The impact of smoking on use of hospital services: The Busselton study

Abstract

Objective: To use the Busselton study to estimate the impact of smoking on hospital use in Australia. Methods: People on the electoral roll in Busselton in 1978 were invited to attend a survey where smoking habits were measured. 3,230 people aged 25 to 79 at the survey were followed until 1994 by record linkage to death records and to the Western Australian Hospital Morbidity Data System. Hyper Poisson regression was used to estimate ratios of hospital separation rates and mean bed day usage according to the 1978 smoking status. Results: Compared with never smokers, current smokers had hospital separation rates that were 1.32 (95% confidence interval 1.16-1.49) times higher and used 1.40 (1...