AHRQ Delivery System Research: Study Snapshot
The Agency for Healthcare Research and Quality awarded Integrated Healthcare Association and the RAND Corporation a demonstration and evaluation grant to develop and implement a bundled payment program, including health plans, hospitals, ambulatory surgery centers, physician organizations, physicians, and technology vendors. This study snapshot presents lessons learned from the demonstration.
By M. Susan Ridgely, David De Vries, Kevin Bozic, and Peter S. Hussey.
Traditional fee-for-service (FFS) payment systems create little or no incentive to coordinate care across providers and settings, and contribute to overuse of services. The Affordable Care Act promotes payment reform by encouraging the testing of various new models, including bundled payments. Under this approach, a group of providers receives a fixed payment from participating health plans. The payment is designed to cover the average cost of a defined "bundle" of services related to a procedure or course of treatment. In addition to the treatment or procedure, the bundle may include standard care needed before and afterward and a defined "warranty" period during which the provider is responsible for the costs of treating associated complications. The goal is to eliminate duplicative and unnecessary care, encourage greater coordination of care across settings, and reduce complications, readmissions, and waste. While experimentation with bundled payments has taken place within the public and private sectors, relatively little is known about how to implement them or about their impact on quality and costs. To address this knowledge gap, the Agency for Healthcare Research and Quality awarded Integrated Healthcare Association (IHA) and the RAND Corporation a demonstration and evaluation grant. IHA convened a group of stakeholders to develop and implement a bundled payment program, including health plans, hospitals, ambulatory surgery centers (ASCs), physician organizations, physicians, and technology vendors.
RAND Corporation used qualitative methods (interviews and focus groups) to assess the development and implementation process. RAND planned to use quantitative methods to evaluate the impact of bundled payments on the costs and quality of orthopedic procedures for commercial enrollees. However, lower than expected procedure volume meant that sample sizes were too small to detect the hypothesized effects of bundled payments on cost and quality, so this component of the evaluation was dropped.
- Significant challenges and delays: Despite initial enthusiasm and substantial effort by various stakeholders, the bundled payment demonstration experienced significant challenges and delays. Ultimately, only three of the six health plans and two of the five hospitals signed contracts, as did two ASCs. Patient volume under these contracts totaled only about 35 orthopedic cases over a 3-year period for the two hospitals; volume in the ASC was higher (roughly 100 orthopedic procedures). The interviews and focus groups identified four major factors responsible for these results:
- Difficulties in defining bundles, driven by risk-related considerations: The consensus-building process for developing bundle definitions experienced delays, primarily due to disagreements on how broad the bundle should be. Key issues included which patients would be eligible (i.e., whether certain pre-existing comorbidities and/or risk factors should exclude a patient) and which services should be part of the bundle (e.g., whether to include presurgical and postacute care). Health plans generally wanted inclusive definitions that covered most if not all patients and made providers responsible for a broad array of services over a long period of time. Not surprisingly, hospitals and surgeons wanted to limit their assumption of risk under the pilot. This disagreement over risk became more challenging due to decisions made by program leaders, including not to adjust payments based on risk, create stop-loss provisions, or require the purchase of reinsurance. For orthopedic procedures, the resulting definitions proved too narrow to capture enough volume to make bundled payments viable. Thus, neither hospitals nor health plans had adequate incentive to invest in needed infrastructure, including clinical care redesign (by providers) and automated claims processing (by health plans).
- Lack of trust, driven by competing interests: Participating health plans, hospitals, and physicians tended to be skeptical of each other's motives, particularly early in the process. Lack of trust and transparency might have been the result of a history of aggressive contract negotiations between the parties. Mistrust manifested in significant disagreement about pricing, with health plans expecting reductions compared to historic FFS payments and hospitals expecting to be paid either the same or more (in consideration of the warranty and inflationary pressures).
- Inadequate infrastructure for claims processing: Health plans' existing automated claims systems could not handle bundled payments. Initially, specialized commercial software did not exist. Even when such software became available, the lack of patient volume and the cost to buy and implement the software made its purchase a nonstarter for participating plans. Payments to physicians also became a problem, since hospitals initially lacked the capability to distribute payments under the bundle.
- Time delays and uncertainty related to State regulations: California's prohibition on the "corporate practice of medicine" prevents hospitals from directly employing physicians. This prohibition created concerns among participating plans and hospitals that payments to physicians under a bundled payment contract might violate the statute. IHA retained a law firm to develop a model contract template to address this issue, but this process took significant time. Even with the template, participating plans and hospitals were not certain that regulators would be satisfied, and consequently adopted a "split-bundle" model with two separate payments—one for all professional services and one for all facility services. In addition, the California Department of Managed Health Care had to approve all contracts between plans and providers, a process that took 9 months for the first contract.
- Ongoing enthusiasm and existing infrastructure on which to build: Even with these challenges, participating stakeholders maintain their interest in bundled payments, provided implementation-related problems can be resolved. In addition, the IHA demonstration generated substantial expertise in technical aspects of implementation that can inform future efforts. For example, participating stakeholders ultimately came to consensus on 10 episode definitions that can be adapted to other settings. In fact, the Wisconsin Payment Reform Initiative has, with minor modifications, adopted IHA's total knee arthroplasty bundle definition. In addition, the demonstration enabled development of extensive specifications for historic cost analysis, and highlighted the flaws inherent in the common practice of using retrospective episode groupers to define the prospective episode payment. Finally, the demonstration led to the development and successful deployment of a common contracting template that largely satisfied the contracting parties and State regulators.
The demonstration generated the following lessons for those interested in implementing bundled payment programs:
- Ensure sufficient case volume: Providers and payers will not have an incentive to make the substantial and costly changes needed (e.g., clinical redesign, automated payment systems) unless adequate case volume exists to justify the investment. To ensure adequate volume, focus on common conditions and procedures and limit exclusions.
- Consider including appropriateness determination: Some literature suggests that use of bundled payments may encourage overuse of procedures. To address this possibility, consider use of appropriateness criteria and/or encourage provider use of shared decisionmaking with patients.
- Find mutually acceptable methods for managing risk: Consider use of alternative strategies for reducing provider risk that will not negatively affect patient volumes, such as risk-adjusted payments, stop-loss provisions, and provider reinsurance.
- Keep initial bundle definition simple: As noted, creating bundle definitions proved to be a complicated and lengthy process. While simple definitions may be less satisfying conceptually, they provide a realistic, achievable starting point.
- Use automated solutions to administer payments: Automated systems for administering bundled payments are being developed, and using them can eliminate the need for manual adjudication (an approach that becomes cost-prohibitive when used on a large scale). The cost of automated systems will likely come down as product offerings proliferate and higher case volume allows the spreading of fixed costs over a larger base.
- Change benefit design to steer patients to providers: Providers had a limited incentive to participate in the IHA demonstration, in part because no mechanism existed to steer patients to them. Changes in benefit design, such as reduced out-of-pocket payments for patients treated by a participating provider, could increase provider enthusiasm for the program.
Williams T, Robinson J. Bundled episode-of-care payment for orthopedic surgery. Integrated Healthcare Association Issue Brief 9. Oakland, CA: IHA; September 2013.
Other IHA Publications: https://www.iha.org/resources
Episode Definitions: http://www.iha.org/episode-definitions.html
Contract Templates: http://www.iha.org/bundled-payment-contract-templates.html
For additional information on AHRQ Delivery System Research visit our Web site or contact Michael Harrison at Michael.Harrison@ahrq.hhs.gov.
M. Susan Ridgely, David De Vries, Kevin J. Bozic, and Peter S. Hussey. Bundled payment fails to gain a foothold in California: the experience of the IHA Bundled Episode Payment Demonstration. Health Affairs 2014 Aug;33(8);1345-52.
Wesley Kary. Bundled Payment Episode and Gainsharing Demonstration white paper. Oakland, CA: Integrated Healthcare Association; September 2013.
Based on research funded by ARRA Grant #1R18-HS020098-01.
The findings and conclusions in this document are those of the authors, who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this brief should be construed as an official position of AHRQ or of the U.S. Department of Health and Human Services.