Validity of diagnoses of chronic diseases in general practice: The application of diagnostic criteria

Abstract

Certainty of a diagnosis is not only important for the patient but also for morbidity studies. In the absence of a gold standard, agreement with diagnostic criteria is often the best approach in measuring the certainty of a diagnosis. The agreement with diagnostic criteria has been studied for 5 chronic diseases (hypertension, chronic ischemic heart disease, diabetes mellitus, chronic nonspecific lung disease and osteoarthritis) in 7 general practices with a total practice population of 23,534 persons. Agreement with diagnostic criteria is operationalized into 3 categories. For each chronic disease a diagnostic quality measure per general practitioner is computed. Retrospective data have been collected in the practices on 2295 diseases in 1989 patients. Two-thirds of the diagnoses were made in general practice. The agreement with the diagnostic criteria for the cases diagnosed in general practice is high, ranging from 96% true positive cases in diabetes mellitus to 58% in chronic nonspecific lung disease. The highest rate of false positive cases is 4%. On the level of general practitioners diagnostic qualities vary from 62 to 96% true positive cases for the different diseases. The variation in diagnostic quality between general practitioners is substantial. The prevalence rates for the 5 chronic diseases are lower after adjustment by only including true positive cases. Diagnoses of the 5 chronic diseases recorded in general practice are generally valid with low numbers of false positive cases.