Citation

Holahan CJ, Brennan PL, Schutte KK, Holahan CK, Hixon JG, Moos RH. J. Stud. Alcohol Drugs 2017; 78(3): 435-441.

Affiliation

Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, California.

Copyright

(Copyright © 2017, Alcohol Research Documentation, Inc., Rutgers, The State University of New Jersey)

DOI

unavailable

PMID

28499111

Abstract

OBJECTIVE: Research on late-middle-aged and older adults has focused primarily on average level of alcohol consumption, overlooking variability in underlying drinking patterns. The purpose of the present study was to examine the independent contributions of an episodic heavy pattern of drinking versus a high average level of drinking as prospective predictors of drinking problems.

METHOD: The sample comprised 1,107 adults ages 55-65 years at baseline. Alcohol consumption was assessed at baseline, and drinking problems were indexed across 20 years. We used prospective negative binomial regression analyses controlling for baseline drinking problems, as well as for demographic and health factors, to predict the number of drinking problems at each of four follow-up waves (1, 4, 10, and 20 years).

RESULTS: Across waves where the effects were significant, a high average level of drinking (coefficients of 1.56, 95% CI [1.24, 1.95]; 1.48, 95% CI [1.11, 1.98]; and 1.85, 95% CI [1.23, 2.79] at 1, 10, and 20 years) and an episodic heavy pattern of drinking (coefficients of 1.61, 95% CI [1.30, 1.99]; 1.61, 95% CI [1.28, 2.03]; and 1.43, 95% CI [1.08, 1.90] at 1, 4, and 10 years) each independently increased the number of drinking problems by more than 50%.

CONCLUSIONS: Information based only on average consumption underestimates the risk of drinking problems among older adults. Both a high average level of drinking and an episodic heavy pattern of drinking pose prospective risks of later drinking problems among older adults.

Language: en