A note from the Editor-in-Chief,
In this month’s Guest Editorial, Clinical Orthopaedics and Related Research® Senior Editor Paul A. Manner MD, FRCSC tackles a question that is both philosophical and very practical: What is value as it pertains to orthopaedic surgery? Considering the imprecision in our approaches to measuring both quality and cost, one’s perception of value—and so the response to Dr. Manner’s question—may depend largely on what one’s stake is in that answer.
Dr. Manner has written extensively on healthcare economics, including numerous times in our Editor’s Spotlight/Take 5 column[8-10]. After reading his Guest Editorial, I suggest also reading this month’s Editor’s Spotlight/Take 5, where Dr. Manner talks to Stephen B. Murphy MD about his 2018 John Charnley Award-winning paper, “Analysis of US Hip Replacement Bundled Payments: Physician-initiated Episodes Outperform Hospital-initiated Episodes”.
— Seth S. Leopold MD
There is no shortage of discussion of value-based health care. Michael E. Porter MBA, PhD of the Harvard Business School  and the late Uwe Reinhardt PhD  of Princeton have been sounding the call for some level of coherence between clinical results and the cost of care since the turn of the century. In CORR®, we’ve run Value-based Healthcare, a quarterly column from Kevin Bozic MD, MBA on how to get more bang for the buck for more than 5 years now [7, 18]. And it’s hard to argue against it. After all, who wants to be the one opposed to value? Clinicians, hospitals, and payers have all put a great deal of time, effort, and money into making health care pay more attention to perceived value.
Without becoming overly philosophical, the problem is one of definition. Terms are concepts, and to answer such questions as, “What is value?” we must find out what we mean by value. Here, we are on shaky ground. We typically define value as quality (or outcomes), divided by cost. Essentially, what we are trying to measure is a form of efficiency, or the cost of care associated with a specified level of quality. However, this quickly becomes nebulous. There is no standard for how to measure value even for a given condition, and still less on how to compare providers. And comparing the worth of interventions for different disease entities is even shakier.
Presumably, your perception of value depends on who you are: A patient, a payer, or a government.
The patient’s perspective on value should receive the most weight. Certainly, as a patient, I want the best results regardless of the resources needed, particularly if I’m insulated from the true cost of care, as most patients are. But even here, measuring outcomes is surprisingly difficult. The quality-adjusted life year (QALY) is the coin of the realm in studies of the economics and provision of health care [11, 13, 14]. Simply put, one QALY represents a year in perfect health, a QALY of zero equates to death, and a given burden of disease or illness results in a QALY somewhere in between. The QALY can be measured by surveying patients directly and asking them (for example) how much time they’d trade in their current condition for a year of perfect health, which is an example of time-tradeoff modeling, or indirectly with tools such as the EuroQol-5 Dimension or the SF-36. Unfortunately, all current outcomes instruments have problems, from lack of sensitivity for specific health conditions to lack of context. Further, value sets are determined from relatively small groups of the general, healthy population and may not reflect the opinions of actual patients with specific conditions. And the outcomes tools we use are intended to assess only those factors that directly result from healthcare; we need to remember that patients do more than go to the doctor’s office. Their needs include not just health-related endpoints, but also social support, financial resources, as well as access to education and technology. By relying solely on measures such as the SF-36, which are designed to assess generic quality of life rather than what individual patients may desire, we put patients at risk from interventions which are overvalued, and by denying those which are undervalued.
If you’re a payer rather than a patient, your thoughts on the topic of value may be quite different. Part of this stems from the fact that measuring cost is somewhat easier than assessing quality. When the Centers for Medicare & Medicaid Services (CMS) looks at how hospitals perform, they base their purchasing on safety (as measured by metrics on infection metrics), clinical care (using 30-day mortality for pneumonia, myocardial infarction, and heart failure as a proxy), efficiency and cost reduction (both absolute and trends in Medicare spending per beneficiary), and the patient experience of care (which relies on the familiar Hospital Consumer Assessment of Healthcare Providers and Systems Survey) . The problem, though, is that many of these outcomes depend on external factors such as social support, housing, access to transportation, which have only tenuous connections to what individual practitioners or health care systems can address. And determining “cost” has its own challenges. In orthopaedic surgery, arthroplasty is a tempting focus for cost analysis because it’s expensive, common, and relatively reproducible. But when Akhavan and colleagues  compared methods, they found that a total joint cost about USD 13,000 with time-based accounting (which tries to measure true resource use and is how a typical business manages its affairs), and USD 24,000 with traditional accounting (where cost is assumed to be a fixed fraction of charge) . Clearly, a hospital would assert that arthroplasty costs the latter, while a payer would prefer the former. To be blunt, it’s impossible to have a meaningful discussion on cost versus quality if the parties involved cannot get any closer on the “easy” part of measuring cost for a single treatment. And, again, somewhat lost in the argument over cost is the patient’s perspective: What about the patient with an uncommon problem, or with a complication? These patients are not well served by a single-minded focus on cost.
With this in mind, how have value-oriented payment programs fared? The short answer: Not well. The Federal Government has tried several approaches through the Medicare program, but few have made any difference at all. The Participating Heart Bypass Center Demonstration project (https://innovation.cms.gov/Medicare-Demonstrations/Medicare-Participating-Heart-Bypass-Center-Demonstration.html) was a bundled-care program where the entire cost of care was paid as a lump sum. Although it achieved cost reduction, there was almost no effect on outcomes. Shared savings programs, such as the Physician Group Practice Demonstration, Premier Hospital Quality Incentive Demonstration, and the Medicare Home Health Pay for Performance Demonstration showed little or no effect on Medicare expenditures and small improvement in care processes .
Many have touted the potential of Accountable Care Organizations (ACOs), where reimbursement is tied to both cost and performance on quality metrics. Under CMS guidelines, institutions can opt to enter different formats or tracks, which include increasing levels of risk versus reward. As one might expect, providers tend to like the idea of reward, but shy away from risk. The result? Because only 5% of ACOs followed the high-risk tracks, CMS actually lost USD 39 million from ACOs in 2016 because it had to pay USD 691 million in bonuses . In addition, the groups that entered the ACO program early (either because they were “true believers” or simply better prepared) have done substantially better . Thus, the ACO concept may not be practical for most providers. In short, sharing risk between payers and providers makes sense in theory, but simply doesn’t seem to work very well, regardless of design.
Finally, does watching the pennies result in the dollars taking care of themselves? The fundamental concept of value-based health care is that overall costs will drop as the cost of care for the majority of patients is driven down. This may be optimistic. Clinicians know that a small number of patients seem to occupy a large portion of their time and effort—and this is mirrored in what their care costs. In the United States, 1% of the population accounted for 23% of total health care expenditures, and 5% of the population accounted for 50% . The “healthy half” of patients with the lowest expenditures accounted for only 3% of spending total spending. Further, although plenty of attention is paid to big-ticket items such as joint surgery, the top-five most-costly medical conditions in terms of health care expenditures  were heart disease, trauma-related disorders, cancer, mental disorders, and chronic obstructive pulmonary disease and asthma. Therefore, spending time and effort on “controlling” the costs of routine care for healthy patients, whether it takes the form of appropriate use criteria, eliminating small variations in practice, or shortening length of stay, even if it represents better care, may not change what we spend on healthcare.
What should be done? If reducing cost is the goal, it seems sensible to focus on the sickest patients with the most-expensive conditions, since their care requires the most resources. There is nothing wrong with seeking value; unfortunately, current models aren’t delivering it. Why? First, as I mentioned, the metrics chosen to measure both quality and cost are probably too crude to be of much use. Second, there often is a long delay between when care is provided and when quality information becomes available; imagine trying to change a patient’s blood pressure medications when the most recent reading is 6 months old. Finally, tying compensation to outcomes in the presence of multiple influences that are beyond provider control (such as care from different providers, patient response, and comorbidities) seems unfair. To convince providers of the utility of value-based programs, data needs to be readily available, and to reflect actual clinical problems and interventions. Value-based provider payment design should reward providers and organizations who create approaches that produce the greatest patient benefit in health outcome, clinical quality, and patient experience.
The solution—for most patients—may be one of two similar options: Episode-based payments, such as bundled care for arthroplasty, makes sense when there is a clearly defined beginning and end to treatment. Similarly, reference-based pricing (where the payer sets a maximum amount they will reimburse for certain medical services) removes financial uncertainty for both patient and provider. The last decade has seen a profound shift in physician employment from small private groups to large hospital-based organizations. Whether one gets better care in the latter is an open question, but it’s clearly more expensive . There is no reason why routine care needs to be performed in a hospital system—perhaps it’s time to put physicians back into the community and look at on-site health centers for employers and their employees. Changing this would require, for example, that payers refuse to pay hospitals more than they would pay an independent physician or free-standing facility for a given services—if financial incentives are better aligned, behavior might follow. Finally, an honest appraisal of what care really costs, with a clear understanding of what outcomes are realistic, would go a long way towards providing real value.
1. Akhavan S, Ward L, Bozic KJ. Time-driven activity-based costing more accurately reflects costs in arthroplasty surgery. Clin Orthop Relat Res. 2016;474:8–15.
3. Centers for Medicare & Medicaid Services. The hospital value-based purchasing (VBP) program. Available at: CMS.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/HVBP/Hospital-Value-Based-Purchasing.html. Accessed November 26, 2018.
4. Cohen SB. Statistical Brief #455: The Concentration of Health Care Expenditures and Related Expenses for Costly Medical Conditions, 2012. Rockville, MD: Agency for Healthcare Research and Quality; October 2014.
5. Conrad DA. The theory of value-based payment incentives and their application to health care. Health Serv Res. 2015;50:2057–2089.
7. Leyton-Mange A, Andrawis J, Bozic KJ. Value-based healthcare: A surgeon value scorecard to improve value in total joint replacement. Clin Orthop Relat Res. 2018;476:934–936.
8. Manner PA. Editor's spotlight/take 5: The Affordable Care Act decreased the proportion of uninsured patients in a safety net orthopaedic clinic. Clin Orthop Relat Res. 2018;476:921–924.
9. Manner PA. Editor's spotlight/take 5: Universal health insurance coverage in Massachusetts did not change the trajectory of arthroplasty use or costs. Clin Orthop Relat Res. 2016;474:1086–1089.
10. Manner PA. Editor's spotlight/take 5: What does a shoulder MRI cost the consumer? Clin Orthop Relat Res. 2017;475:576–579.
11. McKenna SP, Wilburn J. Patient value: its nature, measurement, and role in real world evidence studies and outcomes-based reimbursement. J Med Econ. 2018;21:474–480.
12. Murphy WS, Siddiqi A, Cheng T, Lin B, Terry D, Talmo CT, Murphy SB. 2018 John Charnley award: Analysis of US hip replacement bundled payments: Physician-initiated episodes outperform hospital-initiated episodes. Clin Orthop Relat Res. [Published online ahead of print November 13, 2018]. DOI: 10.1097/CORR.0000000000000532.
13. Pandya A. Adding cost-effectiveness to define low-value care. JAMA. 2018;15;319:1977–1978.
14. Pettitt DA, Raza S, Naughton B, Roscoe A, Ramakrishnan A, Ali A, Davies B, Dopson S, Hollander G, Smith JA, Brindley DA. The limitations of QALY: A literature review. J Stem Cell Res Ther. 2016;6:334.
15. Porter ME. What is value in health care? N Engl J Med. 2010;363:2477–2481.
16. Reinhardt UE. Economics. JAMA
18. Vetter TR, Uhler LM, Bozic KJ. Value-based Healthcare: A novel transitional care service strives to improve patient experience and outcomes. Clin Orthop Relat Res. 2017;475:2638–2642.