia, during 1983-92. A rationale for the study is provided in Chapter 1. In particular, the focus on measures of tobacco smoking as the principal study factor is justified, and decisions made in relation to other putative study factors are explained. Australian lung cancer mortality rates are examined in Chapter 2. Those of 20-44 year old women ceased rising in 1986, before overtaking male rates, despite higher proportions of smokers and lower mean age of commencement of smoking in subsequent cohorts of women born after the 1940s. To investigate this, lung cancer incidence in Australia during 1982-95 is analysed in Chapter 3. The birth cohort trends in incidence of squamous cell carcinoma and small cell lung carcinoma are consistent with a reduction in risk following the introduction of filter-tip cigarettes. Lung cancer rates of young women are also relatively high in New Zealand. The hypothesis tested in Chapter 4 is that this is an outcome of cool climate living. With the exception of the higher-than-predicted rates of young Tasmanian women, the regional variation in lung cancer incidence of 20-44 year olds in Australia and New Zealand was well explained by differences in smoking prevalence. A case-control study was undertaken to explain the high lung cancer rates of young Tasmanian women. Chapter 5 reports on reliability studies that established that agreement was high and random error was low for measurements of lifetime exposure to cigarette smoking. In a study of gender differences in smoking exposures in the source population, Chapter 6 reports that the female controls had lower exposures to tobacco smoking than the male controls. The differences were most pronounced for subjects born in the early 1940s, who constitute the bulk of the age-matched controls. The results of the case-control analysis of tobacco smoking exposures are presented in Chapter 6. In contrast to the previous excess of female cases, when the members of the cohort were younger, 62% (99/160) of cases during 1994-97 were men. The attributable proportions for tobacco smoking were 86% (men) and 87% (women). Two a priori hypotheses — that the women (1) commenced smoking from a younger age, or (2) had higher smoking-related risk of lung cancer by virtue of their gender — were rejected as explanations of the previous excess of female cases in the cohort. Further case-control analyses in Chapter 8 showed that the explanation was not greater female exposures to ETS, occupational carcinogens or diets low in p-carotene and other carotenoids. Relative risks of lung cancer were found to be very high for long-term users of the oral contraceptive pill who were long-term smokers, but there\ud were too few women in the high risk category for that to be the explanation. The possibility that the excess of female cases of lung cancer among young adults in Tasmania was a chance occurrence is raised in the summary presented in Chapter 9