A systematic review of validated methods for identifying suicide or suicidal ideation using administrative or claims data

摘要: PURPOSE: As part of the Mini-Sentinel pilot program, under contract with Food and Drug Administration, an effort has been made to evaluate validity algorithms useful for identifying health outcomes interest, including suicide attempt. METHOD: Literature was reviewed how well medical episodes associated these events could be identified in administrative or claims data sets from USA Canada. RESULTS: Six studies were found include sufficient detail assess performance characteristics algorithm on basis International Classification Diseases, Ninth Revision, E-codes (950-959) intentional self-injury. Medical records death registry information used validate classification. Sensitivity ranged 13.8% 65%, positive predictive value range 4.0% 100%. Study comparisons are difficult interpret, however, as differed substantially many important elements, design, sample, setting, methods. Although varied widely, two located prepaid plans reported that database codes charts achieve good agreement. CONCLUSIONS: Insufficient exist support specific recommendations regarding a preferred algorithm, caution should exercised interpreting clinical pharmacological epidemiological surveillance research rely measures suicide-related outcomes. Copyright © 2012 John Wiley & Sons, Ltd. Language: en