Journal Club Alcohol and Health: Current Evidence March-April 2005

Featured Article High-Risk Alcohol Consumption and Late-Life Alcohol Use Problems Moos RH, et al. Am J Public Health. 2004;94(11):

Study Objective To examine… –the prevalence of at-risk drinking, as defined by various guidelines, and –the association between alcohol consumption and related problems …among older adults

Study Design Mail and telephone surveys at baseline and 10 years later 1291 non-abstinent, community- dwelling older adults

Assessing Validity of an Article about Prognosis Are the results valid? What are the results? How can I apply the results to patient care?

Are the Results Valid? Was the sample representative? Were the subjects sufficiently homogeneous with respect to prognostic risk? Was follow-up sufficiently complete? Were objective and unbiased outcome criteria used?

Was the sample representative? The sampling procedure is not reported in the article. If sampled appropriately, subjects would represent people who… –were aged 55–65 years at recruitment, –were from Northern California, –had telephones, and –had contact with the outpatient healthcare system.

Were the subjects sufficiently homogeneous with respect to prognostic risk? Few data are provided regarding the risk of alcohol problems. Those who had never consumed alcohol were excluded. –92% of subjects had consumed alcohol in the past 2 years.

Was follow-up sufficiently complete? Follow-up was 93% at 10 years. Although this proportion seems “high,” results could be biased if loss to follow-up was biased. –Example: Loss of the lightest drinkers (if they had MORE problems than the light drinkers who were not lost) might imply that the observed findings overestimate problems associated with drinking.

Were objective and unbiased outcome criteria used? Although self-report is a valid measure to determine consumption, it is not clear whether the specific measure used in this study was validated. Alcohol problems were assessed with a validated survey tool.

What are the Results? How likely are the outcomes over time? How precise are the estimates of likelihood?

How likely are the outcomes over time? Prevalence of risky drinking differed depending on the thresholds defined by various guidelines. –23% to 50% among women –29% to 45% among men Those who exceeded limits were more likely to have alcohol-related problems (e.g., difficulties with relationships and functioning) both at study entry and follow-up. –These problems were more prevalent in men.

How likely are the outcomes over time? (cont.) Both men and women reduced consumption after 10 years. The limit of no more than 7 drinks per week/3 drinks per day offered the best combination of sensitivity and specificity in predicting alcohol-related problems in both men and women. –At this cut-off, 16% of women and 34% of men had alcohol use problems at follow- up.

How precise are the estimates of likelihood? Odds ratios and confidence intervals are not provided. The P value for the increased risk of later problems associated with exceeding limits at baseline (versus not exceeding limits) is <0.01.

How Can I Apply the Results to Patient Care? Were the study patients and their management similar to those in my practice? Was the follow-up sufficiently long? Can I use the results in the management of patients in my practice?

Were the study patients similar to those in my practice? Based on the information provided in the paper, this question is difficult to answer. However, the results may be applicable to community-dwelling adults of the age studied.

Was the follow-up sufficiently long? 10 years is appropriate to detect differences in alcohol-related problems based on consumption.

Can I use the results in the management of patients in my practice? Despite the limitations noted, these data do support the use of currently recommended national guidelines for risky and less risky drinking that do not differ by sex. Physicians should counsel patients to avoid exceeding these limits and can provide data from this study as one reason to do so (to avoid the risk of subsequent alcohol-related problems).