Outcomes research using the electronic patient record: Beth Israel Hospital's experience with anticoagulation

Using data captured as part of the routine care of outpatients taking the oral anticoagulant warfarin, we described variation in recording reasons for anticoagulation, selecting target International Normalized Ratio (INR) ranges, and performing coagulation blood tests. Laboratory results were directly captured by or entered into an Anticoagulation Flowsheet, a computer program which is fully integrated with our Online Medical Record (OMR). We studied the 177 patients with flowsheets between October 1993 and January 1995. 90% had a reason for anticoagulation entered; 29 different target INR ranges were entered. For patients with a target INR of 2.0-3.0, the mean number of weeks between blood tests, after a test which was in range, was three weeks (standard deviation 1.7 weeks, range one to twelve weeks). We conclude that routinely collected data contained in an electronic patient record (EPR) can be a rich resource for describing and evaluating clinical practice. We also address several limitations to using EPR data: validity of EPR information, lack of coded information, and imperfect capture of clinician thought processes.