Introduction

Involvement of patients in health care decisions is associated with increased patient satisfaction, quality of care, and reduced use of resources.¹ Shared decision-making (SDM) is defined as “an approach where clinicians and patients share the best available evidence when faced with the task of making decisions, and where patients are supported to consider options, to achieve informed preferences.”² When making clinical decisions based on best available evidence, the values of the patient are incorporated by discussing the choice process, different options, and consecutive decisions.^2,3 Use of SDM involves patients in an active conversation about their needs, values, and preferences.^1,4 Guided by the health care team's experience, patients and/or their families are better informed about treatment options and potential benefits, risks, and costs. SDM with decision aids has been associated with better outcomes, including better medication adherence⁵ and improved patient knowledge and perception of risks when making treatment decisions.⁶

Consequently, SDM training should be implemented in physician education. Successful SDM training provides participants with a solid understanding of the concept of SDM and the communication skills necessary to use it.⁷ Unlike the notion of informed consent, SDM concepts may not be familiar to many practicing physicians and are not yet widely adopted in patient encounters.^8,9 Although continuing professional development may promote SDM in clinical practice,¹⁰ there are few SDM training programs available for internists, and their effectiveness has yet to be evaluated.^1,11 We developed and assessed the impact of a brief SDM training program for senior internal medicine residents.

Methods

The study was conducted at 3 teaching institutions in Switzerland, the 900-bed University Hospital in French-speaking Geneva, and the 550-bed Triemli City Hospital and 260-bed Waid City Hospital in German-speaking Zurich. The background and study design were presented to 39 senior internal medicine residents, and 27 (69%) agreed to participate. All were involved in general internal medicine patient care on inpatient units and in the emergency department.

The study used a single group, pretest-posttest crossover design. SDM training consisted of a 2-hour workshop and pocket material for use in daily practice following the workshop. The impact of the intervention was measured through objective assessment by MD evaluators during standardized patient (SP) encounters, participants' self-assessments, and SP assessment of resident performance. The postintervention measurement took place no sooner than 2 months after the intervention. We used Elwyn and colleagues' framework for the definition, elements of training,¹² and assessment of SDM.¹³ We also considered other sources, such as the integrative model of SDM,¹⁴ the MAGIC Program,¹⁵ and other competencies for SDM.^16,17

Results

Of the 27 physicians in the study, 11 were from Geneva and 16 were from Zurich (participant characteristics provided as online supplemental material). All participants were internists, with experience as senior residents ranging from 6 months to 16 years. None had prior training in SDM; the majority had previous training in communication skills and evidence-based medicine. Gender ratio and mean age were similar for participants and non-participants (female 50%, mean age 36.8 years).

A total of 54 data points were analyzed from pre- and postintervention testing. There was a mean of 74 days between testing sessions (range 63–104), and the workshop took place no longer than 3 days after preintervention testing.

Postworkshop, mean (SD, range) objective ratings of SDM completeness improved from 65 (13, 37–84) to 71 (12, 48–89), a relative increase of 9.2% (ES 0.53), and participants reported being more comfortable with related knowledge (ES 1.42) and skills (ES 0.91), and practicing SDM (ES 0.96). Self-rated communication skills were unchanged. Participants reported they more often applied SDM concepts in practice (ES 0.71), but did not use the associated supporting tools and decision aids. SPs rated the overall competence in discussing the treatment decision as high, without significant change between the 2 sessions (ES 0.19). SPs felt more comfortable with how participants discussed treatment options posttraining (ES 0.44).

The average length of consultations did not differ statistically. Before the training, encounters ranged from 12.9 to 20.7 minutes (mean 17.5); postintervention they ranged from 13.0 to 27.0 minutes (mean 18.4). All outcome data are shown in the table.

table Outcomes for Evaluation of Shared Decision-Making Training Program

View Fullscreen

Discussion

A brief training intervention improved the competencies of senior internal medicine residents for shared decision-making with their patients, as measured by objective SDM completeness ratings and participant self-reports. Of note, participants involved patients in decision-making at a remarkable high degree at baseline (percent score of 65), potentially reflecting senior residents' acquisition of SDM competencies through clinical experience, and training in communication skills and evidence-based medicine. In a systematic review of 33 published studies,¹⁸ only 1 study using a preintervention assessment reported a comparably high OPTION total score at 68.¹⁹ Postintervention, 2 studies showed high average OPTION total scores of 63 and 50, respectively,^20,21 lower than the scores in our study. Of all health care providers rated with OPTION, none of the studies included in the systematic review assessed hospital-based internists.¹⁸ Studies mostly included general practitioners, and their mean score of 23 was lower than that of medical specialists, who more frequently had average scores ≥ 25. The approach to SDM may differ between hospital and outpatient care, due to differences in length and context of the medical encounter. General practitioners know their patients for longer time periods, and may complete a SDM approach over several encounters. The context of the hospital-based encounter may require physicians to involve the patient in decision-making at a single time.

Brief interventions can have a positive effect, as shown by an increase in teacher effectiveness after focused training²² and improved SDM performance of internal medicine residents in an ambulatory context.²³ In our study, the improvement was moderate, potentially reflecting a ceiling effect due to participants' high degree of patient involvement before the intervention. Longitudinal data suggest that SDM training resulted in improvement that was sustained over time.^20,24

From an economic perspective, there may be concerns that SDM use will result in longer patient contact times and be regarded as “nice to have, but difficult to afford.” Our data showed an increase in contact time of less than 1 minute (5% of total time) after the training, with no statistical difference. This is a relevant point in the discussion around the cost and benefits of improving patient communication and involvement.

Limitations of our study include use of a multicomponent intervention that makes it impossible to assess which aspect was responsible for the improved SDM competence. In addition, the modified OPTION scale to assess completeness of the SDM process was not validated. Finally, referring to Kirkpatrick's model,²⁵ we used objective measures for the impact of training on learning and performance, but data on changes in practice were self-reported.

Future studies are needed to examine which interventions work best, and whether the pocket card alone may increase residents' SDM abilities.

Conclusion

A brief intervention improved the objective SDM competencies of senior residents in internal medicine, improved evaluations of patient-resident encounters, and promoted self-reported use of SDM in practice. This approach may be useful in training junior residents and physicians in internal medicine and other disciplines as well as medical students.