Objectives: In 2015, CMS implemented reimbursement for non-face-to-face chronic care management (NFFCCM) for beneficiaries with multiple chronic conditions, including diabetes. This analysis estimated the association between NFFCCM and utilization of inpatient, outpatient, and emergency services.
Study design: We implemented a doubly robust estimator using propensity score matching in a regression context to compare eligible patients who used NFFCCM (n = 282) with eligible patients who did not use NFFCCM (n = 26,759).
Methods: We tested 4 definitions of treatment: having any NFFCCM encounters and having 1 NFFCCM encounter per month, per 2 months, and per 3 months. Two-tailed statistical inference testing was conducted at the 5% level. We examined the utilization differences among patients with diabetes 65 years and older using merged electronic health records for 4 health systems in Louisiana from the Research Action for Health Network database in 2013 through 2018.
Results: We found NFFCCM was associated with increased utilization of care in the outpatient setting by 0.056 visits per month (95% CI, 0.027-0.086) and with lower utilization in the inpatient setting (-0.024 visits per month; 95% CI, -0.038 to -0.010) and in the emergency department setting (-0.017 visits per month; 95% CI, -0.031 to -0.003). Alternative specifications of treatment showed similar associations.
Conclusions: CMS implementation of reimbursement codes for NFFCCM, and subsequent utilization of that reimbursement by health systems, was associated with a shift in patient utilization from high-cost settings (inpatient and emergency department) to a lower-cost setting (outpatient office).