Purpose: To investigate whether duration of follow-up influences the association between alcohol and cause-specific mortality in a prospective cohort study with only baseline assessment of alcohol intake.
Methods: In a cohort of 14,223 men and women participating in the first investigation of the Copenhagen City Heart Study between 1976 and 1978 and followed until 2001, we assessed whether the association between alcohol and mortality was modified by duration of follow-up. The 24 years of follow-up were divided into four intervals, and Cox survival analyses were conducted separately for these four succeeding 6-year periods of follow-up.
Results: The authors found differences in the predictive values of alcohol and beverage-specific types of alcohol depending on duration of follow-up both in terms of all-cause mortality and death from coronary heart disease and cancer. The apparent protective effect of a moderate alcohol consumption on coronary heart disease attenuated during prolonged follow-up, whereas high alcohol consumption became associated with higher risk of death from cancer with longer follow-up.
Conclusions: This study accentuates the importance of taking duration of follow-up into consideration when interpreting risk estimates from prospective studies on the association between alcohol and mortality.