Related Experiment Videos
Self-identification and the underdetection of alcoholism
Abstract:
Data obtained from administration of the Michigan Alcoholism Screening Test (MAST) to 66 institutionalized male alcoholics supported the hypothesis that MAST items which reflected alcoholic self-identification were more likely to result in underdetection of alcoholism than the other MAST items. These findings supported the contention that one of the consequences of devising and validating the tests on self-acknowledged alcoholics is the underdetection of cases. Implications for planning and delivery of alcoholism programs are suggested. The process by which alcoholism screening tests are developed and validated is frequently such that those items are selected which differentiate between a grouping of subjects known to be alcoholics (for example, patients on an alcoholism treatment ward) and a comparison grouping of subjects who are presumed to be free of alcoholism. However, it is questionable whether the alcoholism criterion grouping employed in this process is representative of the universe of alcoholics. The very process of becoming part of a known alcoholic grouping generally involves being identified as an alcoholic and referred for treatment by others (family members, physicians, etc.) and/or oneself. The criterion grouping therefore might include only self-identified alcoholics and exclude alcoholics who were able to avoid detection and self-perceptions of having an alcoholism problem. Consequently the items which are included in the screening test by virtue of differentiating between the criterion group and the comparison group might reflect the phenomenon of self-identification with alcoholism rather than a more broadly defined alcoholic condition. Among the possible consequences of this situation is the underdetection of cases since the screening test would tend to detect instances of self-identified alcoholism to the exclusion of non-self-identified alcoholism. In a prior analysis of data described elsewhere it was determined that a particular screening test -the Michigan Alcoholism Screening Test (MAST) - does appear to reflect alcoholic self-identification instead of, or as well as, a more inclusively defined alcoholic condition. However, it remained to be determined whether or not this situation would indeed result in the underdetection of alcoholics since the possibility existed that "self-identified alcoholism" and a presumably more inclusive "condition of alcoholism" were in fact coterminous categories. This paper deals with the latter issue.