Orthopaedic surgeons have had relatively little exposure to shared decision-making tools in their practice. First popularized in a 1982 report issued by the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research1, shared decision making is generally viewed as an embodiment of the principle of patient-centered care, with shared decisions between patients and providers often facilitated by decision and communication aids. Charles et al.2 identified the characteristics of shared decision making as involving the patient and provider, both parties participating in the treatment decision-making process, requiring information sharing, and both parties agreeing to the treatment decision made, with the recognition that many of these characteristics are continuous variables. The authors of the present study have previously made important contributions to our understanding of how shared decision making might be incorporated into orthopaedic practice, with both attendant benefits and obstacles to overcome3,4.
Knee and hip osteoarthritis treatment includes both medical and surgical interventions often managed by orthopaedists, and when pain and disability are unresponsive to medical treatment, total joint arthroplasty is frequently recommended. Total joint arthroplasty is appropriately viewed as a preference-sensitive procedure, as the indications are driven by the patients’ perception of pain, quality of life, and expectations of improvement offered by the surgery. Utilization rates vary widely on the basis of geography, sex, age, and ethnicity, but patients’ willingness to undergo surgery, patients’ fear of complications, or differences in surgeon practice patterns may also influence this variation4. The authors of this report accurately point out that shared decision-making tools may have gained little favor in orthopaedic practice because of surgeons’ concerns that patients will be less likely to choose surgery, or issues centered around the cost, logistics, and efficiency of implementing shared decision-making models in practice.
Bozic et al. attempt to address some of these concerns in this well-designed randomized controlled trial of shared decision making in candidates for total joint arthroplasty in two academic centers. After initial prescreening, patients were randomized into two groups, an intervention group and a control group. The intervention group was mailed a decision aid consisting of a digital video disc (DVD) and a booklet outlining treatment alternatives in hip and knee osteoarthritis. Each patient in the intervention group was also provided a consultation with a health coach who helped the patient formulate pertinent questions, attended the appointment with the patient, and subsequently provided the patient with a recording of the appointment and a copy of the dictated consultation note. Control subjects received directions to the clinic and a brief handout about hip and knee osteoarthritis and treatment options. Both groups, as well as the surgeons, completed preconsultation and postconsultation surveys. Briefly summarized, the patients in the intervention group were more knowledgeable about risks, benefits, and alternative treatments; were further advanced in their decision-making process; and had more confidence in their questions posed to the surgeon than patients in the control group. Surgeons similarly believed that the intervention group asked more appropriate questions and used their time more efficiently. There was no significant difference in the time spent in consultation between the two groups, nor was there a significant difference in the choice of operative versus nonoperative management between the groups following the consultation—an important finding.
Although the authors are to be congratulated on broadening our perspective on shared decision making with this carefully conducted trial, certain limitations do exist. The methodology necessary for patient selection for this randomized trial does not reflect the usual process of primary care or patient self-referral to the orthopaedic specialist. Considerable resources were expended to limit the patients enrolled in the trial to those considered most appropriate for total joint arthroplasty, and even after 73% of the initial candidates were excluded by the team, another 25% of those patients in the randomized allocation were eliminated by the principal investigator for reasons including inappropriate diagnoses, mental status, and lack of suitability as a surgical candidate. If every patient referred to the orthopaedist for osteoarthritis of the hip and knee was similarly screened, most would agree that it would make for an efficient and pleasurable day in the clinic!
Similarly, the patients enrolled in this study represent an extremely well-educated and affluent group; 88% had at least some college education (30% with postgraduate degrees), and nearly 37% made more than $100,000 annually. One wonders whether shared decision-making tools would work as well (or perhaps better?) in a different population that is not as facile at e-mailing their edited questions to their health-care coach. It also might have been interesting to “blind” the surgeons to the extent that the health-care coach was not allowed in the consultation room, and to encourage both groups to bring a written list of questions to ask the surgeon. To their credit, the authors acknowledge these limitations and realize the opportunity that exists for further efforts in this area.
The use of shared decision making and decision and communication aids in orthopaedic surgery will undoubtedly increase in the future. Is the mix of health-care coach, prescreening nurse personnel, and provider incentives critical to the success of shared decision making? Will it take more than two (physician and patient) to tango? The implementation of such programs will clearly depend on the economics of the accountable care organization era in a changing health-care market, and Bozic et al. are helping to provide the road map.