Journal Club Alcohol and Health: Current Evidence September-October 2005

Featured Article Psychiatric disorders among at-risk consumers of alcohol in the general population Bott K, et al. J Stud Alcohol. 2005;66:

Study Objective To examine the association between… –drinking risky amounts and –psychiatric disorders

Study Design Interviews to assess average alcohol consumption and psychiatric diagnoses Population-based sample of 4074 German adults from 47 communities who participated in a longitudinal study Sample drawn from city records Cross-sectional analyses adjusted for age, sex, and unemployment

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 sample included people from 47 different communities in a distinct region in Northern Germany. It was drawn from community registration files. The method of sample selection is not clear; however, the fact that subjects enrolled in a longitudinal study suggests that they may not have been representative of the general population. –But researchers do report that demographics of the sample were similar to those of the general population.

Were the subjects sufficiently homogeneous with respect to prognostic risk? Drinkers of risky amounts with alcohol use disorders were excluded. All other people in the sample were included in analyses, and therefore were otherwise heterogeneous with respect to risk.

Was follow-up sufficiently complete? The study was cross-sectional and therefore did not have follow-up. All eligible respondents were included in analyses.

Were objective and unbiased outcome criteria used? Diagnoses were determined with a standardized computer-assisted version of the Munich Composite International Diagnostic Interview. Drinking risky amounts was determined by quantity and frequency questions.

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? The study is not prospective. However, the outcomes were more common in people drinking risky amounts (without alcohol use disorders) than in moderate drinkers or abstainers. –Lifetime prevalence of a psychiatric disorder: 37% vs. 28% Affective disorder: Odds ratio (OR) 1.9 Anxiety disorder: OR 1.5 Somatoform disorder: OR 1.6

How precise are the estimates of likelihood? The estimates are precise. –Lifetime prevalence of a psychiatric disorder: 37% (SE* 2.6) vs. 28% (SE 0.8) (P <0.001) Affective disorder: OR 1.9 (95% CI**, ) Anxiety disorder: OR 1.5 ( ) Somatoform disorder: OR 1.6 ( ) *SE, standard error **CI, confidence interval

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? Study subjects reflected the general population of adults in a distinct region of Northern Germany. Therefore, they may not be similar to patients in clinical practice there or elsewhere.

Was the follow-up sufficiently long? The study was cross-sectional; there was no follow-up.

Can I use the results in the management of patients in my practice? The interpretation of results is limited by the study design. –Drinking (or abstinence) could precede psychiatric disorders; psychiatric disorders could lead to drinking (or abstinence). –Causality cannot be determined by the study. –The analyses adjusted for few covariates. Uncontrolled confounding (i.e., characteristics associated with both drinking and psychiatric illness) could explain the reported associations. –Findings may only apply to adults in a specific region of Northern Germany.

Can I use the results in the management of patients in my practice? (cont.) At least in this sample, drinkers of risky amounts without dependence appear to have a higher prevalence of psychiatric disorders. At a minimum, the findings suggest that clinicians should suspect psychiatric disorders in patients drinking risky amounts and vice versa. Whether or not drinking risky amounts leads to these psychiatric disorders should be studied using longitudinal study designs.