How do we know who has had deep vein thrombosis?

An attempt was made to identify all patients diagnosed as having deep vein thrombosis (DVT) in a large teaching hospital during one year. A review of the radiology records showed that 124 (81%) of the 154 venograms performed were for suspected DVT, and that the diagnosis was confirmed in 57 (46%). During the same period, the Hospital Activity Analysis (HAA) records revealed 174 episodes classified as 'phlebitis or thrombophlebitis' in 162 patients. Only 37 of these episodes (21%) had been confirmed by X-ray venography and at least 29 patients were incorrectly classified by HAA. 'Medical' patients accounted for 54% of the episodes of DVT, yet in only 25% of these was the diagnosis confirmed by venography, while 'surgical' DVTs were less frequent but were more often confirmed by venography.