Aggressive Care near the End of Life for Cancer Patients in Medicare Accountable Care Organizations

OBJECTIVES: To compare aggressiveness of end-of-life (EoL) care for older cancer patients attributed to Medicare Shared Savings Programs with that for similar fee for service (FFS) beneficiaries not in an accountable care organization (ACO) and examine whether observed differences in EoL care utilization vary across markets that differ in ACO penetration. DESIGN: Cross-sectional observational study comparing ACO-attributed beneficiaries with propensity score-matched beneficiaries not attributed to an ACO. SETTING: A total of 21 hospital referral regions (HRRs) in the United States. PARTICIPANTS: Medicare FFS beneficiaries with a cancer diagnosis who were 66 years or older and died in 2013-2014. MEASUREMENTS: Outcome measures were claims-based quality measures of aggressive EoL care: (1) one or more intensive care unit (ICU) admissions in the last month of life, (2) two or more hospitalizations in the last month of life, (3) two or more emergency department visits in the last month of life, (4) chemotherapy 2 weeks or less before death, and (5) no hospice enrollment or hospice enrollment within 3 days of death. Analyses were adjusted for demographic and clinical characteristics of beneficiaries and practice characteristics. RESULTS: Compared with beneficiaries not in an ACO, ACO-attributed beneficiaries had a higher rate of ICU admission during the last month of life (37.7% vs 34.0%; adjusted difference = +2.8 percentage points; 95% confidence interval (CI) = 1.0-4.6) but fewer repeated hospitalizations (14.5% vs 15.2%; adjusted difference = -1.7 percentage points; CI = -3.1 to -.3). Other measures did not differ for the two groups. Although the ICU admission rates tended to decrease as ACO-penetration rates increased (P < .01), ACO patients had higher rates of ICU admission than non-ACO patients in both medium and high ACO-penetration HRRs. CONCLUSION: Cancer patients attributed to ACOs had fewer repeated hospitalizations but more ICU admissions in the last month of life than non-ACO patients; they had similar rates of other measures of aggressive care at the EoL. This suggests opportunities for ACOs to improve EoL care for cancer patients. J Am Geriatr Soc 67:961-968, 2019.