Historically, health insurance companies and other payers like CMS made payment to physicians for each service that was required in the course of a given treatment.
With the passage of health care reform (“The Affordable Care Act”), a new program was initiated to bring these services for a given treatment together under one single payment, a “bundled” payment. (For more information on the Bundled Payment Initiative see Bundled Payments for Care Improvement (BPCI) Initiative: General Information).
Shortly after being announced, the initiative also took hold for various procedures and treatments in the commercial (non-Medicare) population as well. These procedure-based bundled payments, defined as a specific amount reimbursed for targeted surgeries such as knee and shoulder replacements, have been a step forward in aligning physician incentives, streamlining administrative burdens, and bringing more focus on healthy outcomes for the patient.
With advances in technology, evolution in data availability, and experience with these bundled payment programs, the next evolution of bundled payments is now possible. The opportunity is to extend beyond the procedure, incorporating pre-operative as well as post-operative services. This evolution is fueled by technology facilitating data collection across multiple sources and enabling analytic capabilities that drive proactive care coordination. Creating comprehensive (not procedural) bundled payment models is likely to ease the economic burden of healthcare through greater efficiency and better outcomes.
Reimagining Care Coordination
Common practice for procedural bundled payments is to cover cost of care from surgery pre-operative care and conclude with discharge planning. This is incomplete. We need to address the whole patient, inclusive of other illnesses, such as diabetes or hypertension, their state of mind, and so on. To address the more complete picture requires more thoughtful and holistic awareness, better use of the available information, and care management approaches other than discharge planning.
Numerous studies have pointed out the impact of social determinants on population health, such as nutrition/diet, activity, and sleep. There is a direct correlation between such drivers as sleep and activity and the prevalence of chronic disease, inflammation, and even the probability of patient compliance.
Examples of relevant articles on these topics can be found on PubMed (diet and hypertension), through the AHA (sleep and heart disease), and the National Institutes of Health (impact of stress on hypertension prevalence).
A procedural bundled payment will have only addressed the current pathology. This may be appropriate for urgent care, but does not address the cause. How can we avoid the urgent situation altogether? How can we help the patient in a way that avoids duplication of clinical and administrative efforts?
To create a more comprehensive care coordination and clinical teaming approach requires moving past current standards of care coordination that rely on sharing information and providing referrals to specialists. The information technology that exists today enables a platform-based business model that integrates and coordinates data exchange across multiple care-partners in a flexible, dynamic manner.
Such a dynamic IT architecture will better support rich data collection and integration. This in turn creates a scalable support system to empower a broad and flexible care team. More emphasis is needed on using such technology to collect and sift through multiple sources of data, as well as to identify clinical and social determinant-based health needs. In addition, the technology enables consistency of care along clinical pathways and facilitates monitoring adherence to the care plan.
In this new care coordination process, the primary care physicians are supported as needed with organizing care teams and coordinating communications between team members to manage both known as well as predicted complications, and address root causes.
Consider a use case for a pre-diabetic hypertensive patient, male, age 45, presenting for shoulder pain. He has no prior history of stroke but is a smoker, works in a manufacturing industry, is married and a father. The diagnosis is confirmed by clinical examination and imaging. The pain persists despite physical therapy and leads to a decision to perform a shoulder arthroplasty (total shoulder replacement).
Care coordination in this case requires additional insight, given that the patient is a smoker and there is a correlation in manufacturing and construction between smoking and alcohol abuse. This may make the patient more susceptible to develop an addiction if certain painkillers are prescribed. Additionally, in the context of a working population where roughly 15% have heart/lung chronic conditions, and of these about 30-40% are dual income parents, there is a reasonable expectation that the patient may be a parent stretched for time.
This could translate into a diet that includes convenience or processed foods, which are known to exacerbate inflammation. Further data collection and analysis may lead to discovery of home or work-based stress that may be impacting sleep. In this case study, the result of the additional insight from further data analysis and care coordination effort is the realization that physical therapy should have been pursued longer and likely be more effective if these other issues are also addressed by the appropriate skilled care giver as part of the coordinated clinical team.
Given the patient’s pre-diabetic condition, care coordination in this case may also require providing pre- operative support that includes nutrition counseling to address the potential and known complications such as fluid and nutrient retention, and infection susceptibility.
In a similar way, post-operative care coordination would address patient transportation needs to and from appointments, medication adherence and education, physical therapy adherence, and additional nutrition/diet counseling. Ideally, there will also be incentives in place (through benefit plan design) to encourage additional activity (e.g. playing ball with his children, gardening, and other exercise) and further education on and modification of diet. For background on typical benefit plan designs see the Kaiser Family Foundation survey.
This level of care coordination may be completed by a centralized care coordination team, such as employed by an Accountable Care Organization (ACO) or similar physician management services / convener organization or may be part of the care management team within a health insurer.
Modeling Comprehensive Bundled Payments
Several modifications in how reimbursement is determined need to be considered as we move away from today’s procedural basis for bundled payment.
Under this new model, the bundled payment encompasses the entire treatment pathway from diagnosis/indicating lab results to “recovery”, however defined (e.g. return to work, return to independent living, etc.). Comprehensive, total cost of care will need to be modeled across multiple care pathways, considering patient-specific complications as well as clinical and biometric risk factors.
The data and analytics exist today to drive these insights. For example, we can analyze patient cohorts to identify the set of circumstances and issues that a coronary bypass graft (CABG) patient is known to have 6-9 months prior to surgery, and how strongly these issues are correlated with complications such as sepsis, onset of pneumonia, post-operative stroke, post-operative depression, etc.
This modeling and analysis needed creates a comprehensive bundled payment that includes all services, counseling and care support into one, prospective payment.
In addition to more complex modeling and analysis, risk adjustment is required for known co-morbidities. It is unreasonable to expect there will be a one-price-fits-all solution. There must be adaptation to consider lower risk patients (e.g. patients with no previous incidents, not diabetic or a smoker) as compared to higher risk patients (e.g. patients over 50, with history of prior strokes, and having a pacemaker).
The next modification is measuring and tracking quality/outcomes-based metrics and providing incentives. The open data platform has a scalable and flexible IT architecture. This infrastructure enables the near-real-time tracking of services rendered against the clinical pathways the care team determined necessary.
Additional incentives should be provided for proactively participating in the clinical teaming effort, rewarding those who are good members and who are openly communicating. Being proactive and maintaining accurate EMR and coding may be considered for an administration bonus since these behaviors reduce redundancy and improve administrative efficiency.
A Quick Comparison
There is a possibility that some patients progress in the pre-op support such that surgery is no longer needed, but rather the support services and remediation activities continue for a longer period of time (e.g. nine months of physical therapy and sleep counseling). This is a scenario where all parties, physicians, patients, and payers benefit.
In the case study, assume $27,000 is paid for a procedural bundle for the shoulder arthroplasty. The underlying actual cost of services would have been considered in setting that price, and we can assume there is a profit margin to the physician as part of that.
Consider instead, that $35,000 is paid under a comprehensive bundle for all the services required pre-op and post-op, including physical therapy, etc. Going one step further, assume $25,000 is spent on physiotherapy, food, sleep and activity-based counseling, but no surgery is performed as symptoms improved with conservative therapy. The lead specialist receives $10,000 despite not having done surgery, but rather the specialist and their staff spent time over-seeing the clinical team.
The patient is recovering with no complications, infections or re-admissions, saving the system the expense associated with waste and redundancy. Further, there are no pre-authorization hassles. A bonus of an extra 20% for the team is likely given these results. In this circumstance, the net margin to the physician is likely the same if not higher than if the surgery was performed. The patient is also likely better off, living a new, sustainable behavioral pattern.
This comprehensive approach has another benefit. Under this model of reimbursement, there is less administration (for example, no pre-authorizations, no benefit eligibility verification, nor amounts to bill or collect). Also, surgeons can increase patient panel sizes without changing office hours.
Oversight of a care team, facilitated by analytics and communications-based technology, is more scalable than performing surgeries.
Procedure-based bundled payments occur at the point when a procedure, such as a surgery, is about to be done. This approach does not address the core reason or cause for why the patient is needing surgery and relies on discharge planning (or transition planning) to resolve patient post-operative questions and recovery concerns.
New care coordination processes that are longitudinal, using a whole patient-based mindset, is a fundamental shift in thinking and heavily leverages the platform-based data and analytics technology architecture available now. The new care team, championed by the specialist responsible for the intended procedure or the fundamental chronic disease, can be supported with analytics and communications to perform care coordination and health sustaining activities that start pre-procedure and continue through recovery/rehabilitation.
The transition to a comprehensive bundle payment is complex and there is a lot of work to be done to realize this vision. It will require collaboration between physicians, health system administration, and insurance companies.
Physicians have incentive to make this change because it removes barriers to delivering care and expands on what care is possible to manage. Health system administrators can appreciate the move to higher margin and more efficient use of their resources. Health insurers that move faster and facilitate this will have better physician networks and more competitive products.
Given the continued economic burden of healthcare there is an urgency to harness this technology and data capability now, to create new clinical pathways and care plans, and in so doing, we can create singular, comprehensive (not procedural) bundled payment models that foster greater efficiency and better outcomes.
About The Author
Mark Jamilkowski has been working with health systems and health insurers for 30 years, with recent emphasis on innovation concerning population health, outcomes-based contracting, and data analytics through his company, G.O Strategic Advisors.