{"doi":"10.1111/jgs.18878","title":"Utilization of transitional care management service among Medicare beneficiaries at high risk of readmission","abstract":"The Center for Medicare & Medicaid Services introduced transitional care management (TCM) service in 2013 to oversee care coordination and reduce readmission during the 30 days following hospital discharge.1 TCM uptake has been low (9.3% in 2016)2 and slowly increasing over time.2, 3 Previous research reported lower utilization of TCM services among non-white and dually eligible populations.4 However, it remains uncertain whether TCM is delivered to Medicare beneficiaries at high risk of readmission, such as those living with frailty or dementia. The objective of this study is to explore the utilization rate of TCM service by beneficiaries' characteristics associated with a higher risk of readmission. In a 5% random sample of 2015–2019 Medicare fee-for-service claims, we identified eligible community discharges from short-term hospitals, psychiatric hospitals, skilled nursing facilities, observational stays, or partial hospitalizations.1 We excluded beneficiaries <65 years old; who died within 30 days post-discharge; who did not have >12 months continuous enrollment for Medicare Part A and B before discharge; and who had major surgical procedures. TCM service was identified using Current Procedural Terminology codes (99,495 or 99,496). Demographic information (age, sex, and race), dual eligibility, census region (Northeast, Midwest, South, and West), type of eligible discharge (short-term hospitals, psychiatric hospitals, skilled nursing facilities, observational stays, and partial hospitalizations), primary discharge diagnosis category (e.g., disease of the circulatory system; diseases of the respiratory system; disease of the digestive system; and other), dementia diagnosis based on Chronic Conditions Warehouse5 algorithm at the time of the discharge, and frailty (claims-based frailty index6 <0.15 [robust], 0.15 to <0.25 [pre-frailty], 0.25 to <0.35 [mild frailty], 0.35 to <0.45 [moderate frailty], and ≥0.45 [severe frailty]) were measured. We examined the rate of TCM use by each characteristic among eligible community discharges. To assess TCM use by a global risk of 30-day readmission, we developed a model using a random-intercept logistic regression to predict 30-day readmission based on all aforementioned variables and compared selected characteristics between those at high risk (≥median risk) versus low risk (