Twenty-one alcoholic patient volunteers in a state'hospital were given a mean of 11 hours of one of three treatments: (a) interpersonal aversion (in which 5 watched his own audiovideo-recorded drinking "confessions" while being systematically derogated by two therapists), followed immediately by systematic desensitization to drinking-related anxieties; (6) interpersonal aversion, followed by a control procedure consisting of friendly interaction; or (c) group discussion. A six-to ninemonth interview follow-up showed that five of the seven traceable 5s given the interpersonal aversion-systematic desensitization treatment had been abstinent, compared with only one of seven treated by the interpersonal aversion-control procedure. This significant difference was supported by changes on the Alcadd Test but not on the MMPI MacAndrew Alcoholism scale. Difficulties in contacting 5s for follow-up precluded complete comparisons with the group discussion 5s.

PROBLEMThere is no satisfactory existing psychometric procedure to assess the severity of an alcoholic person's drinking problem. I n a recent review of more than 150 studies that attempted to delineate specific characteristics of alcoholics on standard personality and other psychological assessment instruments, Primo e 6 ) concluded that little of use has been formulated. Two instruments constructed especially for the assessment of alcoholism-the Alcadd test (18-l9, 20) and the MacAndrew l 6 ) from the MMPI -have fared somewhat better in validity studies, a finding consistent with psychometric evidence (I3) that the most valid assessment devices are those developed and used for a specific purpose.Biographical data, although relied upon heavily to make evaluations in psychiatry and in social work, have not been used widely as an objective basis for measurement and prediction except in personnel situations (33 ). However, what little use they have had in objective clinical assessment(14* 29) has shown their superiority over psychological tests in this regard.A number of researchers have utilized biographical and behavioral data to assess various aspects of alcoholism('* 6 , 'r 21, 23, 28, 31), but there has been no systematic attempt t o develop an assessment device based on biographical and behavioral data. Such an instrument would have a number of uses: in research, by permitting more reliable assessments; in clinical work, as a practical means to assess severity and change; and in theoretical formulations, by giving information about the nature of changes.The present project was designed to develop and validate an index of severity of alcoholism based on self-reported biographical data. A practical definition of alcoholism has been utilized, as suggested by Keller(12), one that refers to a disorder in which repeated drinking causes injury to the drinker's health or to his social or economic functioning. The use of biographical data in assessment is consistent with this definition, because such data provide a practical survey of the degree to which drinking has injured the individual in the past and the extent to which it is continuing to do so. METHOD Since a variety of biographical and behavioral factors related to alcoholism have been reported in the research literature, it was decided to utilize these findings as the empirical basis for the initial development of items and then to use an internal consistency procedure for further refinement.Problem Drinking Schedule, Form I (PDS-I). In a pilot project, 35 biographical items were formulated (some open-ended and some multiple-choice) from several sources of empirical information about alcoholics (2* 3. lo, 25) and were administered as a self-report questionnaire to four alcoholic patients at Mayview State Hospital, Pittsburgh. All reported the PDS-I to be straightforward, interesting and meaningful.PDS-II. A more extensive literature survey resulted in the formulation of 70 multiple-choice biographical items with a reported empirical relationship to alcoholis...

No abstract