The Medicare Bundled Payments for Care Improvement (BPCI) Model 2 was not associated with improved quality performance or outcomes among individuals with heart failure (HF) who were hospitalized, according to results of a study published in JAMA Network Open.
The BPCI program was developed by the Centers for Medicare & Medicaid Services Innovation Center, aiming to lower costs, improve quality of care for patients, and align incentives for physicians and coordination of care. The Model 2 bundled the payment for acute and post-acute care for 30-, 60-, and 90-days post-hospital discharge. The study authors said previous studies demonstrated that bundled payments for care quality had been associated with a decline in cost, without a reduction in quality outcomes for some conditions, including joint replacement.
Investigators used data from the American Heart Association's Get With The Guidelines–Heart Failure (GWTG-HF) registry and Medicare claims, with data being linked between the 2 sources. They compared changes in the process of care and outcomes for those hospitalized with HF whose hospital used BPCI and those whose hospitals did not.
3 Key Takeaways
- The BPCI program, designed to lower costs and enhance care quality, particularly in the post-acute phase, did not demonstrate notable positive impacts on HF-related measures.
- ProcDespite the implementation of BPCI, there were no meaningful changes in achieving various process-of-care measures, except for a decreased likelihood of evidence-based specific β-blocker prescription at discharge.
- While BPCI was associated with a decrease in the odds of in-hospital mortality, it did not significantly influence other secondary endpoints such as length of stay longer than 4 days, referral to cardiac rehabilitation, or discharge to hospice.
Individuals were included if they had a primary HF diagnosis and were admitted to the hospital from November 1, 2008, to August 31, 2018. Race and ethnicity were self-reported, according to the study authors. If there were multiple hospitalizations for a patient, the first hospitalization was used and the others were excluded. Investigators included 197,100 individuals across 562 difference sites in the analysis.
During the study period, there were 8721 individuals hospitalized to 23 BPCI hospitals and 94,530 individuals hospitalized to 224 non-BPCI hospitals, according to the study authors. Those in BPCI hospitals were significantly more likely to be in rural locations, more likely to be in the West, more likely to be in teaching hospitals, and less likely to have a heart transplant center when compared to those in non-BPCI hospitals, according to the results. The investigators found minimal differences in sex, race, ethnicity, age, insurance status, disease category, and comorbidity burden between the 2 types of hospitals.
The study authors reported that there were no meaningful changes in annual HF volume pre- and post-implementation of BPCI between BPCI hospitals and non-BPCI hospitals. Additionally, there were no significant differences in length of stay for those in BPCI and non-BPCI hospitals. However, they noted that the length of stay declined among BPCI and non-BPCI hospitals pre- and post-implementation, with a greater reduction for the BPCI hospitals. Discharge to nursing facilities also declined among BPCI hospitals and increased among non-BPCI hospitals pre- and post-implementation.
In a multivariable regression analysis, the results showed that BPCI was not associated with a change in achieving various process-of-care measures, with the exception of a decreased odds of evidence-based specific β-blocker at discharge, according to the study authors. Primary endpoints that were not associated with change included angiotensin-converting enzyme inhibitors/angiotensin receptor blockers or angiotensin receptor-neprilysin inhibitor at discharge, aldosterone antagonist at discharge, and cardiac resynchronization therapy (CRT) defibrillator or CRT pacemaker placed or prescribed at discharge. They found that the results were consistent across sex, race, and ethnicity, except for HF education, which showed a decrease in odds of at least 60 minutes for those who were Hispanic or Asian.
Furthermore, BPCI was associated with a decrease in odds of in-hospital mortality, but was not associated with length of stay longer than 4 days, referral to cardiac rehabilitation, discharge to hospice or home hospice, or other secondary endpoints. The results were consistent across subcategories, except in-hospital mortality, which showed BPCI was not associated with change in odds for Black individuals or female individuals.
Reference
Oddleifson DA, Holmes DN, Alhanti B, Xu X, et al. Bundled Payments for Care Improvement and Quality of Care and Outcomes in Heart Failure. JAMA Cardiol. 2024. doi:10.1001/jamacardio.2023.5009
