Health care quality: Misdiagnosis at a university hospital in five medical eras

Abstract

Few long-term, autopsy-confirmed studies have been carried out to determine whether modern diagnostic techniques and clinical laboratory procedures have diminished the rate of misdiagnosis in recent years. We extended a previous investigation on the occurrence of misdiagnosis at a German University Hospital in 1959, 1969, 1979, and 1989 by a further decade to 1999/2000. Hospital charts and autopsy records of 100 randomly selected patients who died in hospital in 1999/2000 were reviewed retrospectively for clinical findings, diagnoses, and possible misdiagnoses, and compared with results from the previous four medical eras. If occurring, diagnostic errors were classified as misdiagnosis, false-positive, or false-negative diagnoses. The misdiagnosis rate of 11% in 1999/2000 was unchanged compared to the previous years studied. False-negative diagnoses rose from 22% in 1979 to 34% and 41% in 1989 and 1999/2000, respectively; false-positive diagnoses also increased from 7% in 1989 to 15% in 1999/2000. The most common diagnostic errors in 1999/2000 again involved pulmonary emboli, myocardial infarctions, neoplasms, and infections. New diagnostic procedures such as ultrasound, computed tomography, and magnetic resonance imaging did not reduce the rate of misdiagnosis. By contrast, the patient’s medical history and physical examination continued to play an important diagnostic role, leading to an assumption of the correct final diagnosis in about 75% of cases. The reduction in the autopsy rate from 88% in 1959 to 20% in 1999/2000 was remarkable. Despite ever-improving diagnostic technology, we found no evidence that the rate of misdiagnosis is declining. False-negative and false-positive diagnoses may be on the increase. The precipitous decline in the autopsy rate bodes ill for errando discimus–the ability to learn from our errors.