Abstract

Aims: It has previously been shown that the low necropsy request rate at the University Hospital of the West Indies (UHWI) in Jamaica (35.3%) results primarily from clinicians’ confidence in clinical diagnoses and laboratory investigations. This study aimed to determine the rates of discrepancy between clinical and necropsy diagnoses at the UHWI, because many previous studies from other institutions have shown persistent high rates of discrepancy, despite advances in medical investigative technology over the past several years.

Methods: Data were extracted retrospectively from consecutive necropsies performed at the UHWI over a two year period. The data were analysed to determine the categories and rates of discrepancy, and to determine the relation between discrepancy rates and age, sex, type and number of diagnoses for each patient, hospital service, and length of hospitalisation.

Results: Necropsies were performed on 446 patients; 348 were suitable for further analysis. The overall discrepancy rate was 48.4% and the diagnoses with the highest individual discrepancy rates were pneumonia (73.5%), pulmonary thromboembolism (68.3%), and myocardial infarction (66.7%). Males and older patients were more likely to have discrepant diagnoses. There was a high frequency of discrepancies in patients who died within 24 hours of admission, but there was no consistent relation between length of hospitalisation and discrepancy rate.

Conclusions: The high discrepancy rates documented at the UHWI are similar to those reported globally. This study supports previous attestations that the necropsy remains a vital tool for determining diagnostic accuracy, despite modern modalities of clinical investigation and diagnosis.