| Abstract: | Kleptomania presents difficulties in diagnosis for clinicians. This study aimed to develop and test a DSM‐IV‐based diagnostic instrument for kleptomania. To assess for current kleptomania the Structured Clinical Interview for Kleptomania (SCI‐K) was administered to 112 consecutive subjects requesting psychiatric outpatient treatment for a variety of disorders. Reliability and validity were determined. Classification accuracy was examined using the longitudinal course of illness. The SCI‐K demonstrated excellent test‐retest (Phi coefficient = 0.956 (95% CI = 0.937, 0.970)) and inter‐rater reliability (phi coefficient = 0.718 (95% CI = 0.506, 0.848)) in the diagnosis of kleptomania. Concurrent validity was observed with a self‐report measure using DSM‐IV kleptomania criteria (phi coefficient = 0.769 (95% CI = 0.653, 0.850)). Discriminant validity was observed with a measure of depression (point biserial coefficient = −0.020 (95% CI = −0.205, 0.166)). The SCI‐K demonstrated both high sensitivity and specificity based on longitudinal assessment. The SCI‐K demonstrated excellent reliability and validity in diagnosing kleptomania in subjects presenting with various psychiatric problems. These findings require replication in larger groups, including non‐psychiatric populations, to examine their generalizability. Copyright © 2006 John Wiley & Sons, Ltd. |
