Assessing the severity of hazardous drinking and related consequences among incarcerated women.

BACKGROUND

Recent epidemiological efforts have demonstrated the utility of measuring individual differences in the severity of alcohol use along a single severity continuum marked by alcohol-related problems, symptoms of alcohol dependence, and the social consequences of drinking. Translation of this utility to specialized clinical populations is not assured. The expected inter-relationships among problems, symptoms and consequences, and enhanced sensitivity of combined assessments require confirmation in applied clinical settings.

METHOD

Subjects were 245 incarcerated women who met Alcohol Use Disorders Identification Test criteria for hazardous use of alcohol. Participants were recruited from a statewide adult correctional facility for an ongoing clinical trial testing the effectiveness of brief motivational interviewing on alcohol use and HIV risk behaviors. Participants ranged in age from 18 to 56 (M = 34.1, SD = 8.9), 71.4% were Caucasian, and 65.7% reported <12 years of education.

RESULTS

Analyses suggested that the 6 problems of alcohol abuse, 7 symptoms for alcohol dependence, and 14 alcohol-related social consequences loaded to a single factor (0.38 to 0.85) that formed a continuum of alcohol severity. Contrary to epidemiological studies, physical fights and being arrested were the most prevalent consequences and were associated with lower alcohol severity in this population. Three of the five items that discriminated best between higher and lower alcohol severity were related to familial and relationship consequences.

CONCLUSIONS

Consistent with epidemiological studies, alcohol severity can be measured among incarcerated hazardously drinking women on a single continuum that includes alcohol problems, symptoms, and social consequences. Replication of the expected alignment of problems and symptoms supports the construct validity of the continuum and further challenges the proposed hierarchical structure of abuse/dependence distinction. Large differences in rates of specific consequences and observed effectiveness of tailored social consequence items suggest the benefits of cross-sample validation to improve evaluation of clinical outcomes.