Editor's Spotlight/Take 5: Is There an Association of Patient Mindset and Physician Willingness to Acquiesce to Unhealthy Patient Preferences?

David Ring MD, PhDImagine sitting in a room with your patient who has been referred by another doctor for chronic elbow pain. The patient came prepared with MRI images and a report from the referring doctor that says: “partial tear of the common extensor origin at the elbow.” You start discussing strategies with your patient to help him or her accommodate to the problem, as lateral epicondylitis usually gets better after a year or so [14, 17, 21, 32]. “But doctor, it says I have a tear here.” A few things go through your mind. It’s clear the patient is looking for you to intervene, but there isn’t clear, strong evidence that one injection is better than another or that any type of intervention is better than simply waiting. You do your best to explain this, but the patient insists. “Doctor, the report says there’s a tear, and I want it fixed.” What now? This scenario plays out in every orthopaedic subspeciality clinic every week. Most of our interventions are elective/discretionary/optional, and indications can be twisted to fit the decision: There’s a tear, so let’s do surgery. Insert the diagnosis code for the tendon tear and await authorization for surgical repair. It will be an easy approval, surgery is straightforward, and you’ll get the RVUs you need to meet your quota. But let’s take a step back. Intervening comes with potential harms that may be more severe than waiting out the condition. For corticosteroid injections, these might include atrophy and tendon rupture; chondrocyte-harming effects are well known, as well [11, 16, 28]. If our chosen intervention is surgical, well, no reader here needs to be reminded of those complications, but it’s worth taking a beat to call those complications to mind before we offer those operations to patients whose conditions are likely to resolve if we just give them some time. Let’s take another step back—before this patient’s referral, another doctor decided to order a test. Getting an MRI to find out what’s wrong seems innocuous and potentially definitive. But is an MRI really harmless? The imaging findings may reflect normal age-related changes rather than surgical pathology [9]. Images may overdiagnose fractures and lead to overtreatment [8]. In fact, knee-jerk low-value testing can expose patients to subsequent testing, more invasive procedures, and needless risks [30]. These types of “definitive” tests may lead to downstream harm and unnecessary surgery when patients push for treatment based on a spurious finding [19]. The stressor of having to acquiesce to unhealthy patient requests can diminish the enjoyment of our profession, of treating patients. So, in this month’s Spotlight article [2], David Ring MD, PhD and his team, the Science of Variation Group (SOVG), explore this dilemma. They found that, when pressured by patients, surgeons tend to acquiesce to an unhealthy request by picking something less harmful rather than something that is associated with obvious harms, such as opioids and operative treatment. They also found that physicians may be more susceptible when the patient has more return visits. Making the right choice of protecting patients from unnecessary harm may seem easy when we consider these dilemmas from the comfort of our office chairs, but it’s all too easy to lose track of benefit-harm boundaries during a busy clinic, with the cloud of patient satisfaction surveys constantly hanging over us. So join me as I explore the findings of this paper with Dr. Ring, senior author of, “Is There an Association of Patient Mindset and Physician Willingness to Acquiesce to Unhealthy Patient Preferences?” [2], in the “Take 5” interview that follows. Take 5 Interview with David Ring MD, senior author of, “Is There an Association of Patient Mindset and Physician Willingness to Acquiesce to Unhealthy Patient Preferences?” M. Daniel Wongworawat MD:Most things we do are a balance of benefits and harms. Corticosteroid injections are an example—they come with some benefits and some harms. Might giving in to requests be more likely, and perhaps understandable, if the balance is more in favor of benefit? David Ring MD: When you refer to the harms of corticosteroid injections, I assume you mean the catabolic effects, which can manifest as cutaneous hypopigmentation, subcutaneous fat atrophy, weakening of tendons, and deterioration of articular cartilage. Beyond these iatrogenic forms of physiological harm, there are important forms of harm that we may not always consider, like psychological and financial harms. It’s common for a person to falsely interpret new sensations from degenerative pathophysiology—like osteoarthritis and rotator cuff tendinopathy—as an injury needing repair [12, 18, 22]. The clinician who adheres to evidence and offers an injection knows it has the potential to elicit—at best—a brief and limited alleviation of symptoms. Such injections are often no better, or not much better, than placebo injections [1, 23]. But people tend to take a relatively passive and magical approach to health (e.g., “They’ll know what to do… they’ll fix me”) as opposed to an active and factual approach (e.g., “Help me understand what’s going on and how I can work on it”) [6]. The patient may believe that the injection is necessary to fix the problem—that it is both curative and required. And if the patient believes the injection can cure, there is a risk that the specialist is exploiting the patient’s vulnerabilities by offering it. There is also a potential for other harms, such as a misdiagnosis of the patient’s true preferences (for instance, for treatment better than placebo), a potential reinforcement of false hope, and a missed opportunity to bolster personal health agency. Words and actions that reinforce misconceptions, diminish personal health agency, or leverage vulnerability to magical thinking are all, by definition, harmful. Personal and professional health strategies are best when they are based on evidence and ethics. The evidence we have shows that people want to know if an intervention has a meaningful specific effect, such as antibiotics for infection, and show less interest in treatments that offer mostly nonspecific effects, such as placebo, particularly when the treatments are painful or inherently or potentially harmful [3]. We also know that time alone is a diagnostic test as well as a therapeutic intervention. If there is an important pathophysiology, it will be diagnosed eventually in time, and that waiting period is unlikely to cause harm, since most musculoskeletal treatments address symptoms rather than modify disease. Symptomatic treatments, such as topical and oral analgesics, are nonspecific and readily available with or without a specific diagnosis. Dr. Wongworawat:Some steps might seem innocuous, such as an MRI for nonspecific symptoms. But these may carry hidden harms, such as surgery for incidental findings. How can we better educate our patients about the “harms side” of this balance? Dr. Ring: It’s helpful to be mindful of the limitations of “education,” or the transfer of knowledge. Because human intelligence is, in large part, automatic or intuitive, knowledge alone may not reliably lead to healthful and useful changes. Using MRIs as an example, our first, automatic thought may be, “What’s the harm in taking a look?” The fact that a diagnostic test is associated with potential for harm is counterintuitive in this mental framework, and addressing counterintuitive things often goes beyond simple teaching or educating. That’s the gist of the Nobel-awarded science regarding cognitive bias [15] and nudge theory [29], which suggests that nudges—or interventions that leverage human mental shortcuts while preserving one’s freedom of choice—can be more effective in guiding patients toward safer and more efficient health strategies. For instance, a phrase like, “Most people choose to wait it out when they understand this will go away on its own,” uses social influence to anchor patients on nonintervention as the preferred strategy. And a phrase like, “My hope is that you’ll feel better right away, but I worry about the evidence that shows how shots can actually prolong recovery,” acknowledges and validates a person’s impatience and proactiveness while compassionately conveying that we know from experimental evidence that an injection might undermine the patient’s goals. Moreover, the fact that we found a relative resistance to acquiesce in all of the scenarios, including injection and imaging, suggests to me that specialists are aware of the full range of potential harms. In our study [2], steroid injection for extensor carpi ulnaris tendinopathy and MRI for wrist pain with laxity were regarded as easier to acquiesce to. That’s probably because the potential for psychological and financial harm is less intuitive. It makes sense that the more obvious potential harms—such as an inappropriate opioid prescription and unnecessary surgery—were felt more deeply by surgeons. We are taught not to order imaging for low back pain without “red flags” because doing so has more potential for harm from overdiagnosis and overtreatment of both incidental findings and false positives; the high negative predictive value of the test must be useful [4, 31]. So one of the points of this study was to draw focus to the nontechnical, nonpathophysiological harms. Dr. Wongworawat:Are there specific, practical steps I can take to prevent myself from unnecessarily giving in to a debatable patient request? Dr. Ring: Checklists and other debiasing strategies might help us navigate these stressors [27]. Debiasing strategies are tools we use to catch our mind’s inevitable errors in order to limit harm. Surgeons use a checklist as a debiasing strategy prior to surgery. Checklists are suggested for other aspects of care, such as management of a claim of work injury [10]. Surgeons should construct and practice using a checklist when their patients request a test or treatment. Such a checklist could be implemented at the slightest sense of misgivings about a patient’s request and would help us call attention to any temptation to act contrary to expertise and best practices. A diagnostic checklist might include items such as: What is making me feel uneasy about this request? What aspects don’t feel right to me? To what degree is the patient’s request based in personal health agency and matter-of-fact thinking? To what degree is the request based in misconceptions and false hope? What are the potential harms, including psychological and financial harms? Are there notable potential benefits that outweigh those potential harms? Is there a potential disease-modifying intervention? Will this visit, test, or treatment increase the healthfulness of a person’s regard for their body and its sensations? Will it increase personal health agency? Once aware, the clinician can enlist additional tools in effective communication, guidance, and setting boundaries. For instance, the patient’s goal may start as, “I need to be pain free”—which is a common, unhelpful thought—and arrive at, with our guidance, “I want to maintain my cherished roles and activities.” The latter is nearly always achievable, while the former is often not. These are skillsets that benefit from coaching and practice. Dr. Wongworawat:How do you balance denying potentially unhelpful patient requests against receiving poor scores on patient satisfaction surveys? Dr. Ring: My opinion, based on personal experience and my read of the evidence, is that a small subset of people will get very upset when we set boundaries, perhaps no matter how practiced and skillful our strategies were for establishing trust [7, 20]. Some have suggested that strongly negative patient experiences may, at times, be a manifestation of a personality disorder [13, 25]. One feature that is commonly shared among different personality disorders is the intent to induce strong negative emotions in the clinician, which certainly rings true for the several times I have felt the need to set boundaries. However, overall, negative experience ratings and comments are exceptional in this context [5]. We have found that if we prioritize a patient-specialist relationship based in trust, the experience can be good, even when we delve into sensitive areas, like mental health [24]. In any case, an effective health organization should anticipate a small, but consistent number of poorly rated patient experiences. Each low-rated experience should be considered a “feedback event” that warrants an investigation, just like safety events, because there will always be ways for the individual clinician, their care team, and the organization to improve. Taking a close look will help ensure such opportunities are not overlooked, and the feedback, coaching, and training will help everyone improve. It is also true that this process might often result in bolstering consensus health strategies, including the tools associated with those strategies, such as checklists and communication tactics. An organization might occasionally conclude that the care and clinician/team conduct were exemplary. Dr. Wongworawat:Pointing patients to healthier choices that run contrary to patients’ preconceived notions often takes up more clinic time. How do you recommend overcoming the associated time constraints? Dr. Ring: This goes back to effective communication strategies. A rule of thumb is that if the visit is taking more time than it should, you are probably immersed in an argument rather than a partnership [26]. The patient and specialist may be trying to persuade one another. Often the patient may emphasize the many aspects of their experience that seem to contradict our expertise and support theirs: “I didn’t have any symptoms in this shoulder before…” “And my other shoulder is fine…” Communication strategies that emphasize trust and relationship by putting the diagnosis and care strategy aside initially in order to prioritize the patient feeling heard and understood can shorten visits. If we guide rather than educate, and get the patient talking more than us, it becomes easier to recognize when there is a tussle between the patient’s inner narrative and our expertise. Patients may feel a drive to solve the problem now, and the clinician needs to anticipate and reshape that mindset, as symptoms can rarely be resolved in one visit. When the patient and clinician both agree to work together over time, visits can be more effective and efficient.