Over-reports of recent alcohol consumption in a clinical population: a validity study.

Decisions concerning appropriate treatment in alcoholism programs are often based on the self-reports of the clients. However, few programs have incorporated validation procedures (such as breath tests) since it is generally assumed that alcoholics will deny the extent of their drinking and their subsequent alcohol problems. The self-reports of recent alcohol consumption of sixty-five new entrants to an alcoholism treatment program were validated with breath tests. Five estimates of blood alcohol concentrations were derived by varying elimination rates. For all five estimates, over-reporters comprised a substantial proportion of the total sample (23%-57%) and exceeded the percentage of consistent reporters for those people who had a positive breath test. Correlations between self-reports and breath tests were not significant which indicates that the amount of alcohol consumed does not necessarily relate to reporting behavior. It is concluded that despite the widely held notion that alcoholics deny the extent of their drinking, errors in the direction of over-reporting should be taken seriously and examined more closely. On an individual level, over-presentation of a client's condition may be related to subsequent behavior in a treatment program; on an aggregate level, over-reports may bias the findings of evaluation studies by inflating success rates.