Motivational interviewing for prevention in Swiss family medicine: Opportunities and challenges

Highlights • This study examines the experience of family doctors with motivational interviewing.• Family doctors have different views on the meaning of motivational interviewing.• Implementation varies depending on the features of doctors, patients, and situations.• Motivational interviewing supports family doctors in their usual preventive activity.• Family doctors’ motivational interviewing poses economic and professional challenges.


Introduction
This article presents the results of a qualitative study conducted with 19 family doctors (FDs) who participated in Girasole, a pilot project implemented in Southern Switzerland between 2016 and 2018 to promote motivational interviewing (MI) in family medicine. Participants were trained in MI and invited to use it in their practice to encourage physical activity among patients. The objective of this paper is to present the analysis of the implementation of the intervention, specifically to provide insights into the experiences of FDs' when integrating MI in their daily practice, and to discuss the potential benefits and challenges of this approach in family medicine.
Due to ageing populations and chronic illnesses, primary prevention that promotes healthy behaviours is increasingly necessary (Hajat and Stein, 2018). Primary care plays an important role in this context by assisting patients in identifying their health needs and priorities, and navigating the healthcare system effectively (Wallace et al., 2015).
summarizing, and reinforcing change talk are essential to this approach, as well as behaviour change strategies like setting the scene, assessing importance and confidence, exploring options, and setting goals and a plan (Anstiss, 2009). MI provided by FDs seems promising in terms of efficacy and acceptability: lifestyle/health challenges are addressed by influential, well-known health professionals through a patient-centred approach that promotes a balanced doctor-patient relationship.
MI in primary care has been studied and promoted in many countries (McKenzie et al., 2018;Fatusin et al., 2021;Boom et al., 2022), but a systematic evaluation has proven difficult due to the heterogeneity of the interventions (VanBuskirk & Wetherell, 2014). Generally, there is consensus on the appropriateness of MI in primary care (Rubak et al., 2009;Boom et al., 2022), though debate remains concerning the training of physicians (Dragomir et al., 2019), long-term efficacy on patient behaviour (Thepwongsa et al., 2017;VanBuskirk & Wetherell, 2014;Rubak et al., 2009), and which health professionals are best suited to provide these services (VanBuskirk & Wetherell, 2014). FDs' involvement in particular has been criticised as cost-ineffective ( Van-Buskirk & Wetherell, 2014), and little data exist on their experience with MI or on difficulties in implementing it in their daily practice (Thepwongsa et al., 2017;Boom et al., 2022).

Girasole and its approach
The Girasole pilot project was conceived and promoted by the Health Promotion and Assessment Service (HPAS), a cantonal division responsible for non-communicable disease prevention, health promotion, and health evaluation in Canton Ticino. This Italian-speaking region in Switzerland is characterized by a low rate of physically active individuals and a high frequency of FDs' consultation (Petrini & Roth, 2014;Stanga, 2014). This made Ticino an ideal setting for implementing a pilot project focused on promoting physical activity through MI in primary care. A total of 19 FDs from across Ticino were invited to participate based on their previous experience or interest in health promotion, along with their personal connections with the HPAS. Funding was granted by the Swiss Federal Office of Public Health.
Developed between 2014 and 2016, Girasole combined two programmes previously implemented in French-and German-speaking Switzerland: Paprica, a training course developed by the University Medical Policlinic of Lausanne, aimed at enabling FDs to integrate patient counselling on physical activity into their daily practice (Koutaissoff et al., 2012); and Health Coaching, a programme of the Swiss College of Primary Care Medicine for behavioural counselling and health promotion in primary care (Neuner-Jehle et al., 2013). From Paprica, Girasole adopted the training concept and material on physical activity, while Health Coaching provided the general methodology of MI. Associated materials were developed for use across stages, including a directory of physical activities in Ticino and worksheets for tracking goals, means, and measures of behavioural changes over time. Girasole's communicative approach was characterised by openness, empathy, and support of patient autonomy. FDs were invited to act as "coaches," helping patients define their goals, means, and pace of behavioural change.
The project started in September 2016 with a two-day training session led by a FD specialized in psychosomatic medicine and skilled in MI, utilizing simulated consultations with actors. Following the training, FDs provided positive feedback through a questionnaire assessing its organization, content and relevance. The intervention phase spanned from October 2016 to March 2018. Over the course of 18 months, FDs were encouraged to integrate MI into their practice. They were asked to recruit adults aged 40-75 -a particularly sedentary population in Ticino (Petrini & Roth, 2014;Stanga, 2014) -who were willing to adopt a more active lifestyle. During the intervention phase, the HPAS offered posttraining support through three meetings led by the same trainer for discussing challenges, sharing strategies, and fostering a learning-bydoing approach.

Theoretical and methodological framework
To explore physicians' experience with MI, we conducted a qualitative thematic study following Braun and Clarke (2006). This approach was chosen for its ability to offer a comprehensive and detailed analysis of complex data. Considering MI as a social innovation, we adopted Dearing's theory of innovation diffusion (2009) as a theoretical framework. This theory suggests that five components are key to understanding the diffusion of innovation: the innovation, the adopter, the social system, the individual adoption process, and the diffusion system. We focused on the dimensions of innovation, adopter, and individual adoption process to systematically explore the physicians' perceived characteristics of MI, their readiness to use MI, and their implementation of MI into daily practice. Guided by this framework, we developed semistructured guidelines for qualitative data collection through focus groups and interviews.
This study did not fall within the scope of Art. 2 and Art. 3 of Swiss law on human research and did not require ethics approval. However, all participants provided oral informed consent. The study was also conducted in accordance with the Declaration of Helsinki (World Medical Association, 2013).

Sample
The sample was composed of the totality of the 19 FDs selected to participate in Girasole. The sample represents 5.5% of the physicians in family medicine in Ticino at the time of the intervention in 2017 (FMH, 2022). As a result, the sample is not representative of the average situation of family medicine in Ticino or Switzerland, as young people, women, and doctors who share their practice with one or more colleagues are over-represented (Senn et al., 2016). The characteristics of the participants are summarised in Table 1.

Data collection
Data were collected twice: at the beginning of the intervention phase (Fall 2016) to explore participants' attitudes, expectations, and previous involvement with health promotion in primary care; and at the end of it (Spring 2018) to evaluate participants' experience with Girasole. Based on the participants' availability and geographical proximity, we conducted five mini-focus groups and three individual interviews at the beginning of the project (with all 19 physicians) and four mini-focus groups and four individual interviews at the end (with 16 physicians involved; two had left the project after some weeks, while a third was unable to meet). Mini focus groups (up to 4 people) were privileged to harness group dynamics while gaining in-depth insights in participants' experiences (Krueger & Casey, 2014). Individual interviews were conducted in specific cases due to participant preferences and organizational constraints.
Discussions were based on semi-structured guidelines with open- White/European Geographic area 8 physicians working in an urban area 7 physicians practicing in a semi-urban area 4 physicians practicing in a rural area Kind of medical practice 3 physicians working in solo practice 9 physicians sharing their practice with a colleague 7 physicians working in group practices ended questions (e.g., "Think about the patients to whom you proposed MI this week. How did you decide to offer them MI?"), but participants were encouraged to address any aspects they considered important. The focus groups were facilitated by both authors, whereas the interviews were carried out by the first author. All sessions were conducted in person. Data were audio recorded, transcribed, and anonymised. Additionally, both authors participated as observers in the two-day training session and the three follow-up meetings, taking extensive notes to enrich the narrative dataset. Focus groups, interviews, and follow-up meetings lasted 60-120 min.

Data analysis
Data were analysed thematically following Braun and Clarke's approach (2006), which intends to identify, analyse and report patterns within data through a series of iterative phases: familiarizing with the data by re-reading the transcripts, coding the data, identifying themes, reviewing themes, creating relationships between themes, and writing the report. The first author conducted the coding process inductively, assigning codes to manifest and latent content. Regular meetings were held between the two authors throughout the whole analysis process to discuss coding and interpretation. Disagreements were resolved through discussion and by continuously referencing to the data. At the end of the analysis, a consensual general codebook was developed and utilized to recode all the transcripts with the software Atlas.ti. This process enabled the assessment of the relevance and comprehensiveness of the results, the verification of the data saturation, and the identification of relevant quotations. Results were presented to participants during a final session in Spring 2019, obtaining respondent validation.

Results
We have analysed the implementation of MI by FDs using three dimensions of the innovation diffusion theory (Dearing, 2009): the innovation, the adopter, and the individual adoption process. In the first paragraph (innovation) we explore participants' perception of the Girasole approach. In the second paragraph (adopter) we focus on the FDs' readiness to engage in MI. The third paragraph (individual adoption process) highlights how FDs' concretely incorporate MI in their daily practice. A fourth and last paragraph provides an overview of the findings and proposes a taxonomy of the physicians based on their degree of adherence to the principles of the Girasole approach. Quotes from participants are provided in Table 2 and identified by number. To protect physicians' privacy, the quotations are anonymous.

Innovation
Most participants rated the Girasole approach positively in terms of process of use and effect. They considered it handy, as it was accessible (it could be applied despite the short training), structuring (it improved the current health promotion activities) (#1-2), and flexible (it was adaptable to physicians' and patients' needs) (#3). Some participants thought it was constraining (protocol and tools were sometimes considered artificial and intrusive) (#4-5). Only a few found it incompatible with their daily practice due to workload constraints and a professional identity focused solely on the treatment of illness (#6-7).
Most participants recognized the benefits of the Girasole approach in terms of transforming their professional activity, rather than exclusively focusing on patients' behavioural change: the Girasole approach provided legitimacy to their ongoing health promotion practices (#8), enhanced the gratification of consultations (making them more pleasant and interesting) (#9-10), and fostered generative discussions where new actions could be explored with the patient (#11-12). Some participants perceived the approach as risky (introducing uncertainty into the consultation) (#13), while a minority viewed it as utopian (believing that MI was bound to fail due to the predominance of socio-economic factors on lifestyle) (#14).

Adopter
Two key aspects influenced physicians' adoption of the Girasole approach: their view of health promotion in medical practice and their approach to lifestyle risk factors. Regarding health promotion, all FDs considered it central to their role (#15-16). However, there were varying positions among them: some were willing to prioritize their patients' lifestyles, while others focused more on disease and secondary prevention (#17).
Risk factors were likewise understood differently. One group was determined to address risk factors, emphasizing the role of lifestyle in determining health outcomes (#18-19). A larger group tended to interpret and sometimes justify unfavourable health behaviours, arguing that they were meaningful to the patients (#20). A final group accepted risk factors as resulting from social conditions beyond the control of patients and physicians (#21). These various attitudes influenced both physicians' readiness to participate in the project and their ongoing commitment to it.

Individual adoption process
Physicians adapted the Girasole approach to their professional styles and habits, developing a variety of methods for the identification, recruitment, follow-up organisation, and coaching of patients. Regarding identification, a few physicians made an extended selection: they offered MI systematically by integrating it "routinely" into all consultations, as they believed that every patient could benefit from MI at any given moment (#22). Other physicians made a targeted selection, based on the characteristics of the patient (i.e., their receptiveness or friendliness) (#23-24) or on context (i.e., time) (#25-26).
Regarding recruitment, some physicians scheduled specific MI consultations (#27). Others would improvise, at the risk of not having enough time (#28). Still others briefly introduced MI during regular consultations, but rescheduled appointments ad hoc.
To follow up, some physicians arranged dedicated consultations, despite feeling embarrassed about summoning patients and subsequently charging them for reasons that were both not strictly clinical and not solicited by the patients themselves (#29). Others chose to incorporate follow-up within regular consultations (#30), while some left the initiative to the patients.
As for coaching, we observed three forms of conducting MI. A few participants used "expert" coaching, feeling confident and managing adequately time and MI tools (#31). Most practitioners adopted an "amateur" coaching style, appreciating the approach, but feeling they were lacking training and experience; they used MI selectively with comfortable patients, yet struggled with time and content control during consultations (#32). Others provided a "personalised" coaching approach, seeing MI as a general model but taking liberties in adapting style and materials; they emphasized the physician's authority and charisma over specific techniques (#33).

Taxonomy of physicians
According to their accounts, physicians displayed diverse perspectives and practices regarding MI. A closer examination of the data reveals that the key factor underlying this variation is physicians' adherence to the Girasole approach and its four core principles: importance of health promotion to FDs; legitimacy of FDs' intervention in patients' lifestyles; appropriateness of the motivational approach; and mastery of motivational techniques and tools. Assessing the level of adherence to these principles allows for the classification of participants into four ideal types: resistant, critical, interested, and convinced.
Two physicians can be characterized as resistant: they showed reluctance in addressing health behaviours and patients' lifestyles, instead favouring curative measures or secondary prevention. They considered the Girasole approach utopian and incompatible with their professional identity and abandoned the project after a few weeks. Two other participants, designated critical, showed curiosity in new patient interactions but maintained an asymmetric relationship and did not fully embrace active listening, open questions, and patient empowerment. They believed the Girasole approach constrained their naturalness in patient interactions and adopted a personalised coaching style. The largest group, consisting of 11 interested physicians, were willing to intervene in patients' health behaviours and lifestyles using a motivational approach. However, they faced challenges in applying its tools and techniques due to limited expertise, concerns about patient selection, and time constraints. They felt uncertain about the impact of MI on consultations and adopted an "amateur" coaching style. While intrigued by the approach, they lacked confidence in mastering it, resulting in inconsistent and selective application. In contrast, a final group of four convinced physicians fully embraced the Girasole approach. They found it valuable and transformative, and employed MI with suitable expertise. The taxonomy is summarised in Table 3.

Discussion
Girasole aimed to incorporate MI into primary care to promote physical activity among patients at risk. The qualitative analysis confirmed physicians' interest and adoption of MI in their practice, indicating its feasibility. Previous research supports the importance of addressing patients' health behaviour for FDs and its potential impact on job satisfaction (Boom et al., 2022). However, implementing MI in primary care faces barriers such as time constraints, patient-provider relationship, and patient characteristics (e.g., social background, level of understanding), as highlighted in the literature (Östlund et al., 2015;Reno et al., 2018;Jager et al., 2020;Boom et al., 2022). Our study confirms these findings, demonstrating that diverse attitudes and approaches exist within a group of FDs selected for their interest in health promotion. Therefore, achieving systematic use of MI for health promotion may prove challenging, even among motivated and interested physicians.
Beyond feasibility, our results also show that MI has several potential advantages in family medicine. First, MI provided structure, tools, and legitimacy to ongoing preventive activities. This suggests that more systematic work in health promotion and prevention could help FDs better define their professional identity and social role in contrast to specialist physicians. Second, according to the participants, patients appreciated MI and felt that prevention had a place in consultations. However, this study provides no evidence on the impact of MI on patients' behaviour, which remains difficult to measure outside controlled clinical/pharmaceutical environments.
Some aspects of MI are puzzling. Medicalisation, for instance, has always been a thorn in the side of preventive interventions. Medicalisation stands for "the expansion of medical authority into the domains of everyday existence" (Metzl & Herzig, 2007, 697). In Girasole, physicians proposed health promotion activities to people who presented no symptoms of disease and who did not necessarily ask for prevention advice. Some FDs expressed their discomfort on this point, which raises ethical questions about adopting MI. These are all the more crucial when the physician must emphasise the risks connected to a patient's behaviour. The patient becomes labelled "at risk," which can lead to anxiety, fear, diminished self-esteem, malaise, and stigmatisation (Gérvas et al., 2008). A second problematic point concerns the status and legitimacy that FDs can gain from carrying out systematic health promotion interventions. In this study we observed that the lifestyle recommendations were largely common sense (e.g., avoiding a sedentary lifestyle), as were the methods to achieve these goals (e.g., having a dog to walk). This highlights two weaknesses in FDs' practice of MI: the commonplace of the issues discussed, and the borrowing of techniques from other professional groups (e.g., psychologists or social workers). These weaknesses may question the legitimacy of FDs as healthcare providers. A third issue pertains to the cost-efficiency of FDs in conducting MI. Nearly all participants identified time as a challenge for MI. With a shortage of physicians in primary care, there are arguments in favour of prioritizing curative activities (Gérvas et al., 2008). Involving other and less costly healthcare professionals, such as midwives or nurses (Weingarten and Matalon, 2010), could present a potentially cost-effective alternative. Furthermore, there is a question of whether the promotion of health should be primarily addressed by the community and public policy, focusing on aspects like ensuring the availability of affordable and healthy food. Of course, this is not an either/or debate: community action and public policy can coexist with MI conducted by FDs. However, the efficacy of the latter remains uncertain, as confirmed by the participants who expressed resistance.
This study has some limitations. Since participants were selected based on their interest in health promotion, diversity was limited, even though we were able to partially address this point by interviewing the two doctors who left the project after training. Additionally, FDs were given a questionnaire to provide feedback on the training, but the skills acquired during training were not measured. Finally, data relied on physicians' narratives, but direct observation of MI in the clinical setting and exploration of its impact on patients were not feasible.
Still, this study offers a comprehensive understanding of physicians' implementation of MI. The taxonomy provides a general framework for understanding FDs' positions towards MI and their ways of implementing it. This may be used for statistical studies of the profiles of FDs in relation to health promotion and MI, and for developing tailored training to the different groups. Any further attempt to implement programmes like Girasole should recognise the existence of different individual attitudes and define customised approaches and training.

Conclusion
This analysis of the Girasole pilot project provides new inputs for discussions on the feasibility, efficacy, and acceptability of MI in family medicine. Findings show that MI is a viable activity that offers flexible tools and new ways of interacting with patients to meet the challenges of non-communicable and chronic diseases. The implementation of this approach and the training programmes should consider the diversity of physicians' attitudes towards health promotion and intervention in patients' lifestyle, and their familiarity with MI and its techniques and tools. At the same time, the issues of medicalisation, physicians' status loss, and low cost-effectiveness should not be underestimated.

Sources of funding
This work was supported by the Swiss Federal Office of Public Health (grant no. 16.022197/704.0001-790/2). The sponsor did not have any involvement in the design and realisation of the study or in the writing of the paper.

Table 3
The taxonomy of family doctors based on their adherence to the Girasole approach.