National experience with an electronic medical record for anticoagulation therapy

Excerpt

Introduction Although anticoagulation with warfarin is very common in medical practice, there is great variety in the way anticoagulation therapy is managed, and little is known about optimal management strategies. It has previously been demonstrated that patients in designated anticoagulation clinics have more therapeutic INR measurements than those managed in primary care offices, [1‐3] however many studies are limited by small sample size or single center sampling. There is also great variability among anticoagulation clinics, especially in frequency of INR testing and practitioners managing the clinic. The availability of electronic medical record support for anticoagulation therapy has created a unique opportunity in terms of data analysis and quality assurance. Electronic medical records allow examination of clinical practice on a large scale to assist with the establishment of quality standards and collection of benchmarking data. In this retrospective analysis, we used CoagClinic™ software to examine practice variation in anticoagulation management and the relationship of these variables to INR compliance. Methods System eligibility requirements include a minimum of ten visits per month, and at least one year of continuous use of the application. A total of 130 systems qualified for the study. A total of 1,157,648 encounters involving 119,017 patients were examined. Variables examined included frequency of INR testing, number of providers in the clinic, duration of use of CoagClinic™ software, and scope of practitioner practice. INR compliance rates were calculated using the frequency of results in the therapeutic range and therapeutic range ±0.2. Results Based on the available data, a theoretical “average” anticoagulation clinic would have a patient population of 379 patients and 649 per month, seen by 17 providers in six locations. Nationally, the INR compliance rate utilizing strict range criteria was 52.1%, and 66.3% for INR range ±0.2. Effects on INR compliance rates by factors such as frequency of INR testing, number of providers, duration of clinic activity and type of provider were examined by linear regression analysis or student’s t-test where appropriate. No statistically significant effects by these factors were found with respect to INR compliance rates. Discussion In this analysis, we looked at the variables of testing frequency, number of managing practitioners, duration of use CoagClinic™ software, and primary practitioner scope of practice to determine whether these variables influenced INR compliance. Overall, INR compliance was seen in 66.4% of patient visits (SD = 0.05). This compliance rate was not altered by testing frequency, duration of electronic medical record use, number of providers, or scope of practice of the provider. This suggests that there may be ways to make anticoagulation management more cost-effective, such as decreasing the frequency of INR testing, or establishing more nurse or pharmacist-run clinics, without sacrificing INR compliance. These results also show that nearly 1/3 of patients are found to be out of range of INR compliance at any given time, so there is a need for further investigation in this area to determine whether there are variables which can be adjusted to make anticoagulation management more effective. …