Objective: To determine the accuracy of lung cancer mortality data based on clinical observations in the absence of autopsy and to identify factors affecting the accuracy of diagnosis.
Methods: Admission, pre-autopsy and post-autopsy diagnoses were recorded for 1000 consecutive autopsies in each of two University departments in Budapest with high autopsy rates for persons dying in hospital. In those 87 cases where one or more diagnosis included primary lung cancer, additional data were collected concerning clinical investigations relevant to the diagnosis and the histological type lung cancer, and on smoking habits.
Results: 59% (36/61) of lung cancers seen at autopsy were not detected pre-autopsy, while 50% (25/50) of those diagnosed pre-autopsy were not confirmed at autopsy. Many misdiagnoses arose because patients were too ill to be properly investigated and/or died before investigations could be completed. Accuracy of diagnosis increased with the number of diagnostic techniques applied, but was still far from perfect in the absence of necropsy. Underdiagnosis was commoner in non-smokers and overdiagnosis commoner in smokers.
Conclusions: Without necropsy, lung cancer misdiagnosis is common, especially when modern diagnostic procedures cannot be fully employed. Knowledge of smoking habits may affect diagnostic accuracy.