Several authors from the Brookings Institution recently argued in favor of making Medicare’s Bundled Payment for Care Improvement (BPCI) initiative mandatory. While the principles guiding their recommendations are sound, the recommendations themselves fail to acknowledge five fatal methodological flaws within the BPCI program. Their analysis also overlooks the most logical and reasonable alternative: a physician-focused episode-of-care payment model.
Five Fatal Flaws Of The BPCI Initiative
Each BPCI bundle is triggered by an inpatient stay and in particular by a specific diagnosis-related group (DRG). There are many shortcomings of using DRGs to trigger episodes. Chief among them is that the heterogeneity of procedures and conditions contained in each DRG creates significant variation in the total costs of episodes, making it more difficult to distinguish between warranted and unwarranted variation.
More importantly, when regulators force every episode to be triggered by a hospital stay, they effectively prevent physicians from exercising their judgement in finding the best and most efficient site of service for a given episode. For instance, procedures at some stand-alone surgery centers, in which commercially insured plan members receive complex surgeries, including joint replacements, can be half the price of hospital-based surgeries. Why shouldn’t Medicare beneficiaries and the Medicare Trust Fund take advantage of those options? Many procedures, including total hip replacements, are now performed outpatient, yielding significant savings compared to the inpatient version. Yet the BPCI program penalizes that option. The net result is that it enshrines a mode of caring that may be inefficient, less safe, and provides worse outcomes for patients than otherwise could be.
2. Lack Of Severity Adjustment
The BPCI initiative’s answer to adjusting episodes for the severity of a patient’s condition is to rely on adjustments in the DRG system. These adjustments inflate the price paid to the hospital for a given hospitalization based on the various conditions uncovered during the hospital stay and recorded on the insurance claim. There is no other adjustment. As such, a group of providers taking risk for the management of a patient undergoing a total knee replacement would be paid the same whether or not the patient has, for example, a prior history of heart failure and rheumatoid arthritis. It doesn’t take a genius to figure out what any level-headed hospital leader or orthopedic surgeon would do to prevent getting the short end of the stick on this deal. Worse yet, hospitals can get a higher payment if there are serious complications that occur during the stay.
3. Encouraging Hospitalizations For Acute Exacerbations
The BPCI program includes bundles for patients who have an acute exacerbation leading to a hospitalization for any number of common chronic conditions prevalent in the Medicare population, such as heart failure or obstructive pulmonary disease. Patients who present to the emergency department with some acute event linked to these conditions, if hospitalized, would trigger a bundle. Tying this “feature” of the program to the prior one creates a perfect recipe to admit every patient, especially with moderate acuity, because their treatment during the inpatient and post-discharge will be significantly less expensive than the cost of the target budget. But beyond the perverse financial incentive, the very purpose of payment reform should be to encourage optimal management of patient conditions, and therefore a significant reduction in hospitalizations. The BPCI initiative does the opposite.
4. Weak Ties To Quality Measures
While hospitals enrolled in the BPCI program must report certain quality measures, there are no thresholds required to receive reconciliation payments. In other words, some could be rewarded for reducing cost relative to the target price, even if quality of care deteriorated.
5. Moving Goalposts
The BPCI initiative ignores one of the core principles of an effective payment model, which is to provide a clear and unambiguous target to those who are subject to financial penalties or gains based on their performance. The BPCI program “re-bases” the target every quarter as a function of observed changes in national trends. The shifting target is revealed at least six months after the performance period. Practically speaking, that means you have no idea ahead of time what the target will be and whether you have an opportunity to gain or lose on your performance. It’s not just demoralizing to the participants, it creates a level of uncertainty that health care services researchers have pointed to as a bad payment model design.
Do We Really Want More of This?
Not only do the Brookings authors fail to highlight these fatal flaws, they recommend more of the same, and to broadly mandate it. While there is sound logic to mandating certain payment models to avoid the potential gaming that the authors outline, the potential for gaming is only present because the model is deeply flawed. If the BPCI program offered rigorous adjustment for the severity of patients’ conditions (as we and others have done in our bundled payment programs), created flexibility to initiate episodes outside of a hospital setting, and focused on the management of conditions in addition to procedures, a mandate wouldn’t be needed.
The country cannot afford more of the same—worsening outcomes at higher costs. This is exactly what we would get under Brookings’ policy prescription. There is a better alternative.
There’s A Better Way
When Congress passed the Medicare Access and CHIP Reauthorization Act (MACRA) of 2015, it included the creation of the Physician-Focused Payment Model Technical Advisory Committee (PTAC). The PTAC’s role is to review proposed payment models and make a recommendation about those proposals to the secretary of the Department of Health and Human Services for further action. The recommendations can vary from not pursuing the proposal to a full and immediate adoption. While the secretary is not bound by the PTAC’s decision, in a recent statement current secretary Thomas Price expressed a great interest in receiving those recommendations, mindful that most physicians have very few current alternatives to the Merit-Based Incentive Payment System that is the standard formula for adjusting all Medicare fee schedules.
As it happens, one of the recommendations recently made by the PTAC is to test a wide-ranging episode of care payment model proposed by the American College of Surgeons and supported by many other medical specialties. This model includes the following critical elements missing from the existing BPCI model:
- adjusts for patient severity;
- ties patient outcomes to gains and losses;
- enables physicians to self-assemble the best possible team to manage a patient’s condition, illness, or injury;
- encourages comprehensive teamwork across physician disciplines; and
- encourages physicians to find and use the most efficient and effective course of treatment and site of service for the patient.
As Price mentioned in his public address, there is no silver bullet in payment reform but rather a series of alternatives that should be tested and evaluated to ensure that Medicare beneficiaries and the US public get the best possible outcome. That will not be the case with the current BPCI initiative, and it should not have been mandated, nor should its offspring, the Comprehensive Care for Joint Replacement model and proposed cardiac bundles.
So how should the Trump administration handle Medicare’s current bundled payment programs? At best, it should correct its flaws. At worst, it should simply let them expire. Under no circumstances should it mandate them.
The Trump administration also should embrace the recommendations made by the PTAC and enjoin Medicare to act immediately to test the model proposed by the American College of Surgeons and give the physicians of this country—and the millions of Medicare beneficiaries that depend on them—the ability to thrive under a well-designed and well-conceived alternative payment model.