Infection is the Most Frequent Disease Category of Autopsy-detected Diagnostic Errors in Japan

抄録

A diagnostic error is a failure to establish an accurate and timely explanation of the health problem of a patient or communicate that explanation to the patient. Considering that a diagnostic error may contribute to morbidity and mortality, preventive efforts should be promoted. Therefore, it is important to elucidate the relative frequency of disease category and specific diseases among patients with diagnostic errors. Although various measurements can be used to identify disease category and specific diseases in diagnostic errors, autopsy remains one of the essentials for determining the cause of death. Numerous previous reports of diagnostic errors based on autopsy used Goldman’s criteria . These criteria defined Class I errors as major diagnostic errors that could have changed the patient outcome if known before death and Class II errors as major diagnostic errors that could not have changed the patient outcome even if known before death. An autopsy series from 1960 to 1980 in a single university hospital reported Class I and II error prevalences of 10% and 12%, respectively. Recent studies reported that diagnostic errors have decreased because of advances in clinical skills and diagnostic procedures; however, the prevalence of autopsy-detected diagnostic errors remains high. In 2003, a review in US hospitals estimated that diagnostic error rate was 8.4-24.4%, specifically a Class I error rate of 4.1-6.7%. The most frequent disease categories were cardiovascular, infectious, or neoplastic diseases among the error cases. Thus, autopsy continues to play an essential role in measuring and monitoring diagnostic errors and their disease characteristics. In Japan, several studies have determined autopsy-detected diagnostic errors in the past. In 1963, the prevalence of diagnostic error in a single university hospital was 14.3%. In 2009, the prevalence of Class I errors in the same hospital was 9%. In another single-center study that reviewed 75 autopsy cases from 1983 to 1997, there was discrepancy between clinical and pathological diagnoses in 24 (32.9%) cases. A survey of internal medicine physicians throughout Japan in 1994 reported that the discrepancy between clinical and pathological diagnoses was 11.7%. However, no studies have been conducted in Japan on autopsy-detected diagnostic errors in the 21st century. Japan has had a number of diagnostic advances in technology for laboratory and imaging testing. Results of such study in contemporary Japan could elucidate common issues of diagnostic excellence in developed countries. Therefore, the current study used a large Japanese database of case series to estimate the frequency of diagnostic errors and to describe the relative frequency of its disease categories and specific diseases. The findings of this work will help physicians in modern clinical settings of Japan improve their diagnostic excellence.