T Adair1, D Hoy, Z Dettrick, A D Lopez. 1. School of Population Health, University of Queensland, Herston, Brisbane, Queensland, Australia.
Abstract
BACKGROUND: Global studies of the long-term association between tobacco consumption and chronic obstructive pulmonary disease (COPD) have relied upon descriptions of trends. OBJECTIVES: To statistically analyse the relationship of tobacco consumption with data on mortality due to COPD over the past 100 years in Australia. METHODS: Tobacco consumption was reconstructed back to 1887. Log-linear Poisson regression models were used to analyse cumulative cohort and lagged time-specific smoking data and its relationship with COPD mortality. RESULTS: Age-standardised COPD mortality, although likely misclassified with other diseases, decreased for males and females from 1907 until the start of the Second World War in contrast to steadily rising tobacco consumption. Thereafter, COPD mortality rose sharply in line with trends in smoking, peaking in the early 1970s for males and over 20 years later for females, before falling again. Regression models revealed both cumulative and time-specific tobacco consumption to be strongly predictive of COPD mortality, with a time lag of 15 years for males and 20 years for females. CONCLUSIONS: Sharp falls in COPD mortality before the Second World War were unrelated to tobacco consumption. Smoking was the primary driver of post-War trends, and the success of anti-smoking campaigns has sharply reduced COPD mortality levels.
BACKGROUND: Global studies of the long-term association between tobacco consumption and chronic obstructive pulmonary disease (COPD) have relied upon descriptions of trends. OBJECTIVES: To statistically analyse the relationship of tobacco consumption with data on mortality due to COPD over the past 100 years in Australia. METHODS:Tobacco consumption was reconstructed back to 1887. Log-linear Poisson regression models were used to analyse cumulative cohort and lagged time-specific smoking data and its relationship with COPD mortality. RESULTS: Age-standardised COPD mortality, although likely misclassified with other diseases, decreased for males and females from 1907 until the start of the Second World War in contrast to steadily rising tobacco consumption. Thereafter, COPD mortality rose sharply in line with trends in smoking, peaking in the early 1970s for males and over 20 years later for females, before falling again. Regression models revealed both cumulative and time-specific tobacco consumption to be strongly predictive of COPD mortality, with a time lag of 15 years for males and 20 years for females. CONCLUSIONS: Sharp falls in COPD mortality before the Second World War were unrelated to tobacco consumption. Smoking was the primary driver of post-War trends, and the success of anti-smoking campaigns has sharply reduced COPD mortality levels.