© 1981 Oxford University Press
research-article |
Future Incidence of Lung Cancer: Forecasts Based on Hypothetical Changes in the Smoking Habits of Males
*Finnish Cancer Registry, Liisankatu 21 B, SF-00170 Helsinki 17, Finland.
Hakulinen T (Finnish Cancer Registry, Liisankatu 21 B, SF—00170 Helsinki 17, Finland) and Pukkala E. Future incidence of lung cancer: forecasts based on hypothetical changes in the smoking habits of males. International Journal of Epidemiology 1981, 10: 233–240.
A simulation model was constructed on the basis of which effects of different changes in smoking habits on lung cancer incidence in the future could be quantified directly. A fictitious population, comparable to the entire male population of Finland with respect to smoking habits, was simulated by computer to experience selective general mortality and different lung cancer risks due to different smoking habits and changes in them. The distribution of smoking habits in the population was estimated on the basis of previous country-wide surveys and on published tobacco consumption statistics. The effects of different smoking habits on general mortality and lung cancer incidence were determined on the basis of previous studies.
If the proportion of persons starting smoking is considered to be constant (30% of non-smokers aged 10–14, 15% of those aged 15–19, and 5% of those aged 20–24 years) in each consecutive 5-year period after 1975, and the proportion of smokers stopping in each category of smokers is 20% in the same periods, the age-adjusted incidence rate in the year 2000 would be 46/105. In 1975, the corresponding figure was 71/105. If the proportion stopping is only 10%, the rate in 2000 would be 64/105. If nobody stopped smoking, the figure would be 88/105. Introduction of different alternatives for starting of smoking does not have a material effect on the forecasts up to the year 2000. After that year, however, different alternatives for the proportion of new smokers, the numbers of cigarettes smoked and the age at starting have a considerable effect upon the forecasts. Based on the assumptions made, a postponement of 10 years in starting age has the same effect on lung cancer incidence as cutting the proportions of those starting by one half. Postponement of the starting age by 20 years would eliminate most of the lung cancer cases caused by smoking. In order to prevent a new rise in lung cancer incidence in the early 2000s in Finland, special efforts should be concentrated on persuading youths who started smoking in the 1970s to stop.
Revised 13 January 1981