Citation

Murray RP, Tyas SL, Snow W, Ekuma O, Bond R, Barnes GE. Addict. Behav. 2010; 35(3): 278-281.

Affiliation

Department of Community Health Sciences, University of Manitoba, Winnipeg, MB, Canada.

Copyright

(Copyright © 2010, Elsevier Publishing)

DOI

10.1016/j.addbeh.2009.10.010

PMID

19910124

Abstract

We sought to identify a level of alcohol consumption representing the boundary between health protective and hazardous drinking. The Winnipeg Health and Drinking Survey began in 1990-91 (n=1257). Seven years later, a third wave of interviews (n=785) expanded questions on heavy episodic drinking (HED) and assessed the consumption of >/=3, >/=5, >/=8, and >/=12 drinks at a sitting for each of wine, beer and liquor (equivalent to about 40g, 65g, 105g and 155g of ethanol). Cox proportional hazards models were based on seven years of illness and mortality data following the Wave 3 interview, and were stratified by gender and HED definition. For HED of >/=40g, >/=65g, >/=105g, or >/=155g per occasion, the hazard ratios for morbidity and mortality from all causes were 1.06, 1.09, 1.17, and 1.16 respectively in women, and 1.00, 0.98, 1.02, and 1.02 in men. Most of these hazard ratios were significant in women, whereas none was significant in men. This study did not provide support for a definition of HED that could divide protective from hazardous alcohol consumption.

Language: en