Patient Preference as a Barrier to Needed Care

Oh, yeah. I have a new hip; I’m a new man! If anybody ever tells you that you need a new hip, don’t hesitate.

—Eugene Fama, co-winner of the 2013 Nobel Prize in Economics¹

What if no one tells you so clearly? This was the case for Mr. W., a 54-year-old patient dependent on multiple pain medications, including narcotics. Disabled and no longer working, he walked with a cane and always appeared to be in great deal of pain, though he had no other medical problems. He hardly ever exercised, had a limited social life, and had trouble sleeping. Mr. W. received routine primary care for decades, and had also routinely received prescriptions for his pain meds until his primary care provider ordered new x-rays of his hip to re-evaluate his clinical profile. The tests showed severe osteoarthritis (OA). There was advanced marginal spurring about the joint with subchondral cysts involving the acetabulum and femoral head; a report that is highly consistent with severe hip osteoarthritis (OA). Though he had previously waved off any discussion of joint replacement, this time his doctor educated him about the surgery, addressed his questions about the procedure, and referred him to orthopedics, where the surgeons recommended immediate total hip arthroplasty. The outcome transformed his quality of life and functional abilities and eliminated his need for pain meds.

Why did it take so long for Mr. W. to receive joint replacement when he faced full disability and functional decline? In part, it’s because he belongs to a sizable population who has been reported to harbor low preference for joint replacement in the management of end-stage OA: he is an African American male. Because joint replacement is an elective procedure that is highly preference-sensitive, none of his primary care providers explored further the patient’s expressed lack of interest in surgery. These well-meaning primary care providers simply respected Mr. W.’s opinion and offered him alternative treatments including escalating doses of pain medications. This is not an isolated case. It informs a larger and well-documented national statistic—racial variation in the use of joint replacement.²

This case is a cautionary tale because patient preference is a key ingredient and an antecedent of patient-centered care. Patient-centered care is receiving the well-deserved national attention for its potential to improve quality of care and reduce cost of care. It is commonly operationalized using shared decision-making mechanisms, a practice approach that features prominently in the Affordable Care Act and is promoted in primary care as part of the highly touted Patient-Centered Medical Home.³ Yet as illustrated by Mr. W.’s case, our inability to identify in a timely manner legitimate and well-informed patient preferences regarding treatment choices in day-to-day clinical practice has real-life implications. Several factors challenge our current view of patient preference and may confound our ability to provide authentic patient-centered care. First, patient preferences for medical care are highly variable and not always attributable to respected cultural, religious, or ethnic traditions. A preference for no blood products among patients who belong to the Jehovah Witness religious sect is an accepted patient preference that is familiar to most practicing physicians in the United States. On the other hand, relatively few providers appreciate or even know the fact that minority patients and most specifically African Americans are significantly less likely than non-Hispanic White patients to prefer joint replacement as a treatment option in the management of end-stage knee or hip OA for reasons that are unclear and might not be beyond reproach. It is not simply a matter of risk aversion or lack of trust in the health care system. After all, about 50% of African American patients are willing to undergo joint replacement when clinically indicated and recommended.⁴

Patient preferences for care are not always easily ascertainable in a typical clinical encounter. Physicians have been reported to be less than skilled in soliciting patient preferences or even simply asking what patients want and desire for their care and why. Shared decision-making, a noble approach to care that is fully endorsed by all health care agencies, relies fundamentally on identifying patient preferences and incorporating them into the medical decisions. But shared decision-making is hardly the standard in typical clinical practice today. This is like saying common sense is the best qualification for being a good listener, when in reality common sense is so uncommon, to paraphrase Voltaire.

Emerging research demonstrates that contrary to the prevailing perception, patient preference is malleable. A recent prospective, observational study by Group Health in Seattle showed that orthopedic patients who receive Decision Aids that are designed to educate them about the risks and benefits of joint replacement were more likely to reshape their preferences for the treatment by demonstrating lower utilization rate.⁵ That study did not evaluate the role of race or ethnicity in the observed relationship between use of Decision Aids and joint replacement utilization rate. But a randomized controlled trial of African American VA patients showed knee OA Decision Aids used as an educational tool increases minority patients’ preferences for joint replacement and their likelihood of receiving a referral to orthopedics, when clinically indicated.⁶ The contrasting results of these two studies aside, they both illustrate that patient preference is not a fixed attribute and not always linked to unchallengeable cultural, religious, or ethnic values. Rather, it is an attitudinal disposition that might be acutely sensitive to educational and knowledge-building interventions: a view that is highly consistent with the scant social sciences research and theories, such as the theory of reasoned action.

Providers should not disregard patient preferences for medical treatment, but rather they should evaluate them more carefully and help patients arrive at informed preferences. We should also recognize the marked variations in patient preferences that exist among different patient populations that reflect social and educational positions so that our efforts to educate about medical options are targeted. Patients look to providers and the health care system to meet this objective. In our desire to respect patients’ beliefs and preferences, we must not abdicate our responsibility to educate and advocate for them. Fortunately, a growing array of shared decision-making tools may help the busy clinicians meet this challenge.