Identifying Sources of Inter-Hospital Variation in Episode Spending for Sepsis Care.

OBJECTIVE: To evaluate inter-hospital variation in 90-day total episode spending for sepsis, estimate the relative contributions of each component of spending, and identify drivers of spending across the distribution of episode spending on sepsis care. DATA SOURCES/STUDY SETTING: Medicare fee-for-service claims for beneficiaries (n=324,694) discharged from acute care hospitals for sepsis, defined by MS-DRG, between October 2014 and September 2018. RESEARCH DESIGN: Multiple linear regression with hospital-level fixed effects was used to identify average hospital differences in 90-day episode spending. Separate multiple linear regression and quantile regression models were used to evaluate drivers of spending across the episode spending distribution. RESULTS: The mean total episode spending among hospitals in the most expensive quartile was $30,500 compared with $23,150 for the least expensive hospitals ( P <0.001). Postacute care spending among the most expensive hospitals was almost double that of least expensive hospitals ($7,045 vs. $3,742), accounting for 51% of the total difference in episode spending between the most expensive and least expensive hospitals. Female patients, patients with more comorbidities, urban hospitals, and BPCI-A-participating hospitals were associated with significantly increased episode spending, with the effect increasing at the right tail of the spending distribution. CONCLUSION: Inter-hospital variation in 90-day episode spending on sepsis care is driven primarily by differences in post-acute care spending.

Bundled Payments for Care Improvement Efficacy Across 3 Common Operations.

OBJECTIVE: The aim of this study was to evaluate associations between hospital participation in Bundled Payments for Care Improvement (BPCI) and 30-day total episode and post-acute care spending for lower extremity joint replacement (LEJR), coronary artery bypass graft (CABG), and colec-tomy. SUMMARY BACKGROUND DATA: BPCI has been shown to reduce spending for LEJR episodes largely from reductions in post-acute care. However, BPCI efficacy in other common elective procedures, including CABG and colec-tomy, remains unclear. It is also unknown whether post-acute care spending reductions drive total spending reductions outside of LEJR. METHODS: Retrospective cohort study using 100% Medicare claims data to identify BPCI (312 total) and non-BPCI (1,977 total) acute care hospitals from January 1, 2010 to November 30, 2016 with Medicare-enrolled patient discharges for at least one of the following BPCI episodes: LEJR (454,369 episodes), CABG (107,307 episodes), or colectomy (73,717 episodes). Along with difference-in-differences (DiD) analysis, we constructed generalized synthetic controls in the presence of nonparallel trends to estimate associations between BPCI participation and 30-day total and post-acute care spending. RESULTS: DiD estimates indicated reduced spending for LEJR (-$541.6 [95% confidence interval (CI): -718.0 to -365.3]) and colectomy (-$582.1 [95% CI: -927.3 to -236.8]) but not CABG (-$268.9 [95% CI: -831.5 to 293.7]). Generalized synthetic control estimates indicated reduced spending for LEJR (-$795.3 [95% CI: -10,22.1 to -582.2]) but not colectomy (-$251.3 [95% CI: -997.9 to 335.2]) or CABG (-$257.8 [95% CI: -10,24.6 to 414.8]).Post-acute care comprised 42.6% of LEJR spending reductions and 53.0% of colectomy spending reductions. CONCLUSIONS: BPCI participation was associated with significant spending reductions for LEJR and colectomy but not CABG. We conclude that BPCI has episode-dependent efficacy, largely determined by post-acute care.

Association of Hospital Participation in Bundled Payments for Care Improvement Advanced With Medicare Spending and Hospital Incentive Payments.

Importance: Bundled Payments for Care Improvement Advanced (BPCI-A) is a Centers for Medicare & Medicaid Services (CMS) initiative that aims to produce financial savings by incentivizing decreases in clinical spending. Incentives consist of financial bonuses from CMS to hospitals or penalties paid by hospitals to CMS. Objective: To investigate the association of hospital participation in BPCI-A with spending, and to characterize hospitals receiving financial bonuses vs penalties. Design, Setting, and Participants: Difference-in-differences and cross-sectional analyses of 4 754 139 patient episodes using 2013-2019 US Medicare claims at 694 participating and 2852 nonparticipating hospitals merged with hospital and market characteristics. Exposures: BPCI-A model years 1 and 2 (October 1, 2018, through December 31, 2019). Main Outcomes and Measures: Hospitals' per-episode spending, CMS gross and net spending, and the incentive allocated to each hospital. Results: The study identified 694 participating hospitals. The analysis observed a -$175 change in mean per-episode spending (95% CI, -$378 to $28) and an aggregate spending change of -$75.1 million (95% CI, -$162.1 million to $12.0 million) across the 428 670 episodes in BPCI-A model years 1 and 2. However, CMS disbursed $354.3 million (95% CI, $212.0 million to $496.0 million) more in bonuses than it received in penalties. Hospital participation in BPCI-A was associated with a net loss to CMS of $279.2 million (95% CI, $135.0 million to $423.0 million). Hospitals in the lowest quartile of Medicaid days received a mean penalty of $0.41 million; (95% CI, $0.09 million to $0.72 million), while those in the highest quartile received a mean bonus of $1.57 million; (95% CI, $1.09 million to $2.08 million). Similar patterns were observed for hospitals across increasing quartiles of Disproportionate Share Hospital percentage and of patients from racial and ethnic minority groups. Conclusions and Relevance: Among US hospitals measured between 2013 and 2019, participation in BPCI-A was significantly associated with an increase in net CMS spending. Bonuses accrued disproportionately to hospitals providing care for marginalized communities.

Improving target price calculations in Medicare bundled payment programs.

OBJECTIVE: To compare the predictive accuracy of two approaches to target price calculations under Bundled Payments for Care Improvement-Advanced (BPCI-A): the traditional Centers for Medicare and Medicaid Services (CMS) methodology and an empirical Bayes approach designed to mitigate the effects of regression to the mean. DATA SOURCES: Medicare fee-for-service claims for beneficiaries discharged from acute care hospitals between 2010 and 2016. STUDY DESIGN: We used data from a baseline period (discharges between January 1, 2010 and September 30, 2013) to predict spending in a performance period (discharges between October 1, 2015 and June 30, 2016). For 23 clinical episode types in BPCI-A, we compared the average prediction error across hospitals associated with each statistical approach. We also calculated an average across all clinical episode types and explored differences by hospital size. DATA COLLECTION/EXTRACTION METHODS: We used a 20% sample of Medicare claims, excluding hospitals and episode types with small numbers of observations. PRINCIPAL FINDINGS: The empirical Bayes approach resulted in significantly more accurate episode spending predictions for 19 of 23 clinical episode types. Across all episode types, prediction error averaged $8456 for the CMS approach versus $7521 for the empirical Bayes approach. Greater improvements in accuracy were observed with increasing hospital size. CONCLUSIONS: CMS should consider using empirical Bayes methods to calculate target prices for BPCI-A.

Hospital Participation Decisions In Medicare Bundled Payment Program Were Influenced By Third-Party Conveners.

The Bundled Payments for Care Improvement initiative Advanced Model (BPCI Advanced) is a voluntary Medicare bundled payment model in which hospitals may participate with third-party conveners-private consulting firms that share in the financial risk built into the program. We found that nonteaching and for-profit status was associated with a higher probability of hospital partnership with third-party conveners in BPCI Advanced. Among hospitals participating in at least one inpatient clinical episode, hospitals that partnered with third-party conveners were more likely to select episodes with higher target prices: A $1,000 increase in episode target price was associated with a 1.66-percentage-point increase in the probability of episode participation in BPCI Advanced compared with a 0.72-percentage-point increase for participating hospitals without third-party conveners. Hospitals with third-party conveners also were more likely than those without them to select inpatient clinical episodes with greater opportunities to reduce spending on postacute care and readmissions. These findings have important implications for understanding the role of private consulting firms in the program and for planning potential program modifications in the future.

Target Prices Influence Hospital Participation And Shared Savings In Medicare Bundled Payment Program.

To reduce episode spending for twenty-nine predefined clinical conditions, the Centers for Medicare and Medicaid Services (CMS) implemented the voluntary Bundled Payments for Care Improvement (BPCI) Advanced Model program in 2018. Under this program, hospitals gain or lose revenue depending on their episode spending relative to target prices set by CMS. The relationship between target prices and hospital participation in BPCI Advanced is unknown, as are the financial implications for CMS. Using Medicare claims, we estimate that each $1,000 increase in target prices increased the probability of participation by 0.78 percentage points across all episodes. We then used Medicare claims before the start of BPCI Advanced to evaluate mean reversion, or the tendency for episode spending at individual hospitals to move closer to average episode spending over time, especially for hospitals having higher target prices. Hospitals with spending that was 10 percent more than target prices at baseline could expect spending to decline by 7.43 percent in the performance period, hospitals with spending that was 20 percent more saw spending decline by 9.80 percent, and hospitals with spending that was 30 percent more saw spending decline by 11.93 percent. Our findings suggest that CMS will end up paying substantial bonuses to hospitals that resulted from mean reversion rather than from meaningful reductions in costs.

A Comparison of Estimated Cost Savings from Potential Reductions in Hospital-Acquired Conditions to Levied Penalties Under the CMS Hospital-Acquired Condition Reduction Program.

BACKGROUND: The Hospital-Acquired Condition Reduction Program (HACRP) from the Centers for Medicare & Medicaid Services (CMS) reduces Medicare payments to hospitals with high rates of hospital-acquired conditions (HACs) by 1% each year. It is not known how the savings accruing to CMS from such penalties compare to savings resulting from a reduction in HACs driven by this program. This study compares the reported savings to CMS from financial penalties levied under the HACRP with savings resulting from potential reductions in HACs. METHODS: Using a random sample of 20% of Medicare claims data (January 1, 2009-September 30, 2014), the research team evaluated the association between HACs and 90-day episode spending (adjusted to 2015 dollars), then estimated potential annual savings to CMS if there was a relative decrease in incidence of all HACs by 1%-20%. These savings were then compared to the actual collected HACRP penalties reported by CMS in 2015. RESULTS: All HACs were associated with significant increases in total 90-day episode spending, ranging from $3,183 for iatrogenic pneumothorax to $21,654 for postoperative hip fracture. The total estimated savings to Medicare from potential reduction in all HACs ranged from $2.2 million to $44 million per year, an amount much lower than the $361 million in penalties levied on hospitals per year for HACs. CONCLUSION: The penalties levied under the HACRP far exceed the potential cost savings accruing from a 1%-20% reduction in HACs that might result from hospitals' efforts in response to the program.

Evaluation of US Hospital Episode Spending for Acute Inpatient Conditions After the Patient Protection and Affordable Care Act.

Importance: Under the Patient Protection and Affordable Care Act (ACA), US hospitals were exposed to a number of reforms intended to reduce spending, many of which, beginning in 2012, targeted acute care hospitals and often focused on specific diagnoses (eg, acute myocardial infarction, heart failure, and pneumonia) for Medicare patients. Other provisions enacted in the ACA and under budget sequestration (beginning in 2013) mandated Medicare fee cuts. Objective: To evaluate the association between the enactment of ACA reforms and 30-day price-standardized hospital episode spending. Design, Setting, and Participants: This policy evaluation included index discharges between January 1, 2008, and August 31, 2015, from a national random 20% sample of Medicare beneficiaries. Data analysis was performed from February 1, 2019 to July 8, 2020. Exposure: Payment reforms after passage of the ACA. Main Outcomes and Measures: 30-day price-standardized episode payments. Three alternative estimation approaches were used to evaluate the association between reforms following the ACA and episode spending: (1) a difference-in-difference (DID) analysis among acute care hospitals, comparing spending for diagnoses commonly targeted by ACA programs with nontargeted diagnoses; (2) a DID analysis comparing acute care hospitals and critical access hospitals (not exposed to reforms); and (3) a generalized synthetic control analysis, comparing acute care and critical access hospitals. Supplemental analysis examined the degree to which Medicare fee cuts contributed to spending reductions. Results: A total of 7 634 242 index discharges (4 525 630 [59.2%] female patients; mean [SD] age, 79.31 [8.02] years) were included. All 3 approaches found that reforms following the ACA were associated with a significant reduction in episode spending. The DID estimate comparing targeted and untargeted diagnoses suggested that reforms following the ACA were associated with a -$431 (95% CI, -$492 to -$369; -2.87%) change in total spending, while the generalized synthetic control analysis suggested that reforms were associated with a -$1232 (95% CI, -$1488 to -$965; -10.12%) change in total episode spending, amounting in a total annual savings of $5.68 billion. Cuts to Medicare fees accounted for most of these savings. Conclusions and Relevance: In this policy evaluation, the ACA was associated with large reductions in US hospital episode spending.

Changes in hospital safety following penalties in the US Hospital Acquired Condition Reduction Program: retrospective cohort study.

OBJECTIVE: To evaluate the association between hospital penalization in the US Hospital Acquired Condition Reduction Program (HACRP) and subsequent changes in clinical outcomes. DESIGN: Regression discontinuity design applied to a retrospective cohort from inpatient Medicare claims. SETTING: 3238 acute care hospitals in the United States. PARTICIPANTS: Medicare fee-for-service beneficiaries discharged from acute care hospitals between 23 July 2014 and 30 November 2016 and eligible for at least one targeted hospital acquired condition (n=15 470 334). INTERVENTION: Hospital receipt of a penalty in the first year of the HACRP. MAIN OUTCOME MEASURES: Episode level count of targeted hospital acquired conditions per 1000 episodes, 30 day readmissions, and 30 day mortality. RESULTS: Of 724 hospitals penalized under the HACRP in fiscal year 2015, 708 were represented in the study. Mean counts of hospital acquired conditions were 2.72 per 1000 episodes for penalized hospitals and 2.06 per 1000 episodes for non-penalized hospitals; 30 day readmissions were 14.4% and 14.0%, respectively, and 30 day mortality was 9.0% for both hospital groups. Penalized hospitals were more likely to be large, teaching institutions, and have a greater share of patients with low socioeconomic status than non-penalized hospitals. HACRP penalties were associated with a non-significant change of -0.16 hospital acquired conditions per 1000 episodes (95% confidence interval -0.53 to 0.20), -0.36 percentage points in 30 day readmission (-1.06 to 0.33), and -0.04 percentage points in 30 day mortality (-0.59 to 0.52). No clear patterns of clinical improvement were observed across hospital characteristics. CONCLUSIONS: Penalization was not associated with significant changes in rates of hospital acquired conditions, 30 day readmission, or 30 day mortality, and does not appear to drive meaningful clinical improvements. By disproportionately penalizing hospitals caring for more disadvantaged patients, the HACRP could exacerbate inequities in care.