Multidimensionality of the alcohol withdrawal symptom checklist: A factor analysis of the alcohol withdrawal symptom checklist and CIWA-Ar
Pittman, B., Gueorguieva, R., Krupitsky, E., Rudenko, AA., Flannery, B., & Krystal, JH. (2007). Multidimensionality of the alcohol withdrawal symptom checklist: A factor analysis of the alcohol withdrawal symptom checklist and CIWA-Ar. Alcoholism, Clinical and Experimental Research, 31(4), 612-618. https://doi.org/10.1111/j.1530-0277.2007.00345.x
Background: This study evaluated the factor structure of 2 scales for measuring the severity of the alcohol withdrawal syndrome (AWS): a self-rated scale, the Alcohol Withdrawal Symptoms Checklist (AWSC), and an observer-rated scale, the Clinical Institute Withdrawal Assessment—Alcohol, Revised (CIWA-Ar). Methods: Alcohol-dependent male inpatients [n=127, age: 43.0±9.7 (mean±SD) years] were recruited from an inpatient treatment unit. Both measures of AWS were assessed repeatedly during the initial week of sobriety in these patients. An exploratory factor analysis was applied to the data collected on the first study day. Results: Five independent factors accounted for 64% of total variance in the AWSC: autonomic arousal, depression, nausea and vomiting, alcohol craving, and tension/anxiety. Three items (abdominal pain, hallucinations, confusion) could not be included in the analysis due to insufficient variance. Three dimensions identified for the CIWA-Ar (autonomic arousal, nausea and vomiting, tension/anxiety) were also captured by the ASWC. Total AWSC scores correlated well with total CIWA-Ar scores (r=0.72), supporting validity of the AWSC. Lower correlations between total CIWA-Ar and the 5 factors (r=0.32–0.52) suggested that the CIWA-Ar and AWSC factors had discriminative value. Conclusions: Self-rated measures of AWS could play an important role in complementing observer-rated measures in clinical and research settings. In this sample, the AWSC appeared to identify multiple dimensions of AWS with face validity for clinical relevance.