BACKGROUND: Surveillance systems capturing instances of self-directed violence (SDV) continue to lack uniform nomenclature and classification methodology. AIMS: To apply and compare two retrospective surveillance approaches to youth experiencing SDV presenting to two urban hospitals with Level I Trauma Centers.
METHOD: Two suicide attempt surveillance methods where retrospectively applied to our SDV cohort: (a) a rigorous method facilitated by medical record review and application of standardized classification; and (b) a common surveillance method conducted by systematic queries of suicide attempt key terms and diagnosis codes among hospital databases.
RESULTS: Rigorous surveillance identified 249 patients attempting suicide. The common method's querying suicide attempt in the chief complaint field had a high positive predictive value and specificity; however, sensitivity was low. LIMITATIONS: Authors were unable to determine whether all SDV encounters during the study timeframe were identified for initial screening owing to the hospital's lack of a uniform nomenclature or classification system.
CONCLUSION: Results showed underreporting of suicide attempt cases, inadequate sensitivity and specificity in common surveillance methods, and skewed demographic representation compared with the rigorous surveillance method. This study elucidates the negative impact of inconsistent SDV nomenclature including impeding effective patient identification, treatment, surveillance, and generalizable research.