Abstract
Background: Most prescriptions for sedative–hypnotics are written by family physicians. Given the influence of preceptors on residents’ prescribing, this study explored how family physician preceptors manage sleeping problems.
Methods: Family physician preceptors affiliated with a postgraduate training program in Alberta were invited to participate in this mixed-methods study, conducted from January to October 2021. It included a quantitative survey of preceptors’ attitudes to treatment options for sleep disorder, perceptions of patient expectations and self-efficacy beliefs. Participants indicated their responses on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.” Respondents were then asked whether they were interested in participating in a semistructured qualitative interview that elicited preceptors’ management of sleep disorder in response to a series of vignettes. We analyzed the quantitative data using descriptive statistics and the qualitative interviews using thematic analysis.
Results: Of the 76 preceptors invited to participate, 47 (62%) completed the survey, and 10 were interviewed. Thirty-two survey respondents (68%) were in academic teaching clinics, and 15 (32%) were from community clinics. The majority of participants (34 [72%]) agreed they had sufficient expertise to use nondrug treatment. Most (43 [91%]) had made efforts to reduce prescribing, and 45 (96%) felt able to support patients empathically when not using sleeping medication. The qualitative data showed that management of sleeping disorder was emotionally challenging. Participants hesitated to prescribe sedatives and reported “exceptions” to prescribing, many of which included indications within guideline recommendations. Participants were reluctant to change a colleague’s management.
Interpretation: Preceptors were confident using nonpharmacologic management to treat sleep disorder and hesitant to use sedative–hypnotics, presenting legitimate use of sedatives as exceptional behaviour. Acknowledging social norms and affective aspects involved in prescribing may support balanced prescribing of sedative–hypnotics for sleep disorder and reduce physician anxiety.
Overuse of sedatives is associated with substantial health care costs and patient harm, such as falls and cognitive impairment.1–3 It is estimated that, at any given time, about 12% of Canadians use sedatives, increasing to 16.5% among older adults.4 Consequently, avoiding inappropriate use of sedative–hypnotics is a health care priority in Canada.5,6 Despite many efforts to reduce prescribing, the numbers of patients prescribed sleeping medication has remained relatively unchanged since 2013.4 To date, most efforts to reduce prescribing have targeted physicians, often emphasizing risks of inappropriate prescribing. Guidelines on the management of insomnia stress patient vulnerabilities to tolerance and dependence, and encourage physicians to use nonpharmacologic approaches as first-line management. 7,8 More recently, attention has turned to the challenge of deprescribing9 or avoiding prescribing in the first place, through use of public messaging such as the Choosing Wisely campaign.5
One area that has received relatively little attention is how physicians learn to prescribe, particularly in the workplace.10 This is important, as most postgraduate teaching and learning is service-based: residents work with and learn from their clinical preceptors. Family physician preceptors are powerful role models and key influencers of tomorrow’s prescribers of sedative–hypnotics. The development of medical learners’ clinical practice, including prescribing practice, is influenced by socialization through both formal and informal instruction,10,11 such as occurs with role models and mentors. Systematic reviews exploring the management of insomnia12 and the use of sedative–hypnotics by family physicians13 showed that these care providers have ambivalent attitudes toward prescribing sedative–hypnotics. Physician attitudes are important, as they influence confidence in practice, the ability to respond empathically to patient requests, and how physicians prescribe for different patient groups.14,15 Learners’ attitudes, beliefs and behaviour can change as a direct result of interactions with the teacher.16 A previous study exploring quetiapine use showed the importance of preceptor prescribing practices: participants noted the influence preceptors had on their prescribing after graduation.17 The preceptor–learner relationship has also been identified as a factor in the formation of prescribing behaviour for antibiotics18–20 and insulin.10
As far as we are aware, preceptor prescribing practices for sedative–hypnotics have not been examined, particularly in family medicine. This is important, as family physicians as a group are responsible for initiating and maintaining most prescribing in the community, including in nursing homes and long-term care facilities, where sleeping medication is often prescribed.21,22 In the present study, we aimed to develop an in-depth understanding of how family medicine preceptors manage and prescribe for sleep disorder in clinical practice. Our research question was “How do family physician preceptors in a family medicine residency program in Calgary, Alberta, manage sleep disorders in clinical practice?”
Methods
Study setting and design
This study was carried out in the Family Medicine Residency program at the University of Calgary from January to October 2021. Seventy-six family physician preceptors based in academic teaching clinics (n = 46) or community-based family medicine clinics (n = 30) provide clinical teaching supervision through this program to 70 family medicine residents each year.
This was a mixed-methods study consisting of a quantitative survey and a qualitative interview. Mixed-methods designs build on the individual strengths of qualitative and quantitative data, while supporting the contextualization of findings to make them more usable for stakeholders.23 We used an explanatory sequential (2-phase) design,24 first collecting and analyzing quantitative data, and then conducting interviews to help interpret the quantitative data and explore the relation between family physician attitudes and practices.
The research team consisted of 2 academic family physicians (M.K., M.O.), 2 pharmacists working in academic clinics (T.H., S.K.), a psychologist (T.H.) and a doctoral student (S.C.).
We carried out the study in accordance with the Checklist for Reporting Results of Internet E-surveys (CHERRIES).25
Data sources
Quantitative
We adapted a tool devised by Creupelandt and colleagues14 to explore behavioural factors affecting young doctors’ prescribing practice for benzodiazepines. We piloted this tool with 4 family physicians and modified it to better reflect the Canadian clinical setting (Appendix 1, available at www.cmajopen.ca/content/11/4/E637/suppl/DC1). The final survey elicited participant demographic characteristics and contained 14 questions exploring family physicians’ attitudes toward treatment options (5 questions), their perceptions of patients (2 questions), their self-efficacy beliefs (3 questions) and their prescribing self-efficacy (4 questions). Participants indicated their responses on a 5-point Likert scale ranging from “strongly disagree” to “strongly agree.”
Qualitative
We investigated preceptor attitudes using a semistructured qualitative interview guide that invited participants to outline their management in response to a series of vignettes. We developed vignettes26 to reflect common clinical scenarios, informed by the team’s clinical experience and the research literature, which indicates that physicians perceive some patients as more “deserving” of medication, whereas others are less likely to receive a prescription.13,27 We piloted the interview guide with 2 clinical preceptor colleagues (not interviewed for the study), and no changes were deemed necessary. A copy of the interview guide is presented in Appendix 2 (available at www.cmajopen.ca/content/11/4/E637/suppl/DC1).
Recruitment and data collection
All family physician preceptors affiliated with the urban Family Medicine Residency program, University of Calgary were eligible to participate. A letter outlining the study was emailed to the preceptors by the residency program; 2 reminders were sent at an interval of 2 weeks. The letter included a consent form and a link to the survey, hosted on the University of Calgary’s Qualtrics site. After respondents completed the survey, they were asked whether they were interested in participating in a qualitative interview. To incentivize participation, survey respondents were offered the opportunity to enter a draw for an Apple watch and were offered $100 to participate.
Interviews were conducted by 2 pharmacists (T.H. and S.K.), who were known to some of the participants as clinical team members. All participants who volunteered were interviewed (convenience sampling). Interviews were conducted via Zoom to accommodate public health measures during the COVID-19 pandemic, at a time convenient to participants. In keeping with procedures of semistructured interviewing, the interviewers used the schedule as a guide and did not adhere strictly to the order of questions but, rather, were flexible, following new ideas as raised by participants and using prompts to delve further as necessary.28 Interviews lasted between 40 and 70 minutes.
Interviews were audio-recorded and transcribed verbatim by a professional transcription service. Data collection continued until the research team felt that no new information was being added in interviews and that no additional interviews were required.29
We separated the participants’ sociodemographic data from interview data and anonymized the transcripts to protect participant identity. The identifying key was devised and retained by a member of the administrative staff.
Data analysis and integration
We calculated quantitative descriptive statistics for the survey data. To help interpret survey findings, we grouped the responses “disagree” and “strongly disagree” together. We analyzed the qualitative data thematically30 using NVivo version 12 software. Transcripts were read independently by M.K. and S.C. to identify preliminary codes. The preliminary codes were then reviewed by the entire team, and each team member independently applied codes to 2 interview transcripts. The team met to refine codes, and an inductive codebook was generated. S.C. and M.K. then coded the entire data set using a mix of deductive and inductive coding. Initial subthemes and themes were discussed within the team and refined iteratively through a series of team meetings until the final thematic structure was agreed on and no new themes were identified. Analysis was enhanced by team reflexivity, drawing on the diverse backgrounds of the research team members.
We used the survey findings to help interpret and deepen qualitative data analysis. Aggregated responses from the survey questions informed the codes used to explore the interview data. We further explored for areas of convergence and divergence between the 2 data sets.
Ethics approval
This study was approved by the University of Calgary Conjoint Health Research Ethics Board (REB20–0451). We provided all participants with consent forms detailing the length of time of their participation, the purpose of the study and the investigators involved. All participants provided consent before participating. Any required collection of personal information for the study (e.g., demographic information) adhered with the University of Calgary’s data protection policies and was outlined to participants on the consent form provided before participation.
Results
Of the 76 family physician preceptors affiliated with the urban Family Medicine Residency program, 47 (62%) completed the survey, and 10 family physicians were interviewed. Thirty-two survey respondents (68%) were in academic teaching clinics, and 15 (32%) were from community clinics. The participants’ demographic characteristics are shown in Table 1.
Demographic characteristics of participating family physician preceptors
Survey data
Practitioner confidence in managing sleep disorder using nonpharmacologic approaches was generally high (Table 2). Thirty-seven participants (79%) disagreed that nondrug treatment of sleep problems needs to be supported with medication, and 5 participants (11%) believed the advantages of sleep medication outweigh the disadvantages. Forty-two participants (89%) disagreed or strongly disagreed that management of sleep problems was the role of other professionals. Almost three-quarters of respondents (34 [72%]) agreed or strongly agreed they had sufficient expertise to use nondrug treatment, and most (33 [70%]) did not feel overwhelmed managing patients with psychosocial problems.
Survey responses
The results indicated widespread efforts to reduce prescribing of sleep medication (43 participants [92%]), with broad success (33 [70%]). Almost all respondents (45 [96%]) felt able to support patients empathically when not prescribing sleep medication.
Participants’ perceptions of patients’ expectations diverged. Twenty-one participants (45%) responded neutrally to the statement that patients would be dissatisfied if they did not prescribe medication, 11 (23%) agreed or strongly agreed with this statement, and 15 (32%) disagreed or strongly disagreed. Twenty respondents (43%) agreed or strongly agreed that it was difficult for family physicians to motivate patients to choose nonmedicine treatment, whereas 21 (45%) disagreed or strongly disagreed with this statement.
Interview findings
We identified 3 overarching themes from the interview data: preceptors’ general approach to managing sleeping problems, preceptor hesitancy prescribing medication (which we termed “sedative wariness”) and preceptors’ “exceptions” to their general approach (Table 3; Appendix 3, available at www.cmajopen.ca/content/11/4/E637/suppl/DC1).
Qualitative analysis of family physician preceptors’ management of sleeping problems
General approach to managing sleep disorder
Sleeping problems were viewed as complex, with a wide differential. History-taking, including “delving into the story behind the insomnia,” was key. Practitioners described this process as exploring the “bigger picture,” to “get a sense” of things or looking “in the background,” as well as identifying patients’ main concerns in relation to the impact of sleep difficulty and their expectations around sleep. Participants emphasized the importance of acknowledging patients’ distress. Treatment was described as challenging, necessitating a highly individualized approach to care. As noted by 1 participant, this made prescribing decisions for sleep disorder different from those for other disorders such as hypertension. Treatment was not considered “black and white” but, rather, required comfort with uncertainty.
Sedative wariness
Most participants promoted nonpharmacologic approaches to sleep disorder, including sleep hygiene and cognitive behavioural therapy for insomnia (CBT-I), as first-line treatment. Participants reported using a wide suite of educational resources, including customized patient information leaflets, to support nonpharmacologic management.
Participants hesitated to initiate drug therapy, while acknowledging they perceived that patients often expected them to do so. They described their attitudes toward medication in affective terms (e.g., “not loving” or “hating” sedatives). Only 1 participant openly acknowledged prescribing sedatives, describing their approach as “old school” in their openness to prescribe medication as a first-line treatment despite being aware of colleagues’ hesitancy to prescribe them. There was an ethical dimension alongside the emotional one whereby participants justified “rule-breaking” as “doing the right thing for the patient.”
When exceptions become the norm
Although participants stated explicitly they aimed to use non-pharmacologic management approaches and were wary of using medication, all reported they did, at times, prescribe sedatives. These were presented as deviations from routine care, collectively termed “valid exceptions.” Exceptions were nevertheless consistent across interviewees, representing a social norm. This included prescribing sedatives for short-term use, for patients presenting with acute grief or stress reactions, or in situations in which the physician carefully balanced risks and harms as part of harm-reduction strategies. Physicians’ descriptions of their prescribing in such situations was in keeping with guidelines, yet they presented this prescribing as concessionary and atypical, and expressed discomfort with “breaking the rules.” Rather than rely on diagnostic criteria, physicians often justified these decisions based on experience and personal knowledge of the patient.
A second exception to sedative prescribing was treating a colleague’s patient. Participants hesitated to change a colleague’s management. Management for this group of patients was presented as a “bridge” or “Band-Aid” that could be uncomfortable.
Synthesis of quantitative and qualitative results
Our results showed distinct areas of agreement between the quantitative and qualitative data. Interview data supported the survey findings that physicians are confident in managing sleep disorder and are committed to nonpharmacologic approaches as first-line treatment.
In interviews, participants evinced wariness and discomfort around prescribing sedatives alongside commitment to patient empathy and recognizing patient distress. Tension in balancing these factors may have contributed to the survey participants’ responses to questions regarding whether they believed patients would be dissatisfied if they did not prescribe medication, and whether participants found it difficult to motivate patients to choose nonmedicine treatment. All interview participants recognized that medication could play a role in managing sleep disorder, particularly in acute settings, but were hesitant to acknowledge this, describing prescribing in such situations as “exceptions.”
Interpretation
Our findings indicate that family physician preceptors in our setting emphasized an approach to sleep disorder management that starts by understanding the problem; sleep disturbance was viewed as a symptom, not a diagnosis. Preferred management involved patient education and nonpharmacologic treatments such as sleep hygiene and CBT-I. Participants navigated patient expectations, adopting shared decision-making approaches, on a case-by-case basis. However, participants were hesitant to prescribe medication, even when legitimately indicated, and they used holding strategies to manage colleagues’ patients.
To interpret our findings further, we reflected on the social history of sedative and hypnotic use. Although there was widespread use of this class of medications in the 1980s, contemporary literature and policy emphasize the inherent risks, with a focus on preventing or avoiding initiation of therapy and deprescribing.2,9 Although our participants expressed self-efficacy in management as individuals, their prescribing was also affected by social norms. There was a shared understanding that prescribing was something to be avoided, which resulted in emotional dissonance for participants as they attempted to practise patient-centred care.
One possible interpretation of our data is that family physicians have become avoidant of sedative–hypnotics, even in cases in which they may have a helpful role. This phenomenon has been suggested in the case of opioids.31,32 Social taboo around prescribing sedatives was noted in an earlier UK study, in which a reluctance on the part of older family physicians to be interviewed on the topic was observed. The authors suggested that this “may reflect the blame, shame and lingering responsibility felt by longer-practising family physicians.” 27 That study also showed that current practitioners felt “a sense of responsibility for avoiding the risks associated with past benzodiazepine use.” Balancing the risks and benefits of treatment while accounting for the patient’s context is at the heart of evidence-based medicine.33,34 Although our data suggest that is how family physician preceptors approached the management of sleep disorder, there was also an unintended collusion of medication avoidance.
Our findings draw attention to emotional and social aspects of prescribing, which can conflict with guidelines. There was an implicit corollary between being a “good physician” and avoiding sedative use. For some physicians, this social norm, as re-enforced by monitoring of prescribing35 and censure of colleagues, resulted in prescribing wariness or avoidance. This meant that some patients who could have benefitted from short-term use of such medication were denied treatment. If all sedative–hypnotic use is labelled harmful by default, patients may be denied medication that could ease distress in the acute setting36 or offset risks in situations of harm minimization. Critical reflection of prescribing dilemmas, which allows for exploration of social norms, expectations and empathic patient care, could enable a deeper sense of self-efficacy, identified as foundational for prescribing.13,14
Several participants perceived patients as expecting medication and felt it was challenging to dissuade them from wanting medication. However, research shows that, if patients are made aware of the risks, they are open to alternative approaches, particularly when providers are supportive.3,37 More open dialogue may thus in fact help relieve some of the emotional burden or dissonance that practitioners may experience. A complexity view of prescribing could explore prescribing praxis to inform educational interventions with a view to improving prescribing in family medicine residency programs.
The findings of this study will be used to inform resident training on the use of sedatives and hypnotics in our program.
Limitations
The study was set in Calgary, where 84% of primary care physicians practise in the patient medical home model of care,38 with access to behavioural health consultants and CBT-I. Barriers to extended services, such as access to mental health supports, are often cited as a reason why physicians are more likely to prescribe medication than to refer to other services. 13 The majority of our participants worked in academic teaching clinics, which may not have similar time pressures as community practice. Also, because all our participants were teaching physicians, our findings cannot be extrapolated to represent prescribing practices of all family physicians. Furthermore, our qualitative data are limited to physicians working in academic teaching clinics.
Our study would benefit from replication in other provinces, where clinical services may differ, and inclusion of broader community physician representation. Interviews were conducted by pharmacists known to the participants, which may have led to social desirability bias. The vignettes used drew on our clinical experience and may not reflect the experience of physicians outside our setting. Furthermore, despite our use of vignettes, our participants’ self-reported behaviour may have differed from actual behaviour when unobserved.
Conclusion
Family physician preceptors in our setting were overall confident in their use of nonpharmacologic management of sleep disorder, adopting evidence-based approaches as first-line treatment. Most were hesitant to prescribe sedative–hypnotics. Legitimate use of sedatives was presented as exceptional behaviour, when it was actually the norm across participants. Interventions targeting individual physician’s attitudes, complemented by a balanced approach to permitted use of medication, might result in more open discussion on the use of this group of medication. In this way, fundamental challenges of sedative prescribing, which are emotional and social, and are rarely addressed in guidelines, could be discussed rather than avoided.
Acknowledgement
The authors are grateful to the participants for sharing their time and experience.
Footnotes
Competing interests: None declared.
This article has been peer reviewed.
Contributors: Martina Kelly, Maeve O’Beirne and Todd Hill conceived of the study. Martina Kelly led the study design, with contributions from Sarah Cheung, Maeve O’Beirne, Todd Hill, Trudy Huyghebaert and Shelly Keller. Martina Kelly, Sarah Cheung, Todd Hill and Shelly Keller collected the data. Martina Kelly and Sarah Cheung led data analysis, with contributions from Maeve O’Beirne, Todd Hill, Trudy Huyghebaert and Shelly Keller. Martina Kelly and Sarah Cheung drafted the manuscript. All authors interpreted the data, revised the manuscript critically for important intellectual content, approved the final version to be published and agreed to be accountable for all aspects of the work.
Funding: This study was supported by the Addiction and Mental Health Strategic Clinical Network, Alberta Health Services.
Data sharing: Data used for this study cannot be shared. Any questions can be addressed to the corresponding author.
Supplemental information: For reviewer comments and the original submission of this manuscript, please see www.cmajopen.ca/content/11/4/E637/suppl/DC1.
This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) licence, which permits use, distribution and reproduction in any medium, provided that the original publication is properly cited, the use is noncommercial (i.e., research or educational use), and no modifications or adaptations are made. See: https://creativecommons.org/licenses/by-nc-nd/4.0/
References
- © 2023 CMA Impact Inc. or its licensors
In this issue
Article tools
Related Articles
Cited By...
- No citing articles found.