P N Furness1, I Lauder. 1. Department of Pathology, University of Leicester, Leicester Royal Infirmary, United Kingdom.
Abstract
AIMS: To obtain a crude estimate of the rate at which consultant histopathologists become aware of errors in their work, and to gather information about the circumstances under which errors are made. METHODS: A postal questionnaire was sent to all consultant histopathologists (1021) in the United Kingdom. RESULTS: There was a 19.8% response to the questionnaire (202 returned). 119 pathologists reported 244 errors, 82 from within the preceding year. 42% of these errors had affected patient management. Pathologists usually blamed human error or excessive workload; however, data analysis did not reveal an excess of responses from laboratories with heavy workloads. A disproportionately large proportion of errors appear to be made in samples from the lymphoreticular system and by locum consultant pathologists. CONCLUSIONS: This study undoubtedly suffered from reporting bias, but the figures suggest that a typical pathologist probably becomes aware of having made a serious diagnostic error approximately once per year. A larger, more systematic study of the circumstances under which errors are likely to occur appears to be justified but would require significant resources because of the large sample size required.
AIMS: To obtain a crude estimate of the rate at which consultant histopathologists become aware of errors in their work, and to gather information about the circumstances under which errors are made. METHODS: A postal questionnaire was sent to all consultant histopathologists (1021) in the United Kingdom. RESULTS: There was a 19.8% response to the questionnaire (202 returned). 119 pathologists reported 244 errors, 82 from within the preceding year. 42% of these errors had affected patient management. Pathologists usually blamed humanerror or excessive workload; however, data analysis did not reveal an excess of responses from laboratories with heavy workloads. A disproportionately large proportion of errors appear to be made in samples from the lymphoreticular system and by locum consultant pathologists. CONCLUSIONS: This study undoubtedly suffered from reporting bias, but the figures suggest that a typical pathologist probably becomes aware of having made a serious diagnostic error approximately once per year. A larger, more systematic study of the circumstances under which errors are likely to occur appears to be justified but would require significant resources because of the large sample size required.
Authors: I A Cree; W Guthrie; J M Anderson; M P Holley; D Hopwood; D S Sanders; D M Parham; S Lang; J Lang; J S Beck Journal: Pathol Res Pract Date: 1993-05 Impact factor: 3.250