Introduction

Active participation by patients in the clinical decision-making process is central to patient-centered care and can be achieved through shared decision-making (SDM) [1]. SDM has the following characteristics: (1) the clinician and patient are both involved in the decision-making process, (2) both parties exchange information with each other, (3) treatment preferences are exchanged, and (4) the final treatment decision is agreed upon by both clinician and patient [2].

These elements have been formalized by Elwyn et al. into what has come to be known as the traditional “three talk model” framework [3]. This model states that patient autonomy is built on providing high-quality information and supporting the deliberation process by processing the information and arriving at an informed treatment decision. This is achieved through three steps: the choice talk, the option talk, and the decision talk. In the choice talk, the clinician makes the patient aware that there is a choice to be made between several options (in a revised version of this model this talk is referred to as a team talk to emphasize the goal of inviting the patient as a partner in the decision-making process [4].) Next, the option talk consists of the clinician providing information on these various (treatment) options. Finally, in the decision talk, the options are examined in light of the patient’s preferences and a decision is made together when the patient is ready.

Recently, methods such as the SHARE approach have been developed to make the SDM process more intuitive for practitioners [5]. The SHARE approach converts SDM into a five-step process: Seek the patient’s participation, Help the patient understand their options, Assess their values and preferences, Reach a decision together, and Evaluate the decision. Although the elements of the SHARE approach correspond largely with existing models such as the three-talk model, using a five-step approach is beneficial for training purposes as it assists clinicians in remembering and applying the core elements of SDM [6].

Yet, while these models seek to simplify the SDM process and despite positive intentions, half of clinicians do not put SDM training into practice [7]. SDM implementation is impeded by numerous barriers, both on the clinician side and the patient side. The biggest barriers cited by clinicians are (in order of frequency): time pressure, lack of applicability due to patient characteristics, lack of applicability due to the clinical situation, a perception that patients do not want to participate in decision-making, and unwillingness to ask patients about their preferred level of participation [8]. Patient-reported barriers center on logistical factors (such as a perception that clinicians have busy schedules and a perceived lack of continuity between different clinicians/departments) and consultation factors (certain patient characteristics, trust, a perceived power imbalance [9].)

Despite these barriers, a few institutes have been successful in implementing SDM in clinical practice. In Denmark, the Center for Shared Decision Making at Vejle Hospital in the region of Southern Denmark (www.cffb.dk) was one of the forerunners of a national initiative to bring policymakers, health care professionals, and patient organizations together to make healthcare more patient-centered [10]. With a focus on improving patient-centeredness in oncology, the Center successfully implemented and evaluated SDM, supported by patient decision aids (PDAs) in paper format, in the care process for several cancer patient groups, to the extent that it is now being used as a model for wider regional implementation. Evaluations of the Center’s SDM initiative have shown an increase in patient engagement with no significant increase in consultation length in SDM groups compared to usual care in lung cancer and breast cancer, as well as significant reductions in decisional conflict and decisional regret for patients undergoing lung cancer diagnostics [11–13]. In addition, in a Danish national survey of patient experiences, patients rated Vejle Hospital’s oncology center above the national average on most fronts, particularly clinician efforts to involve patients in decisions [14]. Till date, the perspectives of the clinicians that practice at Vejle hospital have not yet been examined and this remains a gap in the Center’s existing evaluation of the program [15]. Understanding which aspects of the SDM implementation process clinicians find challenging and which strategies help overcome these challenges can add to our overall understanding of how to design and execute implementation initiatives that motivate clinicians to adopt these new ways of working.

The purpose of this study is to gain insight into practitioners’ experiences of the introduction of a SDM initiative into their workflow at Vejle Hospital, where ‘practitioners’ in this context refers to individuals involved in the SDM initiative either as a clinician, nurse, team leader, or researcher. Our study was guided by three research questions centered on: (i) practitioners’ initial attitudes towards SDM as a concept; (ii) their experience of the implementation process and factors that influenced implementation; and (iii) perceived outcomes. Such insights add to the current understanding of the motivational mechanisms that affect SDM implementation and may help inform similar implementation efforts in other contexts and regions.

Methods

Study population

Participants were identified through KDS who was leading the Center and recruited by means of purposive sampling [16]. The main criterion was that the participant should be well-acquainted with SDM, either through direct clinical practice or general awareness by virtue of having a strategic/leadership role. In order to obtain a range of perspectives, both oncologists and oncology nurses were approached (Table 1). Ten participants agreed to be interviewed; five oncologists from four specializations, and five participants with a nursing background. The sample is reflective of the Center’s strategy of applying SDM in a wide range of disease areas (aside from what are typically considered preference-sensitive conditions) and for smaller decisions along the treatment trajectory, such as nutritional decisions made with nurses. Seven participants (four oncologists and three nurses) had undergone the Center’s SDM training and were applying SDM with paper PDAs in routine practice. Aside from their clinical experience, three of the participants played leadership roles in SDM implementation; two leaders were an oncologist and nurse pair based within the Center and responsible for implementation and evaluation efforts.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Participant characteristics.

https://doi.org/10.1371/journal.pone.0259844.t001

Results

The following section presents the themes that emerged under each of the three research questions: practitioner attitudes towards SDM, factors that influence implementation, and perceived outcomes. Our findings revealed that participants’ attitudes regarding SDM prior to its implementation at Vejle Hospital ranged from skeptical to receptive. Four themes emerged under factors that influenced SDM implementation: raising awareness, leadership, reinforcement of new SDM behaviors and habits, and flexibility in PDA delivery. Strong leadership was related to the themes of raising awareness and reinforcement. Participants noted the effects of the SDM implementation on their experience of the clinical process and how measurement methods may need to be adjusted to ensure that SDM implementation is sustained. These themes and the related subthemes are summarized in the thematic map in Fig 1 and elaborated upon below.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Thematic map of themes and subthemes for each research question.

Solid lines indicate links between themes and subthemes, while dashed lines indicate relationships between themes.

https://doi.org/10.1371/journal.pone.0259844.g001

Discussion

Although healthcare is becoming increasingly patient-centered, SDM is yet to become the norm in most countries [22]. Even when clinicians are enthusiastic about patient engagement, misperceptions about what constitutes SDM, what distinguishes it from informed decision-making, and a lack of organizational support can thwart implementation efforts [23]. As a result, SDM and PDA uptake may fail to be sustained over time [24]. Within this context, we studied SDM implementation at Vejle Hospital, Denmark, where, under the leadership of the Center for Shared Decision Making, SDM and PDAs have been implemented with considerable success [11–13]. The purpose of this qualitative study was to examine practitioners’ experiences of the Center’s efforts to introduce SDM and PDAs into clinical practice. In doing so, we provide insights into the factors that may play a role in the hospital’s sustained success in implementing SDM. Our analysis revealed challenges to SDM spanning the pre-implementation to post-implementation phase and the strategies that were used to address them. These include overcoming clinician resistance, ensuring that training and tools are successfully put into practice, and facilitating sustained implementation efforts over the long-term through appropriate performance metrics and continuous communication between the various stakeholders (clinicians, nurses, team leaders, and researchers).

Frequently cited objections to SDM include doubt about its applicability to the clinician situation or the individual patient and perceptions that it will disrupt busy workflows [25,26]. These perceptions were present in the pre-implementation phase in our study and can result in resistance towards adopting a different way of conducting consultations [27]. There was a pronounced difference in our results in how clinicians and nurses viewed patient participation; nurses were enthusiastic about SDM due to their first-hand experience seeing the consequences of not taking patients’ wishes into account in treatment decisions, whereas for clinicians this shift happened after working with patients on developing the PDAs and was reinforced through repeated application of SDM principles. Previous findings from the Center have shown that pre-implementation OPTION scores indicated lower levels of SDM [11]. Communicating this disparity to clinicians may be the first step in shifting attitudes, as clinicians’ perception that SDM improves their practice makes sustained implementation more likely [28].

Interestingly, while time constraints are the most frequently cited barrier to SDM in the literature, this objection was not present in our results. A previous study conducted at the Center which assessed the impact of SDM and PDAs on consultation length found no significant increase, and in fact a decrease in variability of consultation lengths [12]. The authors speculate that the structured nature of the SDM consultations owing to the PDA format makes consultations more standardized in terms of procedure. The results of similar studies using in-consultation PDAs suggest that this format may be ideal for clinicians concerned about time constraints [29,30].

Another factor in increasing clinician motivation was empowering clinicians to take the lead in learning and training each other in SDM, particularly with minimal direct involvement from management. The Center has previously described its approach to creating ownership through involvement and this approach extended to its SDM training course, which is based on the Train-the-trainer model in which clinicians become trained in training their colleagues, thus becoming ambassadors for SDM [15,20]. This approach has been associated with positive results in various contexts, with the importance of a clinical ambassador or champion being emphasized in multiple studies [31–34]. Currently, there is no consensus in the literature regarding which training approach works best for teaching SDM in a clinical setting; measuring training effectiveness is a complex endeavor because there is no generally accepted best measure of SDM [35]. Training evaluation metrics tend to focus on the level of knowledge clinicians gained and indirect outcome measurements such as impact of the training on patient satisfaction and quality of the clinical interaction, while neglecting clinicians’ evaluations and satisfaction of their training [36]. While outcome measures are important metrics, it is also necessary to assess clinician satisfaction with the training itself, as positive appraisal has a significant impact on the motivation to learn from the training and put its principles into practice [37].

Moving from training to regular practice also requires measures to facilitate the development of new habits, provision of appropriate resources, and social support [38]. Previous findings revealed that even in spite of positive intentions, only half of clinicians put SDM training into practice [7]. Our findings suggest that the form and function of PDAs played a key role in this regard. A number of our participants considered the PDA to be as much a supporting tool for clinicians as it was for patients. Using a tangible PDA that is structured according to the steps of the SDM model can help reinforce the SDM training through repetition. Evidence from the Center shows that between two groups of clinicians who had received SDM training, the level of patient engagement and SDM measured by OPTION scores was significantly higher when a PDA was used [11]. This suggests that while training on its own may improve clinicians’ knowledge and skills to an extent, the repetition of using the five-step PDA may help solidify the contents of the training till it becomes routine [39]. There is preliminary evidence of PDAs being used as ‘scaffolding’ to facilitate SDM consultations, and certain PDA formats such as decision cards are specifically designed to function as collaborative tools rather than purely information tools [40,41]. However, we are unable to draw firm conclusions from this result as the overwhelming majority of PDAs are designed for patients to use outside the consultation [42].

Although PDAs are increasingly being offered in a digital format, preliminary evidence suggests that, given a choice, patients tend to use paper PDAs more frequently than digital PDAs and rate them higher in terms of overall satisfaction [43,44]. While both formats increase patient knowledge and users of digital PDAs tend to perform the values clarification function at a higher rate, the limited evidence still points to a digital divide in which there is a preference for print-based PDAs. Practitioners aiming to use PDAs as a supporting tool for SDM would therefore benefit from investigating their patients’ needs and preferences regarding decision support format as well as considering what role PDAs would play in the clinical workflow, i.e. whether they are deployed purely as an informational tool outside the consultation or used within the consultation to guide the SDM process.

It is also important to note that tools such as PDAs, while useful as supporting material, cannot compensate for the relational aspect of the SDM process. The core of SDM is effective communication between patients and clinicians [45]. Furthermore, it may not be possible to have a PDA for each clinical situation, and events such as the COVID-19 pandemic, in which some consultations must be conducted remotely, further underscore the need for empathic patient-centered communication [46,47]. This extends beyond the patient-clinician interaction to the way patients interact with the healthcare system as a whole. Steps taken to build patients’ trust ensure that SDM initiatives do not operate in isolation at the point of decision-making but become part of a culture of care that prioritizes the patient at all points of the pathway [48]. According to our participants, Vejle hospital’s ‘Patient First’ vision was the impetus for its SDM initiative, suggesting that SDM was a means to an end rather than an end in itself. This distinction has been emphasized by several authors, and empirical evidence highlights the importance of a shared vision, i.e. a specific ‘end’, in effective implementation [49–51]. This theme has not been extensively explored in the SDM context and could be a valuable area for future research.

The importance of leadership in sustaining implementation efforts over time was emphasized by multiple participants in our study. Leadership that is unsupportive of change can be a barrier to implementation, and leaders play a key role in shaping the environments that favor sustained application of training and PDAs over the long-term [52]. Alongside the regular SDM training for clinicians, the Center also developed a 3-hour training program geared towards leaders to acquaint them with the principles of SDM to help foster a positive implementation environment, according to K. Dahl Steffensen, MD, PhD (written communication, June 2021). Such initiatives can support the creation of a “learning health care system” in which performance metrics are aligned with the organizational vision for SDM implementation [53]. For instance, involving patients through SDM can shift previously ‘traditional’ treatment patterns based on clinical guidelines to more individualized patterns based on patient preferences; some patients may prefer less intensive treatments, some may choose to forgo certain treatments altogether, or otherwise deviate from guidelines [54]. Health care systems must account for this. Clinical departments that are evaluated using activity-based metrics, such as number of procedures performed or the number of consultations conducted, may have misaligned incentives that hinder SDM implementation [55]. This issue is increasingly on the policy agenda in several countries around the world, including Denmark, although the evidence of current value-based models is limited [56]. While hospital managers and team leaders must be aware of such models and their implications within their own contexts, wider actions at the policy level are needed to put such changes into motion.

Conclusion

This study describes practitioners’ insights and experiences regarding SDM implementation in their practice. We found that prior to implementation, many clinicians harbored a degree of skepticism about the value and practicality of SDM. Under factors that facilitated SDM implementation, four main themes emerged: raising awareness, leadership, reinforcement, and flexibility. Systematically measuring the level of patient involvement prior to SDM implementation helped make clinicians aware of the gap between current practice and an ideal SDM process. Supporting clinicians by means of role-play training and in-consultation PDAs helped in reinforcing the new habits associated with SDM so that implementation could be sustained. Sustainability was reinforced by creating an online platform where generic PDA templates could be customized by clinicians and/or nurses, allowing for greater flexibility and adaptability to different clinical contexts. Strong leadership was instrumental in driving these changes.

In conclusion, putting SDM principles into practice successfully requires clinicians to learn new behaviors and actions. This process may be facilitated by combining a top-down and bottom-up approach in which clinician leaders guide SDM initiatives while empowering clinicians to create supporting tools and learn from each other.

Supporting information

Acknowledgments

The authors thank the staff at the Center for Shared Decision Making for facilitating this study as well as the participating clinicians and nurses who volunteered their time and insights.