Exploring culinary medicine as a promising method of nutritional education in medical school: a scoping review

Background Dietary modifications are considered a first-line intervention for chronic disease management, yet graduating doctors still report not feeling competent to counsel patients on their diet. Research has focused on methods to address this shortfall in physician competency, including culinary medicine. Culinary medicine is an approach to education that involves hands-on food and cooking learning experiences to equip participants with tools for improving the nutrition behaviour and health of their future patients. Despite positive findings in the efficacy of these interventions, they differ markedly in approach and target, which therefore fails to provide adequate evidence that could serve to guide future culinary medicine interventions. Objective A scoping review to synthesize the existing literature on culinary medicine interventions that are offered during medical training. Methods Online databases were used to identify literature published prior to April 2022 that involve a hands-on culinary medicine component to nutrition and examine academic impact, feasibility and acceptability. Results Twenty-four studies met the eligibility criteria. Despite promising gains in nutrition knowledge, confidence and high acceptability of the programs, large variations exist in delivery method, setting, and course content between programs. There is a lack of program cost reporting and long-term follow up of participants, inconsistent evidence for improved nutrition attitudes amongst participants, as well as geographically limited adoption of such programs. Conclusions The findings of this research demonstrate a clear increase in interest in the use of hands-on culinary medicine programs as educational tools, evidence of feasibility in implementation, and improved student nutritional knowledge, skill and counseling compared to a traditional didactic curriculum. The quality of culinary medicine research studies is increasing and the aims of research are narrowing to focus on how culinary medicine can positively impact medical education. The findings from this review will aid in legitimising culinary medicine as an effective delivery method of nutritional education in medical programs. Supplementary Information The online version contains supplementary material available at 10.1186/s12909-022-03449-w.

of patients conducted by health professionals [2]. As a result, nutrition care that supports patients to eat well is considered as an essential, first line step in preventing and managing chronic disease [1]. However, doctors report not feeling equipped to provide adequate nutrition care to patients, despite their acknowledgement that nutrition is a useful and necessary part of patient care [2,3].
Culinary medicine is an emerging field of education related to meal planning, preparation and cooking skills in combination with counseling practices to promote health behaviour change in patients/clients [4]. Interest in culinary medicine has rapidly increased among many medical schools across the United States and other countries as a potentially low-cost, high impact strategy to equip students with practical nutrition skills that can be employed for nutrition care in clinical settings [5]. Several pilot culinary medicine programs have occurred [5] which theorise that culinary medicine initiatives completed in undergraduate medical programs may produce doctors that feel more confident providing nutrition care to their patients [6,7]. These programs have demonstrated significant and positive impact on medical student's attitudes, knowledge, and competencies with practical, hands-on culinary skills and nutrition knowledge, which ultimately supports their patient counseling [8]. Many aspects of culinary medicine pilot programs are unique and wide differences exists in structure, duration, and setting. While these programs provide a wide range of suitable information, they are not helpful for recommending a standardized implementation of culinary medicine in medical programs. A recent scoping review by Asher et al., 2021 explored literature that provides culinary and nutrition training to and by health, education, and culinary professionals. This study demonstrated that the feasibility and outcomes of culinary medicine programs warrant further investigation as there is still insufficient evidence of the impact these programs on practitioners to promote health behaviour changes in patients [4] and the viability of culinary medicine [9]. This scoping review aims to synthesise the existing literature on culinary medicine offered to medical students within their four years of medical school education.

Overview
This review critically synthesised literature examining culinary medicine programs offered to medical students in the setting of a four year medical program. The review was informed by PRIMSA methodological guidelines. The protocol was registered with PROSPERO (ID: CRD42020210766) and the reporting followed PRISMA guidelines [10]. This scoping review was initially registered with PROS-PERO as a systematic review. This review was changed into a scoping review because it offers a wide exploration of culinary medicine research and its characteristics and gaps, rather than answer the specific question of whether culinary medicine programs cause, and adequately assess, change in medical students' attitudes, skills and knowledge of nutrition.

Search strategy
An electronic literature search was conducted initially in June 2020 with the support of an academic librarian. A second electronic literature search, identical to the initial search, was conducted in April 2022 to provide a more comprehensive review, before project completion. Computer-based searches were conducted within MED-LINE (EBSCOhost), SCOPUS, Web of Science, ProQuest (central), Taylor and Francis online, and SAGE journals. The search string entered for MEDLINE was ("Culinary Medicine" OR "Nutritional Medicine" OR "Culinary Education" OR "Food as Medicine") AND ("medical") AND ("Student" OR "Program" OR "Education" OR "Training") with the "scholarly (peer reviewed) journals filter selected. The search strategy was adapted to each database to accommodate for variation in database search engine function. Filters to exclude articles that were not peer-reviewed were used if the database provided that feature. The search strings and filter settings for each database used have been provided in Additional file 1.

Selection criteria
All full-text, peer-reviewed journal articles published prior to 22 April 2022 were eligible for screening. This eligibility criteria was enlisted to ensure that this scoping review synthesizes the existing literature on culinary medicine that explores the feasibility of culinary medicine programs, which may offer evidence-based advice for future interventions.
• Population: Medical students, or medical graduates if they received the intervention during medical school. • Intervention: Any form of culinary medicine education (curriculum or elective) received during medical school. Culinary medicine was considered to be used if the article referred to the activity in that way, or if a practical/hands-on cooking related teaching experience occurred. • Comparison: Studies must have had a control group (difference between groups) or baseline measures (report a difference over time). • Outcome Measures: Changes in medical students' nutritional knowledge, skills, or attitudes were considered relevant.
Preliminary title and abstract screening occurred on 50 records by importing the details into Microsoft Excel and screening by LB, TR, KS, LA, and JT to establish interrater reliability. Interrater reliability using the selection criteria was 98% and deemed sufficiently high to progress to screening.
All records were then imported to Covidence for screening and duplicates were removed upon import. The titles and abstracts for all articles were independently screened twice, by any two of the researchers (TR, KS, LA, LB and JT), facilitated by the Covidence software system. Full-text articles were retrieved and screening to be included in our final review was performed by one researcher and verified by a second researcher (TR, KS, LA or JT). Conflicting screening results were resolved during regular team meetings. Additional articles were identified via ancestry searching and were manually added for screening using the same procedure facilitated by Covidence. The screening process is outlined in Fig. 1 using the PRISMA 2020 diagram [10].

Data charting
Data charting was performed independently by one researcher (TR, KS, LA, or JT) and confirmed by another researcher (TR, KS, LA, JC, or JT). Variables for which data was sought included study characteristics such as setting, population, aim of research, and description of program, "course description", intervention, outcomes, and notable findings (Table 1 and 2). Trends in data were also sought in the format of a yes/no table. Quantitative and qualitative studies were included in this scoping review and were analysed using meta-synthesis, an interpretation of results to offer novel information regarding the viability of culinary medicine in medical education.

Quality appraisal
The methodological quality and reporting of each study was assessed independently, in duplicate using the CASP tool, version 2020 [11]. CASP has specific appraisal checklists designed for different study designs, both within the study and across studies, making it an appropriate tool for this scoping review. Agreement was reached on all of the appraisal items. With regards to satisfying each CASP criteria, papers were assigned a score of 0 for 'yes' , 1 for 'can't tell' and 2 for 'no' . The CASP scores were a sum of the individual criterion scores for each paper. Where scores differed, discrepancies were resolved through team discussions.  To explore an alternative way to provide nutrition education without adding hours to the formal curriculum by (1) surveying student perceptions regarding current nutrition education, (2) surveying student interest in attending a nutrition elective, (3) selecting how the elective could best be delivered, and (4) running and assessing participants' reactions to a short experimental version of the elective 3 session elective (2 hands-on culinary medicine sessions and 1 didactic session in between-total hours NR); The first culinary session dedicated to knife skills and culinary basics; the didactic session covered coronary artery disease and nutritional preventative measures followed by 2 clinical cases in the form of an essay assess-   Open ended questions with prompts Across two pilot cohorts, retention rates and attendance rates were more than 89% and 96% respectively, with mean recommendation scores of 6·25 and 6·67 out of 7, indicating good feasibility and acceptability. Notable methodologies include use of peer-reviewed competency scales for participants self-evaluations. Significant improvements in students' counseling competencies in nutrition and obesity reported. No significant or mixed results reported for attitudes to nutrition counseling, culinary techniques and personal dietary habits  of a dietician's role on a care team, personal beliefs on how physicians' nutrition can impact patient outcomes, cooking skills, and perception of the importance of nutritional education in medical programs; chi-squared analysis and calculated odds ratios reported with Yates correction At baseline most students (92%) strongly agreed or agreed that counseling patients on food choices is essential, but only 29% felt confident in their ability to counsel patients about food choice. and just over half (54%) were familiar with the Mediterranean diet. Post intervention 92% of students felt comfortable discussing nutrition with patients (OR = 26·8), 97% felt familiar with the Mediterranean diet (OR = 25·59) and 93% were confident they understood the role of a dietitian (OR = 23·3). Post intervention, there was a significant increase in the number of students feeling confident in the kitchen (OR = 32·6), and a significant decrease in students who believed healthy eating is expensive or time consuming (OR = 0·43). reporting to think healthy eating is expensive. At baseline students also reported poor understanding of the role of a registered dietician and only 54% of students were familiar with the Mediterranean diet. After the intervention, 92% of students reported feeling confident in counseling patients on nutrition and 93% of students reported to understand the role of a registered dietician. After the intervention, there was a significant increase in the number of students feeling confident in kitchen skills, and a significant decrease in students reporting to think healthy eating is expensive Compared to the pre-course survey, students who responded "strongly agree" in questions related to nutrition counseling in the post-course survey were 26·5 to 31·3% higher (p < 0·05)In the post-course survey (n The introductory 3 sessions were well received and repeated in subsequent years, with Grocery Shopping on a Budget being the favorite. The experiential nature of the learning, particularly cooking activities and those that provided a patient-level experience, and a focus on health habits were the favorite aspects of the program. Many commented that they enjoyed activities and spending time with their peers outside of the medical classroom. Overall, the curriculum was well received with students desiring to continue activities beyond the first year  of participation in all required activities, with students earning 0 points (absent or disruptive group behavior), 1 point (no, minimal or inappropriate participation) or 2 points (effective participation). Students also submitted reflective essays (between 400 and 500 words) that were graded by a "Team Action Group" composed of 7 faculty members with students earning from 0 to 3 points for their essays. Students from all 3 disciplines used a Likert-type scale to provide feedback for the course, ranking items that reflected how the course prepared them for interprofessional learning and patient interactions surrounding nutritional counseling

Results
A total of 2289 articles were screened by their title and abstract. 2249 articles were identified through database searches using the culinary medicine search string created for each database, 168 articles found from these databases were removed by Covidence as duplicates. 208 articles were identified through ancestry searching and were screened by their title and abstract but none were found to be suitable to follow up for full-text review. A total of 54 articles were reviewed as a full-text. Of these articles, 24 met the criteria and were included in the scoping review.
None of the included papers explicitly reported the complete cost of running these programs. In only one paper, the cost of cooking materials, in addition to the license for Tulane's culinary medicine curriculum were reported [38].

Quality appraisal
The mean CASP score was 9.56 out of a possible 20 points for qualitative studies (median = 10, range = 7-13) and 4.13 out of a possible 24 (Median = 4, range = 0-7) for cohort studies. The most frequent items that were not achieved amongst the cohort studies included that the authors did not identify all important confounding factors (9/15) [39, 48-53, 58, 59] and the cohort study was not recruited in an acceptable way (7/15) [39, 45-47, 50, 53, 54] Amongst the cohort studies, it was also found that the follow up of subjects was not sufficiently complete (6/15 studies) [39,[44][45][46][47]54], with the most common reasons including unclear reporting or high attrition rates of participants. Many of the cohort studies also lacked an acceptable recruitment method (7/15 studies) [39, 45-47, 50, 53, 54], with the most common reasons including unclear reporting, lack of inclusion/exclusion criteria and the reliance of convenience sampling biased towards students with a 'voluntary' or 'elective' interest. Half of the studies failed to report attrition rates (n = 12) [34, 38-40, 42-46, 53, 55, 59] and only two of these studies reported reasons why the participants failed to complete the course [52,54].
The overall poorer quality of the qualitative studies was related to the frequent absence of several CASP items in the studies. The relationship between the researchers and participants was poorly reported in all nine qualitative studies, most frequently due to lack of reporting of any such considerations in the methodology [35,38,[40][41][42][43][55][56][57]. In addition, all included papers displayed poor consideration of ethical issues, either due to failure to mention any ethical considerations or having their ethical approval waived by the institution [56]. All included papers also had an insufficiently rigorous analysis of data, most commonly due to a lack of data presented and analysis performed. The findings of the CASP quality appraisal, with reasons for negative scoring per criterion, are described in Tables 3 and 4.

Discussion
This study is one of the first reviews, along with Patel &Kassam 2021 andAsher et al. 2022, to advance understanding of current opportunities and obstacles for culinary medicine within the context of medical, or other health care professional education, by examining the impact of culinary medicine in medical programs using a systematic approach [4,18]. In recent years, a clear interest in the use of culinary medicine as an education tool has been demonstrated in the literature. Between the initial search conducted in June 2020, and a second identical search conducted in April 2022, 11 new papers were published that sought to evaluate culinary medicine teaching as a viable option to teach, or support the teaching of, nutrition in medical school [35, 41, 43, 48-53, 58, 59]. Many pilot and trial implementations indicate promising efficacy towards improving medical students' nutritional knowledge, skills and attitudes [34,44,48,52,54,56,57,59].
The majority of included studies were published in the last half decade and achieved some statistically significant outcomes that promote a hands-on method of nutritional education over traditional didactic methods to educate students to help combat chronic disease as future physicians. Current literature is in line with the growing interest in culinary medicine programs, citing its emergence to the inadequacies of conventional education in combatting the rising burden of chronic disease in the healthcare sector [6,32]. This review demonstrates clear interest in the use of hands-on culinary medicine initiatives, as an educational tool and replacement for the standard curriculum of medical students. Given the relative recency of the published research, this review provides incentive for medical educators to continue to innovate and implement culinary medicine initiatives into medical education. Papers published before 2020 generally reflect poorer quality of studies, large ranges in sample size with varied results, and choice of study designs (mainly qualitative, pilot-studies). In more recent publications, the pilot programs have demonstrated greater quality and have mostly reported quantitative data with a case study design, or a mixed methods study design, allowing for both quantitative assessment of student improvement pre-and post-intervention, as well as qualitative data obtained in the form of free-form evaluation from participants after the completion of the intervention [35, 48-53, 58, 59]. All included papers in this review were published in North America, which may suggest that the increasing interest in culinary medicine appears limited in geographical footprint. However, this focus on North America highlights an opportunity for broader international involvement to legitimise the culinary medicine curriculum as a relevant approach for nutrition training in medical education. It also brings to light the important consideration of cultural relevancy and potential need for modification when international protocols eventuate.
Eighteen of the culinary medicine programs included in our review had a didactic component, with 14 programs utilising pre-coursework and 12 programs also  utilising a case-based group learning format in addition to the hands-on component. Problem based learning, lecture sessions (live or recorded), and pre-coursework are commonplace methods of delivery in medical school curricula and have been proven as an effective means for educating medical students [60,61]. Hence, it seems reasonable that these delivery methods would be adequate for delivery of culinary medicine teaching in combination with hands-on kitchen sessions, providing further merit to the feasibility and implementation of hands-on culinary medicine programs. However, there were also several variations both in the delivery method (i.e., presession quizzes, homework, community involvement, multidisciplinary faculty), setting (i.e., faculty resources, community resources), and aspects of culinary medicine taught (knowledge, skills, attitudes). These variations in culinary medicine programs have been well documented in the literature, leading to a myriad of considerations necessary to planning an effective intervention [14].
Given that the few studies which acquired data across multiple universities as part of their curriculum [34,44,45,55] all sourced their curriculum from Tulane University's Cooking for Health Optimization with Patients (CHOP) curriculum, there is an evident scarcity of a globally accepted standardised curricula. This further exemplifies the need to provide clear objectives to guide future interventional studies and for medical schools interested in utilising culinary medicine to integrate the key principles adopted by schools, with established curricula, such as those from Tulane University [19].
None of the included studies reported the total cost of running the program or performed any cost analysis. Only one study reported the costs of cooking materials, and license for Tulane's culinary medicine curriculum [38]. While previous studies have attempted to address this gap in the literature by directly inviting program directors to comment on program costs/funding among other aspects, the majority of directors elected to provide approximate round figures without further cost breakdown, or simply elect to classify program cost as confidential [7]. Regardless of cost-benefit or cost-effectiveness analyses, lack of reporting for simple expenditures in most interventions makes it challenging to assess whether culinary medicine initiatives can be integrated into existing medical programs as the costs associated with personnel, facilities and equipment/consumables may vary significantly between programs and location. While this is likely due to most studies being pilot studies, a medical qualification is already among the most expensive university programs available to students worldwide, and costs continue to increase at the expense of accessibility [5]. Given the scarcity of outcome data and the intrinsic difficulty in comparing learning outcomes in the context of future benefit to patients combined with the ethical and epistemological difficulty in performing a cost benefit/effectiveness analysis in the setting of medical education, it is recommended that at minimum future studies provide data relating to the monetary costs; allowing the possibility of cost-benefit meta-analyses to be performed once higher quality data related to outcomes are published [5].
In studies which assessed students' nutrition competencies, significant improvements were reported when compared to traditional teaching methods. The key difference was that culinary medicine curricula offers handson practical components and case-based learning as shown in Table 1. These practical culinary components are analogous to lab-based experiential learning, allowing students to apply theoretical knowledge to simulate realworld patient cases and to "learn by doing",-an effective method for clinical knowledge translation [62,63]. Other possible reasons for the improved competency outcomes arise from multidisciplinary faculty and community service components that together helped to increase students' understanding in applying practical nutrition and dietary advice in a community context [64]. Given that physicians play a key role in advising patients in nutrition, enhanced nutrition counseling skills will be effective for improving long term health outcomes for patients struggling with diet, weight loss, diabetes, and other chronic health problems [65]. If increased nutrition counseling competencies gained through culinary medicine programs can be maintained into vocational practice, future doctors will be better equipped to address the incoming burden that chronic diseases poses for healthcare systems [66].
While it is promising that increased confidence in culinary skills and nutrition knowledge were demonstrated through participation in culinary medicine programs, none of the included studies sought longer-term followup with participants, limiting current understanding of the enduring effects of these interventions. This is an important consideration. Four out of 9 of the qualitative studies included in this review had cross-sectional methodologies, which under the Levels of Evidence Framework, form the lowest level evidence in establishing causation, but provide a basis for future study designs to incorporate stronger forms of evidence [67]. Since the ultimate objective for culinary medicine programs is, positive clinical practice outcomes, it would be ideal for future studies to provide more robust evidence when investigating the impact of nutritional competencies gained within a culinary medicine program correlating or translating to better nutrition counseling in practice. Longitudinal studies would be required that track participants postgraduation and involve suitable controls for comparison, which could be a challenging aspect of study design.
The findings of surveys on students' attitudes towards nutrition counseling showed mixed results. Two studies identified negative findings, which while initially surprising, were partially explained by examining the recruitment methods used. In Pang et al. (2019) suitable participants were assessed by application essays, allowing course conveners to select participants based on perceived interest, resulting in potential selection bias. Furthermore, the majority of studies reviewed were electives. Together, these two factors indicate strong selection bias, that of course conveyors selections and participants' personal inclinations, and resulted in high positive pre-course scores on attitudes towards nutrition and nutrition counseling and statistically non-significant results. These negative findings when analysed in tandem with post-course improvements in nutrition counseling competencies suggest that students' individual attitudes towards nutrition is not the largest factor contributing to their lack of nutritional knowledge to effectively counsel patients about diet. Nonetheless, future primary studies assessing culinary medicine programs should attempt to mitigate inherent selection bias and confounding results. Mitigation could potentially be achieved by implementing culinary medicine as a non-selective component with randomised participants and consequent non-participants as controls.

Strengths and limitations of this review
The included papers in this study were collated from multiple databases through keyword search strings to yield results relevant to the research question, which has the potential be too narrow in scope. Screening of 2289 relevant papers was performed manually through Covidence after establishing 98% interrater reliability for an initial 50 results. Whilst interrater reliability was high, the potential for exclusion of relevant papers for final review does exist to some extent. Given that culinary medicine is a relatively recent development in the context of medical pedagogy, all the primary research findings were the results from pilots and on-going trials with limited longitudinal data and differing methodologies. Despite the best efforts to make data comparable by overlaying a systematic approach on qualifying studies, interpretation of results necessarily requires extrapolation to draw relevant pedagogical and clinical conclusions. These subjective effects were mitigated, in-part, by citing peerreviewed papers that supported our analysis. Another consequence of disparate measurement methods between existing studies is the inability to apply quantitative statistical methods in a meaningful manner, such as regression analysis and heterogenicity calculations. Specific challenges to overcome this limitation include accounting for duplicity of results for papers involved in a same trial and determining which outcome performance measures to use. Moreover, publication biases could obfuscate true effects of culinary medicine interventions if significant results are published more often than nonsignificant findings. Furthermore, as many of the results in these trials are likely to have an impact on the continued institutional support of these pilot programs, there is a risk of confirmation biases from investigators. Although study quality was not used to exclude papers used for our review, we have modified the CASP protocol to obtain a numerical rank of study quality (Tables 3 and 4). Several steps within the CASP protocol requires assessors' judgement to determine if a criterion has been met and this process is inherently subjective and may vary between individuals.
While our search was performed in English, which is inherently biased towards the selection of papers produced in Anglophone countries or papers with English translations embedded, we recognise that there may be relevant papers in other languages that have been unintentionally excluded. Tangential to this issue is the unintended isolation of papers reviewed to only the United States of America and Canada, which raises the question of generalisability of our findings towards medical teaching institutions in other countries.
Despite these limitations, this is one of the first scoping reviews of the literature on culinary medicine programs that detail existing culinary medicine program components, methods and results. Through our analysis, future investigations will be able to rely on a consolidated paper to determine appropriate study designs, types of data collection and analytical methodologies.

Conclusions
This paper identified a small but notable body of literature describing culinary medicine programs that have been implemented in medical schools and analysed their findings. Our results indicate that culinary medicine programs are a good initiative to pursue and can be delivered in a hands-on way that provides potential future clinical benefits for students and patients. Culinary medicine programs appear to be a feasible replacement for traditional didactic nutrition education and may be more effective than traditional didactic methods at improving student competency. Yet, despite growing interest in establishing culinary medicine curricula and positive pilot and trial data, there is still a lack of strong evidence to claim that culinary medicine programs are superior to traditional nutrition education in medical school. A lack of standardization between culinary medicine pilot studies further hinders comparability and the ability to extrapolate benefits.