Nutrition and Culinary Education in Food is Medicine Interventions. A Scoping Review

Abstract

Background-

Food is Medicine (FIM) interventions leverage food-based therapies to address diet-related conditions and disparities. Although nutrition and/or culinary education are recommended, the details and extent of these activities remain unclear.

Objective-

To evaluate five key educational characteristics within FIM interventions: components, delivery format, setting, frequency, and educators.

Methods-

Using PRISMA guidelines, a scoping review was conducted of FIM interventions (medically tailored meals, groceries, or produce) that included an educational component. CENTRAL, CINAHL, Embase, PubMed, and SIREN were searched from January 2010 to May 2025. Two investigators independently assessed texts for inclusion and extracted data using standardized methods. Descriptive analysis and narrative synthesis identified patterns and strategies related to educational methods.

Results-

Of 5,703 articles reviewed, 100 met inclusion criteria, mostly from the U.S. (n=91) and focused on adults (n=84). 41.0% of studies provided one educational component, 49.0% provided 2–3, and 10.0% provided 4+. Among ten major educational components, printed materials (43.0%), cooking classes (35.0%), and individual counseling (32.0%) were most common. Most were delivered in person (55.3%), followed by virtual (23.9%). Educational settings included virtual platforms (24.0%), clinics/hospitals (22.4%), and community locations (10.4%). Frequency of activities varied from one-time to available on-demand; weekly (18.6%) and monthly (17.5%) were most common. Educators were most often RDNs (23.6%), followed by clinicians (11.3%) and community-health workers (7.7%); 42.6% were unspecified.

Conclusions-

These novel results provide a comprehensive characterization of educational components of FIM interventions. Findings highlight common patterns, structural gaps, and opportunities to strengthen the design, implementation, evaluation, and reporting of nutritional and culinary education within FIM.

Introduction

Poor diet is a leading cause of death and disability in the United States,¹ where fewer than 10% of adults are cardiometabolically healthy.² Food is Medicine (FIM) programs like medically tailored meals (MTM), medically tailored groceries (MTG) and produce prescriptions (PRx) respond to this critical link between nutrition and health by integrating food-based services into health care delivery as part of an individual’s care plan.³ Nutrition and culinary education are common, and sometimes compulsory,⁴ components of FIM programs, and include a wide variety of strategies designed to reinforce dietary behavior change alongside the provision of meals or produce. For example, MTM programs often include opportunities for individual or group counseling by registered dietitian nutritionists (RDN).³ PRx or MTG programs, where participants receive unprepared or minimally processed healthful foods, may include nutrition education as well as culinary education; for example, the Veterans Health Administration PRx pilots utilize Healthy Teaching Kitchens to teach cooking skills to participants.⁵

Only a few studies have examined the scope of nutrition education within FIM programs, and have focused on selected PRx initiatives, such as those funded by the U.S. Department of Agriculture or programs for patients with diabetes.^6–8 However, the full approaches to nutrition education in FIM programs remain understudied and poorly understood, preventing evaluation of the impact of different educational components, settings, frequencies, and educators on behavior change, benefits utilization, disease outcomes, or health equity.⁸ Importantly, current literature also lacks systematic identification and investigation of the role of RDNs in FIM⁹–specifically in educational and counseling components, for which they have extensive training. Thus, better understanding of existing practices is warranted in order to make recommendations for program and research design. Given the accelerating adoption of FIM by both private and public payers, as well as policy makers,^10,11 standardized practices are needed across program delivery – including in delivery of education – to create scaleable and equitable models.³ To address these gaps in knowledge, a scoping review and narrative summary were conducted to examine the components, delivery format, setting, frequency, and educators of nutrition and culinary education across the spectrum of FIM interventions.

Methods

Study Design

This investigation followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines.¹² The research team included a senior researcher (D.M.), a research assistant professor (R.R.), a postdoctoral student (L.M.), two graduate students (N.R., F.P.), an undergraduate student (S.Y.), a librarian (A.L.), and a senior research coordinator (J.R.).

Eligibility Criteria

Studies were selected according to prespecified criteria. The selection included full-text, peer-reviewed articles and scientific abstracts published after 2010 that (a) implemented FIM interventions in healthcare settings (including PRx, MTG, MTM, or other similar programs referred to as Food Pharmacy, Food Rx, or food voucher programs) as a treatment for health conditions related to diet (such as diabetes, hypertension, obesity, pregnancy, or others); and (b) included a structured nutrition and/or culinary education component. For this review, a structured educational component was defined as any intentional teaching activity or teaching materials designed to reinforce nutritional or culinary knowledge or skills, beyond just providing food or routine clinical instructions. This could include a variety of approaches such as printed materials (e.g., handouts, recipes, and booklets), culinary education, nutritionist sessions, grocery store tours, group classes, and one-on-one counseling. These could be delivered virtually or in person, and could take place in any setting, including but not limited to clinics, homes, or community settings. Studies could include participants of any age, sex, race/ethnicity, geographic location, or socioeconomic background. Only studies published in English were included.

Studies were excluded if they focused solely on federal food assistance programs (like WIC, SNAP and/or school meals), except for GusNIP Produce Prescription Program (PPR) initiatives which were included; or on population-level food assistance (such as food banks, food pantries, or local food distribution programs), without a referral from a healthcare entity. For cohorts with multiple publications, only the primary paper with relevant educational content was included; secondary papers without added content were excluded during full-text screening. The review focused on studies published in or after 2010, given the interest in contemporary educational approaches and the rise in FIM in this period.

Information Sources

A systematic review was performed of articles in CENTRAL, CINAHL, Embase, PubMed, or SIREN. The initial search was conducted through March 19, 2024, and it was updated on May 6, 2025. The final search strategies (Supplemental figure 1) were drafted by an experienced librarian (A.L.) and refined through team discussion. The terms included a variety of keyword combinations, such as “Food is Medicine,” “ Medically tailored foods,” “Produce prescription,” “Veggie Rx,” “Food Rx,” “Voucher programs,” and “interventions.” Additionally, specific intervention terms were included, such as “health system,” “community,” “nutrition education,” and “nutrition counseling;” and MeSH terms like “food,” “fruit,” “vegetables,” and “prescription”. To refine and combine search terms effectively, Boolean operators (AND, OR) were applied. The search results were exported into EndNote,¹³ and duplicates removed by a library technician.

Selection of Sources of Evidence

Titles and abstracts of identified studies were screened by one investigator using the Rayyan web app^14,15 for systematic reviews (F.P. assessed 761 articles; S.Y., 819 articles; L.M., 1187 articles; R.R., 110 articles). For any potentially eligible study, the full text article was obtained and reviewed independently and in duplicate by two investigators (each pair reviewed a subset of full-text articles). For any discrepancies during the selection process, a third investigator was engaged to resolve differences and arrive at a consensus.

Data Charting Process

Five investigators developed the data extraction form in Excel,¹⁶ drafted by two investigators (J.R., L.M.) and subsequently refined collaboratively with three other investigators (R.R., D.M., S.Y.). It captured pertinent information on key characteristics of the study and the educational methodologies employed within the FIM intervention. The extraction form was independently tested by two team members to assess its clarity and accuracy; and then data extraction for each study was performed independently and in duplicate by two investigators (pairs of L.M., S.Y., N.R., R.R., and F.P.).

Data Items and Data Abstraction Process

A systematic, structured data abstraction process was used to focus on describing the characteristics of FIM interventions, with special emphasis on the educational components. For each study, information was collected on author, year, publication type, setting, and eligibility criteria (including target population, socioeconomic context, and relevant diet-related conditions). Intervention-level data included study design, type of FIM intervention (MTM, MTG, and PRx); redemption method (e.g., voucher, electronic card, pickup or home delivery); dosage (type of food provision [voucher ($), food box/bag (lbs), and meals], frequency [one-time, day, weekly, bi-weekly, monthly, at each clinic visit, upon request], amount); and duration. Educational components were categorized into ten types, detailed in Supplemental figure 2, including printed materials (e.g., handouts, recipes, booklets, and nutrition guides), cooking classes, workshops, video tutorials, mobile apps, newsletters, counseling (individual or group, in-person or by phone), and field trips. For each component, data were extracted on delivery format (in-person, virtual, or mixed methods), setting (if applicable), frequency, and educator (categorized as RDN, physician/nurse/ trained medical student, community health worker or other community member, chef or culinary instructor, research staff, multidisciplinary team, or not specified). Evaluation of educational components were also extracted, including attendance/participation rate, participant satisfaction, and any other reported evaluation outcomes (free-text capture).

Risk of Bias Assessment

As the aim was to describe components of the provided nutrition and culinary education rather than assess causality of any effect, a risk of bias assessment was not conducted, consistent with the guidelines and practices recommended by various authors.^12,17

Synthesis of Results

A descriptive synthesis of the studies was conducted, organizing the findings into predefined categories. Frequencies and percentages were calculated to summarize key characteristics of the FIM interventions, such as types and number of nutrition education components, delivery format, setting, frequency, and educator. A narrative approach explored common patterns related to the combination of educational components and their implementation.

Results

Study Selection

A total of 6,334 records were identified across all databases. After removing duplicates (n=1,752) and other exclusions (e.g., reviews, non-English publications, animal studies, government documents, etc.) (n=1,706) (Supplemental Figure 3), 2,876 studies remained for screening. Following title and abstract review, 472 full-text articles were assessed for eligibility. Of these, 42 articles were not accessible; among the remaining 430, articles were excluded because they described community food assistance programs not linked to a healthcare entity (e.g., food banks, pantries, local initiatives, n=104); focused on government programs such as WIC, SNAP, or school meals (n=15); evaluated FIM or food-based interventions in healthcare settings but did not report a nutrition or culinary education component (n=170); or were secondary publications from the same study cohort without additional educational content (n = 41). In total, 100 unique studies^18–117 were included in the final synthesis.

Study Characteristics

Most of the studies focused on PRx (n=55),^{20,22,26,28–30,32,33,36,37,40,41,44,45,48–50,52,53,57–60,66–70,72,76,77,81–84,86–88,89(p2),90,92–98,105,106,108–112,115} followed by MTG (n=28),^{18,19,21,23,27,38,39,42,43,46,47,55,56,61,64,71,74,75,79,80,99,102–104,107,113,114,116} and MTM (n=17) (Table 1).^{23,24,31,34,35,51,54,62,63,65,73,78,85,91,100,101,117} Nearly half were conducted in community health centers or clinics (43.0%),^{18–20,24,28,29,38,39,45,47,53,60,66,69,74–78,80–83,86,88,90,92–100,103,103,106,108,109,111,113–115} followed by hospitals (28.0%),^{21,23,25,26,31,32,34,35,40,41,46,50,51,54,55,61,62,65,67,79,84,91,101,104,107,110,116,117} multiple settings (11.0%)^{30,36,37,43,58,59,72,73,85,87,105} and other types of settings. The study design was most often a pre–post intervention without control groups (43.0%);^{21,22,25,26,36,37,41,43,45,47,49,52–54,56,58–60,63,69,70,72,77–79,82,83,85,88,93–96,102,103,105,106,108,109,113–115,117} 30.0% were randomized controlled trials.^{18,19,24,29,32,34,35,39,40,42,50,51,55,57,61,62,64–67,73–75,80,91,99,101,104,110,111} Most interventions targeted adults (84.0%)^{18,20,22–29,31–42,45–47,49–56,59–65,67–69,71–74,76,78–97,99–101,103,105–107,109–117} and individuals experiencing food insecurity (52.0%).^{19,20,22–28,31–33,36–44,46,50,51,54,56,58,59,61,62,64,65,67,71,72,74–76,78–81,84,89,90,101,103,106,107,110,116,117} The most frequently addressed health conditions were multiple diet-related conditions (37.0%)^{18,22,26–28,33,36,37,43–45,47,49,51,54,56,59,62,63,68,72,75,76,79,81,84,85,88,92–94,97,98,103,113,115,116} and diabetes (26.0%).^{20,24,25,29,34,39,40,50,52,53,61,65,67,71,82,83,89,90,95,96,102,107,109–111,114}

Table 1. Key characteristics of Food Is Medicine interventions that included nutrition or culinary education, identified in a scoping review (n=100).

	n (%)
Type FIM intervention a
Produce Prescription (PRx)	55 (55)
Medically Tailored Groceries (MTG)	28 (28)
Medically Tailored Meals (MTM)	17 (17)
Implementation site
Community health center or clinic	43 (43)
Hospital	28 (28)
Multisiteb	11 (11)
Otherc	10 (10)
Primary health care clinic	6 (6)
Mobile clinic	1 (1)
Pharmacy	1 (1)
Study design
Pre-and post-test without controls	43 (43)
Randomized controlled trial	30 (30)
Qualitative Study	9 (9)
Feasibility Study	11 (11)
Quasi-experimental with controls	6 (6)
Cross-sectional study	1 (1)
Eligibility criteria
Target population
Adult	84 (84)
Household or family level worksd	12 (12)
Child	4 (4)
Socioeconomic status
Food insecure	52 (52)
Low income	21 (21)
Participation in assistance programse	14 (14)
Not specified	7 (7)
Uninsured	6 (6)
Diet-related condition
Multiple diet-related conditionsf	37 (37)
Diabetes	26 (26)
Overweight/Obese	13 (13)
High-risk pregnancy	7 (7)
Hypertension	6 (6)
Cancer	3 (3)
Cirrhosis	2 (2)
Heart failure	2 (2)
HIV	2 (2)
Chronic kidney disease	1 (1)
Depression	1 (1)

^aIn studies with more than one type of FIM interventions, classification was based on the primary intervention.

^bMultisite refers to studies implemented across multiple locations with different settings.

^cOther refers to any structure not included in the previous categories (e.g., farmers’ markets, senior centers, Red Cross, or not specified).

^dHousehold or family-level interventions target the household as a unit rather than focusing on individuals.

^eParticipation in assistance programs encompasses programs such as SNAP, Medicaid, and FQHCs.

^fMultiple diet-related conditions include studies with participants affected by more than one condition such as hypertension, diabetes, prediabetes, hyperlipidemia, cardiac disease, or high-risk pregnancy; participants were not required to have all listed conditions.

Abbreviations: HIV: Human Immunodeficiency Virus; SNAP: Supplemental nutrition assistance program; FQHC: Federally qualified health center.

Supplemental Table 2 provides detailed characteristics for each individual study. The operationalization of the FIM interventions is summarized in Figure 4, including type, redemption method, quantity, frequency, and duration. MTM were most often home-delivered, with an average provision of two meals per day for 6 months. MTG supplied food boxes or funding support for grocery purchases, including through home delivery, pickup, or retail shopping, with a mean value of $77/ month for 6 months. PRx studies included vouchers (mean value $53/month) or food boxes (average weight 57 lbs/month) for a mean of 6 months, redeemed through voucher systems, in-person pickup, or home delivery. Amounts and frequency of food provision varied substantially across the studies.

Figure 4. Characteristics of Food Is Medicine interventions across studies included in a scoping review. FIM interventions were grouped into: Produce Prescription, Medically Tailored Groceries, and Medically Tailored Meals.

^aComponents summarized include redemption method, type of food provision, amount, and frequency. Values in parentheses indicate the number of studies classified under each category.

^bAverage amounts and ranges were standardized per month and calculated only for studies with sufficient data (12 for MTG, 41 for PRx, and 13 for MTM).

^cFor MTG, the amount of food boxes provided was reported across studies either in pounds (lbs) or U.S. dollars ($); therefore, average values were calculated and presented to represent both reporting units.

Abbreviations: FIM: Food is Medicine; MTM: Medically Tailored Meals; MTG: Medically Tailored Groceries; PRx: Produce Prescription; EDC: Electronic debit card; lbs: pounds; mo: month; NS: Not specified; $, U.S. dollars.

Nutrition Education Components

Nutrition education components varied across interventions (Table 3). Ten common components were identified, most frequently printed materials (57.0%),^{18–21,24,25,27,31,38–40,42,44,45,47,49,50,52–56,56–61,67,70,71,75,76,79,82,83,86,88–96,99–107,113,114,116} cooking classes (35.0%),^{25–30,36,37,42,43,45,49,52,53,58,64,67,69,70,72,73,78,82,90,92,93,96,97,102,103,108–110,112,115} and one-on-one counseling (32.0%);^{18,23,25,33–35,39,42,44,50,51,53,56,58,60,63,66,69,73,74,77,79–82,84,85,91,104–107} and least frequently mobile apps (6.0%)^{21,62,66,75,80,114} and field trips (5.0%).^{46,48,69,88,110} Among MTM programs, one-on-one was the most frequent (47.0%).^{23,34,35,51,63,73,85,91} MTG interventions commonly used printed materials (82.1%),^{18,19,21,25,27,38,39,42,47,55,56,61,71,75,79,99,102–104,107,113,114,116} while PRx programs also favored printed materials (50.9%).^{20,40,44,45,49,50,52,53,57–60,67,70,76,82,83,86,88–90,92–96,105,106} Figure 5 presents the structural characteristics of the educational components included in the interventions. Education was most often delivered in person ^{18,19,22–30,32,33,36–39,41,42,44–48,50,52–61,63,64,66,68,69,72,73,77,79,82–84,86,88,91–96,98,99,101–103,105,106,108–110,113–116} with a predefined schedule established at the beginning of each intervention. While clinics were the most frequent implementation sites,^{18,23,23,26,27,29,32,33,38,39,47,53,56,60,64,66,69,73,77,79,82,84,86,91,101,102,105,106,108,113,115,116} community settings,^{22,24,28–30,68,73,82,88,92,94,96,98,109,110} and grocery stores/farmers’ markets were also utilized. ^{44,46,48,52,59,69,88,110} Sessions were predominantly led by RDNs^{23,24,34,35,37,39,41,46,47,49–51,53,62–65,69,73,78,79,84,85,90,91,104,107,109,113,116,117} or other clinicians, including physicians, nurses, and/or trained medical students,^{20,22,32,33,43,44,48,58,81,82,86,96,103,106,111} although some interventions incorporated multidisciplinary teams^{23,24,36,42,56,58,66,71,72,77,81,83,95,110} and community-based educators.^{18,23,33,39,46,52,58–60,67,80,87,92} Educational content ranged from the use of informational materials to counseling and interactive interventions. Additional details on the educational components in each study are available in Supplemental Table 4. Evaluation of educational components was rarely reported and, when mentioned, typically focused on attendance or satisfaction. ^{28–30,43,44,51,51,59,82,83,86,106–109,115}

Table 3. Types of nutrition education provided by FIM intervention^a included in a scoping review.

	Total n=100	MTM n=17	MTG n=28	PRx n=55
Types of nutrition education provided b	n (%)	n (%)	n (%)	n (%)
Printed materials	43 (43)	6 (35.3)	23 (82.1)	28 (50.9)
Cooking classes	35 (35)	2 (11.8)	7 (25)	26 (47.3)
One-on-one counseling	32 (32)	8 (47.1)	10 (35.7)	14 (25.5)
Group counseling	17 (17)	-	5 (17.9)	12 (21.8)
Workshops	17 (17)	1 (5.9)	2 (7.1)	14 (25.5)
Telephone counseling	13 (13)	5 (29.4)	2 (7.1)	6 (10.9)
Video tutorials	9 (9)	1 (5.9)	4 (14.3)	4 (7.3)
Newsletters	6 (6)	2 (11.8)	4 (14.3)	-
Mobile apps	6 (6)	1 (5.9)	4 (14.3)	1 (1.8)
Field trips	5 (5)	-	1 (3.6)	4 (7.3)
Total educational interventions per study c
1 intervention	41 (41)	12 (70.6)	7 (25)	22 (40)
2 interventions	33 (33)	2 (11.8)	12 (42.9)	19 (34.5)
3 interventions	16 (16)	2 (11.8)	6 (21.4)	8 (14.5)
4 interventions	8 (8)	1 (5.9)	2 (7.1)	5 (9.1)
5 interventions	2 (2)	-	1 (3.6)	1 (1.8)

^aIn studies with more than one type of FIM interventions, classification was based on the primary intervention.

^bPrinted materials included handouts, recipes, booklets, or nutrition guides. Cooking classes included group-based cooking sessions. Workshops included structured group nutrition education sessions. One-on-one and group counseling included counseling delivered in person or by telephone. Video tutorials included pre-recorded instructional videos. Newsletters included print or electronic nutrition communications. Mobile apps included smartphone-based nutrition applications. Field trips included visits to grocery stores or farmers’ markets

^cIndicates the number of educational strategies applied in each study, grouped as in this table. Since some studies used multiple educational strategies, the total count may exceed the number of studies.

Abbreviations: FIM: Food is Medicine; MTM: Medically Tailored Meals; MTG: Medically Tailored Groceries; PRx: Produce Prescription.

Figure 5. Characteristics of educational components included in Food is Medicine interventions across studies included in a scoping review.

^aEducational components are summarized across five principal elements: (1) the content of nutrition education provided; (2) the profile of the educator responsible for delivering the intervention; (3) the frequency of the educational component; (4) the location where it was implemented; and (5) the delivery format to participants.

^bValues in parentheses indicate the number of studies that included each component. The total number exceeds the number of included studies, as several implemented more than one educational strategy.

^cBar lengths are scaled proportionally to the number of studies in each category.

Implementers of Educational Components

Educational strategies were implemented by a diverse range of personnel, with variability by educational component and in reporting across studies (Figure 6). In 9 out of 10 components, over 20.0% of studies did not specify who led the intervention,^{18,23,26–29,37,38,40,42,45,47,49,50,52–57,60–62,66–70,74–76,80,88–90,92–94,96–102,105,107,109,110,112,114} with the highest rates in mobile apps (100.0%),^62,66,80,114 workshops (58.8%),^{28,29,37,47,52,68,88,97,109,112} printed materials (56.4%),^{18,27,38,40,45,49,50,53–57,60,61,67,70,75,76,88–90,92–94,99–102,105,107,114} and video tutorials (55.6%).^{62,75,92,110,114} Only one-on-one counseling had implementers reported in more than 90% of studies.^{18,23,23,33–35,39,42,44,50,51,53,56,58,63,66,69,73,77,79–82,84,85,91,104,106,107} Among studies that provided this information, RDNs were most often responsible for telephone counseling (53.8%)^{24,62,65,73,107,116,117} and one-on-one counseling (50.0%).^{23,34,35,39,50,51,53,63,69,73,79,84,85,91,104,107} Physicians, nurses, and medical students were more commonly involved in field trips (20.0%),⁴⁸ newsletters (16.7%),¹⁰³ and less frequently in one-on-one counseling (15.6%).^{33,44,81,82,106} Multidisciplinary teams commonly delivered field trips (20.0%),¹¹⁰ workshops (17.6%),^72,81,110 and cooking classes (14.3%).^{23,36,58,72,110} Culinary instructors were only identified in cooking classes (11.4%)^{30,43,108,115} and in one case of printed materials (1.8%).¹⁹ Community health workers and other community members implemented workshops and group counseling (17.6% each),^{46,58,60,80,87,92} and one-on-one counseling (12.5%).^18,23,58,80 Research staff were responsible for newsletters (16.7%),⁹¹ video tutorials (11.1%),¹⁰² and printed materials (1.8%).⁹¹

Figure 6. Implementers of each educational component within Food is Medicine interventions across studies included in a scoping review.^a,b,c.

^aBars represent the number of studies implementing each type of educational content.

^bBar segments within each bar represent the type of provider delivering the content.

^cTotal exceeds the number of studies because many interventions included multiple strategies.

Discussion

This comprehensive scoping review characterizes how nutrition and culinary education components have been integrated within FIM interventions. The results identified common patterns, significant gaps in the literature, and further opportunities in the field. First, the review had to exclude 170 published studies of FIM interventions because they contained no mention of nutrition or culinary education, quantifying a stark lack of inclusion, standardization, and/or reporting of this critical component of how FIM is conceptualized and should be operationalized.¹¹⁸ Second, while the effectiveness of the provided education in FIM was outside the scope of this review, the finding that only 16 studies described whether or how the educational components were evaluated (and for those reported, assessments were generally restricted to attendance rates or participant satisfaction) is of concern.^{28–30,43,44,51,51,59,82,83,86,106–109,115} This dearth of evaluation of effectiveness precludes the ability to determine which educational strategies most effectively alter behaviors and downstream clinical outcomes within FIM interventions. Third, while RDNs were commonly noted as primary educators, their specific involvement was documented in less than a quarter of studies, indicating substantial room for leveraging their unique skills. In 2025, the Academy of Nutrition and Dietetics published a Food as Medicine Strategic Roadmap calling for greater integration of RDN services into FIM, including developing sustainable payment methods, demonstrating effectiveness, and increasing interprofessional awareness and education;¹¹⁹ consistent with calls for competencies, accreditation standards, and skills most applicable to achieve these goals.¹²⁰ Similarly, among studies incorporating culinary education, only four reported the participation of chefs or culinary instructors, suggesting that professionals with culinary expertise remain underrepresented in FIM education programs.^{30,43,108,115} Their involvement could help align meal preparation and food choices with individual preferences and cultural norms, potentially improving both engagement and long-term adherence to healthy eating.

In this analysis of 100 studies, distinct trends emerged by intervention type. For example, more intensive educational approaches (e.g., 1:1 counseling) were more common with MTMs, consistent with these programs’ more clinically complex populations.¹²¹ Although these patients receive approximately half of their meals in the program, nutrition education remains essential to help them manage the rest of their diet and to sustain healthy eating behaviors once the program ends. More frequent use of printed materials, mobile applications, newsletters, and video tutorials was observed in MTG and PRx programs, where participants may have greater capacity and flexibility to prepare and cook their own meals and therefore benefit from more skill-building, hands-on cooking classes, and field-based activities.³

In addition to differences between FIM programs, consistent structural patterns were identified across all FIM modalities. For example, in-person approaches dominated, typically involving scheduled sessions in clinical settings such as hospitals or healthcare centers. This suggests general integration of educational activities within healthcare delivery models, which could support continuity of care and treatment adherence.¹²² On the other hand, a need for in-person education may also introduce barriers to participation like transportation. Less commonly utilized approaches were documented that extended education to community settings, farmers’ markets, grocery stores, and homes, facilitating practical skill acquisition in everyday contexts—an approach previously demonstrated to improve adherence and sustained dietary behavior changes.^80,123 Despite the growing availability of technology, digital tools such as mobile apps, e-newsletters, and video tutorials were less utilized, highlighting an opportunity for further development. This limited use may relate to the relatively recent adoption of digital technologies in clinical education settings.

The finding that nearly 60% of interventions incorporated two or more educational strategies—most commonly printed materials, counseling, or interactive formats such as cooking classes or workshops—suggests moderate recognition of the need to engage participants through multiple learning approaches and modalities. Prior work indicates that effective education for chronic disease management extends beyond information dissemination, requiring long-term support, patient empowerment, accessible delivery formats, and trained personnel attuned to individual needs.¹²⁴ Nutrition education is in demand: in a national survey of 3,009 U.S. adults, more than half of respondents —and nearly 70% of those experiencing food insecurity—were interested in receiving nutrition counseling as part of their health care.¹²⁵

At the same time, a broad range in counts of different educational strategies was identified, highlighting the heterogeneity of FIM education and the accompanying challenges in comparing clinical outcomes across studies. For example, one 4-month PRx study addressing multiple diet-related conditions included five components (printed materials, video tutorials, cooking classes, workshops, and telephone counseling),⁹² while another 6-week MTG program simply included a recipe with each food box.¹²⁶ Maintaining healthcare interest and momentum in FIM as the field matures will require greater rigor in both implementation and evaluation of nutrition and culinary education, including assessment of uptake and effectiveness of different components to understand barriers and facilitators to participant engagement.

This review has several strengths. It adhered to PRISMA guidelines and used independent duplicate reviews for study selection and data extraction, enhancing validity and reproducibility. Standardized definitions facilitated systematic classification and comparison of educational strategies and their components. Studies were included that considered a broad range of diet-related conditions and from various regions, broadening generalizability. It was more comprehensive than any other previous review of FIM interventions, which have been more narrow in scope, often limited to a single modality (PRx or MTM),^7,8,127,128 a specific clinical condition,^8,128,129 or a U.S.-based context,^7,8,127,129 and frequently restricted to adult populations,^8,128,129 GusNIP-funded programs,⁷ or older searches.^{7,8,127–129} For example, in four of five prior reviews of FIM programs, less than 20 studies were included, and nutrition education was commonly treated as an “additional component” rather than assessed in detail.^8,127–129 Prior descriptions of educational materials in FIM reviews were generally inconsistent and lacked important context—such as delivery format, setting, frequency, and educator—to allow meaningful understanding of the approaches. These works noted a predominance of low-intensity activities, such as printed materials or recipe booklets.^{7,8,127–129} Compared to these earlier reports, the present analysis identified and included a much larger number of studies (n=100), assessed multiple FIM modalities across diverse populations and regions, and documentated key delivery characteristics, common patterns, and specific gaps.

Potential limitations should be considered. This review was limited to peer-reviewed, indexed publications in English and did not include grey literature. This may have led to the omission of some initiatives and limits generalizability to non-English-speaking settings, but increases transparency and reproducibility. Due to the descriptive scope, risk-of-bias assessments were not conducted; this approach aligns with established guidance for descriptive reviews and does not compromise internal validity.^12,130 Details of reporting varied across studies, especially regarding the frequency and duration of educational components and educator characteristics. To help address this and capture as much meaningful information as possible, a standardized extraction protocol and definitions were used.

Conclusions

This scoping review provides a comprehensive characterization of educational components in FIM interventions. The findings highlight common patterns, structural gaps, and key opportunities to strengthen the design, implementation, evaluation, and reporting of nutritional and culinary education within these programs. Future efforts should focus on standardizing reporting guidelines for educational components in FIM, as well as systematically evaluating how educational elements influence behavior change—in this way shifting from description to assessment and helping to craft a more generalizable framework to inform program design, implementation, and policy development.

Research Snapshots.

Research Question

What are the key characteristics and implementation patterns of nutrition and culinary education across Food is Medicine interventions?

Key Finding

In this scoping review of 100 studies (mostly from the U.S. and focused on adults), printed materials (43.0%), cooking classes (35.0%), and individual counseling (32.0%) were the most common nutrition education components, while mobile apps (6.0%) and field trips (5.0%) were the least frequent. Education delivery was primarily in-person (55.3%), with RDNs (23.6%) and clinicians (11.3%) as the most frequent educators, though 42.6% of studies did not specify implementers. Educational frequency varied, with weekly (18.6%) and monthly (17.5%) sessions being most common.