Impact of different definitions on estimates of accuracy of the diagnosis data in a clinical database

Computerized medical databases are increasingly used for research. The infl uence of different definitions of the accuracy of matching on the estimated accuracy of diagnosis data was assessed in a database of visits to a publi c pediatric clinic. Differences between definitions involved 1) unit of ana lysis. 2) number of diagnoses required to match per visit, and/or 3) whethe r database contents are required to match the medical record or medical rec ord contents are required to be matched in the database. Overall, 90% of di agnoses in the database (391/435) were accurately coded relative to the med ical record. Alternatively. 77% of diagnoses listed in the medial record (3 91/506) were accurately coded in the database. When individual visits were used as the unit of analysis, estimates of accuracy using six definitions r anged from 65% to 92%. The most appropriate definition to use for estimatin g accuracy of diagnosis data likely depends on the purpose of the study. Us e of two or more such definitions may enhance portrayal of the accuracy of diagnosis data. (C) 2001 Elsevier Science Inc. All rights reserved.