The Bundled Payment for Care Improvement (BPCI) demonstration models were introduced as one of the nearly 3 dozen innovation models to be tested and evaluated by the Centers for Medicare and Medicaid Innovation (CMMI). The general concept is to “bundle” the payment for all of the patient care services by various providers over an episode of illness. This concept should seem somewhat familiar to you. It might be easier to think about if you conceive of a bundle as just a bigger version of the Diagnosis-Related Group (DRG) we already know. Currently, hospitals receive a single, prospectively determined, dollar amount for all the services they provide during an inpatient stay, based primarily on diagnosis that brought the patient to the hospital. The new bundled payments will work in a similar manner. Medicare appears to be moving toward a single predetermined bundled payment amount that includes all the services provided during the 3 days prior to hospitalization, all the services provided at the hospital, and all the services provided for up to 90 days post-discharge from the hospital. It’s like a “super” DRG – a single all-inclusive payment for the full episode of illness, rather than just the geometric mean length of stay associated with an inpatient stay.
Until last month, there were only four basic Bundled Payment Models approved for testing, they included:
- Model 1: Acute Hospital + Physician Services (Retrospective Payment)
- Model 2: Acute Hospital + Physician Services + Post-Acute Providers + Outpatient Services
- Model 3: Post-Acute Providers + Physician Services + Outpatient Servicesth
- Model 4: Acute Hospital + Physician Services (Prospective Payment)
In July 2015, CMS introduced an additional Bundle Payment Model. CMS issued a Proposed Rule for public comment with a description of a new “mandatory” bundled payment model developed specifically for lower extremity joint replacement cases (MS-DRGs 469 & 470) called the Comprehensive Care for Joint Replacement. You are sure to learn more about this in the very near future, so I will not address it in this post.
I believe there are many potential benefits associated with the experimentation of bundle payment models; chief among these is the enhanced coordination of multiple providers across the continuum of care, each of whom will now be more concerned about what happens to a patient once the patient leaves their particular setting – less emphasis on the “silos of care”. If implemented as planned, we should hope to see a true transformation in the way healthcare providers work together to deliver care to their patients. We should see a greater degree of the communication among providers, seamless handoffs of treatment plans and mutually established clinical and functional goals that would go with the patient from setting to setting, shared root cause analyses to discover and correct the underlying factors that lead to excessive readmission rates, and greater sharing of quality and outcome metrics among the various providers. In my opinion… all of this is really good stuff!
Now that we have had time to gain some actual experience with bundles, we might want to ask ourselves whether they are advancing the larger, longer-range goal of moving from volume to value. Bundles represent an initial learning opportunity on a much longer path. Providers for the first time will focus on the longer time horizon covering the full episode of the illness. In addition to the benefits mentioned above, they might also learn about sharing financial risks and rewards. Providers should openly accept the opportunities and challenges presented while experimenting with bundle payments – learning what works, as well as what does not work so well. However, bundle payments may not be an end in and of themselves. I view them as a transitional step, a way to get providers, payers and patients ready for the more expansive alternative payment models that will surely follow.
Let me explain my thinking on this…
By definition, BPCI requires a short-term acute care hospital inpatient stay to initiate each bundled payment. Given this, hospitals may very well choose to maintain their current approach to revenue generation, believing that more bundles are better – incentivizing some hospital CEOs and CFOs to stay focused on increasing inpatient volumes and capturing more market share from their competitors. Lofty ideas about managing population health, increasing the focus on prevention and wellness, developing systems and processes to maintain the health of patients with multiple co-morbid chronic conditions to avoid exacerbations that might require hospitalization, etc. are not relevant in the BPCI world. The unintended effect of implementing BPCI in the short-term is that some acute care hospitals may pursue higher inpatient hospital utilization, not less!
Therefore, I would encourage everyone to utilize the BPCI demonstration projects to test the water in longitudinal care, but recognize it is a short-term bundle. Let us learn everything we can about designing an optimal plan of care for our patients that covers the full course of care over of a longer 90+-day episode of illness. Apply everything we can learn about better communication and coordination among all the various levels of care; greater involvement of patients and families in the care decisions; develop meaningful indicators of quality, both from the standpoint of clinical outcomes as well as measures based on patient reported outcomes; and learn as much as we can about managing financial risks. There should be no question that the move from traditional Fee-for-Service reimbursement to Bundled Payment for an episode of illness represents both a significant and dramatic change. However, once we feel comfortable having adapted to some of these more fundamental building blocks essential to virtually all alternative payment models, we should keep our eyes focused on the future… a future where the financial incentives in healthcare truly foster delivering greater value and not, even unintentionally, greater volume.
Please note that guest blogs from Guiding Committee and Work Group members represent the views of the individual authors and do not represent official positions of the Guiding Committee, Work Groups, CAMH, or CMS.