Evaluation of a shared decision-making communication skills training for physicians treating patients with asthma: a mixed methods study using simulated patients

An essential aspect of patient-centered care is shared decision-making (SDM). SDM describes the process in which clinicians and patients make a joint treatment decision based on the best current medical evidence and patient preferences. The process of SDM requires that both the patient and physician are aware of the decision-making situation, mutually share information, and identify themselves as equal partners in negotiating clinical decisions [1].

Despite strong international advocacy for SDM and increasing implementation efforts, SDM is not yet the norm in routine care [2, 3]. Although SDM can be applied in most decision-making situations, [4] it is especially relevant in the treatment of chronic and potentially life-threatening diseases, where multiple evidence-based treatment options exist, and treatment decisions might have long-term consequences for the patient [5, 6].

Bronchial asthma is a prevalent chronic respiratory disease. Current clinical guidelines for the management of patients with asthma recommend SDM as key principles in asthma management. [7] SDM also benefits the quality of life for pediatric patients with asthma, improves asthma control, [8] and delays exacerbations of asthma attacks [9]. Moreover, SDM improves pharmacotherapy adherence and clinical outcomes of adult patients with poorly controlled asthma, [10] and patient commitment to therapy [11]. Despite promising evidence and integration in clinical guidelines, patient-centered communication and SDM continue to be poorly implemented in the care of asthma patients [12, 13].

Implementation research intents to understand what kind of health care interventions work in “real world” settings, how they work, and why. Also, implementation research investigates approaches to improve these interventions [14]. Implementation research on SDM has recently shown that interventions on multiple system levels, including different stakeholders, are needed for a change of care that fully embraces SDM [3]. However, strategies to foster the implementation of SDM usually focus on clinician-mediated or patient-mediated interventions [15, 16]. SDM training also addresses the attitudes and consultation behaviors of clinicians, which are important factors in the advancement and uptake of SDM [17]. Still, many existing SDM training programs have not been evaluated, and there are few studies that have assessed the outcome of SDM training on clinician SDM behaviors [18, 19].

Currently, there is poor quality evidence on the efficacy of interventions that aim to improve clinician SDM behaviors, mainly because of the heterogeneity of SDM training programs, [18, 19] and a lack of consensus on how to assess the adoption of SDM [15, 16]. Studies that have investigated the concordance between different perspectives on SDM have provided inconsistent results. [20‐25] Therefore, dyadic or triadic SDM measurement approaches including the perspectives of the clinician, the patient, and an independent observer are gaining increasing attention [26‐28].

Quantitative analysis of the effectiveness of interventions to improve the SDM behaviors of physicians continues to produce inconclusive evidence [15]. This can be partly explained by a lack of consensus on training content, [29] and on evaluation outcomes. [17] However, explorative qualitative studies on the perspectives of clinicians might provide helpful information [30] to design or adapt effective SDM communication skills training. Mixed methods study designs combine the historically distinct approaches of quantitative and qualitative research and are relevant to provide data to address complex health care interventions, [31] including SDM communication skills training. Recently, a mixed methods study assessed the perceptions of clinicians and the barriers towards the adoption of SDM after training. The results provided valuable input for the design of future training and implementation strategies [32]. Implementation research has highlighted the importance of eliciting the perspectives of stakeholders [33] to foster the implementation of SDM, [34] and clinicians are key stakeholders in this process.

A mixed methods study was undertaken to analyze the impact of shared decision-making (SDM) communication skills training for physicians treating patients with asthma on the SDM behaviors of physicians, and to gain information about its reception and needs for modification. Study participants were predominantly highly experienced male physicians working full-time in outpatient care. The results from the quantitative evaluation phase showed increased results from all three perspectives, the physician, the patient, and the observer, after the training. The average SDM observer-rated OPTION⁵ score, the primary outcome, remained relatively low when compared with the relatively high physician and simulated patient SDM ratings. The results from the qualitative evaluation phase revealed that most physicians reported a change of attitudes and behaviors, and mainly approved the training and its components. Physicians considered simulated patient consultations, including subsequent feedback and video-analysis, most beneficial. Moreover, they suggested the integration of real or more realistic simulated patient consultations or the inclusion of actual patients with chronic asthma.

The results from the quantitative evaluation phase showed that physicians improved their SDM behaviors after the training sessions, indicating the potential for SDM communication skills training to support the implementation of SDM in practice. Physicians and simulated patients judged the physician consultation behaviors as highly participatory. Lack of blinding to the study conditions and social desirability bias may explain these high ratings, but the average scores of the SDM-Q-9 and the SDM-Q-Doc in this study are comparable to those found in previously published studies, [25, 46] indicating a common and reproducible finding. Similar to previous studies, [21, 25] OPTION⁵ scores were relatively low compared to the physician and simulated patient SDM ratings, which indicated a lack of consistency in measuring SDM from different viewpoints. However, rather than interpreting this as a problem and aiming to reduce these inconsistencies, systematic integration of all three perspectives may provide the most accurate picture of the SDM process [23].

In the qualitative evaluation phase, physicians voiced their overall satisfaction with the training and stated they would recommend it, especially to interested and younger colleagues. Participating physicians appreciated the opportunity to receive diverse multi-perspective feedback to analyze their own communication behaviors during video-analysis. They also valued the opportunity to watch their colleagues perform consultations. Although a few physicians expressed some discomfort in the training process, most physicians considered these training components to be beneficial and educational. Physicians reported changes in attitudes and behaviors reflecting relational skills, which is one of two core SDM competency categories agreed on by an international group of experts [29]. There were few criticisms of the training, which may be due to the sample of voluntary participants, and the fact that one of the trainers conducted the coaching interviews. This fact is likely to have introduced a social desirability bias. However, a few physicians expressed reluctance to incorporate SDM in daily practice as they considered it incompatible with the limited time in their busy practice schedules, which is a common argument against the use of SDM [47, 48].

Our study has several notable limitations and strengths. First, an easily accessible sample of physicians were recruited who were willing to participate in the study. This meant that it was likely that the study sample largely consists of physicians with a particularly positive attitude towards SDM training, [49] and who might not have been representative for physicians treating patients with asthma in outpatient care in Germany [50]. Moreover, most participating physicians were male, and this may have introduced a gender bias in the results of this study. The study sample was relatively small despite nationwide recruitment through Mundipharma GmbH, and there was no information regarding the reasons for non-participation in the SDM training or the study, and because of this, physician selection bias was likely. Due to the convenience sampling strategy, we did not calculate a power-analysis prior to the study, but performed a post-hoc power-analysis. Second, there was a lack of study blinding of the physicians, simulated patients, and observers, which might have introduced measurement bias due to the Hawthorne effect and social desirability bias. Observer study blinding would have been possible if the study analysis had been limited to analysis of the transcript data, but it was considered to be important to use videotapes for the OPTION⁵ ratings. Third, coaching interviews were conducted by one of the trainers, which might have introduced bias. Moreover, no behavioral theory was used for the quantitative analysis of the interviews, which is another limitation of this study. Fourth, the use of videotaped consultations with student volunteers as patient surrogates for quantitative evaluation limited the external validity of the findings as transference to routine practice was not measured.

However, the use of simulated patients is also a strength of this study as it allowed investigation of physicians’ SDM performance of the physicians under equivalent circumstances, [20, 51] increasing the internal validity of the study. Another important strength of the study was the combination of a quantitative and qualitative evaluation in one study, which allowed comparison of the measured effects of the training with perceptions of the involved physicians, to obtain information regarding effective training components and any needs for modification or change. Another possible strength of the study was that change in the SDM behaviors of the participating physicians was evaluated from three perspectives, which followed the recommendation of a triadic approach to measuring SDM (observer, physician, patient) [26, 28, 52]. Finally, this study also included established measures with promising psychometric properties [41‐43, 53].

This study showed that SDM communication skills training for physicians treating patients with asthma has the potential to improve SDM performance, but would benefit from using real patient consultations. This study combined quantitative and qualitative methods, which allowed for a more comprehensive evaluation of the training compared to one methodological approach alone. Results from both evaluation phases indicate the potential to initiate change in the attitudes and behaviors of physicians. However, physician-reported changes in interviews do not imply implementation of SDM but reflect relational skills, a category of SDM core competencies considered essential by experts in the field [29]. It is conceivable that physicians need opportunities to reflect more on their daily practice and learn basic patient-centered communication skills before aiming to implement SDM [37]. Ongoing training including small group sessions, video analysis of real patient consultations, more opportunities for role-play, audit and feedback sessions at the practice sites could improve the transfer of newly acquired skills to daily clinical practice [54]. These ideas were also discussed by some physicians in this study. The introduction of patient-mediated interventions, including decision aids or initiatives such as the ‘Ask 3 Questions’ campaign [55] could complement the training. To achieve implementation in routine care, complex SDM interventions targeting both healthcare providers and patients have so far resulted in the most promising results [16].