ABSTRACT

From planting to distribution, fresh produce can be contaminated by humans, water, animals, soil, equipment, and the environment. Produce growers play an essential role in managing and minimizing on-farm food safety risks. Because of an increase in public awareness about produce safety, farmer food safety education has become an important research and extension topic. This review article summarizes findings by researchers who have evaluated produce growers' food safety knowledge and attitudes and the effectiveness of food safety educational programs for growers. A search of on-line databases, journal archives, conference abstracts, and reference lists of relevant studies was conducted to locate peer-reviewed articles on produce growers' food safety knowledge and behavioral changes. Study selection criteria included publications in English, publication between 2000 and 2019, and a focus on one of six topics: handling of agricultural water, soil amendments, domesticated animal and wildlife management, worker health and hygiene, food safety plans and record-keeping, and cleaning and sanitation. Forty-three published articles were included in the analysis. Handling of agricultural water and soil amendments were the two topics least understood by growers, whereas worker health and hygiene were the best understood. Food safety educational interventions were evaluated in 13 studies, and most studies used in-person workshops and self-reported pre- and postintervention knowledge assessments. Most reported increased knowledge, some reported improved attitudes and perceived behavioral control, and only four reported behavioral changes. Because of small sample sizes, many studies did not include a statistical analysis of the differences between pre- and postintervention survey results. This review article provides insights and guidance for the development of food safety education for produce growers.

HIGHLIGHTS
• Handling of agricultural water and soil amendments were the topics least understood.

• Behavior change was the matrix least often evaluated.

• On-going training is important to the retention of food safety information.

• On-going training can improve confidence for implementing food safety practices.

The farm environment provides many opportunities for fresh produce to become contaminated during preharvest, harvest, and postharvest activities. Produce was reported to contribute to 46% of the foodborne illnesses and 23% of the deaths in the United States between 1998 and 2008 (33). In 1998, the U.S. Food and Drug Administration (FDA) (49) published a guidance for the produce industry aiming at minimizing microbial food safety hazards for fresh fruits and vegetables. These good agricultural practices (GAPs) have served as the basis for voluntary on-farm food safety strategies for many years. However, in 2011, the Food Safety Modernization Act (FSMA) was signed into law and directed the FDA to enact seven food safety rules. Of these, the Produce Safety Rule (PSR) sets regulatory standards for the safe growing, harvesting, packing, and holding of raw agricultural commodities for human consumption (50). However, FSMA-related food safety training and education for produce growers was needed to support the transition and compliance with the new regulation. Therefore, a variety of educational initiatives were developed targeted to large and small growers to help deliver educational programs, increase food safety knowledge, and provide technical support.

Agriculture extension educators have evaluated the impact of a variety of educational interventions (including GAPs training and Produce Safety Alliance grower training) on growers and the outcomes from participation in these programs. Various methods can be used to evaluate and measure program outcomes: needs assessments, focus groups, surveys, case studies, semistructured interviews, and observations. Evaluations of knowledge, behavior, and attitude are used as indicators representing needs, baseline program information, outcomes, and the impacts of the programs. However, approaches to overall program evaluation have been highly variable among educational interventions, making it challenging to determine the overall impacts of produce safety programs. Most food safety educational programs are based on the transfer of knowledge from educator to produce grower. However, knowledge does not always lead to changes in behavior (12, 45), and that factors that influence behavior do not always have a linear relationship (45).

Researchers have assessed grower knowledge of, attitudes toward, and implementation of best practices for growing, harvesting, and storing produce to minimize the risk of microbial contamination (17, 24, 38, 44), the influences on grower decision making regarding on-farm food safety (34, 47), and resource and information needs to help implement best practices for compliance with regulations (20, 35, 46). Many of these assessments have focused on growers in a region or state and have considered farm size based on the PSR exemptions. However, little has been done to compare the overall impact of these interventions on food safety practices and thus to identify ways to improve the interventions.

An understanding of the on-farm practices of produce growers will aid in the development of educational programming that promotes knowledge, attitudes, and behavioral changes. The purpose of this review was to summarize the assessments that have been conducted among the grower community to provide insights and guidance on the development of future grower food safety education programs. This summary was conducted (i) to assess produce growers' attitudes, knowledge, and self-reported and observed on-farm food safety practices, (ii) to evaluate the effectiveness of previous grower food safety education interventions, and (iii) to provide recommendations for future food safety education for produce growers.

## METHODS

A search of on-line databases, journal archives, and conference abstracts was conducted in September 2019 to obtain published a group of food safety studies that included produce growers. Databases and archives used to conduct the search included the Educational Resources Information Center (ERIC), Google Scholar, Purdue University Library, Agricola, the Food Protection Trends archive, the Journal of Extension archive, the Food Control archive, the Journal of Food Science Education archive, the Foodborne Pathogens and Disease archive, the International Food Microbiology archive, and the International Association for Food Protection conference abstract book. The search key words were [“farm” or “farmers” or “growers” or “produce”] and [“education” or “food safety”]. The current literature of grower food safety education is limited. Key words such as “evaluation” and “fruit and vegetable growers” were not used because they were too specific to identify enough articles for the review. The studies were included when they met all the selection criteria: (i) the participants were produce growers, (ii) researchers conducted a needs assessment, evaluated growers' attitudes, knowledge, or on-farm food safety practices, or evaluated the effectiveness of food safety educational interventions for growers, (iii) primary data were collected through use of a survey, a focus group, interviews, or observations, (iv) the study report was published in English, and (v) the study report was published between 2000 and 2019. Duplicate studies by the same research teams who evaluated the same education interventions but published the reports in different journals were not included. The references of selected studies were examined to identify studies missed with the previous on-line search.

All studies were reviewed, and the findings were summarized in six categories: handling of agricultural water; soil amendments; management of domesticated animals and wildlife; worker health, hygiene, and training; cleaning and sanitation; and food safety plans and record keeping. The information from each study was summarized in Excel (version 16, Microsoft, Redmond, WA). Study methodology, sample size and characteristics, identified challenges, and preferred food safety delivery formats were also recorded for further analysis. Two trained researchers coded and verified the data: one entered the information in Excel, and the other verified the information.

Although the protocol of the study was carefully developed, several limitations remained. In the studies included, growers' knowledge, attitudes, and practices were measured and reported in various ways, making it difficult to compare and compile the results across studies. In some studies, grower knowledge was measured with a 5-point Likert scale, and in others grower knowledge was measured with a correct response rate for knowledge questions. The number of published studies on related topics also was limited, so the sample size for this review was relatively small. Not all published studies included were peer reviewed. The abstracts from the International Association for Food Protection conferences that were included provided only limited information about study methodology and findings. Therefore, the results of this review may not be representative of all produce growers, although it can serve as a guide for future work.

## RESULTS AND DISCUSSION

A total of 43 published studies from six countries were selected, including 34 journal articles, two government reports, and seven International Association for Food Protection conference abstracts. The majority of the studies (86%) were conducted in the United States, 5% were done in Canada, and the remainder were done in the United Kingdom, Trinidad and Tobago, Brazil, and Iran (Table 1). The number of studies concerning on-farm food safety has been increasing since 2000; over half of the studies (56%) were published after 2015, and only 5% were published between 2000 and 2004. Among all the studies, 93% included surveys as the primary data collection method, followed by interviews (9%), observations (7%), and focus groups (2%). In 56% of the studies, measurements were based on GAPs, 21% focused on the PSR, and 2% focused on on-farm food safety practices. The remainder did not specify which food safety regulations or principles were used. Between 2000 and 2018, questions in most studies (73%) were based on GAPs, whereas after 2018, most questions were based on the PSR (69% of studies). The PSR was signed into law in 2011 as a part of the FSMA, but implementation did not begin until 2016. Most of the studies were conducted with small- and medium-scale produce growers (60 and 40%, respectively); only 30% of studies included large-scale growers. In a few studies (30%), food safety educational interventions were evaluated among produce growers. Of those studies, three included the use of theoretical models: Bennett's hierarchy and the theory of planned behavior (31, 45, 47).

TABLE 1

Characteristics of selected produce grower food safety education studies (n = 43)

Many growers perceived the implementation of on-farm food safety principles as beneficial. Over 80% of growers agreed that following on-farm food safety principles helped reduce the risk of cross-contamination on produce (23). However, growers lacked confidence in their ability to follow these safety principles (40). The percentage of compliance with recommended principles differed across studies as was affected by growers' knowledge and attitudes. In this review, growers' knowledge, attitudes, and practices of on-farm food safety principles were further evaluated.

### Handling of agricultural water

Agricultural water refers to the water used during production, harvest, and postharvest such as irrigation water and wash water, which can spread pathogens through direct contact with produce. Many growers lacked knowledge of how to safely handle agricultural water. Their knowledge scores on water safety were 59 to 68%, much lower than the 80% value defined for proficiency (37, 38). Growers' self-reported knowledge related to the type of water tests required, water testing frequency, and analysis of test results was especially limited (35, 46).

Growers had misperceptions relating to agricultural water safety. In two needs assessment studies, many growers did not recognize agricultural water as a source of foodborne pathogens. Ivey et al. (23) found that less than half of produce growers strongly agreed that irrigation water and wash water were potential sources of contamination, and <20% of growers believed that limiting contact between irrigation water and the edible parts of crops would be very important for reducing cross-contamination. Hultberg et al. (22) found that 66% of growers thought that irrigation water would not spread pathogens to crops.

Growers' self-reported sources of agricultural water varied by the intended use of the water, farm size, and farm location. When water was intended to be used in postharvest activities (e.g., washing), well water (44 to 88% of growers) and municipal water (9 to 31%) were commonly used, whereas only a few growers (0.4 to 5%) used surface water (21, 39, 48, 52). Production water is defined as the water used in contact with produce during growing activities (e.g., irrigation, fertigation, and crop sprays) (50). Well water (37 to 59% of growers), surface water (e.g., ponds, springs, rivers, or streams) (8 to 59%), and municipal water (3 to 17%) were three main sources of production water (3, 4, 9, 13, 21, 22, 39, 52). Some growers relied on natural rainwater without using production water (4, 22). Large-scale growers were more likely than small-scale growers to use production water (4) and were more likely to use well water as production water (34). Farm location might be another reason for the difference between the source of production water. Fifty-nine percent of 297 New England growers reported using surface water, mostly from ponds (38%) (13), whereas only 8% of 246 Minnesota growers reported using water from ponds and streams for irrigation (22).

Testing of agricultural water is critical for ensuring water safety. Many growers (51 to 88%) reported testing their water sources for contamination (1, 43, 51), 27 to 40% reported using tested water for production (9, 13, 21), and 39% reported using tested water as wash water (21). However, even when growers tested the water, the testing frequency was lower than the PSR requirements of at least once per year for well water and five times per year for surface water (4). Growers were more aware of the safety concerns for postharvest water than for production water and believed the treatment of postharvest water would be more effective than the treatment of production water for reducing cross-contamination (23). Thus, growers were more likely to treat postharvest water than to treat production water (9, 22, 23, 52).

### Soil amendments

Soil amendments are any materials (chemical, biological, and physical) that are added to soil to improve its properties and support the growth of crops. Biological soil amendments of animal origin are considered treated when they have been adequately processed via a scientifically validated procedure to eliminate microorganisms. When these amendments have not been processed to eliminate the microbial risks or have been contaminated, they are considered untreated (50). Because of elevated microbial food safety risks, manure handling and application were the most common concerns about soil amendment practices reported in the selected studies. Growers had limited knowledge of soil amendment topics, especially regarding biological soil amendments of animal origin such as composted manure (35, 37). Growers self-reported that their current knowledge level was low about types of soil tests, testing frequency, time interval for application, and treated versus untreated soil amendments (46).

Self-reported practices of manure usage varied; in some studies about half of the growers (43 to 60%) used manure as a soil amendment on crops (13, 21, 28, 44), and others (8 to 17%) reported low manure usage (4, 9, 51). Farm size may have contributed to this difference, as indicated by the demographic information available. The majority of growers reporting a higher percentage of manure usage were small-scale farms with <5 acres (2 ha) (28, 44), whereas most large-scale farms (10 to 1,000 acres [4 to 405 ha]) reported a lower percentage of manure use (4, 9). Both untreated raw manure and compost were commonly used. Untreated raw manure has significant microbial risks compared with compost that has been treated to eliminate pathogens. In a report for the U.S. Department of Agriculture, Economic Research Service, only 5% of 4,618 U.S. produce growers used untreated raw manure and 7% used compost (4), which were relatively low rates compared with those in other studies: 16 to 58% for raw manure and 29 to 66% for compost (13, 36, 44, 48). A few studies included further investigation of the compost used. Most growers (74 to 80%) who used compost reported that the processing time and temperature of their compost met validated standards (4, 36).

To reduce the food safety risks from manure, validated manure treatments and time intervals between application and harvest are two important factors for eliminating pathogens. Hultberg et al. (22) reported that 66% of 246 small-scale growers properly treated manure to reduce pathogen levels, whereas Pires et al. (36) reported that only 28% of 594 organic growers from small- and medium-scale farms treated manure before application. Regarding the time interval between manure application and harvest, the majority of growers in the study by Pires et al. incorporated a 90- to 120-day waiting period (36), and 69% of growers in the study by Hultberg et al. waited >120 days before harvest (22). Approximately 6 to 27% of growers who used manure waited <90 days between application and harvest (21, 36), which is not enough time to achieve pathogen reduction. When raw manure contamination was observed, only 26% of growers took corrective action to reduce the risks (36).

Proper manure storage is essential to prevent cross-contamination of produce. The best practices for manure storage include preventing runoff and wind drift from manure to the production field, packing and storage area, and water sources. The most common manure storage for growers was a manure pile (36). Growers were aware of the importance of risk management for manure storage; 74% of vegetable growers self-reported that the manure stored close to the production fields was contained to prevent cross-contamination (22). However, observations revealed that only 27% of the farms stored manure properly (17).

### Management of domesticated animals and wildlife

Animals, both domesticated animals and wildlife, can carry and transfer pathogens to produce in the production and postharvest phases. Limiting animal access, monitoring animal intrusion, and tracking adjacent land uses can reduce cross-contamination risks. Restricting the access of domesticated animals and wildlife by using fencing, netting, or deterrents decreases possible fecal contamination in production fields and water sources. Monitoring animal intrusions by checking for feces, tracks, or trampling helps growers assess cross-contamination risks and determine whether corrective action is needed. Understanding adjacent land uses, such as animal farms or landfills, allows growers to identify and avoid possible contamination from the surrounding environment.

Growers had a relatively clear understanding of animal management, but many did not agree that certain preventive measures are important. Growers' knowledge about animal and pest control topics was 86% as reported by Pivarnik et al. (38), with mastery defined as 80%. Growers self-reported a moderate understanding of the prevention of cross-contamination from animal feces to produce (35). However, Ivey et al. (23) reported that <40% of the vegetable growers believed it was important to separate animal farms from production fields (38% of growers) and to use barriers to limit animal access (35%).

Because animals may enter production fields, many growers reported taking measures to limit animal access. In the study by Harrison et al. (21), 52% of growers indicated that domesticated animals could access their production fields. In the study by Marine et al. (28), >80% of growers admitted that wildlife entered their production fields during the growing season. The majority of growers (68 to 97%) reported taking measures to prevent animals from entering production fields (13, 22, 51, 52), and 93% kept animals and pests out of storage and packing areas (22). Preventive measures taken by growers to restrict access by wildlife included trapping, hunting, building fences, chemical repellents, deterrents, and domestic guard dogs (4, 6, 25, 28).

Many growers monitored their fields for animal intrusions. In the study by Astill et al. (4), 70% of 4,618 growers reported monitoring their fields, whereas in the study by Adalja and Lichtenberg (1) 47% of 394 growers reported monitoring. Growers who ran large-scale farms were more likely to report monitoring of animal intrusions than were small-scale growers (4).

Uses of land adjacent to produce fields also can contribute to cross-contamination. Produce can be contaminated directly or indirectly through contaminated agricultural water or soil. Many growers tried to minimize the risks by growing produce far from animal operations and preventing animals from contacting water sources. However, their practices might have been affected by the types of produce they grew. Eighty-two percent of vegetable growers in the study conducted by Hultberg et al. (22) indicated that their production fields were located far from animal operations. However, in the study by Wright et al. (51) 54% of apple growers reported animals grazing on adjacent land, and >20% of the produce growers in the study by Bihn et al. (9) noted the presence of animal farms, landfills, or other operations within 1 mi (1.6 km) of their production fields. To prevent cross-contamination between animals and water sources, most growers prevented livestock from contacting the sources of irrigation water (13, 22).

### Worker health, hygiene, and training

Humans can carry many pathogens such as Shigella, Shiga toxin–producing Escherichia coli, Salmonella, hepatitis A virus, norovirus, and Cyclospora. Workers in direct contact with fresh produce can introduce contamination via hands, clothing, feces, saliva, and blood. Ensuring worker health, hygiene, and proper training can help reduce the risks of cross-contamination. Overall, growers understood worker health, hygiene, and training topics and saw the importance of workers' food safety awareness. The grower knowledge score for the worker health and hygiene topic was higher than the mastery level of 80% (38). They self-reported moderate understanding about the worker training topic (35). Growers understood the importance of making sure workers are knowledgeable about food safety (38), and >80% of growers believed their workers were at least somewhat knowledgeable about food safety topics (23).

Workers who are sick can transmit pathogens easily; thus, they should not be in close contact with fresh produce. Most growers (78%) were aware that visibly ill workers should not handle produce (13). In the study by Hultberg et al. (22), 83% of 246 vegetable growers indicated that they kept sick workers from contacting produce, and 74% of 27 farms in the on-farm visits actually engaged in that practice (17).

Many growers pay close attention to worker hygiene, especially hand washing practices. Over 70% of growers reported requiring workers to wash their hands before working (22, 52). Growers understood the importance of hand washing, but the hand washing frequency of their workers was inconsistent. In the study by Soon and Baines (45), almost all participants washed their hands before harvesting and packing produce and after using the bathroom, but only 60% of workers washed their hands after coughing and sneezing. To maintain worker hygiene, most growers (83 to 96%) self-reported that they provided cleaning facilities for their workers, such as hand washing and bathroom facilities (1, 6, 13, 52). However, only 66% of the 226 small- and medium-scale growers in the study by Harrison et al. (21) stated that their cleaning facilities for workers were near production fields and packing areas. In the study by Laury-Shaw et al. (25), only 61% of 41 small-scale growers provided hand washing facilities with soap and running water for workers within 0.25 mi (0.4 km) of the work area.

Worker training is essential to ensure the implementation of food safety practices. Seventy-nine percent of growers reported that their workers had received proper food safety training (52). Many growers offered training to their workers (1), including training in general food safety (47 to 50% growers) (4, 13, 25), raw manure handling (83%) (36), hygiene practices (78%) (22), and sanitation (41%) (21). No growers reported providing trainings on water management.

### Cleaning and sanitation

Cleaning and sanitation are critical steps for preventing produce cross-contamination from food contact surfaces such as containers, equipment, and tools. Pivarnik et al. (37, 38) reported that growers lacked knowledge about postharvest handling practices. However, many growers self-reported proper cleaning and sanitation practices. The majority reported cleaning (86 to 93%) and sanitizing (84%) harvest containers (1, 13, 22, 28). In the study by Marine et al. (28), significantly more growers in 2013 reported sanitizing harvest containers compared with growers in 2010. Marine et al. suggested that the difference could be due to increased grower knowledge as a result of the training provided by the university agriculture extension programs between 2010 and 2013. Practices use for maintaining the cleanliness of transportation containers also were studied. Harrison et al. (21) reported that only 33% growers cleaned containers used to transport produce to markets. Among those who cleaned containers, 46% used water and detergent for cleaning; other cleaning methods included the use of bleach solution, soapy vinegar water, and the combination of sunlight and rainwater. Many growers also reported cleaning and sanitizing other food contact surfaces, including equipment (55 to 100% of growers) (1, 13, 23, 51, 52) and harvesting tools (67%) (22). However, Astill et al. (4) found that although growers reported they both cleaned and sanitized containers, equipment, and tools, sanitization occurred less often than did cleaning.

Researchers' observations of on-farm practices differed. Hamilton et al. (17) visited 27 produce farms and observed that >80% of the farms kept harvest containers and tools visibly clean. However, Ellis et al. (15) observed nine produce farms and da Cruz et al. (14) observed one produce farm where little or no cleaning and sanitization of harvest containers, equipment, and tools occurred. Ellis et al. also found that equipment was most commonly cleaned by using only water without detergent or sanitizer. This method is cause for concern because containers, equipment, and tools can become a source of contamination when they are not cleaned and sanitized properly.

### Food safety plans and record keeping

A farm food safety plan helps growers identify risks in their operation, establish a plan to address the identified risks, and maintain records of information related to on-farm food safety practices. The FSMA PSR regulation does not require farms to have a food safety plan, but accurate record keeping is useful for tracking farm practices, dealing with audits, and addressing buyers' questions. Growers' record keeping practices differed by the type of information recorded. Most growers kept records of crops (harvest and/or shipping) (76 to 77%) (1, 52), application dates of soil amendments (33 to 92%) (1, 13, 36), information about soil amendments (type, source, and time interval between application and harvest) (67 to 94%) (36, 39), water testing results (41 to 45%) (1, 13), and worker training information (33%) (13). Records were less often kept regarding animal intrusion (21% of growers), flooding (14%), crop testing (11%), soil amendment testing (6%), and water treatment (methods and monitoring results) (3 to 4%) (1).

Many growers had a food safety plan, but related policies and procedures were not always in a written form. Over half of growers (60%) reported they had a complete food safety plan (11). However, only a few growers (3 to 21%) prepared written food safety policies and procedures (25). The most common written policies and procedures concerned produce growing and handling (28%), hand washing (21%), and surface cleaning (18%), and the least common written policies and procedures concerned worker clothing (3%) (25, 28).

### Motivators for and barriers to implementation of food safety practices

Growers were motivated to comply with on-farm food safety principles for two reasons: meeting customer requirements and feeling responsible for producing safe food products. The majority of growers were motivated to implement food safety practices and to obtain certifications required by their customers, especially wholesale retailers (6, 47). Some growers implemented food safety practices out of a sense of responsibility to ensure product safety and to protect their customers (6, 45, 47).

Growers also faced many barriers to compliance with food safety principles. Three main barriers were identified in the studies: (i) growers perceived compliance costs to be burdensome; (ii) growers had limited time to implement recommended practices; and (iii) growers lacked knowledge and resources. Compliance costs, including the cost of certification, required tests, and other farm investments, were ranked as one of the greatest barriers to implementation of food safety practices (5, 24, 30, 44). Becot et al. (6) estimated that the average cost of receiving GAP certification for small- and medium-scale farms in Vermont was between $2,599 and$3,983. Bovay et al. (10) estimated that FSMA PSR annual compliance costs represented >6% of the value of the annual produce sales of small and very small farms. The percentage for smaller farms was higher than that for larger farms. Costs were especially overwhelming for small-scale growers, and some growers were concerned that they might be unable to cover the investment (24, 44).

A lack of time was the second major barrier to implementation of food safety practices. Growers self-reported that farm work prevented them from having enough time to learn about food safety principles or to implement recommended food safety practices. Development of an on-farm food safety plan and keeping records were emphasized as time-consuming practices (5, 24, 44, 47). As reported by Astill et al. (5), many growers experienced audit fatigue because of various customer food safety requirements.

Growers also self-reported limited knowledge of food safety principles (30, 47) and lack of resources, including qualified food safety personnel, testing laboratories, and multilanguage food safety training resources (5, 34). A lack of qualified food safety consultants, auditors, and testing labs for water samples increased the difficulty of compliance with recommended practices because growers needed to invest more time and money to find those resources (5). Parker et al. (34) conducted semistructured interviews with 31 small- to large-scale growers and identified the need for food safety training resources in relevant languages and in formats that are culturally appropriate for workers. This barrier was greater for medium- and large-scale growers because they usually hired more workers from culturally diverse backgrounds. Language differences increased the difficulties associated with training workers, thus limiting grower ability to comply with food safety principles.

### Educational interventions

In 13 studies, educational interventions were implemented and evaluated (Table 2). The evaluation results were reported in seven journal articles and six conference abstracts published between 2012 and 2019. Some research teams published evaluation articles on the same educational intervention in different journals. For this review, we did not count those repeated evaluations as separate educational interventions. In 10 studies, researchers evaluated in-person-only interventions, in 1 study researchers evaluated a video-only intervention, and in 2 studies researchers evaluated hybrid interventions: a combination of both in-person and on-line multimedia interventions. In three of the studies, researchers stated explicitly that they used theoretical learning models in the interventions (31, 45, 47). Other studies may have incorporated theoretical learning models, but this fact was not stated explicitly. All 13 studies included pre- and postintervention surveys as evaluation tools, and in 6 studies, researchers distributed delayed postintervention surveys 3 to 12 months later to evaluate the long-term effectiveness of the interventions. In one study, a focus group was used for evaluation of the intervention.

TABLE 2

Changes in attitudes, knowledge, perceived behavioral control, and behavior after educational interventions (n = 13)a

TABLE 2

Extended

Although educational interventions targeted behavioral change, not all evaluations measured behavioral change directly. Knowledge change was commonly used as an indicator to assess the effectiveness of an intervention. However, many theoretical learning models, such as Bennett's hierarchy (7) and the theory of planned behavior (2), suggest that knowledge is not the single influential factor of behavioral change. People's attitudes and confidence in their ability to perform a certain behavior (perceived behavioral control) also contribute to their behavioral change intention, which can be a predictor of actual behavioral change. Ten evaluation studies indicated increased knowledge (18, 19, 26, 27, 29, 31, 32, 42, 45, 47), four indicated improved attitudes toward food safety (31, 42, 45, 47), four indicated improved perceived behavioral control (11, 31, 45, 47), six measured behavior change intention (11, 18, 25, 31, 45, 47), and only four indicated behavioral change (8, 11, 25, 47). Because of limited sample sizes or survey design constraints, not all reported improvements were statistically significant. In five studies, statistically significant knowledge increases were found immediately after interventions (18, 26, 27, 31, 45). In one study, significant improvement in attitudes and perceived behavioral control was found after the educational intervention. Nayak et al. (31) reported the mean scores of attitudes toward GAP-related statements, and grower confidence in their ability to implement GAPs increased significantly after the intervention. In only one study was behavioral change found in the immediate postintervention survey. Bihn et al. (8) reported that 48% of growers in the postintervention survey had developed over half their farm food safety plan, which was 35% more than the preintervention survey. When developing an evaluation plan for educational interventions, measurements of behavioral change should be included. Behavioral change models and other conceptual models help predict behavioral change and describe the impact of educational interventions.

A longitudinal approach allows researchers to gain a better understanding of the factors that influence actual behavioral change and promote retention of intervention effectiveness after a certain time period. Six studies incorporated a longitudinal approach (8, 11, 18, 25, 32, 47), and the results were reported for four studies. In two studies, behavior enhancement was noted at the delayed postintervention survey, indicating the long-term effectiveness of the interventions (8, 25). Bunning et al. (11) and Tobin et al. (47) reported that the delayed postintervention survey revealed that more growers actually worked on their food safety plans than had expressed intention to do so in the immediate postintervention survey. However, the delayed postintervention survey also revealed that fewer growers conducted a self-audit and applied for third-party certification than had expressed intention to do so in the immediate postintervention survey (47). The effectiveness of educational interventions also can fade over time. Tobin et al. found that at 6 months, postintervention growers' attitudes, knowledge, and perceived behavioral control decreased from those expressed immediately postintervention.

To better deliver food safety information to growers, some researchers evaluated growers' preferred information delivery formats and their trusted sources. Checklists, fact sheets, handouts (printed and on-line), classes (in person and on-line), and on-line videos were desirable education methods (35, 45, 46). When delivering large quantities of information in videos, researchers suggested developing “bite-size” materials that delivered only one concept or topic per video, which was easier for growers to digest (41, 45). In terms of growers' trusted sources for food safety information, university representatives were ranked as the top source of information (23, 36). Farmer peers or educators with farming experience were also considered credible and trustworthy (16, 46).

Future educational programs should be tailored to meet the educational needs of produce growers and address compliance barriers. Program changes could be made to significantly increase knowledge specifically in agriculture water, soil amendments, and management of domesticated animals and wildlife. Shorter and more routine educational interventions may better establish stronger and more trusted relationships that more effectively increase knowledge and improve the adoption of improved food safety practices. Educational programs should be peer reviewed and consistent with guidance and regulatory documents. Program content should also emphasize topics most beneficial to growers, which would be based on needs assessments from previous studies or primary data from focus groups.

This review covered produce growers' knowledge, attitudes, and practices toward on-farm food safety principles, identified growers' motivators and barriers to implementing food safety practices, and evaluated the effectiveness of various educational interventions. Most produce growers were aware of cleaning and worker training topics. However, they had limited knowledge about the food safety risks associated with agriculture water and soil amendments. Major barriers identified for grower compliance with the food safety principles were costs, time, and limited knowledge and resources. In 13 studies, researchers evaluated food safety educational interventions, and most studies included in-person workshops and self-reported pre- and postintervention knowledge assessments. In most studies researchers reported increased knowledge, in some studies researchers reported improved attitudes and perceived behavioral control, but in only four did researchers report behavioral changes. Because of the small sample sizes in these studies, many did not include statistical analysis of the differences between pre- and postintervention survey results. This review provides insights and guidance on the future development of food safety education for produce growers.

## ACKNOWLEDGMENTS

This work was supported partially by the National Institute of Food and Agriculture, U.S. Department of Agriculture, Hatch Projects (S1077-1016049 and S294- IOW05522), the Food Safety Outreach Program (2018-70020-28851), the Local Food Promotion Program (AM180100XXXXG124), and the Indiana State Department of Agriculture (A337-19-SCBG-18-006).

## SUPPLEMENTAL MATERIAL

Supplemental material associated with this article can be found online at: https://doi.org/10.4315/JFP-20-320.s1

## REFERENCES

REFERENCES
1.
A.,
and
Lichtenberg
E.
2018
.
Implementation challenges of the Food Safety Modernization Act: evidence from a national survey of produce growers
.
Food Control
89
:
62
71
.
2.
Ajzen,
I.
1991
.
The theory of planned behavior
.
Organ. Behav. Hum. Decis. Process
.
50
:
179
211
.
3.
Allard,
S.,
Suri
M.,
Solaiman
S.,
Callahan
M. T.,
White
C.,
Goeringer
P.,
Pee
D.,
Haymaker
J.,
Hashem
F.,
May
E.,
Parveen
S.,
Kniel
K.,
Sharma
M.,
Micallef
S. A.,
Goldstein
R. R.,
and
Sapkota
A.
2019
.
Comparison of mid-Atlantic grower perceptions and nontraditional irrigation water realities: a conserve study, P2-160
.
J. Food Prot.
82
(Suppl. A):
215
.
(Abstract.)
4.
Astill,
G.,
Minor
T.,
Calvin
L.,
and
Thornsbury
S.
2018
.
Before implementation of the Food Safety Modernization Act's produce rule: a survey of US produce growers. Economic information bulletin 194
.
U.S. Department of Agriculture
,
Economic Research Service,
Washington, DC
.
5.
Astill,
G.,
Minor
T.,
Thornsbury
S.,
and
Calvin
L.
2019
.
US produce growers' decision making under evolving food safety standards. Economic information bulletin 210
.
U.S. Department of Agriculture
,
Economic Research Service,
Washington, DC
.
6.
Becot,
F. A.,
Nickerson
V.,
Conner
D. S.,
and
Kolodinsky
J. M.
2012
.
Costs of food safety certification on fresh produce farms in Vermont
.
Horttechnology
22
:
705
714
.
7.
Bennett,
C.
1975
.
Up the hierarchy
.
J. Ext
.
13
:
7
12
.
8.
Bihn,
E.,
Wall
G.,
Newbold
E.,
and
Schmit
T.
2015
.
Evaluation of a multi-day good agricultural practices training and farm food safety plan writing workshop, T6-07
.
J. Food Prot
.
78
(Suppl. A)
:
47
48
.
(Abstract.)
9.
Bihn,
E. A.,
Smart
C. D.,
Hoepting
C. A.,
and
Worobo
R.
2013
.
Use of surface water in the production of fresh fruits and vegetables: a survey of fresh produce growers and their water management practices
.
Food Prot. Trends
33
:
307
314
.
10.
Bovay,
J.,
Ferrier
P.,
and
Zhen
C.
2018
.
Estimated costs for fruit and vegetable producers to comply with the Food Safety Modernization Act's produce rule. Economic information bulletin 195
.
U.S. Department of Agriculture
,
Economic Research Service,
Washington, DC
.
11.
Bunning,
M.,
Sullins
M.,
Zander
A.,
Follman
R.,
Wells
J.,
Wolsey
P. S.,
Wall
G.,
and
Vanderpool
T.
2015
.
Tools for improving on-farm food safety workshops and survey of food safety costs, P1-141
.
J. Food Prot
.
78
(Suppl. A)
:
115
116
.
(Abstract.)
12.
Clayton,
D. A.,
Griffith
C. J.,
Price
P.,
and
Peters
A. C.
2002
.
Food handlers' beliefs and self-reported practices
.
Int. J. Environ. Health Res
.
12
:
25
39
.
13.
Cohen,
N.
2005
.
Farm food safety practices: a survey of New England growers
.
Food Prot. Trends
25
:
363
370
.
14.
da Cruz,
A. G.,
Cenci
S. A.,
and
Maia
M. C. A.
2006
.
Good agricultural practices in a Brazilian produce plant
.
Food Control
17
:
781
788
.
15.
Ellis,
J. D.,
Strohbehn
C. H.,
and
Henroid,
D. H.
Jr.
2005
.
Assessing on-farm food handling practices of Iowa-grown produce and eggs in regard to food safety
.
Food Prot. Trends
25
:
758
761
.
16.
Ganpat,
W.,
N.,
Walter
S.,
Roberts
L.,
Nandlal
J.,
and
Smith
N.
2014
.
Compliance with good agricultural practices (GAPs) by state-registered and non-registered vegetable farmers in Trinidad, West Indies
.
Food Secur
.
6
:
61
69
.
17.
Hamilton,
K. E.,
Umber
J.,
Hultberg
A.,
Tong
C.,
Schermann
M.,
Diez-Gonzalez
F.,
and
Bender
J. B.
2015
.
Validation of good agricultural practices (GAP) on Minnesota vegetable farms
.
Foodborne Pathog. Dis
.
12
:
145
150
.
18.
Harrison,
J.,
Boyer
R.,
and
Harrison
M.
2015
.
Enhancing the safety of locally grown produce—outcomes and challenges of food safety education for small-scale farmers and market managers, P1-142
.
J. Food Prot.
78
(Suppl. A)
:
116.
(Abstract.)
19.
Harrison,
J.,
Boyer
R.,
Harrison
M.,
and
Pethel
M.
2018
.
Effectiveness of online versus face-to-face produce safety training for farmers and farmers market managers, P2-31
.
J. Food Prot.
81
(Suppl. A)
:
162
.
(Abstract.)
20.
Harrison,
J. A.,
Critzer
F. J.,
and
Harrison
M. A.
2016
.
Regulatory and food safety knowledge gaps associated with small and very small food businesses as identified by regulators and food safety educators—implications for food safety training
.
Food Prot. Trends
36
:
420
427
.
21.
Harrison,
J. A.,
J. W.,
Harrison
M. A.,
Cannon
J. L.,
Boyer
R. R.,
and
Zehnder
G. W.
2013
.
Survey of food safety practices on small to medium-sized farms and in farmers markets
.
J. Food Prot
.
76
:
1989
1993
.
22.
Hultberg,
A.,
Schermann
M.,
and
Tong
C.
2012
.
Results from a mail survey to assess Minnesota vegetable growers' adherence to good agricultural practices
.
Horttechnology
22
:
83
88
.
23.
Ivey,
M. L. L.,
LeJeune
J. T.,
and
Miller
S. A.
2012
.
Vegetable producers' perceptions of food safety hazards in the midwestern USA
.
Food Control
26
:
453
465
.
24.
Jackson,
C.,
Larcher
D.,
Goodrich-Schneider
R.,
Gravani
R. B.,
Bihn
E. A.,
and
Schneider
K. R.
2007
.
Determining the effect of good agricultural practices awareness on implementation: a multi-state survey
.
Food Prot. Trends
27
:
684
693
.
25.
Laury-Shaw,
A.,
Strohbehn
C.,
Naeve
L.,
Wilson
L.,
and
Domoto
P.
2015
.
Current trends in food safety practices for small-scale growers in the Midwest
.
Food Prot. Trends
35
:
461
469
.
26.
Lepper,
J.,
Danyluk
M.,
Chapin
T.,
Krug
M.,
McEgan
R.,
Harder
A.,
Narine
L.,
Goodrich
R.,
Langford
T.,
and
Saha
J.
2019
.
Food safety education and outreach for Florida farmers, P1-85
.
J. Food Prot.
82
(Suppl. A)
:
103
.
(Abstract.)
27.
Mahmoud,
B. S.,
Stafne
E. T.,
Coker
C. H.,
Bachman
G. R.,
and
Bell
N.
2016
.
Implementing good practices programs to encourage production of high-quality, safer produce in Mississippi
.
J. Ext.
54
(4)
:
4IAW5.
28.
Marine,
S. C.,
Martin
D. A.,
A.,
Mathew
S.,
and
Everts
K. L.
2016
.
Effect of market channel, farm scale, and years in production on mid-Atlantic vegetable producers' knowledge and implementation of good agricultural practices
.
Food Control
59
:
128
138
.
29.
Mathiasen,
L.,
Morley
K.,
Chapman
B.,
and
Powell
D.
2012
.
Using a training video to improve agricultural workers' knowledge of on-farm food safety
.
J. Ext.
50
(1)
:
a6
.
30.
Narine,
L. K.,
Harder
A.,
and
Danyluk
M.
2019
.
Floridian producers' concerns about the Food Safety Modernization Act
.
Food Prot. Trends
39
:
237
244
.
31.
Nayak,
R.,
Tobin
D.,
Thomson
J.,
R. B.,
and
LaBorde
L. F.
2015
.
Evaluation of on-farm food safety programming in Pennsylvania: implications for extension
.
J. Ext.
53
(1)
:
a9
.
32.
Pahl,
D.,
T.,
Narrod
C.,
Walsh
C.,
Martin
D.,
and
Micallef
S.
2015
.
Evaluating the effectiveness of food safety trainings by determining factors that influence grower adoption of food safety certification practices, P1-145
.
J. Food Prot.
78
(Suppl. A)
:
117
.
(Abstract.)
33.
Painter,
J. A.,
Hoekstra
R. M.,
Ayers
T.,
Tauxe
R. V.,
C. R.,
Angulo
F. J.,
and
Griffin
P. M.
2013
.
Attribution of foodborne illnesses, hospitalizations, and deaths to food commodities by using outbreak data, United States, 1998–2008
.
Emerg. Infect. Dis
.
19
:
407
415
.
34.
Parker,
J. S.,
Wilson
R. S.,
LeJeune
J. T.,
and
Doohan
D.
2012
.
Including growers in the “food safety” conversation: enhancing the design and implementation of food safety programming based on farm and marketing needs of fresh fruit and vegetable producers
.
Agric. Human Values
29
:
303
319
.
35.
Perry,
B.,
Shaw
A.,
Johnsen
E.,
Enderton
A.,
Strohbehn
C. H.,
and
Naeve
L.
2019
.
Assessment of Midwest growers' needs for compliance with the Food Safety Modernization Act Produce Safety Rule
.
Food Prot. Trends
39
:
212
217
.
36.
Pires,
A. F. A.,
Millner
P. D.,
Baron
J.,
and
Jay-Russell
M. T.
2018
.
Assessment of current practices of organic farmers regarding biological soil amendments of animal origin in a multi-regional US study
.
Food Prot. Trends
38
:
347
362
.
37.
Pivarnik,
L. F.,
M. S.,
Leydon
N.,
and
Gable
R. K.
2006
.
New England home gardeners' food safety knowledge of fresh fruits and vegetables
.
Food Prot. Trends
26
:
298
309
.
38.
Pivarnik,
L. F.,
Richard
N. L.,
Wright-Hirsch
D.,
Becot
F.,
Conner
D.,
and
Parker
J.
2018
.
Small- and medium-scale New England produce growers' knowledge, attitudes and implementation of on-farm food safety practices
.
Food Prot. Trends
38
:
156
170
.
39.
Rangarajan,
A.,
Pritts
M. P.,
Reiners
S.,
and
Pedersen
L. H.
2002
.
Focusing food safety training based on current grower practices and farm scale
.
Horttechnology
12
:
126
131
.
40.
Rezaei,
R.,
Mianaji
S.,
and
Ganjloo
A.
2018
.
Factors affecting farmers' intention to engage in on-farm food safety practices in Iran: extending the theory of planned behavior
.
J. Rural Stud
.
60
:
152
166
.
41.
E.,
and
Ellis
J. D.
2010
.
Food tube: coverage of food safety issues through video
.
J. Food Saf
.
30
:
162
176
.
42.
Shaw,
A. M.,
Strohbehn
C. H.,
Naeve
L. L.,
Domoto
P. A.,
and
Wilson
L. A.
2015
.
Knowledge gained from good agricultural practices courses for Iowa growers
.
J. Ext.
53
(5)
:
rb3.
43.
Simmons,
K.,
Harrison
M. A.,
Hurst
W. C.,
Harrison
J.,
Brecht
J.,
Schneider
K.,
Simonne
A.,
and
Rushing
J.
2007
.
Survey of food defense practices in produce operations in the Southeast
.
Food Prot. Trends
27
:
174
184
.
44.
Sinkel,
D.,
Khouryieh
H.,
J. K.,
Stone
M.,
and
Shen
C.
2018
.
Knowledge and implementation of good agricultural practices among Kentucky fresh produce farmers
.
Food Prot. Trends
38
:
111
121
.
45.
Soon,
J.,
and
Baines
R.
2012
.
Food safety training and evaluation of handwashing intention among fresh produce farm workers
.
Food Control
23
:
437
448
.
46.
Strohbehn,
C. H.,
Enderton
A. E.,
Shaw
A. M.,
Perry
B. J.,
Overdiep
J.,
and
Naeve
L.
2018
.
Determining what growers need to comply with the Food Safety Modernization Act Produce Safety Rule
.
J. Ext.
56
(7)
:
rb1.
47.
Tobin,
D.,
Thomson
J.,
LaBorde
L.,
and
R.
2013
.
Factors affecting growers' on-farm food safety practices: evaluation findings from Penn State Extension programming
.
Food Control
33
:
73
80
.
48.
Tong,
C. B.,
Schermann
M. A.,
Diez-Gonzalez
F.,
and
Rossbach
J.
2017
.
Food safety risks of leafy greens from small-acreage farms in Minnesota
.
J. Ext.
55
(4)
:
RIB4.
49.
1998
.
Guidance for industry: guide to minimize microbial food safety hazards for fresh fruits and vegetables
.
50.
2020
.
FSMA final rule on produce safety
.
Standards for the growing
,
harvesting, packing, and holding of produce for human consumption
.
51.
Wright,
J. R.,
Sumner
S. S.,
Hackney
C. R.,
Pierson
M. D.,
and
Zoecklein
B. W.
2000
.
A survey of Virginia apple cider producers' practices
.
Dairy Food Environ. Sanit
.
20
:
190
195
.
52.
Young,
I.,
Rajic
A.,
Dooh
L.,
Jones
A. Q.,
and
McEwen
S. A.
2011
.
Use of good agricultural practices and attitudes toward on-farm food safety among niche-market producers in Ontario, Canada: a mixed-methods study
.
Food Prot. Trends
31
:
343
354
.