Produce Food Safety and Farm to School Initiatives: School Nutrition Personnel Perspectives

The farm to school initiatives aim to support and promote the use of local foods in child nutrition programs (USDA FNS, 2021). Farm to school initiatives entail various agricultural products sourced from local or regional producers, including fresh fruits and vegetables; raw poultry, pork, and beef; milk; eggs; and seafood. Farm to school activities have grown in frequency and have diversified over the years (Colasanti et al., 2012; Feenstra & Ohmart, 2012). The Farm to School Census shows that 65.4% of school food authorities who responded to the study reported participating in farm to school activities in the 2018–2019 school year (United States Department of Agriculture [USDA] Food and Nutrition Service, 2021). The motivation to use local foods included support for the local economy, access to fresher and high-quality food, and the ability to purchase in small quantities (Izumi et al., 2006; Izumi et al., 2010; Joshi et al., 2008; USDA, 2021).

Study participants representing school nutrition programs have also expressed concerns with the challenges of coordinating and serving local products such as cost, reliability of supply, seasonality of products, food safety, quality, and delivery (Colasanti et al., 2012; Izumi et al., 2006; Janssen, 2014; Pinard et al., 2013; USDA, 2021). Other barriers include federal and state procurement regulations, availability of products, and lack of local producers (Colasanti et al., 2012; Izumi et al., 2006; USDA, 2021).

At the same time, growers and producers are equally challenged when providing local products to schools. Berkenkamp (2012) identified distribution, seasonality of products, and food safety as challenges faced by growers and producers when distributing their product to schools. Most (57 of 86) of the farmers were very or somewhat interested in developing a food safety plan for their farm. Growers noted that ability to adequately clean produce before delivering it to the school and food safety requirements as issues that exacerbate their difficulty in establishing farm to school connections (Berkenkamp, 2012). Roche et al. (2015) noted that activities that connected producers and buyers along with food safety training for the farmers and producers can help overcome many obstacles, including seasonality, availability, and food safety concerns.

While school nutrition programs use a variety of products provided by local producers and growers, fruits and vegetables are the most common staple in the majority of programs with farm to school initiatives (USDA, 2021). Approximately 73% of schools with a farm to school initiative reported purchasing local fruits and 71% reported buying local vegetables (USDA, 2021). Further, because these products are often served fresh, there is no kill step to minimize microbial load. In 2021, fresh produce was identified in 153 of the159 foodborne illness outbreaks tracked to schools, colleges, and universities (Center for Disease Control and Prevention National Outbreak Reporting System Dashboard, n.d.). In 2019, Carstens et al. (2019) reported that when comparing 2004 to 2010, there were more produce-related foodborne outbreaks and deaths in the United States from 2010 to 2017. Thus, considering fresh fruits and vegetables are the most commonly used product as part of farm to school initiatives, there is an inherent food safety risk. To the authors’ knowledge limited research related to farm to school and food safety has been conducted. Therefore, the purpose of this study was to identify farm to school program coordinators’ perceptions of produce safety in farm to school activities. This information can be used to bring awareness about factors that can impact produce safety in farm to school activities and guide the development of interventions related to improving farm to school produce safety.

Farm to school initiatives are broad and encompass a wide array of agricultural products. To allow for focused and useful outcomes, fresh fruits and vegetables, except those grown as part of an aquaponics program or in a school garden, were selected as the primary focus of study.

Methods

A three-phase Delphi technique was used for data collection. The Delphi Technique is a qualitative method that allows the researcher to delve into the variables of interest and arrive at a group consensus by providing experts with successive rounds of questionnaires, each building upon the results of the previous rounds of questions (Hsu & Sandford; 2007; Thangaratian & Redman, 2005). Directors of school nutrition programs who participate in farm to school initiatives were included in the study. While local is defined differently by school nutrition professionals, for this study, local was defined as a 150-mile radius of the school district. The research protocol for this study was approved by the University Institutional Review Board.

Sample Selection

School nutrition personnel involved in farm to school initiatives were nominated by school state program coordinators from each of the 50 states. To be nominated, participants had to represent a USDA-supported school nutrition program that obtained fresh fruits or vegetables from a local source for a minimum of the past two years.

Some questions were used for screening purposes while others to gather general information about the sample. A Qualtrics screening questionnaire was sent to those individuals nominated by the states. The screening questionnaire asked participates to elaborate on their participation in farm to school initiatives in the two years before the study, location, their role in the farm to school activities (direct involvement in farm to school activities, not including school gardens or aquaponics), and school district size. All respondents who previously participated, were participating, and/or were directly involved in farm to school activities were invited to take part in the Delphi study.

Instruments

Three questionnaires were used, one for each round of data collection. To ensure questionnaires yielded quality responses, a pilot test of the first-round questionnaire and methods was conducted with respondents not participating in the main study. Round two and three instruments were carefully reviewed, but not pilot tested, because subsequent rounds of data inquiry are based on the responses to the previous questionnaire. Therefore, it is difficult to pilot test the subsequent questionnaires when pilot-test participants would have no knowledge of the preceding responses.

The round one questionnaire consisted of seven open-ended questions related to of the safety of produce in farm to school activities, challenges/barriers to ensuring produce safety in farm to school activities, past successes in farm to school activities, and sources or information used or needed to implement or support farm to school produce safety activities.

The round two questionnaire consisted of six closed-ended questions. Participants were asked to rate their level of confidence in the overall safety of produce, local produce, and produce procured from vendors; how often they used various sources of information about farm to school produce; GAP certification; the impact of several items on farm to school produce safety, and how often various sources of information related to produce safety were utilized.

The round three questionnaire included six closed-ended questions. Questions were related to sourcing; GAP certification; and an importance ranking question related to the safety of farm to school produce; and informational sources for farm-to school activities.

Data Collection

Data were collected through three successive rounds following a Delphi process (Hsu & Sanford, 2007; Thangaratian & Redman, 2005). All rounds of data were collected using the Qualtrics system. The initial email invitation for round one included instructions for survey completion, informed consent, and a survey link. The survey was closed three weeks after the initial email. Data were then analyzed, and the second round of questions formulated.

To maintain a connection with the sample, a follow-up email was sent to participants three weeks after the close of the initial survey to inform them that the second survey would be arriving soon. The communication methods for the second- and third-phase questionnaires mirrored those used for the first phase. The second questionnaire was sent seven weeks after the close of the first questionnaire. Two weeks after the close of round two, the third-round questionnaire was sent.

Data Analysis

Five researchers specializing in food safety analyzed responses to each of the opened-ended questions by working independently to categorize responses, using pre-determined categories, and then compared results to reach consensus. Based on responses, the following categories emerged: overall status of food safety in farm to school activities, farm to school produce safety by source, and information used by participants for farm to school produce safety.

For data analysis of rating, scoring, and ranking questions for rounds two and three, the raw data set was imported from the Qualtrics survey system into SPSS to compute frequencies, percentages, and means. Open-ended responses were reviewed for categories and specific patterns as noted in the review process for round one.

Results

Response Rate and Sample Description

A total of 120 nominations from 22 states were received. All USDA FNS regions, except the Western Region, were represented. Of the 120 nominations sent, 80 participants completed the screening questionnaire, and 78 participants were invited to participate in the first round of the Delphi process. A total of 40 participants completed the first round of questions, 35 participants completed the questions for round two, and questions to round three were completed by 32 participants.

Demographic and operational characteristics of those who participated in the initial round are presented in Table 1. Most participants represented small (less than 2,500 students) or medium (2,500–19,999 students) school districts; similar to findings from previous research (Sauer et al., 2018; Roberts et al., 2020; Sauer et al., 2021). The majority of the respondents were school nutrition directors (n=35), and more than half (n=23) had been employed in school foodservice for 5–15 years.

Nearly all of the districts (n=38) indicated active participation in the USDA and/or state-sponsored farm to school initiative. Of the total food procured through the farm to school initiatives, fruits and vegetables represented the highest estimated annual percentage of the district’s purchases (excluding school gardens and aquaponics). When asked where local fresh fruits and vegetables were purchased, 32 reported buying directly from farmers, 24 from vendors, and eight from other sources, such as community gardens or food hubs. Half of the participants accepted donations from local farmers, most of which involved fresh fruits and vegetables. A total of 18 respondents reported receiving funding to support produce-related farm to school activities within the past two years.

Table 1. Demographics and operations information for school districts represented in the study (n=40)

Frequency		Frequency
District Size		Participation in USDA and/or State-sponsored Farm to School Program Initiative
Small (<2,500)	14
Medium (2,500–19,999)	21
Large (20,000–39,999)	01	No	02
Mega (≥40,000)	04	Yes	38
Role		Has received funding to support
Director	35	farm to school activities
Manager	02	No	21
Supervisor	02	Yes	18
Other (Dietetic coordinator)	01	No response	01
Years in School Foodservice		Source of local fresh fruits and
Less than 5 years	02	vegetables
5–15 years	23	Directly from farmer	32
16–25 years	10	From a vendor	24
More than 25 years	05	Othera	08
Receive donated products from a local source
No	20
Yes	20
aCommunity garden (1), Cooperative (1), food hub (3), DoD (1), local farm store (1), and local greenhouse (1)

Produce Safety in Farm to School Activities

Food safety has been identified as one of the challenges for school nutrition programs to use local foods (Colasanti et al., 2012; Izumi et al., 2006; Janssen, 2014; Pinard et al., 2013; USDA, 2021). In this study, more than half of the participants (n=29, 72%) mentioned that produce coming from local farmers is safe. Though, participants also referred to several factors that should be considered to ensure produce safety (Table 2). Factors most frequently mentioned by participants were farm size (n=9, 22%), GAP certification (n=6, 15%), cleanliness of incoming produce (n=6, 15%), and internal school refrigeration (n=6, 15%). When talking about GAP certification, one participant mentioned: “communicating with our local farmers about the importance of the condition of the produce when it arrives, as well as the condition of the packaging has been crucial. Many of our local farmers have become GAP certified” (P4). Another respondent referred to the importance of the cleanliness of the product:

Most challenges stem from the fact that most items grow in dirt. They may come to the school dirty. This is very different from the ‘prewashed’ and bagged items in retail and through distributors. This requires proper receiving, cleaning, and storage. Another challenge would be the possible truck/vehicle the items are delivered or picked up in. You see very little refrigerated transportation from the farm. (P40)

In the study conducted by Berkenkamp (2012), similar results were found. Farmers noted that the ability to clean the product before delivery was one of the factors that make it difficult for them to deliver to school programs.

Table 2. School nutrition operators’ perceptions: Factors related to ensuring produce food safety in farm to school initiatives (n=40)

Factor	Frequencya
Farm size	9
GAP Certification	6
Cleanliness of incoming produce	6
Internal school refrigeration capacity	6
Refrigeration of delivery vehicles from farms	5
School nutrition staff’s knowledge about produce safety	4
Communication between farmers and operators	4
Distance between farm and school	3
Cleanliness of delivery vehicles from farms	2
Documentation and paperwork	2
Farmers’ knowledge of school food safety requirements	1
aTotal does not equal 40 because of responses to multiple factors

Participants rated the impact of factors identified in ensuring the safety of produce in farm to school activities (Table 3). Most of the factors were reported as having some positive or significant positive impact on the safety of produce. Knowledge of school nutrition staff about produce safety (4.74 ± 0.56), farmers’ knowledge of school food safety requirements (4.60 ± 0.74), communication between farmers and operators (4.54 ± 0.70), and internal school refrigeration capacity (4.54 ± 0.85) were rated as the most important, with more than half of the participants rating the factors as having a significant positive impact. Farm size was rated as the least important (3.17 ± 0.89), with more than half of the participants rating it as having no impact.

Table 3. Level of impact on factors related to ensuring farm to school produce safety (n=35)

Factors	Frequency
	Significant Negative Impact (1)	Some Negative Impact (2)	No Impact (3)	Some Positive Impact (4)	Significant Positive Impact (5)	Mean ± SD
School nutrition staff’s knowledge about produce safety	0	0	2	5	28	4.74 ± 0.56
Farmers’ knowledge of school food safety requirements	0	1	2	7	25	4.60 ± 0.74
Communication between farmers and operators	0	2	2	6	25	4.54 ± 0.85
Internal school refrigeration capacity	0	1	1	11	22	4.54 ± 0.70
Cleanliness of delivery vehicles from farms	0	2	5	8	20	4.31 ± 0.93
Cleanliness of incoming produce	1	2	3	9	20	4.29 ± 1.04
Documentation and paperwork	0	3	4	12	16	4.17 ± 0.95
Distance between farm and school	0	2	11	10	12	3.91 ± 0.95
Refrigeration of delivery vehicles from farms	1	2	8	14	10	3.86 ± 1.00
Farm size	1	4	22	4	4	3.17 ± 0.89

Based on the impact rating, participants were then asked to rank the importance of the factors on ensuring farm to school produce safety (1=most important to 11=least important). Farmer’s knowledge of food safety requirements (2.66 ± 1.68) and school nutrition staff knowledge about food safety (4.34 ± 2.50) were ranked as the most important factors. In a previous study, farmers noted that school food safety requirements represent a challenge for them to be able to deliver the product and make farm to school connections (Berkenkamp, 2012). When exploring these responses further, most participants ranked these as their top three choices of the 11 options. Distance between the farm to school (9.47 ± 1.11) and farm size (10.31 ± 1.15) were ranked as the least important with the largest number of participants ranking these between nine and eleven (least important).

When asked about the influence of GAP certification in their purchasing decision, most participants (n=25, 78%) indicated that GAP certification was likely or extremely likely to influence their decision to purchase produce from local farms. Most of the participants (n=18, 56%)) also indicated that the safest produce came from local farms with GAP certification, while 12 of them consider produce coming from farms with or without GAP certification to be equally safe. One participant expressed the challenges faced when farmers were not GAP certified:

Produce food safety (lack of GAP certifications) does not permit some (smaller) growers to work with distributors that provide school districts with produce. Lack of GAP certification can also limit who is permitted to work with processors (smaller growers without GAP certification cannot work with processors to get us products we need—sliced beets for beet chips, diced butternut squash, etc.). (P38)

 Farm to School Produce Safety by Source

Participants were asked about their confidence level related to the source that would provide the safest produce. Most participants reported little to no confidence in the safety of produce regardless of the source (n=28, 80%), safety of local produce obtained directly from vendors (n=32, 91%), and the safety of produce obtained directly from local farms (n=32, 91%).

When asked to choose the safest way to source their produce, the majority of the participants (n=19, 54%) indicated produce is equally safe no matter the source it came from. Five of the participants suggested local produce obtained directly from a distributor was the safest option; four participants rated produce obtained directly from local farmers as the safest option, and three participants selected non-local produce obtained directly from the distributor as the safest option.

Sources Used by Participants for Farm to School Produce Safety

A list of most frequently used sources was gathered to develop a clear understanding of which sources participants utilized in developing their farm to school programs. The sources most frequently reported to be used (Table 4) were the USDA (3.67 ± 1.27), state agency (3.67 ± 1.11), and the Institute of Child Nutrition (3.31 ± 1.26). More than half of the participants reported using these sources often or always. Sources used least frequently were the National Farm to School Network (2.78 ± 1.07) and state and cooperative extension or universities (2.78 ± 1.07). Only six participants reported using other sources.

Table 4. Frequency of sources used for farm to school produce safety

Resource	Frequency					Mean ± SD
	Never (1)	Rarely (2)	Sometimes (3)	Often (4)	Always (5)
USDA (n=33)	3	3	6	11	10	3.67 ± 1.27
State Agency (n=33)	2	3	6	15	7	3.67 ± 1.11
The Institute of Child Nutrition (n=32)	5	2	7	14	4	3.31 ± 1.26
SNA (n=34)	3	8	11	8	4	3.06 ± 1.15
Local Health Department (n=32)	5	10	7	4	6	2.88 ± 1.36
State cooperative extension or universities (n=32)	5	6	13	7	1	2.78 ± 1.07
The National Farm to School Network (n=32)	8	5	11	6	2	2.66 ± 1.23
Othera,b (n=6)	4	1	1	0	0	1.50 ± 0.84
aGA organics (1), State Farm to School (1), United produce (1), no additional information provided (3) bOther was an option available to participants where they could enter the resource(s) that they found useful and indicate how frequently each was used.

Participants were asked to rank the helpfulness of the utilized sources of information in supporting their farm to school produce safety activities (1=most helpful to 7=least helpful). The USDA (2.21 ± 1.32), state agency (2.68 ± 1.55), state cooperative extension/universities (2.8 ± 1.72), and the Institute of Child Nutrition (2.95 ± 1.36) were reported as the most helpful.

When asked about the sufficiency of available resources, most participants (n=33, 94%) somewhat agreed or strongly agreed that there were sufficient informational resources to ensure food safety when implementing produce-focused farm to school activities. Specific resources that were determined to be lacking included packaging requirements for transportation and GAP certification requirements. Other responses alluded to the need to have information more readily available and information that clarifies differences in health code and other regulations among different jurisdictions.

When asked about information that is lacking which would support produce food safety in farm to school activities, participants provided topics such as food safety standards (for the program and the farmers, n=5), cleaning procedures (n=3), protocols on specifications (n=3), procurement (n=2), certification (n=2), bid templates (n=2), detailed information about suppliers (n=1), requirements for farm/plants (n=1), and success stories from other programs (n=1). Eleven participants (34%) expressed that there was enough information available.

Participants were asked to elaborate on what types of programmatic support they needed to implement and maintain safe produce in their farm to school activities. Responses regarding food safety support needed were described by topics such as offering training and certification (n=12), providing financial support (n=6), having consistent safe food safety rules (n=6), building connections (n=5), supplying proper tools for cleaning and preparing vegetables (n=2), and facilitating support for local farmers (n=2). According to Roche et al. (2015), activities that link producers and purchasers, combined with education about food safety, can assist in overcoming challenges, such as seasonality, availability, and food safety concerns.

Conclusions & Applications

The purpose of this study was to identify farm to school program coordinators perceptions about the food safety of produce in farm to school activities. This is, as far as we are aware, the first study on the perceptions of product safety in farm-to-school activities among school program coordinators. A three-round Delphi methodology provided an understanding about produce safety in school nutrition programs with farm to school activities.

Produce coming from local farmers was perceived as safe by most of the respondents. Farm size, GAP certification, the cleanliness of incoming food, and internal school refrigeration were the factors that participants brought up the most related to ensuring produce safety in farm to school activities. Participants rated several key factors as having a positive or significantly positive impact on the safety of produce. These factors included the knowledge of school nutrition staff about produce safety, farmers’ familiarity with school food safety requirements, effective communication between farmers and operators, and the internal school refrigeration capacity. Notably, these aspects were deemed crucial by more than half of the participants, underscoring their significance in ensuring the safety of produce in farm-to-school programs. It was also highlighted that the farm size was considered the least important factor, with most participants perceiving it as having no substantial impact on produce safety. This suggests that the scale of the farm may be a less influential determinant of safety, when compared to other factors evaluated in the study.

Of note is the lack of confidence in the safety of produce held by those who participated in the study. Overall, participants were only slightly confident in the safety of produce, regardless of the source. Participants were not confident at all in the safety of produce obtained from vendors, nor were they confident in the safety of produce from local vendors. These concerns emphasize the need for comprehensive strategies and practices to ensure the safety of locally sourced produce and enhance child nutrition program’s confidence in locally grown produce.

Most participants suggested that GAP certification would influence their decision to purchase produce from local farms, and that the safest produce came from local farms with GAP certification. GAP certification was also identified as a challenge or barrier when ensuring produce safety. GAP certification is key in the decision-making process for buying local. Farmers and programs should be trained in the details and requirements of GAP certification. Future research on GAP certification status in local farms and GAP requirements in states could have a positive impact on farm to school procurement.

To enhance the safety of produce in farm to school programs, stakeholders should prioritize strengthening the knowledge of school nutrition staff, fostering farmer awareness of food safety requirements, facilitating effective communication between stakeholders, and improving internal school refrigeration capacity. Findings from this study can serve as a valuable resource for policymakers, educators, and practitioners involved in farm to school initiatives, ultimately promoting safer and healthier food choices for child nutrition programs. Further research and efforts in these areas are essential to continue improving the safety and quality of produce in farm to school programs.

Farm to school produce safety information and updates should be routed through the USDA, state agencies, and the Institute of Child Nutrition as a priority. The most common sources used to implement and maintain farm to school produce safety activities were the USDA, state agencies, and the Institute of Child Nutrition. The USDA, state agency, state cooperative extension/universities, and the Institute of Child Nutrition were sources deemed as most helpful. Interestingly, the local health department was only used as a source of information rarely and their resources were ranked as the least helpful among all sources of information. These findings suggest that while a foundation exists for food safety in farm to school initiatives, there is room for improvement, particularly in addressing the identified gaps and enhancing accessibility to essential information. Further efforts in these areas may contribute to better food safety practices within farm to school programs.

Child nutrition programs and state agencies can develop training programs on how farm to school produce safety could address some of the concerns identified in this study. Training on farm to school produce safety should focus on those topics identified by participants as influencing produce safety such as GAP certification, farmers knowledge on food safety requirements, and school nutrition staff knowledge about produce safety. Training should also be stratified and customized for different audiences such as distributors, school nutrition employees, and farmers.

This study has some limitations that need to be considered when interpreting the results. The sample size was small, caution should be used when generalizing results. This study aimed for a better, in-depth understanding of school nutrition operators’ perceptions related to produce food safety; data presented provides valuable information that can be used for future research in identifying food safety barriers in farm to school initiatives. A notable limitation of the study is self-reported data, which are possibly influenced by selective memory, attribution, and exaggeration. Although the methods included different recruitment methods and attempts at representation from all USDA FNS regions, responses from the Western Region were not gathered. The Delphi technique requires continued commitment from participants who are being asked a similar set of questions multiple times. Drop out between rounds can also happen especially when the time between rounds is extended. This was minimized by keeping participants informed of the process and when to expect each successive set of questions.