Shared decision making (SDM) is an interactive communication practice in which physicians and patients work together to make decisions informed by the best available evidence and weighted according to specific values of the patient.1 SDM has recently been touted as the meeting point between evidence-based medicine and high-value care and a potential cost-lowering intervention.2,3 The Affordable Care Act has mandated that medical care include “shared decision-making programs between patients and physicians which incorporate the patient’s preferences and values into the medical plan.”4 The Accreditation Council for Graduate Medical Education (ACGME) echoes this sentiment, requiring residents to perform and incorporate SDM into their practice as an internal medicine milestone for effective patient communication.5 Despite these requirements, practitioners across all specialties find implementation of SDM into their daily practice difficult because of many barriers, including lack of SDM training.1,6–9
Despite the known benefits of SDM, multiple studies have demonstrated poor use of this practice by clinicians.6,10,11 Training aimed at practicing physicians has been shown to improve knowledge, attitudes, and use of SDM.11–14 However, efforts directed toward physicians-in-training are scarce.7,15,16 Further, these studies did not evaluate participants’ communication skills in real-time clinical practice. Thus, we developed, implemented, and evaluated an innovative curriculum in SDM that uses a seven-step approach to SDM and a standardized patient (SP) case. The goal of the curriculum was to increase knowledge of and positive attitudes toward SDM, as well as use of SDM communication skills among senior internal medicine residents.
An innovative curriculum in SDM was developed for and implemented with postgraduate year (PGY) 3 internal medicine and PGY4 medicine–pediatrics residents across four outpatient continuity clinics in the University of Pittsburgh Medical Center internal medicine residency program during their final six months of training, from January through June 2015. This group of residents had already participated in a longitudinal patient–doctor communication curriculum throughout their training, which included basic interviewing skills and motivational interviewing. We used a pre–post study design to determine the effectiveness of this curriculum (see Figure 1 for overview). The curriculum evaluation received exempt status from the University of Pittsburgh Institutional Review Board.
We administered a precurriculum survey to determine residents’ baseline knowledge of and attitudes toward SDM. An extensive literature review of basic elements of SDM and prior methods for evaluating SDM knowledge and attitudes was used to create survey questions.11,14,17,18 This survey was vetted by a panel of medical education and communication experts at the University of Pittsburgh and piloted with general medicine fellows prior to distribution. We distributed the survey on the first day of the ambulatory rotation (see Supplemental Digital Appendix 1 at http://links.lww.com/ACADMED/A494). Demographic information was collected—namely, gender, continuity clinic site, prior SDM training, and future care goals. We assessed knowledge through four multiple-choice questions on the basic elements and benefits of SDM. Residents were also asked to report perceived barriers to implementing SDM and were asked to rank the importance of and perceived confidence with elements of SDM on a four-point Likert-type scale (1 = not confident/important to 4 = very confident/important). Residents also ranked the importance of SDM in the practice of high-value and evidence-based care using the same Likert-type scale. The pre and post versions of the survey used many of the same questions, but we only collected demographic information on the pre survey.
SDM curriculum and implementation
In the first of four weeks of the ambulatory rotation, we gave residents an audio recorder and instructed them to record a minimum of one and up to six full outpatient continuity clinic encounters in which they counseled patients on initiating longer-term medications, lifestyle modifications, or cancer screening. The patients did sign a consent form at the time of the recording stating that recordings would be used as an educational tool and reviewed with other physicians as part of their communication training. Residents were asked to hand in their recordings one week later at the start of the four-hour SDM curricular session. During this session, the residents learned the key elements of SDM including a brief, evidence-based overview of the benefits of SDM in relation to patient satisfaction, adherence, and providing high-value care.14,17,19–22 Group size ranged from four to six residents per month.
The residents were introduced to the novel seven-step model outlined in Figure 2. We opted for a physician-directed model based on the findings of a Cochrane review stating that interventions directed at the physician improved SDM use.12 This model was derived after review of available models with an emphasis on the essential themes of SDM including discussing options, assessing patients’ values, and negotiating a decision.1 On the basis of a prior study that documented concern among junior physicians about patients misinterpreting SDM as physician incompetence, we felt it was important to include explaining equipoise, meaning that there are multiple available choices without a clearly correct answer to patients, as one of our seven steps.6 A circular model in which the order of the steps is fluid was created to allow for more flexibility compared with prior linear models in hopes that this model could be used in multiple clinical settings.
Following the didactics, residents practiced these seven steps and communication skills with an SP. SP cases are a well-established and effective tool for teaching communication skills to medical residents.23 A case was developed to reflect current guidelines in statin therapy as primary prevention of coronary artery disease. This two-part case involved a middle-aged female who was instructed to discuss her abnormal cholesterol results received during an employment physical. Residents were asked to explain the risk of vascular disease and to counsel the patient on treatment options, including lifestyle modifications and statin therapy for a 10-year atherosclerotic cardiovascular disease risk of 8% on the first visit, and a risk of 15.6% calculated on the patient’s following visit one year later. We piloted the case with a group of faculty with training in teaching communication.
The SP sessions were structured to encourage an iterative reflective process that incorporated time-outs; self-reflection; and feedback from peers, faculty, and the SP based on learner-centered teaching methodology.24 This practice allowed one resident to interview the SP while the other residents served as active observers. Residents were asked to identify a “take-home point” or a learned skill they would implement into their clinical practice immediately following their encounter with the SP. The residents took turns interviewing the SP. Each resident had the opportunity to practice with the SP either in visit 1 or 2. Following this session, residents were again asked to record outpatient continuity clinic encounters in which they counseled patients on initiating a longer-term medication, lifestyle modifications, or cancer screening and were instructed to bring these recordings to the final week 4 SDM review session.
During this two-hour SDM debriefing session, residents were invited, though not required, to review both sets of their clinical audio recordings with the group and receive formative feedback on their communication skills. Residents also participated in a guided small-group discussion of the barriers they faced to using SDM and brainstormed solutions to those barriers. Residents completed the post survey of knowledge of and attitudes toward SDM at the conclusion of this session.
Evaluation and statistical analysis
We evaluated the curriculum in two parts: first by pre- and postcurricular surveys of knowledge of and attitudes toward of SDM, and additionally via pre- and postcurricular audio recordings of clinical SDM encounters. The survey responses were confidential; the pre and post surveys were linked to the clinical recordings using a unique identifier. We used descriptive statistics to describe baseline characteristics of the sample and perceived challenges to the use of SDM. Because differences in knowledge and attitude scores from pre to post were not normally distributed, we assessed these changes using the nonparametric Wilcoxon signed-rank test.
SDM skills were evaluated using audio recordings of SDM encounters in the residents’ continuity clinics. We evaluated the recordings using the validated Observing Patient Involvement in Decision-Making (OPTION) scale.19 This scale has been validated in a prior study with a high degree of internal consistency (a Cronbach alpha of 0.79).25 We selected the OPTION scale for this study because it is designed for observation of SDM in real-time clinical scenarios. Two reviewers (J.R. and S.M.) blinded to the timing of the recording in relation to the curriculum scored the recordings using the OPTION scale. The interrater reliability was calculated after reviewers reviewed a subset of the recordings independently, with an intraclass correlation of 0.87 (95% CI = 0.61–0.97).
Linear mixed-effects models were fit to compare pre- and postcurriculum OPTION total and individual scores, as multiple recordings existed per resident and not all residents submitted both a pre and post recording. In each model, we included a fixed effect for “time” (pre vs. post) and added a random effect for resident to account for repeated measurements made on the same resident. All models were fit via maximum likelihood estimation and converged without any problems. Diagnostics revealed that residuals were approximately normally distributed, thus satisfying the normality assumption for linear mixed modeling.
A total of 36 out of 38 eligible PGY3 and PGY4 internal medicine and medical pediatric residents participated in the curriculum (participation rate = 94%). Of these, 47% (16 out of 36) were female.
Survey data: Changes in knowledge and attitudes
A total of 32 of the 36 participants completed pre- and postcurriculum surveys (response rate 88%). Median knowledge scores increased significantly from 75% (3/4 questions) to 100% (4/4 questions) correct (interquartile range = [75%, 100%], P < .01). The greatest improvements were seen in the residents’ ability to define equipoise (37%–84% correct, or 11–26 out of 32 residents) and to identify essential elements of SDM (76%–96% correct, or 24–30 out of 32 residents).
Residents also demonstrated significant improvement in overall perceived confidence (median composite score pre 2.87, post 3.0, P < .01) and overall importance of SDM (median composite score pre 3.14, post 3.5, P < .01) as well as for each component skill (see Figures 3 and 4). There was also significant improvement in residents’ understanding of the importance of SDM to the practice of high-value care (median score pre 3, post 4, P = .02). There was nonsignificant improvement in the residents’ understanding of SDM as related to the practice of evidence-based medicine (median score pre 3, post 3.5, P = .23).
Clinic recordings: Change in SDM communication skills
A total of 44 recordings (31 precurriculum and 13 postcurriculum) were reviewed and scored on the OPTION scale. There was nonsignificant overall improvement in total OPTION score after completion of the curriculum (mean score pre 17.29, post 19.13, difference in scores 1.84, P = .27).
While reviewing the recordings, it became apparent that the five international medical graduate (IMG) residents who submitted recordings seemed to take a different approach to patient-centered communication. For example, while the clinical scenarios the IMG residents recorded were similar in type to those that other residents submitted, they tended to use more directive language that could come across as paternalistic. When we stratified results for U.S. versus IMG status, there was significant improvement in total score for U.S. graduates with a mean difference of 5.15 (P = .01). There was significant improvement in U.S. graduate residents’ ability to elicit a decision from the patient and discussing pros and cons with a mean difference of 0.74 (P = .01) and 0.74 (P = .04), respectively. There was also a trend toward improvement in the residents’ ability to list options and explore the patient’s expectations and concerns (see Table 1). For the IMG residents, the mean difference in total score was not significant (−3.3 points, P = .13).
Through the development of our unique SDM curriculum based on a seven-step model for SDM and a novel SP case, we were able to demonstrate improvement in residents’ knowledge of and attitudes toward SDM in this pilot study at a single institution. The use of SPs has been well documented as a useful method for teaching communication skills.24,26 Although role-play and critical-care-based SP cases have been used in similar curricula, to our knowledge our SDM training is the first to use and pilot a primary-care-focused SP case.15,16,27,28
Our curriculum is unique in that it includes a skills component and evaluation of real-time clinical practice. In initial clinical recordings prior to completion of the SDM curriculum, residents scored a mean OPTION score of only 17.3 out of 48 possible points. This indicates that senior internal medicine residents at our institution may not regularly incorporate SDM into their clinical practice. Participants in this study struggled with explaining equipoise, discussing pros and cons of options, and eliciting a decision from the patient. We were able to demonstrate improvement in residents’ SDM skills use in real-time clinical practice after participation in the curriculum. Although similar findings have been demonstrated among attending physicians, this is the first resident curriculum pilot, to our knowledge, to evaluate the impact of SDM skills training in direct patient encounters.29
Improvement was seen in the total OPTION score for U.S. graduates, with the most significant improvement in the U.S. graduate residents’ ability to discuss pros and cons and elicit a decision. Improvement was also seen in the areas of listing options and discussing patients’ expectations and concerns, but, perhaps because of the limited number of recordings available, this improvement did not reach statistical significance.
Interestingly, the five IMG residents in this study did not demonstrate improvement in SDM skills despite improvement in knowledge, confidence, and importance. Although investigating a difference between U.S. and IMG residents was not a focus of this pilot, there are some potential explanations for this anecdotal difference. First, there is great variability in communication training in foreign medical schools, with limited training in skills such as patient-centered communication and motivational interviewing.30,31 Second, there is cultural variation in medical decision making, with many cultures taking a more physician-centered approach. Third, the patient-centered focus of SDM may be particularly challenging for IMG residents.32 Given these well-described differences, future work is required to better understand whether there are unique educational needs for this residency population in the area of SDM and patient-centered communication.
The results of our pilot study should be interpreted with the following limitations in mind. First, because it was a single-institution study, generalizability may be limited. The total number of clinical audio records available for study was small, and not all residents who participated in the curriculum provided a recording. We suspect that although not formally investigated, the low number of recordings was due to the complex logistics involved with recording in clinic. Also, while VA-based residents were invited to participate in the curriculum and the formative feedback aspects of evaluation, we were not permitted to include clinical recordings from the VA in the evaluation of the curriculum for research purposes. There may be some selection bias in the residents who turned in recordings and in the patient encounters chosen to record. To minimize this patient selection bias, we asked residents to record multiple encounters in hopes that we would receive a more accurate depiction of individual communication skills. Finally, though we were able to demonstrate short-term improvement in residents’ knowledge, attitudes, and communication skills regarding SDM, further study is required to assess the long-term impact of our curriculum.
Although residents’ perceived understanding of the relationship of SDM to high-value care improved with the training, their perceived understanding of how SDM relates to evidence-based medicine did not. This curriculum did not provide formal training on decision aids or other decision tools. Next steps from our group include providing formal training on the use of decision aids and a discussion of risk calculations with patients, with the goal to improve residents’ understanding of the relationship between evidence-based medicine and SDM. Another important future consideration is involving an end point regarding the patient’s perspective, such as surveying the patients about their experience with SDM in the clinical setting.
We developed a novel curriculum to teach University of Pittsburgh internal medicine residents SDM using didactic teaching with a seven-step model of SDM, SP practice, and direct feedback on recorded patient encounters. Our curriculum pilot resulted in improvement in a small group of residents’ ability to identify appropriate clinical situations for SDM, knowledge of equipoise, attitudes toward the practice, and use of SDM communication skills. We demonstrated in our single institution that SDM can be taught and assessed in a discrete time frame, which can foster development of an important ACGME communication milestone for internal medicine residents.
Acknowledgments: The authors would like to acknowledge Dr. Doris Rubio, Director of the Center for Research on Health Care Data Center at the University of Pittsburgh School of Medicine, for her statistical support and analysis. The authors would also like to thank Julie Mastro for her administrative assistance with this project.
1. Godolphin W. Shared decision-making. Healthc Q. 2009;12(spec no patient):e186–e190.
2. Hoffmann TC, Montori VM, Del Mar C. The connection between evidence-based medicine and shared decision making. JAMA. 2014;312:1295–1296.
3. Frosch DL, Moulton BW, Wexler RM, Holmes-Rovner M, Volk RJ, Levin CA. Shared decision making in the United States: Policy and implementation activity on multiple fronts. Z Evid Fortbild Qual Gesundhwes. 2011;105:305–312.
4. Prabhakar AM, Harvey HB, Platt JT, Brink JA, Oklu R. Engaging our patients: Shared decision making and interventional radiology. Radiology. 2014;272:9–11.
5. Green ML, Aagaard EM, Caverzagie KJ, et al. Charting the road to competence: Developmental milestones for internal medicine residency training. J Grad Med Educ. 2009;1:5–20.
6. Zeuner R, Frosch DL, Kuzemchak MD, Politi MC. Physicians’ perceptions of shared decision-making behaviours: A qualitative study demonstrating the continued chasm between aspirations and clinical practice. Health Expect. 2015;18:2465–2476.
7. Légaré F, Politi MC, Drolet R, Desroches S, Stacey D, Bekker H; SDM-CPD Team. Training health professionals in shared decision-making: An international environmental scan. Patient Educ Couns. 2012;88:159–169.
8. Fowler FJ Jr, Gerstein BS, Barry MJ. How patient centered are medical decisions? Results of a national survey. JAMA Intern Med. 2013;173:1215–1221.
9. Braddock CH 3rd, Edwards KA, Hasenberg NM, Laidley TL, Levinson W. Informed decision making in outpatient practice: Time to get back to basics. JAMA. 1999;282:2313–2320.
10. Hoffman RM, Elmore JG, Fairfield KM, Gerstein BS, Levin CA, Pignone MP. Lack of shared decision making in cancer screening discussions: Results from a national survey. Am J Prev Med. 2014;47:251–259.
11. Légaré F, Witteman HO. Shared decision making: Examining key elements and barriers to adoption into routine clinical practice. Health Aff (Millwood). 2013;32:276–284.
12. Légaré F, Ratte S, Stacey D, et al. Interventions for improving the adoption of shared decision making by healthcare professionals. Cochrane Database Syst Rev. 2010;(5):CD006732.
13. Légaré F, O’Connor AM, Graham ID, Wells GA, Tremblay S. Impact of the Ottawa Decision Support Framework on the agreement and the difference between patients’ and physicians’ decisional conflict. Med Decis Making. 2006;26:373–390.
14. Bieber C, Nicolai J, Hartmann M, et al. Training physicians in shared decision-making—Who can be reached and what is achieved? Patient Educ Couns. 2009;77:48–54.
15. Simmons L, Leavitt L, Ray A, Fosburgh B, Sepucha K. Shared decision making in common chronic conditions: Impact of a resident training workshop. Teach Learn Med. 2016;28:202–209.
16. Yuen JK, Mehta SS, Roberts JE, Cooke JT, Reid MC. A brief educational intervention to teach residents shared decision making in the intensive care unit. J Palliat Med. 2013;16:531–536.
17. Caldwell JG. Evaluating attitudes of first-year residents to shared decision making. Med Educ Online. 2008;13:10.
18. van der Horst K, Giger M, Siegrist M. Attitudes toward shared decision-making and risk communication practices in residents and their teachers. Med Teach. 2011;33:e358–e363.
19. Pellerin MA, Elwyn G, Rousseau M, Stacey D, Robitaille H, Légaré F. Toward shared decision making: Using the OPTION scale to analyze resident–patient consultations in family medicine. Acad Med. 2011;86:1010–1018.
20. Young HN, Bell RA, Epstein RM, Feldman MD, Kravitz RL. Physicians’ shared decision-making behaviors in depression care. Arch Intern Med. 2008;168:1404–1408.
21. Makoul G, Clayman ML. An integrative model of shared decision making in medical encounters. Patient Educ Couns. 2006;60:301–312.
22. Elwyn G, Frosch D, Thomson R, et al. Shared decision making: A model for clinical practice. J Gen Intern Med. 2012;27:1361–1367.
23. May W, Park JH, Lee JP. A ten-year review of the literature on the use of standardized patients in teaching and learning: 1996–2005. Med Teach. 2009;31:487–492.
24. Fryer-Edwards K, Arnold RM, Baile W, Tulsky JA, Petracca F, Back A. Reflective teaching practices: An approach to teaching communication skills in a small-group setting. Acad Med. 2006;81:638–644.
25. Elwyn G, Edwards A, Wensing M, Hood K, Atwell C, Grol R. Shared decision making: Developing the OPTION scale for measuring patient involvement. Qual Saf Health Care. 2003;12:93–99.
26. Barrows HS. An overview of the uses of standardized patients for teaching and evaluating clinical skills. Acad Med. 1993;68:443–451.
27. Hoffmann TC, Bennett S, Tomsett C, Del Mar C. Brief training of student clinicians in shared decision making: A single-blind randomized controlled trial. J Gen Intern Med. 2014;29:844–849.
28. Morrow CE, Reed VA, Eliassen MS, Imset I. Shared decision making: Skill acquisition for year III medical students. Fam Med. 2011;43:721–725.
29. Elwyn G, Edwards A, Hood K, et al; Study Steering Group. Achieving involvement: Process outcomes from a cluster randomized trial of shared decision making skill development and use of risk communication aids in general practice. Fam Pract. 2004;21:337–346.
30. Dorgan KA, Lang F, Floyd M, Kemp E. International medical graduate–patient communication: A qualitative analysis of perceived barriers. Acad Med. 2009;84:1567–1575.
31. Pilotto LS, Duncan GF, Anderson-Wurf J. Issues for clinicians training international medical graduates: A systematic review. Med J Aust. 2007;187:225–228.
32. Jain P, Krieger JL. Moving beyond the language barrier: The communication strategies used by international medical graduates in intercultural medical encounters. Patient Educ Couns. 2011;84:98–104.
Supplemental Digital Content
© 2018 by the Association of American Medical Colleges