Orthopaedic surgeons all start out as physicians, and before that, as medical students. Even before that, we wrote essays on medical school applications, in which we conveyed greater or lesser degrees of ardor on the topics of empathy, compassion, and caring. But I’d venture to say that none of us—even those who score in the single digits on the Jefferson Scale of Physician Empathy —suggested in our application essays that the kind of insurance a patient has should to determine whether that patient receives surgery for a serious injury.
And yet, many orthopaedic practices screen patients based on insurance. I suspect that the many groups that accept better-paying kinds of insurance and decline the rest do so with the belief that patients thus decanted will find their way to so-called safety-net hospitals for care. It’s reasonable to wonder how many patients actually do, and how many others miss the safety net. Even at hospitals that “accept” patients with low-paying insurance or none at all, it’s fair to wonder to what degree funding may factor in the decision to provide expensive, discretionary interventions, like surgery on closed clavicle fractures.
We may never know the answers to these questions with certainty. The next-nearest thing would be asking those questions in an all-payer database that is large enough to control for the most-important confounding variables—like age, gender, fracture location, and socioeconomic status, and a patient’s self-reported race—and see whether patients with private insurance were more likely to undergo surgery for a fractured clavicle than were patients who had insurance that did not pay as well (or who had no insurance at all).
A study in this month’s Clinical Orthopaedics and Related Research® did just that, and found that, even after controlling for relevant confounding variables like income and fracture location, patients with private insurance were far more likely to undergo clavicle-fracture surgery than patients without insurance or than patients with Medicare or Medicaid . The effect sizes were large, and the confidence intervals were tight. This well done and well reported study, by Irfan H. Ahmed MD, and his multi-disciplinary team at Rutgers University New Jersey Medical School, raises serious concerns about access to care, healthcare disparities, and surgical decision-making.
A reader may look for uncontrolled-for confounding variables; (s)he may even find a few. However, the findings here are so stark that the onus now is on those who believe this important part of the healthcare system is fair to prove it.
Certainly there is no reason to believe a finding like this would be limited to patients with clavicle fractures; this likely is a bellwether for a broader trend in our specialty (and probably across medicine more generally). As such, it’s worth the time of all readers of CORR®, whether or not they work on injured shoulders.
Join me as I go behind the discovery with Dr. Ahmed, senior author of “Is Insurance Status Associated with the Likelihood of Operative Treatment of Clavicle Fractures?” in the Take 5 interview that follows.
And if you’re wondering how orthopaedic surgeons generally scored on the Jefferson Scale of Physician Empathy that I mentioned in the opening paragraph , the answer is “not very well”: We were 11th out of 12 specialties surveyed. At least we edged out anesthesiology. Something to think about.
Take 5 Interview with Irfan H. Ahmed MD, senior author of “Is Insurance Status Associated with the Likelihood of Operative Treatment of Clavicle Fractures?”
Seth S. Leopold MD: Congratulations on this thought-provoking study. We’d all hate to imagine that a patient’s type of insurance would determine his or her care. Please offer readers some “non-sinister” explanations for your main findings.
Irfan H. Ahmed MD: I’m sure there are many “non-sinister” reasons. Some of the more-obvious ones include missing some uninsured patients who met operative criteria and chose to not undergo the surgery if offered or had it done much later due to unavoidable circumstances. Our study included hospitals as well as surgicenters, but there is a possibility some locations may have been missed. In addition, non- or under-insured patients might be worried about the potential financial implications (other than the direct costs such as the surgeon’s fee) and choose not to undergo surgery.
Dr. Leopold:By way of alternative, are there any potential explanations for your findings that trouble you? And what explanations do you believe are most plausible for the fact that even after controlling for a variety of important demographic and other variables, better insurance was associated with a greater likelihood of surgery being performed?
Dr. Ahmed: Not surprisingly, the “sinister” explanation troubles me. If a patient presents with an open fracture or life/limb threatening injury, I believe the overwhelming majority would manage those injuries and not look at insurance status. But in borderline situations, where we are more likely to notice a likelihood of operative treatment, I do believe a patient’s insurance status affects the treatment given to patients.
To the second point, I personally believe demographic and unconscious bias plays a role in how patients get treated, but we need more studies to prove this. Better insurance has its obvious effect on how physicians may treat patients and this could overcome the bias the provider may have.
Patients who have insurance are more likely to be operated on regardless of indication. Those who have injuries/conditions that could be treated either way are more likely to have surgery. Patients without insurance are more likely to have non-surgical treatment regardless of indication. In fact, some patients without insurance and/or those with Medicaid who may benefit from surgery or have borderline indications are likely to be treated non-surgically as well.
It’s difficult to determine whether patients with insurance may be having surgery for an injury that is not indicated or patients without insurance are not having surgery when in fact it is indicated.
Dr. Leopold:How might we change the dynamic here, and who has the authority to make the changes that would help patients most?
Dr. Ahmed: That’s a difficult question to answer. I think this may have to come from the physicians themselves which I understand is difficult to monitor and/or implement.
As physicians, we know that having oversight from such entities as the American Board of Orthopaedic Surgery could have negative effects such as added time and paperwork for the physician as well as the fact that deciding whether or not to operate is an objective variable.
Any such change will be at a cost. We may decrease the frequency with which unnecessary/borderline surgery is performed but the cost of administrative burden will increase.
Dr. Leopold:If poorer insurance (and perhaps more generally, deprivation) is associated with the likelihood of particular medical and surgical recommendations, it doesn’t take much imagination to recognize how this might harm patients. But if our profession makes no changes, what risks and harms might accrue to our specialty, and to medicine more generally?
Dr. Ahmed: I think the risks and harms already are affecting our patients. There are studies [1, 2, 4] showing certain minorities and poorer ethnic groups are less likely to undergo surgery or receive similar care. As physicians, we have to become more involved in the healthcare debate and expanding access to more affordable health care. Our debate shouldn’t necessarily focus on reimbursement but on indications and the good health of our patients.
Dr. Leopold:Patients who self-identified as black and those who self-identified as Hispanic were less likely to undergo surgery, and—please correct me if I’m wrong—the findings about racial differences in surgical utilization were independent both from those about insurance status as well as socioeconomic status. That suggests the likelihood that the findings about race and surgical usage in your study were caused by other factors. What explanations can you offer for the findings you made about race, and how might we verify your contentions in future studies?
Dr. Ahmed: That’s correct. The finding that patients who were black or Hispanic were less likely to undergo surgery were independent from insurance status. One explanation could be the type of insurance the patient had and how that affects the physician and the patient. “Better” insurance may mean better reimbursement and this again could affect how the patient is treated. Patients with Medicaid, for example, will likely have a higher deductible or less coverage, which will have a major effect on whether one chooses to undergo surgery.
1. Dodwell E, Wright J, Widmann R, Edobor-Osula F, Pan TJ, Lyman S. Socioeconomic factors are associated with trends in treatment of pediatric femoral shaft fractures, and subsequent implant removal in New York state. J Pediatr Orthop. 2016;36:459-464.
2. Gundle KR, McGlaston TJ, Ramappa AJ. Effect of insurance status on the rate of surgery following a meniscal tear. J Bone Joint Surg Am. 2010;92:2452-2456.
3. Hojat M, Gonnella JS, Nasca TJ, Mangione S, Vergare M, Magee M. Physician empathy: Definition, components, measurement, and relationship to gender and specialty. Am J Psych. 2002;159:1563-1569.
4. Laditka JN, Laditka SB. Insurance status and access to primary health care: Disparate outcomes for potentially preventable hospitalization. J Health Soc Policy. 2004;19:81-100.
5. Congiusta DV, Amer KM, Merchant AM, Vosbikian MM, Ahmed IH. Is Insurance Status Associated with Likelihood of Operative Treatment of Clavicle Fractures. Clin Orthop Relat Res. 2019 [Published online ahead of print June 6, 2019]. DOI 10.1097/CORR.0000000000000836