Methods to improve diagnostic accuracy in a community mental health setting

Objective: This study determined the extent to which adding structured proc edures improved diagnostic accuracy for outpatients with severe mental illn ess in a community mental health setting. Method: The Structured Clinical Interview for DSM-III-R (SCID) was used to interview 200 psychiatric outpatients. A research nurse reviewed medical re cords and amended the SCID diagnoses accordingly. A research psychiatrist o r psychologist reviewed the diagnostic data and interviewed each patient to verify or further modify the previous findings. Diagnostic outcomes at eac h step of the procedure were compared to determine whether adding additiona l data improved diagnostic accuracy. The additional time required for each element of the diagnostic procedure was also assessed. Results: Kappa comparisons of the different diagnostic levels showed that a dding additional data significantly improved accuracy. Diagnoses rendered b y combining the SCID and review of the medical record were the most accurat e, followed by the SCID alone, and then diagnoses made by psychiatrists dur ing routine care. In addition, the SCID alone identified five times as many current and past secondary diagnoses as were documented routinely in patie nts' charts. Conclusions: Combining structured interviewing with a review of the medical record appears to produce more accurate primary diagnoses and to identify more secondary diagnoses than routine clinical methods. The patients' knowl edge of their diagnoses was limited, suggesting a need for patient educatio n in this setting. Whether use of structured interviewing in routine practi ce improves patient outcomes deserves further study.