Abstract 143A: Does Insurance Type Predict Quality of Anticoagulation Care in a Boston Safety Net Hospital System?

Abstract

Background: Boston Medical Center (BMC) is the primary safety net hospital for Eastern Massachusetts and has a diverse patient population with diverse insurance types. Such types include commercial and public insurance (Medicare and Medicaid) and Free Care, limited coverage funded by money distributed to safety-net institutions to care for uninsured patients. In 2006, the state expanded Medicaid eligibility and began offering Commonwealth Care, comprehensive subsidized coverage with retail pharmacy benefits, for uninsured patients. The impact of these insurance types on individual cardiovascular conditions has not been studied. Venous thromboembolism (VTE), comprised of deep venous thrombosis and pulmonary embolism, is a condition whose clinical course is dependent on high quality anticoagulation care including easy access to medication and providers. Time in the therapeutic range (TTR), the percentage of time a patient spends with INR between 2 and 3, has emerged as the preeminent way to measure quality of anticoagulation care. In this study, we compared quality of anticoagulation among different insurance types. Methods: Using clinical data, we identified adults aged 18 to 64 with a new episode of VTE diagnosed in the years 2003 to 2010 at BMC or its affiliated health centers. To be eligible for inclusion, each patient had to have an ICD-9-CM code for VTE and an INR test in the month following VTE diagnosis. We computed TTR using all INR values from diagnosis to 12 months according to the Rosendaal method. We then measured the mean TTR for each of six insurance categories based on primary insurance at time of diagnosis. Using multiple linear regression, we adjusted measurements for sex, age, race, language preference, area poverty, VTE type, recent surgery, and number of Elixhauser comorbidities. Results: We identified 1099 patients with VTE. Twenty-three percent had commercial insurance, 37% Medicaid, 16% Medicare, 4% Commonwealth care, 18% Free Care, and 2% other. Mean TTR was 39.3%. Patients with Free Care and Commonwealth Care had similar TTR compared to those with commercial insurance. Patients with Medicaid, Medicare or other insurance had significantly lower TTR, compared to those with commercial insurance. Conclusion: Quality of anticoagulation care was low in this population. Residual confounding such as from healthy worker effect may account for higher TTR in patients with commercial insurance. In future work we plan to expand measurement of insurance effects to patients receiving anticoagulation for indications other than VTE and adjust our measurements for temporal bias