The present study examined the validity of four scales which attempt to systematically characterize alcoholic patients on a continuum of dependence severity: the Rand Dependence scale, the Severity of Alcohol Dependence Questionnaire (SADQ), the Last Month of Drinking Withdrawal scale, and the Last Six Months of Drinking Impaired Control and Dependence scale. Forty patients in an inpatient alcohol rehabilitation program were administered these four scales in addition to two separate assessments of alcohol consumption and two measures of psychological problems. In a factor analysis of all eight scales, the four dependence scales loaded on one factor, indicating that the dependence measures were highly intercorrelated. This was interpreted as evidence for the high convergent validity of the dependence measures. The dependence scales also showed good discrimenant validity, since the alcohol consumption and psychosocial problem scales did not load on the dependence factor but load on the two other factors. Predictive validity of the dependence scores was then examined. The SADQ and the Rand scores showed a significant positive correlation with amount of detoxification medication consumed in the hospital. The dependence scores did not precdict 6 month post‐treatment abstinence or rapidity of reinstatement of heavy drinking after a post‐treatment slip. Further refinement of the construct of dependence and/or better assessment methods may be necessary to improve the predictive utility of dependence questionnaires.
|Original language||English (US)|
|Number of pages||7|
|Journal||British Journal of Addiction|
|State||Published - Feb 1986|
All Science Journal Classification (ASJC) codes
- Medicine (miscellaneous)