Episodes of care and bundled payments, a sustainable approach

Written by David A. Janiec, Director of Payment Strategy, Rothman Institute | August 23, 2019 | Print  |

This article is a portion of a book titled "Challenges, Risks and Opportunities in Today's Spine World " edited by Stephen Hochschuler, MD, Frank Phillips, MD, and Richard Fessler, MD. You can find links to the previous chapters at the end of this article.

The healthcare industry is clearly in the middle of a learning period, the goal of which is to understand how value-based care functions differently from fee-for-service care and the benefits a shift may provide. While this comment may surprise some practitioners who are not looking beyond the horizon, make no mistake that defined bundled payment programs are a temporary practice space, and the industry is in the middle of a learning curve for the future of healthcare. The next 8-10 years will be a period of folding in more and more clinical activity until enough information can be gleaned to shape a final value-based model. Early activity in bundled programs demonstrates the development of increased use of evidenced-based medicine, a better understanding of costs, improvements in care coordination, and higher quality patient care. It is for this reason that it is extremely important not only to participate in these value-based models for educational and financial benefit, currently, but also to look with a critical eye to ensure a sustainable final model evolves.

First and foremost, a bundled payment model needs to support proper patient care. The easiest way to succeed in this effort is to ask physicians, not administrators, how this can be accomplished. Physicians provide or direct most of a patient’s care so they need to champion the process. Fee-for-service care fails because it was designed and has been driven by the health insurance industry, and as a result is not driven by patient care but rather transactions and their costs. It is procedure-facing and was structured to reimburse specific activity in a piecemeal fashion. It is a system that incents all clinicians to do more by performing both covered services that are truly a value-add to the patient, as well as many others that could arguably be considered duplicative, superfluous or of low clinical value. This is not to imply that clinicians are intentionally running up unnecessary charges or lacking focus on high quality care, but rather that the incentives are perverse and actually prohibit the ability to coordinate high-value, individual care while still receiving adequate compensation. Health insurers have made weak attempts at defining quality but have had no real measurement, nor provided any reward, for excellent clinical service under fee-for-service contracts.

The best way for physicians to understand what they can offer, is to take an inventory of what they have and, more importantly, how they work. They must assess the processes that are currently in place, understanding where variation exists, and measure quality. For example, there are many different ways to perform the same surgery. How is the proper timing determined? What approach would be best? Which implant(s) will be used? Which instrumentation? What post-discharge therapies will be involved and for how long? How will pharmacy be used and controlled? It becomes easy to see how important it is to arrive at and implement procedural consensus. It is also important to have access to real quality measurements. It is imperative that surgeons understand their rate of complications, reoperations, readmissions, infections, post-acute emergency room visits and so on. Perception is often different from reality so accurate data are important. Defining the most efficient care for a given condition is the starting point. After the required minimum care for a particular procedure has been established, the patient-facing adjustments required for the specific patient being considered can be incorporated into the care plan. Care which takes into account the minimal requirements for treatment of a specific condition that is further tailored to the immediate patient’s need cannot be any more precise or efficient. A thoughtful process will result in the most efficient use of clinical staff time, reduce cost for the health insurer, and limit the time and financial investment for the patient. Everyone wins.

The clinical readiness required to manage the activity in a bundled program cannot be overestimated. After care has been deliberately assessed and protocols are in place, the real discussion of programmatic development can take place. This work involves taking the high quality work that has just been defined during the assessment and placing it into a bundled payment model. While this may sound easy, because after all we have just established that the best care is the most efficient, patient-facing care, it isn’t easy to model for any number of reasons.

Different procedures are more or less complex so not all will follow the same bundled model. As an example, a hip replacement surgery is far less complex than a spinal fusion surgery. A hip replacement procedure most often occurs when the patient’s joint is bone on bone due to advanced arthritis, other conservative treatments have been exhausted, and pain is intense. The surgery requires a fairly standard implant, pain management and possibly some minor post-acute physical therapy. A fusion, on the other hand, can result from degenerative disc disease, spondylolisthesis, instability or any of a number of other diagnoses, and can be performed in a number of ways to arrive at the same satisfactory result. Two different surgeons might decide on different timing for the same surgery, may suggest treating different levels, may use different instrumentation, and so on. Some of these decisions may change the necessity of various types of post-acute care. It’s becomes evident that legitimate variation in treatment adds a wrinkle to the normal controllable variation in a process and bundled structure. While not insurmountable, it requires additional attention.

Any of the aforementioned scenarios can impact success or failure in a bundled model which is contrived for the purposes of learning; therefore, it is important to be sure to build a model that will add value. This is, after all, the idea behind the development of these models: learning how to deliver value-based care. That said, even the most thoughtfully developed bundled payment models may fall prey to unintended consequences, and for this reason many health insurance payers are leaning towards the use of bundled models developed by third parties that have been clinically designed and vetted. Having standardized models ensures industry-wide comparability, enhances universal learning, and helps avoid many of the ascertainable modeling issues. Health insurance payers have as substantial of an investment as clinicians do in developing good learning tools. Both have to learn how to get this right and practice makes perfect: both need data.

None of this absolves a practice from vetting models on its own. Standard models do not address issues pertaining to practice-specific clinical pathways or quality issues that may impact performance. They do not address the mathematics behind the appropriate volume required in a bundled payment program to safeguard against the risk of one case going financially out of control. Based on practice activity and patient demographics, case-level risk adjustment may be necessary. Exclusions for certain comorbid conditions or concurrent diagnoses may be needed to support learning. From a financial perspective, it may be appropriate to limit downside risk entirely, or at least for a period of time. It is, however, important to remember, the idea behind the modeling is to advance the provision of efficient, high quality care, and not exclusively to win financially. The latter is certainly an enticement, but in the development process, it can become easy to be distracted and make decisions that will make attaining high measures without real progress possible; this is not the correct result. Always reflect on whether negotiated language results in a) an appropriate learning curve, and b) longer term knowledge that will ultimately benefit the provision of future value-based care.

As a part of the process, keep in mind that it is absolutely appropriate to ask for data from a health insurance payer before you develop or enter a new bundled payment program. One cannot know what he doesn’t know. Health insurance payers are sitting on a wealth of historic claims information. Internal practice systems data can be used for comparison to payer data. This will instantly provide a sense as to whether the payer has a good handle on producing data. It will also allow the marrying together of some data to better understand areas of programmatic challenge as well as the potential for success. Clinical and financial data have not historically been aggregated to provide insights in healthcare, but they are actually quite easily joined and clearly explain the financial impact of healthcare provided. Understanding how the two interrelate opens the door for analysis and potential change. Do not discount the many ways practice data, both claims data and clinical data, can be combined to provide powerful insights on activity within your practice.

Understand that some of what will be learned from a bundled payment program is that value-based care really is a different proposition, clearly not business as usual. The who and the what of practice management from days gone by may not apply as focus changes from procedure-facing (what I do) to patient-facing (who is in front of me). Patient-facing means understanding the immediate patient: the diagnosis with which he or she has presented, what treatment steps have been taken prior, what needs to be done, what the follow up might be…and oversight of all of this. If the issue is acute, that can translate to a commitment of months. If chronic, the commitment can easily be measured in years. Value is an efficiency measure (quality of care received as defined by the cost to provide it) and will be rewarded accordingly. Process change, reallocation of duties or the addition of staff may become a requirement. As an example, a spinal fusion patient needs support through the entire perioperative experience. A nurse navigator who is introduced preoperatively and follows the patient all of the way through post-operative recovery will earn his or her keep in terms of cost and patient experience many times over. This is a team member who would have been considered a “nice to have” in the past but is now a “need to have” in the present. Bundled program management, and value-based care in whatever format it finalizes, will require new investments to garner returns.

Finally, be prepared to receive a lot of valuable data once a program begins. A dedicated staff member or outside consultant is a good idea as a lot of this work is still a bit manual, at least in the beginning. (Remember there were no systems designed to accommodate any of this.) Buried in the data provided is a lot of information which will be key to understanding areas of strength, as well as areas for improvement. Much of the activity of bundled payment programmatic management involves vetting data that comes from the sponsoring health insurance payer and accounting for savings, but it is also imperative that the data are used for change management. The best information and the most professional reports are both valueless if they stay with the bundled payment management team. Information must be communicated to the clinical teams that are actually in front of patients with easily understood, actionable information, or nothing is accomplished. Also remember bundled payments are a learning process…for everyone. The first set of reports may not be in a format that speaks to those who need to change and may require tweaking. Data formats from every payer are slightly different and some payers may change data formats intra-program. Clinical practice will not be perfect on the macro or micro levels. Both clinical and administrative team members will have room for improvement, but being armed with data is powerful. Find ways to use reporting to foster healthy competition without blaming or shaming teams or individuals. This is a process.

Bundled payment programs and similar Alternative Payment Models will not be around forever. They are exactly as they are named: programs and models. They have not been introduced (and are not being advanced) frivolously. The clear intention is that they are a specific learning opportunity for a limited time. That said, their construct will become an enduring component of what will ultimately be defined as value-based care.

Previous chapters:

Challenges, risks and opportunities in today's spine world

Spine care - Balancing cost with innovation

What are big data and predictive analytics

Predictive Analytics and Machine Learning

The HSS Spine Care Model, Part 1

The HSS Spine Care Model, Part 2

The Rothman Model, Part 1

The Rothman Model, Part 2

The History of Texas Back Institute

Texas Back Institute, Part 2

Private practice vs. hospital employee: Where we are today and why

ASCs: The economics of ASCs

Episodes of care and bundled payments

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies here.

Top 40 Articles from the Past 6 Months