Subjective measures of binge drinking and alcohol-specific adverse health outcomes: a prospective cohort study


Tapio Paljärvi, Department of Primary Care and Public Health, Imperial College London, St Dunstan's Road, London W6 8RP, UK. E-mail:


Aim  To determine the performance of subjectively defined intoxications, hangovers and alcohol-induced pass-outs in identifying drinkers at risk for adverse health outcomes.

Design  Prospective population-based cohort study.

Setting  Working-aged Finnish general population.

Participants  A total of 21 204 alcohol-drinking men and women aged 20–24, 30–34, 40–44 and 50–54 years at baseline who participated in the Health and Social Support (HeSSup) postal survey in 1998.

Measurements  Binge drinking was measured by subjectively defined intoxications/drunkenness, hangovers and alcohol-induced pass-outs. Hazardous drinking was defined according to Finnish guidelines as weekly total intake of >287 g of ethanol for men, and for women > 191 g of ethanol (≥24 and ≥16 standard drinks, respectively). Study participants were followed-up for 7 years for alcohol-specific hospitalizations and deaths. Proportional hazard models and areas under the receiver operating characteristics curves (AUC) were used to analyse the data.

Findings  Of the drinkers, 6.5% exceeded the weekly limit for hazardous drinking, and 1.5% experienced the alcohol-specific end-point during the follow-up. Subjective intoxications, hangovers and alcohol-induced pass-outs all predicted future alcohol-specific diagnoses independently of average intake and of several other potential confounders. In identifying baseline hazardous drinking, subjective intoxications had a superior performance in relation to other subjective measures of binge drinking. In identifying future alcohol-specific hospitalizations or death, subjective intoxications had also the best performance, but this was not significantly different from the other binge drinking measures, or average intake.

Conclusions  Subjectively defined intoxications, hangovers and alcohol-induced pass-outs are population-level proxy measures of at-risk drinking patterns.