Adapting Pain and Opioid Prescribing Continuing Education to the Virtual Environment: Optimizing Program Factors While Navigating Context

Safer Opioid Prescribing (SOP) is a well-established Canadian continuing professional development (CPD) program focused on improving chronic pain care and opioid prescribing practices. It was developed in 2013, launched fully in 2014, and has had well over 1,000 health care professional participants, particularly from primary care, attending from every province in Canada. The program, described in detail elsewhere [1], was conceived as an education-focused policy intervention for the Canadian opioid crisis, and developers and faculty intentionally drew on principles of complex health interventions to inform the design and delivery of the program [2]. Increasingly, complex interventions are characterized as such not only because of their internal constitution but also because of being embedded in dynamic, unpredictable, and complex contexts [3]. SOP has undergone several major adaptations in response to important contextual shifts, such as the release of national clinical practice guidelines for opioid prescribing for chronic pain in 2017 [4] and the increasing emphasis on access to primary care–based opioid use disorder treatment [5]. The present article will focus on a further adaptation to a major contextual shift—namely, the unplanned adaptations to the global COVID-19 pandemic and the resulting physical distancing measures and travel restrictions. SOP was well prepared to adapt fully to virtual delivery. The Ontario opioid epidemic was characterized by disproportionate harms, such as overdose and death, in rural and remote communities [6], many of which may not have access to high-quality, accredited CPD programs without pharmaceutical industry involvement. As such, SOP is delivered in part as an online program that includes a series of three 90- to 120-minute synchronous webinars held monthly in the evenings after typical clinic hours to increase accessibility. A previous implementation evaluation demonstrated that this model is successful in over-reaching to the target audience of rural and remote primary care providers [1]. While being delivered online, the program was innovative in including key evidence-based elements of practice change–oriented CPD, including interactivity, multiple exposures, sufficient duration, and blended learning [7]. The webinar series is then followed by a full-day, in-person workshop that focuses on what faculty considered to be higher-order and challenging clinical skills relevant to chronic pain management and opioid analgesic prescribing. These include key communication skills, such as: 1) counseling around non-opioid therapy in environments where there is little accessibility to such therapies; 2) identifying complications from opioid use and communicating appropriate interventions, such as opioid tapering; and 3) making and communicating a diagnosis of opioid use disorder. In-person delivery of the workshop allows for extensive formal and informal interactivity. This interactivity is a key facilitator of solidifying a community of practice, which in turn helps to drive positive practice change. Thus, the central question in adapting this workshop for virtual delivery during COVID-19 was, “How do faculty build appropriate interactivity to nurture a community of practice and support the development of higher-order communication skills that can be applied in practice?” We identified two types of facilitators for the virtual adaptation: programmatic (relating directly to the program) and contextual (larger-scale factors not directly related to the program). Program faculty and administrators were already very experienced in delivering and administering virtual education. Thus, there were not large attitudinal, technical, or resource-related challenges to overcome. Second, as the webinars were a prerequisite for the workshop, the learners were already primed for virtual learning. Importantly, the webinars were explicitly designed with multiple forms of interactive learning with specific technical instantiations on the virtual platform (Table 1). Forms of virtual interactive learning Forms of virtual interactive learning Webinar-based interactive learning included individual reflection, sharing practice experiences, making clinical decisions, conducting clinical consults, referring to practice tools, examining case studies, and debriefing of responses. For the webinars, learners were able to see the presentation slides, hear the faculty, text chat with the faculty, and respond to poll questions. Faculty and learner video was not turned on, and learner microphones were not turned on in order to facilitate scalable, large-group learning. Thus, the majority of webinar interactivity was mediated by poll question responses and text chats. The workshop includes the same set of interactive learning methods with the addition of clinical role plays, which are focused on building communication skills. The role plays are highly interactive, include minimal supervision from faculty, and are conducted at the end of the workshop when learners have developed the most comfort in the learning environment. To replicate this virtually, we had to make wider use of the available functionality of virtual platforms, including specifically learner audio and video. We identified our primary goal for virtual workshop adaptation as being the development of an environment in which learners had sufficient familiarity and comfort to actively engage in the role plays. Thus, we settled on a stepped and clearly signposted approach to interactivity in the workshop. We began the workshop primarily with poll questions and chats, with which learners were already familiar from the prerequisite webinars. Then, as we reviewed and debriefed case studies around opioid-related complications and harms, we introduced learner discussion through audio and also video. Subsequently, learners were asked to present, through video, challenging cases around chronic pain management and opioid prescribing from their own practices. These cases were debriefed as a group through a combination of text chat and video discussion. Finally, in the last section of the day, we made use of breakout rooms to allow the learners to engage in full-video role plays, with faculty moving between the breakout rooms to offer support. This learning was supported by debriefing with the use of a checklist of key communication factors addressed in each role play. Each role play corresponded to one of the three challenging communication scenarios identified above. This adaptation was made iteratively as faculty reviewed and reflected on evaluative feedback from the learners. In addition to the programmatic factors, we recognized several contextual factors that facilitated virtual adaptation. In Canada, continuing medical education accreditation is overseen by two primary bodies: the College of Family Physicians of Canada and the Royal College of Physicians and Surgeons of Canada. Both colleges offer different grades of accreditation for educational programs, which are based on administrative, educational and ethical standards. For each college, the highest levels of accreditation offer three credits per hour of learning. Though we had previously advocated for such higher levels of accreditation for virtual programs that include high interactivity, Canadian accrediting bodies had restricted higher levels of accreditation to in-person programs. The SOP in-person workshop was accredited at the highest level, for a total of 18 credits. During the pandemic, higher levels of accreditation by both colleges were extended to virtual programs, thus making the virtual workshop an attractive learning option. Second, despite the pandemic’s disruption of normal medical practice, chronic pain care and opioid prescribing likely continued to be high learning priorities. This may, in fact, have become more pressing as the pandemic wore on and reports of skyrocketing opioid-related harms surfaced [8]. Although increasing harms may have been driven primarily by disruptions in street drug markets, the disruption to routine medical care may have been a driving factor as well. The majority of chronic pain care and opioid prescribing is conducted in community-based outpatient settings that, in Canada at least, were typically poorly resourced to adapt to the early pandemic. Community-based providers, in recognizing this, may have then continued to identify chronic pain and opioid prescribing education as a key learning priority. Third, much of the adaptation in primary care was being done virtually. For example, virtual primary care provision in Ontario increased 5600% during the pandemic, constituting 71.1% of all visits during the pandemic, from previously negligible levels [9]. Thus, primary care providers were suddenly delivering a new form of care, reflecting, of course, larger societal shifts in communication. This may have been another important contextual factor driving learners to seek out forms of education that mirrored the kind of care that they were newly providing. To provide some early indicators of the effectiveness of this transition, we compared evaluative data between the 2019 (pre-COVID) and 2020 (COVID) workshops. These evaluative data are routinely and anonymously collected at the end of each workshop and are used by faculty to continuously improve program delivery. No demographic or otherwise identifying information is collected with these evaluative responses. For these reasons, the University of Toronto Research Ethics Board declared the reporting of these data as exempt from ethical review. Three separate workshops were held in 2019, with a total of 40 individual evaluation respondents. Five workshops were held in 2020, with 50 respondents. Two workshops were held in June 2020 and were the first attempts at the virtual adaptation of the workshop, and the remainder were held in October and December 2020. First, we report evaluative responses from Likert scales relating to overall enhancement in knowledge and skills, relevance to practice, adequacy of active learning, and confidence with specific chronic pain and opioid prescribing clinical skills (Table 2). Each of the confidence questions relates directly to the challenging communication skills that are developed through the role plays. All scales range from 1 to 7, where 1 represents “strongly disagree” or “not at all confident,” and 7 represents “strongly agree” or “very confident.” We summarized the ratings with means and standard deviations, and we compared the cohorts with the Wilcoxon rank-sum test, with P< 0.05 considered as statistically significant. We conducted a sub-analysis comparing only the October and December 2020 workshops with the 2019 workshops, as we noted higher evaluative responses in the October and December 2020 workshops than in the June 2020 workshops, indicating improvements in the program over time. All analyses were conducted with SAS 9.4 (SAS Institute Inc., Cary, NC, USA) statistical software. Knowledge, relevance to practice, active learning, and confidence Bold text is to identify the key outcome type addressed by each evaluative question (knowledge and skills, relevance, adequacy of active learning and various kinds of confidence measures). Knowledge, relevance to practice, active learning, and confidence Bold text is to identify the key outcome type addressed by each evaluative question (knowledge and skills, relevance, adequacy of active learning and various kinds of confidence measures). The first important trend is that the mean ratings were consistently lower in the 2020 cohort than in the 2019 cohort, though there were no statistically significant differences in any of the ratings. Notably, differences were smaller in the sub-analysis. There were no statistically significant differences in any of the confidence-in-practice ratings in the overall comparison. The second group of data was in response to the question, “Please indicate which CanMEDS roles you felt were addressed during this educational activity.” Learners could choose as many as applied. CanMEDS-Family Medicine is a national competency framework that forms “an overall picture of the roles and responsibilities of Canadian family physicians along with the competencies required to support their work” [10]. The framework identifies seven specific roles that are often used to guide education planning, including: Family Medicine Expert, Communicator, Collaborator, Leader, Health Advocate, Scholar, and Professional. We compared these role identifications addressed by the SOP workshop between the 2019 and 2020 cohorts with a chi-squared test, which was two sided and where P < 0.05 was considered statistically significant (Table 3 ). Most roles were reported to be addressed less frequently in the 2020 cohort, except for Leader and Health Advocate. There was no statistically significant difference between the two cohorts. CanMEDS roles CanMEDS roles When comparing the evaluative outcomes between the pre-COVID, in-person workshops and the COVID-era virtual workshops, we see no statistically significant differences. These data suggest two important findings. First, at least with respect to the evaluative outcomes included here, we learned that it is possible to make a successful virtual transition of a highly interactive and communications-focused in-person workshop. This is the case specifically for the key challenge of developing communication skills and confidence in challenging clinical scenarios relating to chronic pain management and opioid prescribing. The second finding, however, is that this successful transition was not immediate. The first two iterations of the virtual workshop delivered in June 2020 had lower ratings, particularly with respect to enhancement of knowledge and skills, relevance to practice, and adequacy of active learning. Although adjustments to the program were made between the summer and subsequent workshops, we cannot determine with certainty whether the improvements were due to programmatic or contextual factors. Programmatically, faculty and administrators may have been more experienced with the preparation, timing, and delivery of the program. Although the comparison does control for factors such as program content and number of faculty, data about other factors, such as the number of participants in each session or the specific faculty who delivered each program, were not available for this analysis. Contextually, learners may have become more experienced and savvy with virtual engagement given the broader shifts toward virtual care and learning in the health professions. Besides this uncertainty around the cause of the improvements, the general trend toward lower ratings in the virtual workshops should be noted. It is possible that with more time and experience, these ratings would continue to improve. However, these lower ratings could also signal a small tradeoff in even mature virtually delivered programs between program accessibility and quality. We might expect even larger gaps in programs that do not build virtual interactivity sequentially and through multiple exposures over a period of months. One way to compensate for this tradeoff could be closer integration between various aligned education programs within a jurisdiction. For example, SOP participants have increasingly been linked to available mentorship and Extension for Community Healthcare Outcomes programs. The future of virtual continuing education in opioid prescribing education is unclear. We have demonstrated here that adaptations to virtual learning with high interactivity are possible, but they may require sustained investments in virtual delivery, with thoughtful and clear applications of active learning principles. For SOP specifically, once physical distancing measures and travel restrictions are lifted, it is an open question whether to continue to deliver the workshop virtually, to deliver it in person, or to deliver it as a hybrid model with simultaneous virtual and in-person learners. This evaluation tells us clearly, though, that the decision around delivery format will likely depend more on contextual than on programmatic factors. It will be difficult to sustain virtual delivery of the SOP workshop if, for example, there is a return to pre-COVID accreditation standards where only in-person learning can be accredited for multiple credits per hour. Likewise, concerns around virtual engagement fatigue are real [11]. Although health professionals may be willing to continue to engage in intensive virtual learning in the current environment, it is not at all clear whether this engagement will continue if in-person learning and travel become options again. It is certainly possible that there may be a reaction against virtual engagement, with a pent-up desire to return to in-person engagement. Ongoing attentiveness to the context-dependence of SOP and other programs will continue to be key to navigating these uncertainties . As we have adapted the SOP workshop from in-person to virtual delivery, we have learned several important lessons that will continue to guide this program and may be of value to other pain and opioid CPD developers, faculty, and learners. First, we have learned that successful adaptation is possible and is facilitated by both programmatic factors (such as the existing virtual education expertise) and contextual factors (such as accreditation norms). Second, we learned that clear program logic (understanding how the program works) and clear program principles (e.g., the commitment to incorporate CPD best practices) can help to navigate even rapidly changing contexts such as the COVID-19 pandemic. Third, we learned that iterative adaptation and improvement through changing contexts is supported by the routine and standardized collection of evaluative data that align with program objectives. Finally, we learned that adapting to virtual delivery may involve tradeoffs between improved accessibility and continuity, on the one hand, and overall program effects and quality, on the other . Evaluative data, despite being limited in supporting causal inferences around what specifically drove change, can provide helpful insight into the magnitude of these tradeoffs and thus support decision-making around appropriate program delivery modalities. Funding sources: None. Disclosures and conflicts of interest: AS receives an academic stipend from the University of Toronto as director of Safer Opioid Prescribing. The research of AS is supported by grants from the University of Toronto Department of Family and Community Medicine, Health Canada (Substance Use and Addictions Program), and the Canadian Institutes of Health Research. AS has no disclosures with commercial interests to declare. MH receives an academic stipend from the University of Toronto as faculty for Safer Opioid Prescribing. KH is employed as staff at Continuing Professional Development, University of Toronto, and has received consulting fees relating to opioid prescribing education development via Health Canada (Substance Use and Addictions Program). The authors acknowledge Drs. Kirsten Dixon and Kathleen Doukas for their contributions to workshop adaptation. We thank Dr. Rahim Moineddin for conducting the statistical analysis.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.