The Center for Medicare and Medicaid Innovation (CMMI) announced January 10 a new, voluntary bundled payment model, Bundled Payments for Care Improvement Advanced (BPCI Advanced).
This model is essentially a retooled version of the agency’s first bundled payment model under the Affordable Care Act, known as BPCI. Significantly, this model will qualify as an advanced alternative payment model (APM) for physicians under the Quality Payment Program created by the Medicare Access and CHIP Reauthorization Program of 2015 (MACRA). In creating this model, the agency seeks to promote coordination of care to achieve decreased costs and improved quality of patient care.
Program Structure and Performance Measurement
The BPCI Advanced performance period will begin October 1, 2018, and will end December 31, 2023. Participants will accept financial and performance risk for one or more of 29 inpatient and three outpatient 90-day clinical episodes. Episodes may be initiated by either acute care hospitals or physician group practices (each being known as an “episode initiator”). Similar to the original BPCI initiative, there will be two types of participants: convener participants and non-convener participants. A convener participant is an entity that accepts financial risk on behalf of episode initiators. It can be an acute care hospital, physician group practice, other Medicare-enrolled provider or supplier, or even a non-Medicare enrolled entity. A non-convener participant accepts risk on its own behalf, and can only be an acute care hospital or a physician group practice.
CMMI will measure financial performance by comparing actual spending to a target price, which will be calculated based on participants’ historical Medicare fee-for-service spending on items and services typically provided during a clinical episode minus a “discount” of 3%, which represents built-in savings for Medicare. The agency will be measuring total cost of care, which means that most Medicare Parts A and B expenditures will be included, with very few exclusions. Semiannually, CMMI will compare actual spending on clinical episodes with the target price, then adjust the difference with a quality score based on the participant’s performance on identified quality metrics, to determine whether the participant has overspent or achieved savings. Participants must repay Medicare any overspending, but will receive a payment from Medicare for any savings (with both repayments and bonuses subject to limits).
Tools for Participants
CMMI has provided several useful tools for program participants, based on its experience with prior demonstration models. Participants may share both financial gains and losses with clinicians who provide care during the episode, provided that the participant and clinician have formalized a sharing agreement that meets certain requirements (and subject to the Department of Health and Human Services waiving applicable fraud and abuse laws). These sharing agreements can help participants use economic incentives to engage clinicians in working together to achieve improved financial and quality performance.
In addition, the agency will waive three Medicare payment rules: the requirement of a three-day inpatient stay in order for Medicare to cover a stay in a skilled nursing facility; the telehealth restriction limiting Medicare coverage to rural areas; and also a post-discharge home visit policy that requires direct physician supervision of certain post-discharge services provided in a patient’s home. These payment waivers will allow participants increased flexibility to provide services in the setting that optimizes cost and quality while meeting a patient’s clinical needs. Finally, participants will be required to participate in “learning system” activities designed to facilitate peer learning and information sharing related to performance improvement.
Prognosis for Success
Though not a perfect predictor, some qualified successes in the original BPCI program show the BPCI Advanced model may hold some promise. The most recent evaluation of BPCI Model 2 (the closest in structure to BPCI Advanced) showed that, among the episodes with sufficient participation to allow evaluation, Medicare payments were 4.5% less than what they would have been without BPCI. This was largely due to reduced spending on post-acute care – in particular, a decrease in use of institutional post-acute care and a greater use of home health. On the other hand, there were few statistically significant differences in quality of care. BPCI Advanced may allow even greater flexibility in post-discharge options, which could lead to additional cost savings. In addition, while the original BPCI program measured the quality of care, quality scores did not have an impact on payment. BPCI Advanced incentivizes improved quality through payment: since a lower quality score can increase potential repayment or decrease any bonus paid out, there are stronger incentives for participants to redesign care processes to achieve significant and sustained quality improvements.
In addition, a recent article in the Journal of the American Medical Association noted significant dropoff in BPCI participation. However, BPCI did not qualify as an advanced APM for purposes of the Quality Payment Program. In contrast, BPCI Advanced may be an attractive option because it provides an advanced APM opportunity for a number of specialists, who had limited such opportunities prior to this model. Since physicians can earn a bonus of 5% of their Medicare payments by participating in an advanced APM, this could be a significant incentive. Further, the learning system activities required of program participants may help with participant retention through supports such as technical assistance and information exchange.
Time will tell if this model will achieve the important goals CMMI has laid out. In the meantime, we will continue to monitor new and innovative models to address payment and care delivery.