Recognition of anxiety disorders by family physicians after rigorous medical record case extraction: results of the Netherlands Study of Depression and Anxiety.

OBJECTIVE Previous studies reported low and inconsistent rates of recognition of anxiety disorders by family physicians (FPs). Our objectives were to examine (a) which combination of indications within medical records most accurately reflects recognition of anxiety disorders and (b) whether patient and FP characteristics were related to recognition. METHOD A cross-sectional comparison was made between FPs' registration and a structured diagnostic interview, the Composite International Diagnostic Interview, according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, criteria. Seven definitions of recognition were tested using diagnostic codes, medication data, referral data and free text in medical records. Data were derived from the Netherlands Study of Depression and Anxiety. A total of 816 patients were included. RESULTS Recognition ranged between 9.1% and 85.8%. A broader definition was associated with a higher recognition rate, but led to more false positives. The best definition comprised diagnostic codes for anxiety disorders and symptoms, strong free-text indications, medication and referral to mental health care. Generalized anxiety disorder was best recognized by this definition. Recognition was better among patients with increased severity, comorbid depression and older age. CONCLUSION FPs recognized anxiety disorders better than previously reported when all medical record data were taken into account. However, most patients were nonspecifically labeled as having a mental health problem.