Why the study?
Is low-value testing before low-risk hand surgery associated with increased downstream healthcare use and costs?
Is low-value testing before low-risk hand surgery associated with increased downstream healthcare use and costs?
Answers are grounded in this paper and cite their sources.
Low-value preoperative testing before minor hand surgery triggers a cascade of further unnecessary testing and increased healthcare costs.
There may be no more-important problem in our specialty (and, perhaps, across medicine more generally) than that of overdiagnosis and overtreatment. Chasing indolent conditions and incidental findings causes psychological harm by pathologizing the normal [6], it results in hundreds of billions of dollars in needless costs [4], and most importantly, it injures patients when unnecessary surgical procedures are brought to bear in conditions for which safer alternatives like exercise or even the passage of time are equally effective [9, 10, 12, 13]. The reason I say that this may be our most-important problem is that causing harm where avoiding it is possible takes surgeons as far off mission as we can go. Because of that, I was so excited to read this month’s Editor’s Spotlight article, “Is Low-value Testing Before Low-risk Hand Surgery Associated With Increased Downstream Healthcare Use and Reimbursements? A National Claims Database Analysis” [16]. In it, Robin N. Kamal MD, MBA of Stanford University and his team characterized the degree to which low-value diagnostic testing begets further low-value interventions. Spoiler alert: It’s a high degree, and this is in a population—patients undergoing minor hand surgery—in whom little preoperative testing is indicated to begin with. It’s not difficult to imagine that in patients undergoing larger interventions the amount of tail-chasing in the form of needless work-ups and interventions would be greater still. One of the more-common complaints I hear from reviewers and readers is that large-database studies can only tell us what we already know. I disagree. At their best, these studies teach us things that we can learn no other way. Single-center or even multicenter trials and observational studies lack the numbers and the completeness of follow-up to answer the big questions that studies like this one by Dr. Kamal’s team can address. You may not do hand surgery (I don’t), but this study [16] is worth your attention because it covers so many important, big-picture topics. Read it, and on the way there, join me as I go behind the discovery in the Take 5 interview with its senior author, Dr. Kamal. In our interview, he talks about the problems of overdiagnosis and overtreatment, what we can and should do differently as individuals, and what our professional societies ought to do to help, as well as how the power of big data can bring us the answers we need. Take 5 Interview with Robin N. Kamal MD, MBA, senior author of “Is Low-value Testing Before Low-risk Hand Surgery Associated With Increased Downstream Healthcare Use and Reimbursements? A National Claims Database Analysis” Seth S. Leopold MD:Congratulations on a study that shines a bright light on the problem of overdiagnosis and overtreatment. Please orient our readers by characterizing the problem in whatever ways you find most compelling. How big a deal is orthopaedic overdiagnosis and overtreatment in terms of harms or costs? Robin N. Kamal MD, MBA: Thank you for your interest. As you already mentioned, analyses of large datasets are an opportunity to understand a phenomenon from a perspective that most of us as practicing surgeons rarely get to see. The motivation for this work started with somewhat arbitrary testing policies required by surgery centers, which often are based strictly on age cutoffs and not patient health. Based on this, our group had previously collaborated on work highlighting low-value testing for common hand conditions [5, 8]. I think your readers probably have seen these testing requirements, and most of the time, we probably go along with these requirements as they are the gateway to treating our patients.Robin N. Kamal MD, MBABut what is less often discussed is what happens after: What are the effects of a low-value test on a patient? We found that often the test triggered a cascade that led to a greater adjusted odds ratio of future testing. The per-patient costs from these cascade events was low: For a population of about 70,000 patients, the expenses totaled USD 3.5 million. But this represents only a subset of the US population and a limited number of conditions. One can extrapolate just how this same phenomenon is likely occurring with other low-risk procedures where the testing is feasibly to be similarly of low value (such as knee arthroscopy or shoulder arthroscopy). Additionally, the cost-sharing continues to shift to patients with increasing out-of-pocket expenses because of expanded use of high-deductible plans in the United States. So, while these costs may seem small, for most Americans, they are not. Dr. Leopold:Numbers in this setting can come across as rather abstract. I’m guessing you’ve had some more-personal, perhaps even poignant, experiences in your practice that motivated you to study this problem. Can you share one or two of those? Dr. Kamal: Our group’s work studies ways to be better, either at care delivery (quality), costs of care (value), or decision-making. I’ve had the opportunity to work in private practice surgery centers as well as large academic health systems, and the contrast of value systems and priorities has motivated some of this work as well. I recall at our outpatient surgery center, there were rules based on having stents and needing cardiac workup for certain patients, regardless of procedure or even if the patient was having local anesthesia only. Keep in mind, tests like these require our patients, and sometimes their family members, to take time off of work, pay for parking, pay for a test, and then wait for a test result that they hope will not unmask an unknown problem. It was a clear example of wasteful practice, with no evidence to justify it, particularly when we know some of these hand procedures are being done in surgeons’ offices. We had previously used value stream mapping, where one analyzes every input in a process and assesses how it provides value to the output (or doesn’t). In this case, we studied a high-quality surgical procedure. We collaborated with some of my colleagues on shifting the value analysis upstream to preoperative testing, where, maybe unsurprisingly, the guidance for testing for an older patient receiving same-day spine surgery was the same for someone having a trigger finger release. Dr. Leopold:Sometimes problems like this can feel too big for the individual surgeon to engage with. What can the individual surgeon do differently to minimize unnecessary harm to patients that arises from overdiagnosis and overtreatment? I think most of us think the things we do are well indicated, but in aggregate, you found that this seems not always to be the case. Dr. Kamal: As orthopaedic surgeons, a lot of what drives our day-to-day behavior is rooted in the fact that we enjoy being in the operating room. So, when a patient presents to us and there is either no strong evidence, low-quality evidence, or debatable evidence, it can be easy to rationalize intervening with surgery. With an estimated USD 800 billion annual waste in the US healthcare system, it only makes sense that there is an opportunity within orthopaedic surgery to address these situations where surgical treatment may be unnecessary. Operative treatment of distal radius fractures in an older, low-demand patient or arthroscopy and partial meniscectomy in the setting of knee arthritis are just a couple of examples that come to mind [2, 11, 13]. We can address this by closely examining the care that we would want for ourselves or our family members. I know I scrutinize the care that my family members receive, evaluate what their out-of-pocket expenses are going to be, and whether there are nonoperative treatments that may work. Indeed, we know surgeons treat themselves differently (and often without surgery) than they do patients [14]. Acknowledging we have our own biases and acknowledging we all have an opportunity to improve are good first steps. It is human nature, to some degree, to create a mental model of how we see the world work and to have future events fit within that mental model. Kahneman describes this as “theory-induced blindness,” where after creating a mental model of how we believe a process is occurring, it can be challenging to notice its flaws. I’ve seen this happen to me as a hand surgeon and how my practice has changed. For example, 15 years ago, I was trained to fix almost all displaced distal radius fractures regardless of age and functional demand. Today, there is guidance from strong evidence that low-demand elderly patients can have similar patient-reported outcomes with nonoperative treatment. The American Academy of Orthopaedic Surgeons (AAOS) Clinical Practice Guideline [2] and appropriate use criteria [7] products are examples of attempts to help synthesize contemporary and high-quality evidence to change our practices. Dr. Leopold:When I think about system-level problems that look especially terrible to the lay public, overdiagnosis and overtreatment are high on my list. A key tenet of any profession—and of medicine in particular[3]—is the obligation to self-regulate. Professions that fall short risk facing external regulatory pressures that I believe our professional societies (and most surgeons) would prefer to avoid. What roles do large, national societies like the AAOS, certification entities like the American Board of Orthopaedic Surgery (ABOS), or residency accreditation bodies like the Accreditation Council for Graduate Medical Education have to play in this? Dr. Kamal: First, while the AAOS has made leaps in terms of promoting the collection of PROMs and the creation of a registry that supports improving care delivery, there are no policies on understanding and analyzing our practices to improve unwarranted variation in care, which includes overdiagnosis and overtreatment. While the ABOS ensures competency when we begin our practices, there are no policies that ensure our competency and adherence to evidence that evolves over time. Scrutinizing—and even criticizing—ourselves may be hard in the short term, especially if they lead to uncomfortable conversations and interventions, but it will be better for our profession in the long term. Either the AAOS or ABOS policies that support transparency and scrutiny of our own practices could focus on variation in surgeon practice patterns (like the use of physician-owned ancillaries and operative treatment in spite of strong evidence against surgery) and facility-level variation, such as quality of care from ambulatory surgery centers. Second, while our culture in orthopaedic surgery continues to improve in many ways, national societies have not focused on normalizing self-measurement and transparency as a means to performance improvement. I’ve experienced the anxiety my colleagues feel when I’ve chaired national clinical practice guidelines or performance measures committees. The common fear is that by citing strong evidence that supports nonoperative treatment for a given condition, the patient who is the exception to the rule will not be able to receive appropriate treatment or experience delays, leading to a potential underutilization problem. I’ve heard this argument a number of times at high-level meetings and always find it a little curious that our field does not step up and advocate for all those patients without insurance, or with limited insurance who have difficulty accessing the most basic orthopaedic services. Or how there are rampant inequities in care delivery based on race in our specialty that go unaddressed [1]. Putting the financial drivers to the side, I haven’t experienced a frank discussion about whether we actually have an overutilization problem, or how our field can be an industry leader in addressing inequities in healthcare delivery, and this perhaps is where the culture of our profession has an opportunity (and a duty) to evolve. The AAOS, American Orthopaedic Association, and ABOS are positioned to lead this change if they so choose. Lastly, a lot of my group’s work has been rooted in the experiences of patients. For example, most of our studies are vetted by a patient panel prior to us initiating the study to ensure we are asking a question that is important to patients and their health. Perhaps a way in which we can better self-regulate is by professional societies more actively partnering with or managing a diverse group of patients who can guide our advances in quality measurement, creation of policy, and national-level advocacy. A stronger presence of patients could ensure that when conclusions are drawn or decisions are made for the benefit of patients, such as in the setting of debatable evidence for clinical practice guidelines or performance measures, that these decisions align with patients’ needs and goals, and not those of physicians, who sometimes engage in overdiagnosis and overtreatment. Dr. Leopold:How might the Centers for Medicare & Medicaid Services skillfully engage with the problem of overdiagnosis and overtreatment in orthopaedic surgery? Are there some diagnostic tests or some surgical procedures that should receive additional scrutiny from payers? Dr. Kamal: There are two ways to see the world of healthcare delivery, and they’re not mutually exclusive. The first is a purely economic one, where transactions are occurring between physicians, hospitals, and payers, and payers continue to create increasing barriers to provide payment. In this view, each stakeholder in the “market” is responsible to its shareholders and is always vying to increase its revenues. While a cynical view, there is truth to the fact that as surgeons we have to pay the bills, and the adage “no margin, no mission” remains relevant. The second way to see the world of healthcare delivery is a strictly clinical one, where all stakeholders have patients’ interests at heart, and decision-making is always guided by the best interests of each individual patient. Again, these views are not mutually exclusive, but they do point to obvious conflicts in how finite resources are used to improve the health of populations and individual patients, while also securing revenues to cover the day-to-day expenses. Perhaps one reason it has been socially acceptable to overdiagnose and overtreat is that it is an approach that seems to fit with both of these views on healthcare. By providing more services, we continue to feed the healthcare machine and take care of patients (even if our services may not always be necessary). And when there is debatable evidence, or a debatable clinical practice guideline, we enable low-value care for all by justifying it for the exception to the rule, again, to continue feeding the healthcare machine and meet our fiscal responsibilities. As I mentioned before, the culture of our profession has to change so that professional societies reallocate resources they currently deploy in support of surgeons being able to perform any interventions they wish, in favor of being more selective, and encouraging surgeons to limit their interventions to those that are supported by strong evidence. My opinion is that we appropriately aligned the AAOS Clinical Practice Guideline for Distal Radius Fractures in this way when we highlighted how patient-reported outcomes were similar for operatively and nonoperatively treated distal radius fractures in older patients (usually > 65 years of age). Some of the more-recent clinical practice guidelines, however, such as those addressing knee arthroscopy for degenerative meniscus tears [13], may have not continued this culture shift and I anticipate will come under greater scrutiny in the future. If we embrace a culture of shifting toward high-value care and avoiding unnecessary tests and treatments, practices that overdiagnose and overtreat will lose financially. Professional societies like the AAOS will need to have mechanisms, such as implementation science units, in place to support these practices to transition toward high-value care. Our specialty’s focus to understand which services are high value, and which are simply overdiagnosis and overtreatment, will continue to be accelerated by payers shifting from fee-for-service to value-based plans. And although regional market forces can allow some to delay this transition, it is inevitable. For example, Medicare Advantage (where the plan assumes risk for the total cost of care of the patient) enrolled around 43% of all Medicare beneficiaries in 2021, and Medicare expects that within the next 10 years all plans will place plans at risk for the total cost of care [15]. These types of payment models focus on improving patient and population health with fixed resources and have both upside and downside risk to the health system or plan, disincentivizing overtreatment and low-value care. I expect as these at-risk plans become a larger share of group practice and hospital’s patient populations, our cultural and operational adoption of high-value care in orthopaedic surgery will accelerate.
No takes yet. Share an insight, caveat, or question.
Seth S. Leopold (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: