New Supplemental Benefits and Plan Ratings Among Medicare Advantage Enrollees

Key Points Question Was Medicare Advantage plans’ adoption of supplemental benefits targeting members’ nonmedical and social needs associated with changes in enrollees’ ratings of plan experience? Findings This cohort study compared 388 356 plan ratings of US Medicare Advantage enrollees in plans that adopted an expanded primarily health-related benefit, Special Supplemental Benefit for the Chronically Ill, or both and found that plans that adopted both benefits saw a significant mean increase of 0.22 points (of 10) in plan ratings vs plans that adopted neither. Meaning These findings suggest that more investments in supplemental investments may improve plan ratings and experiences for medically and socially complex Medicare beneficiaries.

This supplementary material has been provided by the authors to give readers additional information about their work.available MA plan benefit, enrollment, and crosswalk data.NOTES eTable 7 presents the unadjusted and adjusted difference-indifference (DID) estimates of the impact of PHRB, SSBCI, or both benefits adoption on individual plan rating expressed as the average plan-rating across treatment and control plans for plans that adopted more than one benefit.N represents the number of person-years in each treatment group.Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated physical and mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket (OOP), premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.All analyses apply survey weights.***p<0.001,**p<0.01,*p<0.05. a Enrollees in plans that adopted two or more SSBCI benefits are removed from the analysis because the trends were not graphically or statistically (p=0.02)parallel prior to adoption in 2021.SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an adjusted event study for plans that adopted a primarily health related benefit (PHRB) in 2019, 2020, or 2021 that did not adopt a SSBCI at any point; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket [OOP], premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.All analyses apply survey weights.

Treatment
eFigure 3. Unadjusted Event Study for Primarily Health Related Benefit (PHRB) Adoption, 2019-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an unadjusted event study for plans that adopted a primarily health related benefit (PHRB) in 2019, 2020, or 2021 that did not adopt a SSBCI at any point; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).All analyses apply survey weights.
eFigure 4. Adjusted Event Study for Special Supplemental Benefits for the Chronically Ill (SSBCI) Adoption, 2020-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an adjusted event study for plans that adopted a Special Supplemental Benefits for the Chronically Ill (SSBCI) in 2020 or 2021 that did not adopt a PHRB at any point; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max OOP, premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.All analyses apply survey weights.
eFigure 5. Unadjusted Event Study for Special Supplemental Benefits for the Chronically Ill (SSBCI) Adoption, 2020-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an adjusted event study for plans that adopted a Special Supplemental Benefits for the Chronically Ill (SSBCI) in 2020 or 2021 that did not adopt a PHRB at any point; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).All analyses apply survey weights.eFigure 6. Adjusted Event Study for Plans that Adopted Both Benefits, 2020-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an adjusted event study for plans that adopted a PHRB and SSBCI in 2020 or 2021 that did not previously adopt either; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max OOP, premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.All analyses apply survey weights.
eFigure 7. Unadjusted Event Study for Plans that Adopted Both Benefits, 2020-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Estimates reflect outcomes from an adjusted event study for plans that adopted a PHRB and SSBCI in 2020 or 2021 that did not previously adopt either; 2020 CAHPS data is unavailable and therefore is not shown.We removed plans that adopted the PHRBs in 2019 only, 2019 and 2020 only, 2019 and 2021 only and plans that adopted the SSBCIs in 2021 only (634 plans, 37% of the sample).All analyses apply survey weights.NOTES eTable 10 presents a linear regression of the relationship between supplemental benefit adoption and an enrollee's self-reported receipt of receiving an additional benefit from their health plan to help manage their chronic condition.***p<0.001,**p<0.01,*p<0.05.

eFigure 1 .
Sample Inclusion and Exclusion Criteria SOURCE Authors' analysis of 2017-2021 Medicare Advantage (MA) Consumer Assessment of Health Care Providers and Systems (CAHPS) survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES: HMO: Health Maintenance Organization; PPO: Preferred Provider Organization; MBSF: Medicare Beneficiary Summary File.We excluded individuals enrolled in Medicare-Medicaid, PACE, MSA, PFFS, PDP, Cost, and employersponsored plans (n=245,807).

eFigure 1 .
Sample Inclusion and Exclusion Criteria eTable 1. Contract-Level, Plan-Level, and Enrollment-Level Adoption of PHRB and SSBCI Benefits eTable 2. Difference-in-Differences Parallel Trends Validation, 2021 eTable 3. Sample Characteristics for Enrollees in Treatment and Control Plans at Baseline, 2017-2019 eTable 4. Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Probability of Non-Response in the MA CAHPS, 2017-2021 eTable 5. Unadjusted and Adjusted Impact of PHRB Adoption on Plan Rating, 2019 eTable 6. Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Plan Rating, 2017-2021 eTable 7. Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Plan Rating for Plans that Adopted Two or More Benefits, 2021 eTable 8. Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Health Utilization and Self-Reported Health, 2021 eFigure 2. Adjusted Event Study for Primarily Health Related Benefit (PHRB) Adoption, 2019-2021 eFigure 3. Unadjusted Event Study for Primarily Health Related Benefit (PHRB) Adoption, 2019-2021 eFigure 4. Adjusted Event Study for Special Supplemental Benefits for the Chronically Ill (SSBCI) Adoption, 2020-2021 eFigure 5. Unadjusted Event Study for Special Supplemental Benefits for the Chronically Ill (SSBCI) Adoption, 2020-2021 eFigure 6. Adjusted Event Study for Plans that Adopted Both Benefits, 2020-2021 eFigure 7. Unadjusted Event Study for Plans that Adopted Both Benefits, 2020-2021 eTable 9. Rate of Extra Benefit Availability Due to Health Condition Across Benefit Cohort, 2017-2021 eTable 10.Relationship Between Supplemental Benefit Adoption and Enrollee Receipt of Extra Benefit Due to Health

eTable 1 .
Contract-Level, Plan-Level, and Enrollment-Level Adoption of PHRB and SSBCI Benefits Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES: The p-values are from linear probability models estimating the interaction of treatment group and year in the pre-period to test if the parallel trends assumption was met.P>0.05 shows there does not appear to be significant differences in parallel trends.CI: confidence interval.Sample Characteristics for Enrollees in Treatment and Control Plans at Baseline, 2017-2019 Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Probability of Non-Response in the MA CAHPS, 2017-2021 Unadjusted and Adjusted Impact of PHRB Adoption on Plan Rating, 2019 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk dat.NOTES eTable 5 presents the unadjusted and adjusted difference-indifference (DID) estimates of the impact of PHRB adoption on individual plan rating expressed as the average plan-rating across treatment and control plans.For all benefits, 2017-2018 is the pre-period and 2019 is the post-period.N represents the number of person-years in each treatment group.Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated physical and mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket (OOP), premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.***p<0.001,**p<0.01,*p<0.05.Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Plan Rating for Plans that Adopted Two or More Benefits, 2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.SOURCE eTable 3.SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES eTable 4 presents the unadjusted and adjusted difference-indifference (DID) estimates of the impact of PHRB, SSBCI, or both benefits adoption on probability of non-response expressed as the odds of non-response across treatment and control plans.For all benefits, 2017-2019 is the pre-period and 2021 is the postperiod.N represents the number of person-years in each treatment group.Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated physical and mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket (OOP), premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.***p<0.001,**p<0.01,*p<0.05.SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES eTable 6 presents the unadjusted and adjusted difference-indifference (DID) estimates of the impact of PHRB, SSBCI, or both benefits adoption on individual plan rating expressed as the average plan-rating across treatment and control plans.N represents the number of person-years in each treatment group.Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated physical and mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket (OOP), premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.***p<0.001,**p<0.01,*p<0.05.SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly Unadjusted and Adjusted Impact of PHRB, SSBCI, or Both Benefit Adoption on Health Utilization and Self-Reported Health, 2021 Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES eTable 8 presents the unadjusted and adjusted difference-indifference (DID) estimates of the impact of PHRB, SSBCI, or both benefits adoption on individual health utilization metrics and selfrated health expressed as the average health utilization metrics and self-rated health across treatment and control plans for plans that adopted more than one benefit.N represents the number of person-years in each treatment group.Fully adjusted models control for individual-level (sex, race/ethnicity, age, original reason for entitlement, dual eligibility, educational attainment, self-rated mental health, receipt of proxy assistance, self-reported hospitalization utilization, living alone, self-reported chronic conditions, census region, inpatient admissions, nursing home admissions, home health visits, chronic conditions, and number of days in 2021 an enrollee was in the plan), plan-level factors (plan type, plan size, SNP status, contract start year, max out of pocket (OOP), premium, deducible, and change in max OOP, premium, and deductible from the prior year, and plan age), and county MA penetration rate.All analyses apply survey weights.***p<0.001,**p<0.01,*p<0.05. a Enrollees in plans that adopted both are removed from the unadjusted analysis because the trends were not graphically or statistically (p=0.02)parallel prior to adoption in 2021 in fully adjusted models.

eTable 9 .
Rate of Extra Benefit Availability Due to Health Condition Across Benefit Cohort, 2017-2021 SOURCE Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data.NOTES Question from the MA CAHPS Survey: Your health plan benefits are the types of health care and services you can get under the plan.In the last 6 months, [MA-Only] did your health plan offer you extra benefits because you have a health condition (like high blood pressure)?Relationship Between Supplemental Benefit Adoption and Enrollee Receipt of Extra Benefit Due to Health Condition, 2017-2021 Authors' analysis of 2017-2021 MA CAHPS survey data and Medicare administrative enrollment data and publicly available MA plan benefit, enrollment, and crosswalk data. SOURCE