Abstract
Background:
Over half of Medicare beneficiaries are enrolled in privately-administered Medicare Advantage (MA) plans, and aortic stenosis predominantly affects older patients. In this study, we evaluated whether differences in care delivery between MA and fee-for-service (FFS) Medicare contribute to cardiovascular outcomes after transcatheter aortic valve replacement (TAVR).
Methods:
Medicare patients who underwent TAVR from 1/1/2018 to 12/31/2022 were eligible for cohort inclusion. TAVR utilization rates in MA and FFS were calculated as the age and sex standardized number of procedures per 10,000 person-years. Patient characteristics were compared between MA and FFS using standardized mean differences (SMD), with an SMD ≥ 10 considered meaningful. The association of MA vs FFS enrollment with 30-day and 1-year outcomes was assessed using logistic and Cox regression models, respectively.
Results:
Among the 317,630 patients (64.4% FFS, 35.6% MA) undergoing TAVR, the mean age was 80.6 ± 7.0 years and 55.1% were male. TAVR utilization rates were similar between MA and FFS, with 11.2 TAVRs/10,000 person-years in FFS compared to 11.7 TAVRs/10,000 person-years in MA by the end of the study. Compared with FFS patients, MA patients were more likely to be Black (5.4% vs 3.0%, SMD: −11.9) and dual-enrolled in Medicaid (13.7% vs. 8.6%, SMD: −16.3). Comorbidity profiles were similar (SMD < 10). In the comorbidity-adjusted model, MA patients had lower odds of 30-day death (adjOR: 0.90, 95% CI: 0.86–0.95). At one year, MA patients had a lower adjusted risk of death (HR: 0.94, 95% CI: 0.92–0.96), all-cause readmission (HR: 0.86, 95% CI: 0.85–0.88), stroke (HR: 0.90, 95% CI: 0.87–0.94), and aortic valve reintervention (HR: 0.86, 95% CI: 0.76–0.96).
Conclusions:
Despite similar clinical profiles and TAVR utilization, MA patients appeared to experience better 30-day and one-year outcomes than FFS patients. Differences in care delivery may contribute, but future studies are needed to understand differences in outcomes.
Keywords: TAVR, MA, FFS, aortic stenosis
Introduction:
Aortic stenosis is the most prevalent valvular heart disease among older adults worldwide, and the burden of disease continues to rise with the aging population.1 In the United States, current estimates suggest that moderate or greater aortic stenosis affects 3.1% of adults 65–85 years old and approximately 7% of individuals over age 80.2 When left untreated, severe symptomatic aortic stenosis is associated with high morbidity and mortality rates as high as 50%.3 Transcatheter aortic valve replacement (TAVR) emerged as a treatment option for aortic stenosis in 2011 and now offers patients faster recovery with overall similar outcomes compared to surgical aortic valve replacement (SAVR).3–8 As of 2019, rates of TAVR for the treatment of aortic stenosis have superseded those of SAVR.9
Since aortic stenosis primarily affects Medicare-eligible adults, it is important to understand how care delivery in Medicare influences patient outcomes. Currently, Medicare-eligible adults have the option to enroll in either traditional fee-for-service Medicare (FFS) or a privately-administered Medicare Advantage (MA) plan – a decision that can have important implications on how and where they receive care.10,11 Because MA insurers receive capitated, risk-adjusted payments for providing care to each beneficiary, MA plans often have structural mechanisms to help control spending. Examples of such mechanisms include incentivizing guideline-based treatments, developing disease management programs, promoting outpatient care delivery, and establishing enhanced post-procedure care coordination systems to reduce disease burden and hospitalizations.12–14 However, MA plans also use preferred provider networks and prior authorizations to regulate spending.15,16 This is in sharp contrast to FFS in which reimbursement is tied directly to the number of services provided to the beneficiary, and beneficiaries have the opportunity to seek care at institutions of their choosing. Prior studies have suggested lower utilization of some healthcare services including emergency department visits, ambulatory surgery and procedures, and hip/knee replacements in MA compared to FFS.17 However, it remains unclear whether the mechanisms designed to control spending in MA lead to underutilization of appropriate treatments.18
Although roughly half of Medicare beneficiaries today receive healthcare through privately-administered MA plans, there are ongoing questions regarding whether the differences between MA and FFS contribute to meaningful differences in the quality of care delivery.19 As overall TAVR utilization is increasing rapidly with expanded FDA approval for use in low-risk patients, TAVR is of particular interest for a comparison of MA and FFS. Furthermore, since patients with severe aortic stenosis undergoing TAVR represent a population with a high burden of comorbidities overall, there are fewer concerns about large unmeasured differences between MA and FFS TAVR patients confounding an informative comparison of these two groups. In this study, we use Medicare claims data to answer three key questions: 1) How does overall TAVR utilization compare in MA versus FFS?; 2) Do the baseline or hospital characteristics of MA patients undergoing TAVR differ from those of FFS patients undergoing TAVR?; 3) Are the 30-day and one-year outcomes after TAVR different for MA vs. FFS beneficiaries?
Methods:
Data Sources and Study Population
The Medicare claims data used in this analysis are subject to a data use agreement and are not publicly available. The study population was identified using MA encounter data and 100% FFS claims. All Medicare patients ≥ 65 years old who underwent TAVR from 1/1/2018 to 12/31/2022 were eligible for inclusion in the study (Figure 1). The following International Classification of Diseases-10 (ICD-10) procedure codes were used to identify the occurrence of TAVR: 02RF3JZ, 02RF3KZ, 02RF38Z, 02RF37Z, 02RF37H, 02RF38H, 02RF3JH, 02RF3KH. Medicare program type was determined using the Master Beneficiary Summary file, and patients were characterized as MA versus FFS patients based on enrollment status during the month of their TAVR procedure. Patients were excluded if they did not have continuous Medicare enrollment for 12 months prior to undergoing TAVR. Patients were also excluded if their insurance status captured in the Master Beneficiary Summary Files in the month of the TAVR procedure was inconsistent with the inpatient claims source (MA vs. FFS) from which the TAVR procedures were identified. Patients who underwent valve-in-valve TAVR, including TAVR in TAVR, were excluded. This study was approved by the Institutional Review Board at Beth Israel Deaconess Medical Center. Medicare claims were accessed through the CMS Virtual Research Data Center (VRDC) secure environment under an existing data use agreement.
Figure 1:
Sample Selection
Figure 1 demonstrates the application of study eligibility criteria to identify the population of interest in Medicare claims.
* For patients who underwent >1 TAVR during the study period, outcomes were described for index procedure
† Patients were excluded if there was inconsistency between insurance status (MA vs FFS) in the Master Beneficiary Summary File and the type of insurance claim filed (MA vs FFS) for the TAVR procedure
FFS = fee-for-service; MA = Medicare Advantage; TAVR = transcatheter aortic valve replacement
Patient Characteristics
Patient age, sex, race, and dual-enrollment in Medicare and Medicaid were obtained from the Master Beneficiary Summary Files. The comorbidities of interest were defined using the Chronic Conditions Warehouse (CCW) algorithm and included the following: prior myocardial infarction (MI), Alzheimer’s disease, atrial fibrillation (AF), chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), heart failure, diabetes, hip/pelvic fracture, ischemic heart disease, depression, stroke/transient ischemic attack (TIA), cancer, anemia, asthma, hyperlipidemia, hypertension, obesity, alcohol use disorders, tobacco use, and peripheral vascular disease.20 Additional sociodemographic factors that were assessed include region (Northeast, Midwest, South, West), social vulnerability index, and rurality (metro, micro, rural). When possible, hospital characteristics including hospital ownership, teaching status, and hospital bed size were identified through linkage between Medicare claims and the American Hospital Association survey. The distribution of MA and FFS patients treated at each hospital with > 50 TAVR procedures was also evaluated. Finally, to evaluate geographic access to TAVR care, the distance in miles between each beneficiary’s residence and the implanting hospital was calculated using the SAS GEODIST function, which measures the distance between the latitude and longitude coordinates of the beneficiary’s 5-digit residential ZIP code and those of the hospital’s 5-digit ZIP code. To reduce concerns about differences in coding practices between MA and FFS driving differences in comorbidity ascertainment, only comorbidities identified during the patient’s index TAVR hospitalization were considered, as has been done in other recent comparisons between MA and FFS.21–24 Similarly, claims-based patient frailty scores were not assessed due to possible differences in coding practices between MA and FFS.
Patient Outcomes
The primary outcomes of interest were all-cause mortality at 30 days and one year, ascertained using the Master Beneficiary Summary File. The following secondary 30-day outcomes were evaluated using claims algorithms that have been validated previously for evaluating TAVR outcomes in MA and FFS: all-cause readmission, acute kidney injury, heart failure hospitalization, stroke, disabling stroke, bleeding, major vascular complications, permanent pacemaker implantation, acute myocardial infarction, transient ischemic attack, and aortic valve reintervention (Table S1).25–28 The secondary one-year outcomes of interest were all-cause readmission, heart failure hospitalization, stroke, disabling stroke, transient ischemic attack, and aortic valve reintervention (Table S1). The hospital length of stay for the index procedure was calculated as the difference between the admission date and date of hospital discharge. Post-hospital discharge destination was also compared between MA and FFS. All outcomes were assessed using Medicare claims through December 31, 2023.
Statistical Analysis
First, to evaluate trends in utilization rates of TAVR in MA and FFS, the age- and sex-standardized number of TAVR procedures per 10,000 person-years of enrollment for each quarter of the study period was calculated among beneficiaries ≥ 65 years old. For this calculation, the numerator was the number of TAVR procedures performed in a given quarter, stratified by MA versus FFS. The denominator, person-years of enrollment per quarter, was derived from the Master Beneficiary Summary File by summing monthly enrollment flags for all Medicare beneficiaries aged ≥65 years. Only months in which a beneficiary was alive and enrolled contributed to the person-years for that quarter. Individuals who switched insurance were counted toward person-years for each insurance based on their monthly enrollment flag. Finally, utilization rates were age- and sex- standardized to the distribution of the overall Medicare population in 2022. Since differences in TAVR utilization between MA and FFS could be driven by differences in utilization of SAVR, trends in SAVR utilization rates were also assessed.
Next, baseline patient and hospital characteristics were compared between TAVR patients enrolled in MA and FFS using standardized mean differences (SMD), presented as a percentage of the pooled standard deviation. An SMD ≥ 10% was considered to represent a significant difference between groups.
Finally, the risks of binary 30-day outcomes (excluding the index hospitalization) were compared between MA and FFS patients using logistic regression. The difference in mean length of stay in MA and FFS was compared using linear regression. For one-year outcomes (excluding the index hospitalization), Cox proportional hazards regression models were used to estimate hazard ratios comparing MA to FFS. The proportional hazards assumption was assessed graphically using log(-log) survival plots. Unadjusted, age and sex-adjusted, comorbidity-adjusted, and fully adjusted models were fit for both 30-day and one-year outcomes. Fully-adjusted models included age, sex, comorbidities, dual-enrollment in Medicaid, region, rurality, and social vulnerability to evaluate whether these factors were contributors to any observed differences between MA and FFS. Procedure year was also included in the comorbidity-adjusted and fully-adjusted models. Hospital characteristics were not included in the fully adjusted model, as differences in hospital characteristics between MA and FFS were hypothesized to be potential mediators of any observed differences in outcomes. For the assessment of one-year outcomes, patients were censored at the time of death or disenrollment from Medicare. As a sensitivity analysis, select in-hospital outcomes (all-cause mortality, aortic valve reintervention, and permanent pacemaker implantation) were assessed in MA and FFS. Throughout the analysis, point estimates with 95% confidence intervals are reported, rather than p-values. All analyses were restricted to patients with complete data on the variables analyzed (complete case analysis); missing values were not imputed. All analyses were conducted using SAS version 9.4 (SAS Institute Inc, Cary, NC).
Results
TAVR Utilization
Of the 347,422 Medicare patients who underwent an initial TAVR procedure between 2018–2022, 317,630 patients (91.4%) met eligibility criteria for study inclusion, of which 204,605 (64.4%) were enrolled in FFS and 113,025 (35.6%) were enrolled in MA (Figure 1). Missing data was minimal for most covariates (Table S2), and outcome data were complete at 30 days and 1 year for all patients except those who died before assessment (censored for death). TAVR utilization, assessed as the age- and sex- standardized number of TAVRs/10,000 person-years during each quarter of the study, was similar in MA and FFS (Figure 2). By the final quarter of the study, there were 11.2 TAVRs/10,000 person-years in FFS compared to 11.7 TAVRs/10,000 person-years in MA. Utilization of SAVR was also similar in MA and FFS throughout the study period (Figure S1).
Figure 2:
TAVR Utilization in Medicare Advantage and Fee-for-Service
Figure 2 highlights age and sex standardized rates of TAVR/10,000 person-years of enrollment among fee-for-service and Medicare Advantage beneficiaries during the study period. Quarterly utilization rates were standardized to the age and sex distribution of the overall Medicare population in 2022. TAVR = transcatheter aortic valve replacement
Baseline Characteristics of MA and FFS Patients Undergoing TAVR
The mean patient age in the overall sample was 80.6 ± 7.0 years, and 55.1% were male (Table 1). Compared with FFS patients, MA patients were more likely to be Black (5.4% vs 3.0%, SMD: −11.9) and dual-enrolled in Medicaid (13.7% vs. 8.6%, SMD: −16.3) (Table 1). MA patients were also less likely to be rural-based (6.9% vs 11.1%, SMD: 14.6) and in the least vulnerable quantile for social vulnerability (18.8% vs. 23.9%, SMD: 12.4). There were no significant differences in the rate of comorbidities between MA and FFS patients (SMD <10 for all comorbidities).
Table 1:
Differences in Patient Characteristics Between MA vs. FFS
| Patient Characteristic | All TAVR Patients (N = 317630) | MA (N = 113025) | FFS (N = 204605) | SMD* |
|---|---|---|---|---|
| Demographics | ||||
| Age (yrs) | 80.6±7.0 (317,630) | 80.3±6.9 (113,025) | 80.8±7.1 (204,605) | 6.9 |
| Male | 174,893 (55.1%) | 61,087 (54.0%) | 113,806 (55.6%) | 3.2 |
| Race | ||||
| White | 291,371 (91.7%) | 101,040 (89.4%) | 190,331 (93.0%) | 12.8 |
| Black | 12,228 (3.8%) | 6,082 (5.4%) | 6,146 (3.0%) | −11.9 |
| Other | 10,763 (3.4%) | 4,815 (4.3%) | 5,948 (2.9%) | −7.3 |
| Dual Enrollment | 32,952 (10.4%) | 15,442 (13.7%) | 17,510 (8.6%) | −16.3 |
| Region | ||||
| Northeast | 71,008 (22.4%) | 24,173 (21.4%) | 46,835 (22.9%) | 3.6 |
| Midwest | 71,178 (22.4%) | 25,924 (22.9%) | 45,254 (22.1%) | −2.0 |
| South | 109,863 (34.6%) | 37,365 (33.1%) | 72,498 (35.4%) | 5.0 |
| West | 63,952 (20.1%) | 24,256 (21.5%) | 39,696 (19.4%) | −5.1 |
| SVI Group | ||||
| Most Vulnerable Quantile | 56,223 (17.7%) | 20,454 (18.1%) | 35,769 (17.5%) | −1.6 |
| Second Most Vulnerable Quantile | 99,992 (31.5%) | 37,659 (33.3%) | 62,333 (30.5%) | −6.1 |
| Third Most Vulnerable Quantile | 89,251 (28.1%) | 32,213 (28.5%) | 57,038 (27.9%) | −1.4 |
| Least Vulnerable Quantile | 70,154 (22.1%) | 21,273 (18.8%) | 48,881 (23.9%) | 12.4 |
| Rurality | ||||
| Metro | 254,790 (80.2%) | 95,453 (84.5%) | 159,337 (77.9%) | −16.9 |
| Micro | 32,287 (10.2%) | 9,762 (8.6%) | 22,525 (11.0%) | 8.0 |
| Rural | 30,404 (9.6%) | 7,789 (6.9%) | 22,615 (11.1%) | 14.6 |
| Comorbidities | ||||
| Prior AMI | 3,138 (1.0%) | 1,202 (1.1%) | 1,936 (0.9%) | −1.2 |
| Alzheimer’s Disease | 2,162 (0.7%) | 826 (0.7%) | 1,336 (0.7%) | −0.9 |
| Atrial Fibrillation | 2,781 (0.9%) | 1,684 (1.5%) | 1,097 (0.5%) | −9.5 |
| Chronic Kidney Disease | 120,266 (37.9%) | 43,725 (38.7%) | 76,541 (37.4%) | −2.6 |
| Chronic Obstructive Pulmonary Disease | 61,994 (19.5%) | 22,912 (20.3%) | 39,082 (19.1%) | −2.9 |
| Heart Failure | 221,631 (69.8%) | 77,620 (68.7%) | 144,011 (70.4%) | 3.7 |
| Diabetes | 117,657 (37.0%) | 44,006 (38.9%) | 73,651 (36.0%) | −6.1 |
| Hip/Pelvic Fracture | 247 (0.1%) | 109 (0.1%) | 138 (0.1%) | −1.0 |
| Ischemic Heart Disease | 221,265 (69.7%) | 77,683 (68.7%) | 143,582 (70.2%) | 3.1 |
| Depression | 21,009 (6.6%) | 8,260 (7.3%) | 12,749 (6.2%) | −4.3 |
| Stroke / Transient Ischemic Attack | 6,335 (2.0%) | 2,414 (2.1%) | 3,921 (1.9%) | −1.6 |
| Cancer | 67,325 (21.2%) | 21,791 (19.3%) | 45,534 (22.3%) | 7.3 |
| Anemia | 87,794 (27.6%) | 30,664 (27.1%) | 57,130 (27.9%) | 1.8 |
| Asthma | 68,610 (21.6%) | 25,444 (22.5%) | 43,166 (21.1%) | −3.4 |
| Hyperlipidemia | 227,323 (71.6%) | 81,239 (71.9%) | 146,084 (71.4%) | −1.1 |
| Hypertension | 286,408 (90.2%) | 101,415 (89.7%) | 184,993 (90.4%) | 2.3 |
| Obesity | 66,806 (21.0%) | 25,009 (22.1%) | 41,797 (20.4%) | −4.2 |
| Alcohol Use Disorders | 3,080 (1.0%) | 1,158 (1.0%) | 1,922 (0.9%) | −0.9 |
| Tobacco Use | 15,035 (4.7%) | 5,761 (5.1%) | 9,274 (4.5%) | −2.6 |
| Peripheral Vascular Disease | 62,598 (19.7%) | 24,572 (21.7%) | 38,026 (18.6%) | −7.9 |
| Hospital Characteristics † | ||||
| Ownership | ||||
| Public/Municipal | 22,827 (8.5%) | 6,757 (7.2%) | 16,070 (9.1%) | 7.2 |
| Private, Non-profit | 226,611 (84.0%) | 80,324 (85.3%) | 146,287 (83.3%) | −5.5 |
| Private, For-profit | 20,425 (7.6%) | 7,117 (7.6%) | 13,308 (7.6%) | 0.1 |
| Teaching Status | ||||
| Major Teaching Hospital | 109,390 (40.5%) | 35,170 (37.3%) | 74,220 (42.3%) | 10.1 |
| Minor Teaching Hospital | 130,964 (48.5%) | 48,576 (51.6%) | 82,388 (46.9%) | −9.3 |
| Non-teaching Hospital | 29,509 (10.9%) | 10,452 (11.1%) | 19,057 (10.8%) | −0.8 |
| Hospital Bed Size | ||||
| <100 | 5,901 (2.2%) | 1,609 (1.7%) | 4,292 (2.4%) | 5.2 |
| 100 to 399 | 85,760 (31.8%) | 31,439 (33.4%) | 54,321 (30.9%) | −5.3 |
| 400 and above | 178,202 (66.0%) | 61,150 (64.9%) | 117,052 (66.6%) | 3.6 |
| Annual Medicare TAVR Volume of Hospital in Which Patient Underwent TAVR | ||||
| <= 50 TAVRs/year | 37,005 (11.7%) | 13,739 (12.2%) | 23,266 (11.4%) | −2.5 |
| > 50 TAVRs/year | 280,431 (88.3%) | 99,107 (87.8%) | 181,324 (88.6%) | 2.5 |
| Distance from Patient Residential Zip Code to Hospital Zip Code (Miles), mean ± SD ‡ | 50.6 ± 173.2 | 39.4 ± 143.5 | 56.6 ± 186.9 | 10.3 |
AMI = acute myocardial infarction; FFS = fee-for-service; MA = Medicare Advantage; SMD = standardized mean difference; SVI = social vulnerability index; TAVR = transcatheter aortic valve replacement
SMDs are presented as a percentage of the pooled standard deviation
94,198 MA patients could be linked to a hospital in the American Hospital Association dataset; 175,665 FFS patients could be linked to a hospital in the American Hospital Association dataset
data available for 175,455 FFS patients and 93,908 MA patients
Hospital Characteristics
Linkage to the American Hospital Association survey was possible for 83.3% (94,198/113,025) of MA patients and 85.9% (175,665/204,605) of FFS patients. Hospital characteristics were similar between linked patients enrolled in MA vs FFS (Table 1). Regardless of MA vs FFS enrollment, most patients underwent TAVR in hospitals that were private and non-profit (>80%), minor teaching (>45%), and >400-beds (>65%). Approximately 88% of patients underwent TAVR at a hospital in which the annual Medicare TAVR volume was >50 TAVRs/year. MA patients traveled a mean distance of 39.4 miles for TAVR, while FFS patients traveled a mean of 56.6 miles (SMD: 10.3). Of the 752 hospitals with >50 TAVR procedures, 6 hospitals served only FFS patients, 7 hospitals served only MA patients, and the remainder had a wide range of distributions of MA and FFS patients (Figure 3).
Figure 3:
Hospital Distribution of Medicare Advantage and Fee-for-Service Patients
Figure 3 demonstrates the percentage of Medicare Advantage and Medicare Fee-for-Service patients treated at each of the 752 hospitals that completed > 50 TAVR procedures. The percentage of MA procedures is demonstrated in red and of FFS procedures is in blue.
MA = Medicare Advantage; FFS = fee-for-service; TAVR = transcatheter aortic valve replacement
Outcomes of TAVR in MA and FFS Patients
After adjusting for comorbidities, there was no difference in the length of stay for patients enrolled in MA and FFS (overall mean: 4.1 days) (Table 2). However, patients enrolled in MA were more likely to be discharged home than FFS patients (adjusted OR[adjOR]: 1.39, 95% CI: 1.35–1.43) (Table 2). In the comorbidity-adjusted model, MA patients had lower odds of almost all 30-day outcomes including all-cause mortality (adjOR: 0.90, 95% CI: 0.86–0.95), all-cause readmission (adjOR: 0.88, 95% CI: 0.86–0.90), and stroke (adjOR: 0.87, 95% CI: 0.81–0.93). The odds of these outcomes remained significantly lower in MA even after adjustment for region, rurality, social vulnerability, and dual-enrollment in Medicaid (Table 3). Results were generally similar in the unadjusted and age and sex adjusted models.
Table 2:
Comparison of Discharge Metrics between MA vs FFS Patients
| All TAVR Patients mean ± S.D. or n (%) n=317630 |
MA Patients mean ± S.D. or n (%) n=113025 |
FFS Patients mean ± S.D. or n (%) n=204605 |
Unadjusted | Age + Sex Adjusted | Comorbidity Adjusted | |
|---|---|---|---|---|---|---|
| Mean Difference or OR* | Mean Difference or OR* | Mean Difference or OR* | ||||
| Length of stay | 4.1±4.7 | 4.1±4.9 | 4.1±4.6 | −0.02 [−0.05, 0.02] | −0.01 [−0.05, 0.02] | 0.002 [−0.03, 0.03] |
| Discharge Destination | ||||||
| Home | 287,521 (90.5%) | 104,119 (92.1%) | 183,402 (89.6%) | 1.35 [1.32, 1.39] | 1.33 [1.29, 1.36] | 1.39 [1.35, 1.43] |
| Short-stay hospital | 864 (0.3%) | 278 (0.2%) | 586 (0.3%) | 0.86 [0.74, 0.99] | 0.86 [0.75, 0.99] | 0.82 [0.71, 0.95] |
| Skilled nursing facility | 23,760 (7.5%) | 6,854 (6.1%) | 16,906 (8.3%) | 0.72 [0.70, 0.74] | 0.73 [0.71, 0.75] | 0.71 [0.69, 0.73] |
| Long-term care facility | 1,328 (0.4%) | 354 (0.3%) | 974 (0.5%) | 0.66 [0.58, 0.74] | 0.66 [0.59, 0.75] | 0.65 [0.57, 0.74] |
| Death/hospice | 4,061 (1.3%) | 1,369 (1.2%) | 2,692 (1.3%) | 0.92 [0.86, 0.98] | 0.93 [0.88, 1.00] | 0.92 [0.86, 0.98] |
FFS = fee-for-service; MA = Medicare Advantage; OR = odds ratio; TAVR = transcatheter aortic valve replacement
Reference Group = FFS
Table 3:
Comparison of 30-day Outcomes in MA vs FFS Patients
| Outcomes | All TAVR Patients n=317630 |
MA Patients n=113025 |
FFS Patients n=204605 |
Unadjusted OR† (95% CI) |
Age + Sex Adjusted OR† (95% CI) |
Comorbidity Adjusted OR† (95% CI) |
Fully Adjusted* OR† (95% CI) |
|---|---|---|---|---|---|---|---|
| All-cause mortality | 6716 (2.1%) | 2254 (2.0%) | 4462 (2.2%) | 0.91 [0.87, 0.96] | 0.93 [0.88, 0.98] | 0.90 [0.86, 0.95] | 0.90 [0.85, 0.95] |
| All-cause readmission | 34763 (10.9%) | 11584 (10.2%) | 23179 (11.3%) | 0.89 [0.87, 0.92] | 0.90 [0.88, 0.92] | 0.88 [0.86, 0.90] | 0.87 [0.84, 0.89] |
| AKI | 21391 (6.7%) | 7606 (6.7%) | 13785 (6.7%) | 1.00 [0.97, 1.03] | 1.01 [0.98, 1.04] | 0.95 [0.93, 0.99] | 0.94 [0.92, 0.97] |
| HF Hospitalization | 5647 (1.8%) | 1896 (1.7%) | 3751 (1.8%) | 0.91 [0.86, 0.97] | 0.93 [0.88, 0.98] | 0.92 [0.87, 0.97] | 0.90 [0.85, 0.96] |
| Stroke | 6462 (2.0%) | 2307 (2.0%) | 4155 (2.0%) | 1.01 [0.95, 1.06] | 1.01 [0.96, 1.06] | 0.87 [0.81, 0.93] | 0.87 [0.81, 0.93] |
| Disabling Stroke | 3682 (1.2%) | 1216 (1.1%) | 2466 (1.2%) | 0.89 [0.83, 0.96] | 0.90 [0.84, 0.97] | 0.78 [0.72, 0.85] | 0.78 [0.71, 0.84] |
| Bleeding | 80065 (25.2%) | 27515 (24.3%) | 52550 (25.7%) | 0.93 [0.92, 0.95] | 0.93 [0.92, 0.95] | 0.91 [0.90, 0.93] | 0.92 [0.90, 0.94] |
| Major Vascular Complications | 14634 (4.6%) | 5030 (4.5%) | 9604 (4.7%) | 0.95 [0.91, 0.98] | 0.95 [0.91, 0.98] | 0.92 [0.89, 0.96] | 0.93 [0.90, 0.97] |
| PPM | 26081 (8.2%) | 8881 (7.9%) | 17200 (8.4%) | 0.93 [0.90, 0.95] | 0.94 [0.91, 0.96] | 0.92 [0.90, 0.95] | 0.93 [0.90, 0.95] |
| AMI | 4629 (1.5%) | 1694 (1.5%) | 2935 (1.4%) | 1.05 [0.98, 1.11] | 1.05 [0.99, 1.12] | 0.95 [0.89, 1.02] | 0.95 [0.89, 1.01] |
| TIA | 1588 (0.5%) | 530 (0.5%) | 1058 (0.5%) | 0.91 [0.82, 1.01] | 0.92 [0.83, 1.02] | 0.83 [0.75, 0.93] | 0.82 [0.74, 0.92] |
| Aortic Valve Reintervention | 467 (0.1%) | 185 (0.2%) | 282 (0.1%) | 1.19 [0.99, 1.43] | 1.18 [0.98, 1.43] | 1.15 [0.94, 1.39] | 1.12 [0.92, 1.37] |
AKI = acute kidney injury; AMI = acute myocardial infarction; FFS = fee-for-service; HF = heart failure; MA = Medicare Advantage; OR = odds ratio; PPM = permanent pacemaker; TAVR = transcatheter aortic valve replacement; TIA = transient ischemic attack
The fully adjusted model included adjustment for age, sex, comorbidities, region, rurality, social vulnerability, procedure year and dual-enrollment in Medicaid
Reference group = FFS
The graphical assessment of the proportional hazards assumption demonstrated acceptable parallelism overall (Figure S2). At one year, MA patients had a lower comorbidity-adjusted risk of all adverse outcomes of interest including all-cause mortality (HR: 0.94, 95% CI: 0.92–0.96), all-cause readmission (HR: 0.86, 95% CI: 0.85–0.88), stroke (HR: 0.90, 95% CI: 0.87–0.94), and aortic valve reintervention (HR: 0.86, 95% CI: 0.76–0.96). Findings were similar in the unadjusted and age and sex adjusted models, and differences in one-year outcomes persisted after adjustment for region, rurality, social vulnerability, and dual-enrollment in Medicaid (Table 4).
Table 4:
Comparison of 1-year Outcomes in MA vs FFS Patients
| Outcomes | MA Patients n=113025 |
FFS Patients n=204605 |
Unadjusted HR† (95% CI) |
Age + Sex Adjusted HR† (95% CI) |
Comorbidity Adjusted HR† (95% CI) |
Fully Adjusted* HR† (95% CI) |
|---|---|---|---|---|---|---|
| All-cause mortality | 11.22% (11.04%, 11.40%) | 11.80% (11.66%, 11.94%) | 0.95 [0.93, 0.97] | 0.97 [0.95, 0.99] | 0.94 [0.92, 0.96] | 0.93 [0.91, 0.95] |
| All-cause readmission | 38.06% (37.78%, 38.35%) | 41.97% (41.75%, 42.19%) | 0.88 [0.87, 0.89] | 0.89 [0.88, 0.90] | 0.86 [0.85, 0.88] | 0.85 [0.84, 0.86] |
| HF Hospitalization | 7.56% (7.40%, 7.72%) | 8.03% (7.91%, 8.15%) | 0.94 [0.91, 0.96] | 0.95 [0.92, 0.97] | 0.91 [0.89, 0.94] | 0.90 [0.88, 0.93] |
| Stroke | 4.33% (4.21%, 4.45%) | 4.44% (4.35%, 4.53%) | 0.97 [0.94, 1.01] | 0.98 [0.95, 1.02] | 0.90 [0.87, 0.94] | 0.89 [0.86, 0.93] |
| Disabling Stroke | 2.43% (2.34%, 2.52%) | 2.78% (2.70%, 2.85%) | 0.88 [0.84, 0.92] | 0.89 [0.85, 0.93] | 0.82 [0.78, 0.86] | 0.81 [0.77, 0.85] |
| TIA | 1.10% (1.04%, 1.17%) | 1.32% (1.27%, 1.37%) | 0.83 [0.78, 0.89] | 0.85 [0.79, 0.91] | 0.80 [0.75, 0.86] | 0.79 [0.73, 0.84] |
| Aortic Valve Reintervention | 0.46% (0.42%, 0.50%) | 0.52% (0.49%, 0.55%) | 0.89 [0.79, 0.99] | 0.88 [0.79, 0.98] | 0.86 [0.76, 0.96] | 0.84 [0.75, 0.94] |
FFS = fee-for-service; HF = heart failure; HR = hazard ratio; MA = Medicare Advantage; TIA = transient ischemic attack
The fully adjusted model included adjustment for age, sex, comorbidities, region, rurality, social vulnerability, procedure year and dual-enrollment in Medicaid
Reference group = FFS
As a sensitivity analysis, in-hospital outcomes were compared between MA and FFS (Table 5). After adjusting for comorbidities, MA patients had lower in-hospital rates of all-cause mortality (HR: 0.92, 95% CI: 0.85–0.99) and permanent pacemaker (HR: 0.78, 95% CI: 0.75–0.81), but similar rates of aortic valve reintervention (HR: 1.00, 95% CI: 0.78–1.27).
Table 5:
Sensitivity Analysis of Select In-Hospital Outcomes
| Outcomes | MA Patients n=113025 |
FFS Patients n=204605 |
Unadjusted HR(95% CI)† |
Age + Sex Adjusted HR(95% CI)† |
Comorbidity Adjusted HR(95% CI)† |
Fully Adjusted* HR(95% CI)† |
|---|---|---|---|---|---|---|
| All-cause mortality | 1182 (1.0%) | 2407 (1.2%) | 0.89 [0.83, 0.95] | 0.90 [0.84, 0.96] | 0.92 [0.85, 0.99] | 0.90 [0.84, 0.97] |
| Aortic Valve Reintervention | 103 (0.1%) | 193 (0.1%) | 0.97 [0.76, 1.23] | 0.96 [0.75, 1.22] | 1.00 [0.78, 1.27] | 0.95 [0.74, 1.22] |
| Permanent Pacemaker Implantation | 8789 (7.8%) | 17452 (8.5%) | 0.74 [0.72, 0.77] | 0.75 [0.72, 0.78] | 0.78 [0.75, 0.81] | 0.78 [0.75, 0.81] |
FFS = fee-for-service; HR = hazard ratio; MA = Medicare Advantage
The fully adjusted model included adjustment for age, sex, comorbidities, region, rurality, social vulnerability, procedure year and dual-enrollment in Medicaid
Reference group = FFS
Discussion
As TAVR is now the predominant treatment for severe aortic stenosis and more than 34 million older adults are now enrolled in MA plans, it is increasingly important to explore differences between MA and FFS to ensure that MA plans are delivering the high-quality, cost-effective care they were intended to provide. In this study, we compared utilization, patient and hospital characteristics and outcomes between MA and FFS patients undergoing TAVR and report three key findings. First, utilization of TAVR was similar in MA compared with FFS, and overall utilization of TAVR increased during the study period. Second, although there were important differences in sociodemographic characteristics between MA and FFS patients undergoing TAVR, there were no observed differences in the prevalence of baseline comorbidities. Third, patients enrolled in MA were more likely to be discharged home following a TAVR hospitalization and less likely to experience adverse outcomes at 30 days and one year—findings that were consistent in analyses that adjusted for sociodemographic factors and comorbidities.
Although we hypothesized that we might find lower TAVR utilization in MA due to prior authorization requirements, we observed similar utilization rates in MA and FFS throughout every quarter of the study period.15,29 These findings are in contrast to prior studies that have suggested lower utilization of healthcare services including ambulatory surgeries and procedures in MA compared to FFS.17,30 It is possible that the presence of clear guidelines regarding appropriate TAVR coverage in the CMS National Coverage Determination contributes to these findings.31 Although it would be informative to evaluate TAVR utilization in the context of disease prevalence (e.g., rates of severe aortic stenosis) in each group, this was not possible in Medicare claims. That said, the presence of similar utilization rates for SAVR and TAVR in MA and FFS supports the idea that treatment patterns for severe aortic stenosis are similar across insurance types.
The differences in sociodemographic factors between MA and FFS patients that we observed among TAVR patients are similar to those that have been seen in prior studies comparing MA and FFS patients. Notably, MA patients were more likely to be Black, metropolitan-based, and dual-enrolled in Medicaid.32,33 These differences are likely attributable to greater market penetration of MA in select communities.11,19 It has previously been challenging to ascertain whether differences in documented comorbidities between MA and FFS patients reflect true differences in patient acuity or differential coding practices, including possible “upcoding” in MA.34,35 In our study, we used only ICD-10 codes from the index TAVR hospitalization to define patient comorbidities, thus mitigating concerns about upcoding in MA, and found similar rates of comorbidities between MA and FFS. Our findings are similar to those of recent studies using only in-hospital claims or non-claims sources to ascertain comorbidities in MA and FFS and suggest that MA and FFS patients are likely not as different as was previously thought.24,36,37
As has been seen previously with MA patients hospitalized for joint replacement, heart failure, and stroke, MA patients undergoing TAVR are more likely to be discharged home and to use less intensive post-acute care services than FFS patients.14,38,39 This finding may be driven by greater emphasis on outpatient care delivery in MA. The persistence of better outcomes in MA compared to FFS in our fully-adjusted model, after adjusting for several sociodemographic factors and measured comorbidities, could suggest better overall care delivery in MA. Since hospital characteristics are generally similar between MA and FFS, this could be driven by more frequent referral to high-performing operators or better post-TAVR care coordination. The possibility of better in-hospital care for MA patients could be supported by the lower rates of select in-hospital outcomes (all-cause mortality, permanent pacemaker implantation) among MA patients compared to FFS patients in our study. An alternative explanation would be that MA plans only offer TAVR to the healthiest subset of their population. This argument is supported by the fact that MA achieves better outcomes than FFS despite enrolling a higher proportion of socioeconomically disadvantaged patients, who historically often experience worse outcomes.
Although our findings add valuable insights on care delivery in Medicare in the context of TAVR, our study should be interpreted in the context of some limitations. First, because we used Medicare claims to compare MA and FFS patients, there were limitations in the granularity of comorbidities that we could adjust for in our analysis. For example, we were unable to account for whether patients were at high-, intermediate- or low-risk for TAVR. It is possible that MA and FFS patients are different in ways that we were unable to measure, and that these differences contribute to differences in outcomes. Furthermore, although we restricted comorbidity ascertainment to index hospitalization ICD-10 codes to mitigate differential upcoding, residual differences in coding intensity between MA and FFS cannot be excluded. Similarly, although we found similar procedure utilization rates in MA and FFS, we were unable to assess utilization in the context of disease prevalence. Outcomes were ascertained using validated claims-based algorithms, and absolute event rates for some outcomes may reflect some degree of measurement error inherent to claims ascertainment. However, any such misclassification is unlikely to differ systematically between MA and FFS patients, and the relative differences in outcomes between groups therefore remain credible. Finally, we aggregated all MA plans together for the purposes of our study. Although this is often done for the analysis of MA patients, there could be differences between MA plans that we are not evaluating in this analysis.
In conclusion, in this large, contemporary study of Medicare patients undergoing TAVR, we found similar TAVR utilization between MA and FFS, and that patients undergoing TAVR in MA and FFS are similar in their baseline comorbidities. However, MA patients undergoing TAVR have better 30-day and one-year outcomes, possibly driven by differences in care delivery. Future studies are necessary to further explore causes of differences in MA and FFS outcomes and identify factors that could further improve outcomes overall.
Supplementary Material
What is Known:
Aortic stenosis is a disease that primarily affects Medicare-eligible adults.
Although over half of Medicare beneficiaries are enrolled in privately administered Medicare Advantage (MA) plans, it is unclear how care delivery in Medicare influences patient outcomes after transcatheter aortic valve replacement (TAVR).
What the Study Adds:
Although we found similar clinical profiles between MA and fee-for-service Medicare patients undergoing TAVR, MA patients appeared to experience better 30-day and one-year outcomes and were more likely to be discharged home.
Utilization of TAVR among MA and fee-for-service beneficiaries was similar.
Although differences in care delivery may contribute to these findings, future studies are needed to better understand drivers of the differences in outcomes.
Disclosures:
- Dr. Lalani receives funding from the NIH T32 HL160522 and reports consulting fees from the Center for Medicare & Medicaid Services, outside of the submitted work.
- Dr. Oseran receives research support from the National Heart, Lung, and Blood Institute (K23HL173636). He serves as a consultant to agilon health outside the submitted work.
- Dr. Kim receives funding from the NIH T32 HL160522.
- Dr. Cohen receives research grant support from Abbott, Edwards Lifesciences, Medtronic, Boston Scientific, CathWorks, Philips, Zoll Medical, Corvia, and JC Medical and consulting income from Abbott, Edwards Lifesciences, Medtronic, Boston Scientific, and Heartbeam.
- Dr. Wadhera is the Principal Investigator of grants from National Heart, Lung, and Blood Institute (R01HL164561, R01HL174549) and the National Institute of Nursing Research (R01NR021686) at the National Institutes of Health, American Heart Association Established Investigator Award (24EIA1258487), and the Donaghue Foundation, and has also received support from the Richard A. and Susan F. Smith Center for Outcomes Research, where he serves as the Associate Director. He also reports serving as a consultant to Abbott Vascular and Chambercardio outside the submitted work.
- Dr. Yeh reports the following disclosures: Research Grants: Abbott Vascular, Boston Scientific, Edwards Lifesciences, Elixir Medical, JenaValve, Medtronic, Siemens; Consulting: Abbott Vascular, Boston Scientific, CathWorks, Centers for Medicare and Medicaid Services, Edwards Lifesciences, Elixir Medical, JenaValve, Magenta Medical, Medtronic; Employment: Special Government Employee, US Food and Drug Administration
Abbreviations:
- FFS
fee-for-service
- MA
Medicare Advantage
- SAVR
surgical aortic valve replacement
- SMD
standardized mean difference
- TAVR
transcatheter aortic valve replacement
References
- 1.Small AM, Yutzey KE, Binstadt BA, Voigts Key K, Bouatia-Naji N, Milan D, Aikawa E, Otto CM, St. Hilaire C, on behalf of the American Heart Association Council on Genomic and Precision Medicine; Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation; and Council on Cardiovascular and Stroke Nursing. Unraveling the Mechanisms of Valvular Heart Disease to Identify Medical Therapy Targets: A Scientific Statement From the American Heart Association. Circulation. 2024;150:e109–e128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brener MI, Chuang ML, Nishimura R, Hamid NB, Snyder C, Thompson JB, Ungarten J, Takahashi K, Poterucha TJ, Elias P, et al. Population Prevalence of Valvular Heart Disease in the United States: The PREVUE-VALVE Study. J Am Coll Cardiol. 2026;87:3227–3239. [DOI] [PubMed] [Google Scholar]
- 3.Makkar RR, Fontana GP, Jilaihawi H, Kapadia S, Pichard AD, Douglas PS, Thourani VH, Babaliaros VC, Webb JG, Herrmann HC, et al. Transcatheter Aortic-Valve Replacement for Inoperable Severe Aortic Stenosis. N Engl J Med. 2012;366:1696–1704. [DOI] [PubMed] [Google Scholar]
- 4.Mack MJ, Leon MB, Thourani VH, Pibarot P, Hahn RT, Genereux P, Kodali SK, Kapadia SR, Cohen DJ, Pocock SJ, et al. Transcatheter Aortic-Valve Replacement in Low-Risk Patients at Five Years. N Engl J Med. 2023;389:1949–1960. [DOI] [PubMed] [Google Scholar]
- 5.Makkar RR, Thourani VH, Mack MJ, Kodali SK, Kapadia S, Webb JG, Yoon S-H, Trento A, Svensson LG, Herrmann HC, et al. Five-Year Outcomes of Transcatheter or Surgical Aortic-Valve Replacement. N Engl J Med. 2020;382:799–809. [DOI] [PubMed] [Google Scholar]
- 6.Popma JJ, Deeb GM, Yakubov SJ, Mumtaz M, Gada H, O’Hair D, Bajwa T, Heiser JC, Merhi W, Kleiman NS, et al. Transcatheter Aortic-Valve Replacement with a Self-Expanding Valve in Low-Risk Patients. N Engl J Med. 2019;380:1706–1715. [DOI] [PubMed] [Google Scholar]
- 7.Reardon MJ, Mieghem NMV, Popma JJ, Kleiman NS, Søndergaard L, Mumtaz M, Adams DH, Deeb GM, Maini B, Gada H, et al. Surgical or Transcatheter Aortic-Valve Replacement in Intermediate-Risk Patients. N Engl J Med. 2017;376:1321–1331. [DOI] [PubMed] [Google Scholar]
- 8.Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, Gleason TG, Buchbinder M, Hermiller J, Kleiman NS, et al. Transcatheter Aortic-Valve Replacement with a Self-Expanding Prosthesis. N Engl J Med. 2014;370:1790–1798. [DOI] [PubMed] [Google Scholar]
- 9.Carroll JD, Mack MJ, Vemulapalli S, Herrmann HC, Gleason TG, Hanzel G, Deeb GM, Thourani VH, Cohen DJ, Desai N, et al. STS-ACC TVT Registry of Transcatheter Aortic Valve Replacement. J Am Coll Cardiol. 2020;76:2492–2516. [DOI] [PubMed] [Google Scholar]
- 10.Agarwal R, Connolly J, Gupta S, Navathe AS. Comparing Medicare Advantage And Traditional Medicare: A Systematic Review. Health Aff (Millwood). 2021;40:937–944. [DOI] [PubMed] [Google Scholar]
- 11.Neuman P, Jacobson GA. Medicare Advantage Checkup. N Engl J Med. 2018;379:2163–2172. [DOI] [PubMed] [Google Scholar]
- 12.Rosenzweig JL, Taitel MS, Norman GK, Moore TJ, Turenne W, Tang P. Diabetes disease management in Medicare Advantage reduces hospitalizations and costs. Am J Manag Care. 2010;16:e157–162. [PubMed] [Google Scholar]
- 13.Ayanian JZ, Landon BE, Zaslavsky AM, Saunders RC, Pawlson LG, Newhouse JP. Medicare beneficiaries more likely to receive appropriate ambulatory services in HMOs than in traditional medicare. Health Aff (Millwood). 2013;32:1228–1235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Huckfeldt PJ, Escarce JJ, Rabideau B, Karaca-Mandic P, Sood N. Less Intense Postacute Care, Better Outcomes For Enrollees In Medicare Advantage Than Those In Fee-For-Service. Health Aff (Millwood). 2017;36:91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jacobson G, Neuman T. Prior Authorization in Medicare Advantage Plans: How Often Is It Used? [Internet]. KFF. 2018. [cited 2025 July 25];Available from: https://www.kff.org/medicare/issue-brief/prior-authorization-in-medicare-advantage-plans-how-often-is-it-used/ [Google Scholar]
- 16.Jacobson G, Trilling A, Neuman T, Damico A, Gold M. Medicare Advantage Hospital Networks: How Much Do They Vary? Available from: https://www.kff.org/affordable-care-act/medicare-advantage-hospital-networks-how-much-do-they-vary/
- 17.Landon BE, Zaslavsky AM, Saunders RC, Pawlson LG, Newhouse JP, Ayanian JZ. Analysis Of Medicare Advantage HMOs compared with traditional Medicare shows lower use of many services during 2003–09. Health Aff (Millwood). 2012;31:2609–2617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fuglesten Biniek J, Sroczynski N, Freed M, Neuman T. Medicare Advantage Insurers Made Nearly 50 Million Prior Authorization Determinations in 2023 [Internet]. KFF. 2025. [cited 2026 Jan 22];Available from: https://www.kff.org/medicare/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2023/ [Google Scholar]
- 19.Freed M, Biniek JF, Damico A, Neuman T. Medicare Advantage in 2024: Enrollment Update and Key Trends [Internet]. KFF. 2024. [cited 2025 Jan 3];Available from: https://www.kff.org/medicare/issue-brief/medicare-advantage-in-2024-enrollment-update-and-key-trends/ [Google Scholar]
- 20.Other Chronic Health, Mental Health, and Potentially Disabling Conditions [Internet]. Chronic Conditions Data Warehouse. [cited 2024 Aug 11];Available from: https://www2.ccwdata.org/condition-categories-other [Google Scholar]
- 21.Meyers DJ, Trivedi AN. Medicare Advantage Chart Reviews Are Associated With Billions in Additional Payments for Some Plans. Med Care. 2021;59:96–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Murrin S. Some Medicare Advantage Companies Leveraged Chart Reviews and Health Risk Assessments To Disproportionately Drive Payments. 2021;Available from: https://oig.hhs.gov/documents/evaluation/2794/OEI-03-17-00474-Complete%20Report.pdf
- 23.Oseran AS, Wadhera RK, Orav EJ, Figueroa JF. Effect of Medicare Advantage on Hospital Readmission and Mortality Rankings. Ann Intern Med. 2023;176:480–488. [DOI] [PubMed] [Google Scholar]
- 24.Landon BE, Anderson TS, Curto VE, Cram P, Fu C, Weinreb G, Zaslavsky AM, Ayanian JZ. Association of Medicare Advantage vs Traditional Medicare With 30-Day Mortality Among Patients With Acute Myocardial Infarction. JAMA. 2022;328:2126–2135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Strom JB, Zhao Y, Faridi KF, Tamez H, Butala NM, Valsdottir LR, Curtis J, Brennan JM, Shen C, Boulware M, et al. Comparison of Clinical Trials and Administrative Claims to Identify Stroke Among Patients Undergoing Aortic Valve Replacement: Findings From the EXTEND Study. Circ Cardiovasc Interv. 2019;12:e008231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Butala NM, Strom JB, Faridi KF, Kazi DS, Zhao Y, Brennan JM, Popma JJ, Shen C, Yeh RW. Validation of Administrative Claims to Ascertain Outcomes in Pivotal Trials of Transcatheter Aortic Valve Replacement. JACC Cardiovasc Interv. 2020;13:1777–1785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Butala NM, Lalani C, Tale A, Song Y, Kolte D, Baron S, Strom J, Cohen DJ, Yeh RW. Use of Claims to Assess Outcomes and Treatment Effects in the Evolut Low Risk Trial. Circ Cardiovasc Interv. 2025;18:e014592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lalani C, Medina F, Oseran AS, Liang L, Song Y, Butala NM, Kazi DS, Cohen DJ, Strom JB, Wadhera RK, et al. Validation of Medicare Advantage Claims for Long-Term Outcome Assessment in Low-Risk Aortic Valve Replacement. Circ Cardiovasc Qual Outcomes. 2025;18:e011991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Schwartz AL, Brennan TA, Verbrugge DJ, Newhouse JP. Measuring the Scope of Prior Authorization Policies: Applying Private Insurer Rules to Medicare Part B. JAMA Health Forum. 2021;2:e210859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Curto V, Einav L, Finkelstein A, Levin J, Bhattacharya J. Health Care Spending and Utilization in Public and Private Medicare. Am Econ J Appl Econ. 2019;11:302–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.NCD - Transcatheter Aortic Valve Replacement (TAVR) (20.32) [Internet]. [cited 2026 Jan 22];Available from: https://www.cms.gov/medicare-coverage-database/view/ncd.aspx?NCDId=355
- 32.Offiaeli K, Meyers DJ, Macneal E, Johnston KJ, Brown-Podgorski B, Roberts ET. Patient-Reported Characteristics Across Dual-Eligible Medicare Advantage Plan Types. JAMA Network Open. 2025;8:e255791. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Aggarwal R, Gondi S, Wadhera RK. Comparison of Medicare Advantage vs Traditional Medicare for Health Care Access, Affordability, and Use of Preventive Services Among Adults With Low Income. JAMA Netw Open. 2022;5:e2215227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kronick R, Welch WP. Measuring coding intensity in the Medicare Advantage program. Medicare Medicaid Res Rev. 2014;4:mmrr2014.004.02.a06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Jacobs PD, Kronick R. Getting What We Pay For: How Do Risk-Based Payments to Medicare Advantage Plans Compare with Alternative Measures of Beneficiary Health Risk? Health Serv Res. 2018;53:4997–5015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Oseran AS, Aggarwal R, Figueroa J, Joynt Maddox KE, Landon BE, Wadhera RK. Prevalence of Chronic Medical Conditions Among Medicare Advantage and Traditional Medicare Beneficiaries. Ann Intern Med. 2025;178:327–335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Oseran AS, Aggarwal R, Wadhera RK. Cardiovascular Risk Factor Management in Medicare Advantage and Traditional Medicare. Circ Cardiovasc Qual Outcomes. 2025;18:e012143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Kumar A, Rahman M, Trivedi AN, Resnik L, Gozalo P, Mor V. Comparing post-acute rehabilitation use, length of stay, and outcomes experienced by Medicare fee-for-service and Medicare Advantage beneficiaries with hip fracture in the United States: A secondary analysis of administrative data. PLoS Med. 2018;15:e1002592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Figueroa JF, Wadhera RK, Frakt AB, Fonarow GC, Heidenreich PA, Xu H, Lytle B, DeVore AD, Matsouaka R, Yancy CW, et al. Quality of Care and Outcomes Among Medicare Advantage vs Fee-for-Service Medicare Patients Hospitalized With Heart Failure. JAMA Cardiol. 2020;5:1349–1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.