The evidence indicates that clinicians who use a single method to obtain patient information regularly draw faulty conclusions.

For instance, Fennig, Craig, Tanenberg-Karant, and Bromet (1994) reviewed the diagnoses assigned to 223 patients as part of usual hospital practice. Clinical diagnoses were then compared with diagnoses derived from a comprehensive multi-method assessment that consisted of a semistructured patient interview, a review of the patient’s medical record, a semistructured interview with the treating clinician, and an interview with the patient’s significant other, all of which were then reviewed and synthesized by two clinicians to derive final diagnoses from the multi-method assessment.

Even though Fennig, Craig, Tanenberg-Karant, et al. (1994) used very liberal criteria to define diagnostic agreement (e.g., major depression with psychotic features was treated as equivalent to dysthymia), the diagnoses assigned during the course of typical clinical practice had poor agreement with the diagnostic formulations derived from the more extensive synthesis of multiple assessment methods.

Overall, after discounting chance agreement, the clinical diagnoses agreed with the multi-method conclusions only about 45-50% of the time. This was true for a range of disorders on the schizophrenic, bipolar, and depressive spectrums.

Because these conditions are treated in decidedly different ways, such frequent misdiagnoses in typical practice suggest that many patients erroneously receive antipsychotic, antimanic, and antidepressant medications.

from Meyer, G. J., Finn, S. E., Eyde, L. D., Kay, G. G., Moreland, K. L., Dies, R. R., Eisman, E. J., Kubiszyn, T. W., & Reed, G. M. (2001). Psychological testing and psychological assessment. A review of evidence and issues. The American Psychologist, 56(2), 128–165.