A new RAND Corporation study on bundled payments in the private sector was published in the March 2021 issue of Health Affairs.¹ The study analyzed data from over 2,000 procedures performed as part of a direct payment program by Carrum Health between 2016 and 2020,² and found significant savings from this bundled payment program, without any significant changes in quality.³ This study adds important evidence to the argument in favor of bundled payments and is especially important because it examines the under-studied area of bundled payment models from commercial payment systems.

While fee-for-service payments⁴ still dominate U.S. healthcare reimbursement models, bundled payments, also known as episode-based payments, offer an alternative payment structure that has received increasing attention in recent years, especially from the Centers for Medicare & Medicaid Services (CMS).⁵ Bundled payments are single payments for all healthcare services corresponding to a specific treatment or condition.⁶ Healthcare providers who accept bundled payments from a payor assume the financial risk for all costs of medical services that exceed the bundled payment amount for the particular treatment or condition.⁷ Bundled payments operate under the assumption that the model will incentivize providers to lower costs and reduce unnecessary services.⁸

The first modern iteration of bundled payments from CMS was the Medicare Participating Heart Bypass Center Demonstration, which took place from 1991 through 1996.⁹ The short-lived test, which involved only four hospitals, showed promising results.¹⁰ The hospitals in the program were able to significantly lower costs related to bypass surgery while simultaneously maintaining quality.¹¹ However, later research into the demonstration project found that cost reductions actually came from nursing management and pharmacy changes.¹² In 2006, bundled payments gained significant attention when Geisinger Health System implemented its “ProvenCare” model, which packaged coronary heart bypass surgery into one bundled price.¹³ The model proved much more successful than originally anticipated, and with extraordinary quality results, including a significantly shorter length of stay for patients in the model.¹⁴

The results of the ProvenCare program helped spur considerable support for more widespread use of bundling. Over the next ten years, beginning with the 2010 Patient Protection and Affordable Care Act (ACA), CMS made significant moves toward bundled payment models.¹⁵ First, CMS created the Bundled Payments for Care Improvement (BPCI) Initiative, which created four broad models of care wherein payments are bundled for a particular type of care.¹⁶ Subsequently, CMS implemented a payment system for joint replacement surgery, called the Comprehensive Care for Joint Replacement (CJR) model, in 2016.¹⁷ Joint replacements are the most common surgery among Medicare beneficiaries,¹⁸ and the cost and quality can vary significantly.¹⁹ This model has been relatively less successful, with recent evidence indicating that the only type of CJR model clinical episode that results in cost savings is lower extremity joint replacement.²⁰ While these savings: were lower than predicted; may be partially explained by participant demographics; and, may hold less benefit for newer program participants, positive early results from bundled payment programs, including overall cost savings to participating hospitals,²¹ encouraged CMS to pursue bundled payment models further and create a new BPCI Advanced Model in 2018.²²

The new RAND study on private bundled payments showed significant cost savings. The mean prices for spinal fusion, joint replacement, and bariatric surgery – the three most common procedures – decreased by 29.1% (from $98,944 to $69,780), 18.4% (from $38,498 to $31,355), and 6% (from $29,225 to $27,625), respectively.²³ Price variation also decreased significantly for all three procedures.²⁴ A reduction in episode prices of $4,229 was observed for all three procedures, with 85% of these savings going to the self-insured employers. However, both employer and patient spending decreased, with patients seeing the greatest reductions in terms of relative costs.²⁵ Further, savings grew over time: prices decreased by $4,402 in the first year of implementing the bundled model and by $6,225 in the second year and thereafter.²⁶ Employers and patients saw a similar rate of decrease over time in their costs, from $3,712 in year one to $5,963 after and from $499 in year one to $550 after, respectively.²⁷ Researchers tested their hypothesis both with and without accounting for patient-level covariates in their model and observed similar results, suggesting that any patient characteristics that were not accounted for and model choice likely did not confound study results.²⁸

Despite the potential cost savings, published research has focused primarily on CMS and public payor programs, leaving private payor bundled payments severely under-studied.²⁹ This subject has been difficult to research mainly due to a lack of coordinated incentives for providers, payors, and patients which has led to numerous implementation issues.³⁰ Payment systems have largely been unable to identify providers willing to participate in a bundled payment system, connect willing providers with payors capable of processing these payments, and encourage patients to utilize these bundled arrangements when receiving care.³¹ While the potential savings would benefit employers, payors, and patients, the issue of building a solid evidence base becomes a double-edged sword because, according to the RAND study’s lead author, “employers don't want to be the first ones to adopt something this new…[but] if employers aren’t clamoring for these types of models and want to pass rising healthcare costs down to employees, there’s not a large incentive for private payers to invest in these plans.”³² Future savings, however, seem to be not just probable, but within reach, for private payors – savings which also will be passed down to employers and patients. As the public market continues to explore new bundled payment models, studies such as RAND’s will be integral in encouraging private payors, employers, and patients to follow.