A scoping review of social determinants of health curricula in post-graduate medical education.

Social determinants of health are responsible for 50% of ill health. The Royal College of Physicians and Surgeons of Canada CanMEDS role of "physician advocate" requires physicians to attain competency in this particular domain, but physician trainees feel this is not well covered in their training programs. This study performed a scoping review of social determinants of health curricula that had been described, implemented and evaluated in post-graduate medical education. A search using MEDLINE(OvidSP) database, with search terms "residency," "curriculum," and "social determinants" with no age, language, and publication date restrictions was done. Researchers identified a total of 12 studies, all from the United States, in internal medicine (n=4), pediatrics (n=4), family medicine (n=2), or multiple (n=2) residency programs. Most curricula (n=8, 67%), were longitudinal, and most contained both patient or community exposure (n=11, 92%) and/or classroom-based components (n=10, 83%). Most (78%) curricula improved participant related outcomes, including exam performance, awareness regarding personal practice, confidence, improved screening for social determinants of health and referral to support services. Program specific outcomes were frequently positive (50%) and included resident satisfaction and high course evaluation scores, high representation of resident and faculty from minority groups, applicability of training to underserviced populations, and improved engagement of marginalized community members. When evaluated, academic outcomes were always positive, and included acceptance of scholarly projects to national conferences, publication of research work, grants earned to support health projects, local or national awards for leadership and community engagement, and curriculum graduates later pursuing related Masters degrees and/or establishing medical practices in underserved areas. Only one study reported a patient-related outcome, with advice provided by health care providers considered by patients to be helpful. Researchers used these results to design recommendations for creation of a post-graduate curriculum to address social determinants of health were provided.


Introduction
The social determinants of health (SDOH) are "the conditions in which people are born, grow, live, work and age, and the wider set of forces and systems shaping the conditions of daily life." 1 These determinants are broad and include income, education, housing, food security, employment and job security, social safety net and inclusion, gender, race, early childhood factors, access to health services, aboriginal status, and disability. In Canada, 50% of ill health can be attributed to SDOH, 2 a greater proportion than biology, genetics, or the health care system. Thus, it is essential that health care providers know how to identify patients at risk of ill health due to SDOH, know strategies and resources to address health disparities, and then advocate for patient access to these resources.
The Royal College of Physicians and Surgeons of Canada developed the "CanMEDS" framework to guide medical education and physician practice. 3 The CanMEDS role of "Health Advocate" includes the physician's competence to identify patients' SDOH. Resident physicians feel that while it is important to include Health Advocacy in medical curricula, it is not currently well covered, perhaps because it is deemphasized relative to other CanMEDS roles. 4,5 Barriers to effectively delivering health advocacy education include the lack of published materials, lack of clarity regarding the physician's role, unclear learning objectives, insufficient role modeling, and lack of a reliable method of assessment of competency. 6 Given the marked impact SDOH have on patient health, the paucity of effective SDOH education impairs a physicians' achievement of competence and inevitably leads to detrimental effects on patient outcomes.
A working group of program directors from Canadian family medicine post-graduate training programs met in 2011 to identify global health competencies for Canadian Family Medicine Training. 7 Similarly, medical students at three Canadian medical schools established e-learning as an effective tool to enhance global health knowledge. 8 However, there remains a paucity of SDOH curricula interventions that are implemented and evaluated at the post-graduate level.
The primary objective of this study was to identify studies in which the intervention of a post-graduate medical education curriculum focused on SDOH has been described, implemented, and evaluated in any patient population with disparities in health status or access. The secondary objective was to generate recommendations to design an effective SDOH curriculum within the CanMEDS framework.

Search strategy of scoping review and study selection criteria
A literature search confirmed that a review with the same objectives had not previously been published. A professional librarian (EU) ran a search in MEDLINE(OvidSP) for articles describing curriculum development centered around the topic of Social Determinants of Health, in post-graduate medical education (Appendix A). Search terms included "residency," "curriculum," "competency based education," "interdisciplinary studies," "problembased learning," "social determinants of health," "health status disparities," "social medicine," "health services accessibility," "health equity," "socioeconomic factors," "medical students," "medical education," "program development," "program evaluation," "evaluation studies," and "validation studies." Due to the lack of studies from outside the United States in the first search, an "additional search" was performed. Search terms included "health advocate," "CanMEDS," "internship," "residency," "curriculum," "competency-based education," "interdisciplinary studies," and "problem-based learning." There were no age group, language, or publication date restrictions for either search (Appendix A). Studies that were references in articles, but not found in the primary search, were hand searched and collected.

Data synthesis and analysis
We included studies if a curriculum intervention to teach SDOH in a post-graduate medical education program was described and evaluated, with outcomes reported. We described studies by author, participants, post-graduate specialty program, and country, intervention (format, content and activities), e63 method of evaluation, and outcomes. Studies were excluded if they did not describe and evaluate an intervention, and if outcomes were not reported.
Studies were summarized in Excel 16.16.7 for Mac. Meta-analysis of data was not possible because of the heterogeneous nature of available studies' interventions and outcomes. All data analysis was performed in Excel 16.16.7 for Mac.

Ethics approval
No direct patient contact was made for this study, and all accessed studies are a matter of public record. Therefore, there was no requirement to apply to Health Science Research Ethics Board for this study.

Results
Search of MEDLINE(OvidSP) yielded 352 references ( Figure 1). Researchers screened references and removed duplicates yielding 29 manuscripts. Researchers added four references by hand searching and reviewing references. We reviewed 33 full manuscripts. There were 20 studies that were excluded, leading to 13 manuscripts describing 12 studies which were included in this review (Appendix B). Studies were excluded because the curriculum was not focused on SDOH (n=7), the study was not at the post-graduate level (n=5), the evaluations or outcomes were not described (n=4), the studies were duplicates (n=3) or a literature review (n=1).
An additional search yielded 333 references, leading to 15 manuscripts reviewed. None of these manuscripts met inclusion criteria, and all were excluded because they were not at the post-graduate level and/or did not describe a curriculum intervention and outcome.

Curriculum format and content
Eight studies' curricula (67%) had longitudinal components that spanned across one or more year of residency ( Table 1, Appendix B). 11,[13][14][15][16][17][18][19]21 These were organized either as blocks, incorporated either into ambulatory rotations or as separate tracks or specialties within a residency program. Three studies (25%) incorporated short interspersed learning components as modules, conferences, or half-day sessions. 10,12,20 One study (8%) used a single-day learning experience. 9 The curriculum content included resident needs assessment, literature appraisal, expert opinion and community input; however, this was too heterogeneous across studies to make any conclusions.

Evaluation methods
All curricula were evaluated using methods from one or more domains ( Table 2, Appendix B). Most studies (75%) focused on participant outcomes such as frequency of SDOH screening, use of resources, referrals made to supporting services, knowledge assessments, measures of resident engagement, selfreflection, e-portfolios, achievement of core competencies, or pre/post surveys that assessed preparedness, skills, attitudes, competence, and knowledge. [9][10][11][12]14,[16][17][18][19]21 The majority (67%) of studies also evaluated program specific outcomes such as course evaluations, surveys of resident and faculty demographics, multisource feedback or analysis of the cost and time to implement the curricula. 11,13-15,17-21 Academic benchmarks were evaluated in three (25%) studies, and included scholarly projects, publications, presentations at conferences, funding from donations or grants, and post-graduate career tracking. 15,16,19,21 Patient-related outcomes were reported in only three (25%) studies, and included the results of patient surveys, the rate of outpatient clinic use and patient care outcomes (which were not further described). 10,12,14

Evaluation outcomes
Studies that evaluated participant-related outcomes often (78%) reported positive outcomes ( Table 2, Appendix B), which included improved competence or knowledge score on exams, increased awareness regarding personal practice, change in attitude or confidence, improved screening for SDOH with increased referral to support services, and increased exposure to mentors and potential career paths. [9][10][11][12][17][18][19]21 Studies that evaluated program-specific outcomes reported positive outcomes in 50% of studies. [13][14][15]18 These included high course evaluation ratings or resident satisfaction, high rates of representation of resident and faculty from minority groups, applicability of training and knowledge to care of underserviced populations, lack of need for additional funding required to implement curriculum, and improved engagement of marginalized community members agreeing to teach, to advise and to host field trips. All (100%) studies that evaluated academic benchmarks showed positive outcomes, with acceptance of scholarly projects to national conferences, publication of research work, grants earned to support health projects, local or national awards for leadership and community engagement, and curriculum graduates later pursuing related Masters degrees and/or establishing medical practices in underserved areas. 15,16,19,21 Only one study (33%) reported a positive patient-related outcome, with advice provided by residents reported by patients to be helpful. 12

Discussion
In Canada, half of all ill health can be attributed to social determinants of health. 2 Almost 15% of Canadians live at or below the poverty line, with marginalized populations disproportionately represented. 22,23 Low income associates with shorter life expectancy, inability to afford prescribed medications, and higher incidence of chronic health issues such as diabetes and hypertension. 22,23 However, physicians lack the training and resource knowledge to address these health disparities. [4][5][6] This scoping review revealed 12 studies of curricula interventions focused on SDOH, with vastly e65 heterogeneous program elements, evaluation methods and outcomes.
The most common curriculum format was longitudinal exposure throughout post-graduate training. This typically incorporated continuity of care clinics, encouraging familiarity and therapeutic alliance with the patient population and allied providers. 20 These curricula developed comfort and knowledge of available resources, with improved understanding of the multiple aspects involved in the care of underserved populations. 13 Curricula that were short and interspersed left trainees with improved understanding of the importance of SDOH, but the intervention was of inadequate duration to gain full understanding and competence on how to address these issues. 20 Studies had varied outcomes and evaluation methods, with most focusing on participant or program specific elements. Most of the emphasis was placed at the level of resident reaction and learning that occurred immediately after completion of the intervention. Only one study repeated evaluation at the end of residency and two years post-graduation 17 to assess for sustainable retention of attitudes, behaviours and knowledge. There were a few studies that assessed change in resident behaviour related to frequency of SDOH screening, use of resources, and referrals made to supporting services. However, this assessment was immediately post-intervention and did not look at sustainability of behaviour change. Only three studies evaluated outcomes related to academic benchmarks. These studies showed success in achieving target objectives such as scholarly work, and post-graduate engagement in leadership, advocacy, and careers focused on vulnerable populations. However, whether these curricula enhanced patient experience is unknown. Patientrelated outcomes were infrequently evaluated, and reported positive results only once. 12 One study sought to evaluate patient outcomes, 14 but authors didn't describe what these outcomes entailed, nor were these outcomes reported in the publication.
Although studies' interventions and outcomes were heterogeneous, there were several recurring themes in successful curricula that had positive academic and participant related outcomes. Curriculum recommendations were extrapolated and generated from these observations. The format of a SDOH curriculum should be organized as a longitudinal experience to enable residents to have repeated exposure to elements within the program. This will solidify understanding, competence, and comfort level required to effectively manage and advocate for patients suffering from health inequities.
The content included within curriculum should vary depending on the needs of the community one serves, and should use multiple sources such as resident and community needs assessment, literature review, and local expert opinion. Programs should all include at least a basic introduction that describes the SDOH and how they impact patient care and health outcomes. Other suggested content items include resources relevant to a patient's socioeconomic status, considerations in the patient-provider interaction, leadership and health advocacy, interprofessional collaboration, and project management.
There should be multiple types of learning activities so trainees have comprehensive exposure to the curriculum content from different perspectives. Programs should include patient or community exposure for hands-on, experiential learning. This should be supplemented and reinforced with either classroom-based and/or independent learning, or for more rigorous programs, a resident research or advocacy project.
Evaluation methods should be comprehensive and should target objectives that investigate participant, patient, program and academic related outcomes. The Kirkpatrick's model of reaction, learning, behaviour, and results 24 can help organize this framework. This would assure study not only of shortterm outcomes such as participant satisfaction and learning, but also long-term change in physician behaviour, physician engagement in vulnerable populations, scholarly achievements, and most importantly the impact on patient care outcome.
There are a number of strengths of this scoping review. Firstly, studies found were from multiple training specialties, including curricula that incorporated several training programs. This increases the generalizability of the data, increasing the likelihood that recommendations apply to a broad range of specialty programs. Secondly, this study is a current and timely review on post-graduate curricula of SDOH, an essential topic as per the Royal College of e66 Physicians and Surgeons of Canada, and indeed a topic that has been identified as insufficiently taught. While study interventions and outcomes were heterogeneous, recommendations could be made based on the best available evidence.
This study does have an important limitation. All studies were performed in the United States. The United States is the 12 th wealthiest country in the world, 25 so less wealthy countries that may suffer greater health disparities are underrepresented.
Physicians continue to lack sufficient training in social determinants of health, an area that has a major impact on patient health. This scoping review provides recommendations for a post-graduate curriculum to extinguish this gap in physician training. While recommendations are made, this review also identifies that further research is required to clarify learning objectives and the physician's role within the domain of health disparities. (i) Standard elective evaluation forms to assess strengths/weakness and perception of educational experience.

Conflicts of interest
(i) High level of resident satisfaction with learning experience, applicability to training, and understanding aspects of care for underserved. (ii) Weaknessess: low patient volume, and lack of on-site preceptor in some clinics