1. Introduction
The World Health Organization (WHO) recommends that an adult should eat at least 400 g of produce every day to mitigate the risk of chronic illness and ensure an adequate daily intake of dietary fiber [
1]. Therefore, many countries encourage fresh produce consumption through campaigns at the national government level. For example, in the United States (U.S.), the Department of Agriculture launched the MyPlate program [
2] and, in the United Kingdom (U.K.), the Department of Health & Social Care [
3] launched a program called the “Change4Life programme” to encourage people to eat more fresh produce and promote weight control. Moreover, consumers prepare less food at home due to a busier lifestyle and perceive fresh-cut produce as being healthy [
4]. These government-led activities, the scientific literature, and consumer needs result in an increasing demand for fresh produce [
5]. For example, the volume sales and dollar sales of fresh produce rose by 8% and 22% from 2010 to 2014 in the U.S. [
6].
Within the fresh produce segment, the fresh-cut produce industry has developed rapidly in the last decade [
7] and has become a substantial segment of the produce industry with multi-billion-dollar sales [
8]. Based on the Statista report, the consumption of fresh-cut produce in the U.S. accounts for approximately 29.5% of fresh produce consumption [
9]. The International Fresh-Cut Produce Association defined fresh-cut produce as “any fresh fruit or vegetable that has been physically altered from its original form but remains in a fresh state” [
10]. According to the data from Supermarket Perimeter, fresh-cut and value-added vegetables, as a category, had a 6.6% increase in dollar sales in 2018 compared with 2017, partly due to convenience in terms of minimizing food preparation time [
11,
12]. Value-added vegetables include dehydrated, jammed, and pickled vegetables, which account for a relatively small portion of processed vegetables [
13].
There has been a correlation between the increased intake of fresh and fresh-cut produce and the number of foodborne disease outbreaks associated with fresh and fresh-cut produce [
4,
14,
15,
16]. According to the U.S. Centers for Disease Control and Prevention (CDC) National Outbreak Reporting System [
17], a total of 47,961 outbreaks were reported, where fresh produce accounts for approximately 60% of the outbreaks (n = 28,686) in the U.S. between 2000 and 2018. In addition, a rising concern from consumers exists about the risks associated with fresh-cut produce [
18]. In particular, the frequency of foodborne disease outbreaks linked with fresh-cut produce has increased, despite the fresh-cut produce being perceived as safe by consumers [
18,
19,
20]. A previous study found that among 606 produce-associated foodborne illness outbreaks, 73.3% of the outbreaks were associated with fresh-cut produce [
21]. Most recently, a
Salmonella Javiana outbreak in February 2020 related to fresh-cut mixed fruit (melons and pineapple) resulted in a total of 165 people infected across 14 states [
22].
Fresh-cut produce has a higher risk of being associated with foodborne pathogen contamination due to additional processing procedures, including washing, cutting, and packaging when compared to other fresh intact produce [
23]. Two possible reasons can explain why fresh-cut produce may carry higher foodborne illness risks compared with fresh produce [
21]. First, processing of the produce (i.e., cutting, peeling, and shredding) can demolish cell surfaces and expose cytoplasm, which may offer another nutrient source for microorganisms than intact produce [
24]. Second, most fresh-cut produce is consumed raw, meaning cooking is not used to inactivate and reduce pathogen loads [
14]. Fresh-cut produce items that are ready to eat can make them carriers of foodborne pathogens if such items are not prepared and handled properly [
16,
18]. Previous studies have found that bagged fresh-cut romaine lettuce can contain equal levels of foodborne pathogens compared to whole leaf lettuce [
15,
25,
26].
Three essential factors that influence fresh-cut produce safety practices in the farm to fork continuum were pointed out by food safety scholars: (1) post-harvest practices in processing facilities, (2) employees’ handling practices in retail facilities, and (3) consumers’ handling practices in domestic kitchens [
27,
28]. Previous studies have mainly focused on the first two factors: post-harvest and employees’ handling practices in retail facilities [
4,
15,
29,
30]. The current study focuses on consumers’ handling practices to alleviate foodborne illness risks. Additionally, although significant gender and generation differences in food safety knowledge and perception have been observed in the contexts of farmers’ market and foodservice operations [
31,
32], their impacts on food-safety-related knowledge and practices in the context of fresh-cut produce need to be examined.
The overarching objectives of the current study were to investigate the knowledge, risk perceptions, and practices (self-reported) regarding fresh-cut produce among U.S. consumers and to determine the relationships among these factors. The results obtained from this study can be used to make targeted consumer food safety tools. Specifically, the present study conducts a national questionnaire to examine (1) U.S. consumers’ food safety knowledge, risk perceptions, and handling practices associated with fresh-cut produce; (2) the effects of demographic factors (i.e., gender and generation) on consumers’ food safety knowledge, risk perceptions, and food safety practices toward fresh-cut produce; and (3) the interaction effect of consumers’ risk perceptions on the relationship between food safety knowledge and food safety practices.
3. Results
3.1. Demographics
Table 1 shows the descriptions of the demographic attributes of the respondents. More than 60% of the respondents identified as women (61.8%, N = 579). The respondents included 511 millennials (54.5%), 224 baby boomers (23.9%), and 202 members of generation X (21.6%). Nearly 40% of the respondents received college degrees or above. Moreover, approximately 50% of the participants had an annual income between USD 35,000 and 79,999 (49.6%, N = 465). Approximately 48% of the respondents purchased and consumed fresh-cut produce two to three times weekly.
A principal component analysis using Varimax rotation was conducted to explore the validity of the measures of food safety practices and food safety risk concerns. The results revealed a two-factor solution accounting for 72.2% of the overall variance. The eigenvalues of the two dominant factors were 4.85 and 2.37, respectively, indicating that the validity of food safety practices and risk concerns is acceptable. All of the factor loadings of food safety practice items on the first factor were above 0.7 and the factor loadings of risk perception items on the second factor were also above 0.7, indicating good measurement validity [
47].
3.2. Consumers’ Knowledge and Practices Associated with Fresh-Cut Produce
Table 2 demonstrates the results of food safety knowledge associated with fresh-cut produce. Only 9.5% (N = 89) of the respondents correctly answered that bagged salad/fresh-cut fruit should be stored above raw meat and poultry in the refrigerator. In addition, 27.0% (N = 253) of the respondents correctly answered the question regarding kitchen surface cleaning and sanitizing, and only 34.9% (N = 327) of the respondents chose to throw away the bagged salad/fresh-cut fruits with no bruises or damage but passed the expiration date.
In terms of food safety practices (
Table 3), adapting the scoring system from previous studies [
37,
48], insufficient food safety practices related to fresh-cut produce are defined as any practice of which the average score is less than 4 (frequently, about 70% of the time). Respondents fell short in the following areas: storing bagged salad/fresh-cut fruits and vegetables on the highest shelf in the refrigerator (mean = 2.28 out of 5), putting a thermometer in my refrigerator to check the temperature (mean = 2.60), throwing bagged salad/fresh-cut fruits and vegetables away if they have been at room temperature for four hours (mean = 3.26), and throwing bagged salad/fresh-cut fruits and vegetables away if they pass the expiration date (mean = 3.56). The finding of food safety practices was consistent with that of food safety knowledge, indicating that consumers do not know or follow the food safety guidance regarding storage and time and temperature controls of fresh-cut produce [
35,
36,
49].
3.3. Consumers’ Food Safety Knowledge and Practices Associated with Fresh-Cut Produce among Demographic Groups
MANOVA was used to compare the differences among social demographic factors, including gender, generation, and their interaction effect on consumers’ food safety knowledge and practices associated with fresh-cut produce. The results of the MANOVA shown in
Table 4 indicated that the overall model was significant in terms of Wilks’ Lambda test (Wilks’ Lambda gender (2, 929) = 10.05,
p < 0.01; Wilks’ Lambda generation (4, 1858) = 49.54,
p < 0.01; Wilks’ Lambda gender*generation (4, 1858) = 10.34,
p < 0.01).
Consumers’ food safety knowledge (F
knowledge (1, 928) = 10.03,
p < 0.01, ηp
2 = 0.01) and food safety practices (F
practice (1, 928) = 4.23,
p < 0.05, ηp
2 = 0.01) associated with fresh-cut produce were significantly different between the men and women respondents.
Table 5 shows the results of the Bonferroni post hoc test. The women (mean = 4.07) had higher food safety knowledge than the men (mean = 3.75) at a 0.01 statistical significance level.
Significant differences also exist in consumers’ food safety knowledge (Fknowledge (2,928) = 45.09, p < 0.01, ηp2 = 0.09) and food safety practices (Fpractice (2, 928) = 27.91, p < 0.01, ηp2 = 0.06) associated with fresh-cut produce among generations. The results of the Bonferroni post hoc test showed that the baby boomer respondents (mean = 3.31) had lower food safety knowledge than the generation X respondents (mean = 4.08) and the millennial respondents (mean = 4.34) at a 0.01 statistical significance level. However, the baby boomer respondents (mean = 3.95) exhibited higher food safety practice levels than the generation X respondents (mean = 3.69) and the millennial respondents (mean = 3.58) at a 0.01 statistical significance level.
Additionally, a significant interaction effect was seen between gender and generation on consumers’ fresh-cut produce handling practices (Fpractice (2, 928) = 14.82, p < 0.01, ηp2 = 0.03). An L-matrix analysis was used to further examine differences in consumers’ fresh-cut produce handling practices among six demographic groups. The results showed that baby boomer men (mean = 4.10) had significantly higher levels of food safety practices than generation X men (mean = 3.78) and millennial men (mean = 3.48). In addition, baby boomer women (mean = 3.80) had significantly higher levels of food safety practices than generation X women (mean = 3.60).
3.4. Consumers’ Food Safety Knowledge and Risk Perceptions Associated with Consumers’ Food Safety Practices
A hierarchical multiple regression analysis was conducted to examine the effects of food safety knowledge and risk perception and their interaction effect on consumers’ food safety practices. Model 1 examined the effects of control variables, including gender, age, income, and education on consumers’ fresh-cut produce handling practices. The result of Model 1, shown in
Table 6, indicated that age (β
age = −0.01,
p < 0.05) and education (
βeducation = −0.07,
p < 0.05) significantly affected consumers’ food safety practices associated with fresh-cut produce.
Model 2 examined the main effects of food safety knowledge and risk perceptions on consumers’ fresh-cut produce after controlling the effects of social demographic factors. After including food safety risk perceptions and knowledge in the regression model, consumers’ education level was not significantly related to their food safety practices (βeducation = −0.03, p > 0.05). The result of Model 2 indicated that when controlling the effects of age, gender, income, and education, food safety knowledge (βknowledge = 0.17, p < 0.05) positively affected consumers’ food safety practices associated with fresh-cut produce significantly. Particularly, in addition to the findings in Model 1, consumers who had more food safety knowledge implemented significantly safer fresh-cut produce handling practices than those who had less knowledge. However, consumers’ risk perception of foodborne illness was not significantly related to their fresh-cut produce handling practices (βrisk = 0.03, p > 0.05).
Model 3 examined the interaction effect of food safety knowledge and risk perceptions on consumers’ fresh-cut produce handling practices. Except for age (β
age = −0.01,
p < 0.05), other social demographic factors were not significantly related to consumers’ fresh-cut produce handling practices in Model 3. The result of Model 3 indicated that food safety knowledge (β
knowledge = 0.16,
p < 0.05) and the interaction effect (β
interaction = 0.07,
p < 0.05) positively affected consumers’ food safety practices associated with fresh-cut produce significantly. Bootstrapping analysis was conducted to assure the robustness of the regression analyses [
50]. The result of the bootstrapping analysis was consistent with multiple linear regression models. The interaction effect between food safety knowledge and risk perceptions had a significant and positive effect on consumers’ food safety practices (β
interaction = 0.07,
p < 0.05, 95% CI: [0.03, 0.09]). As shown in
Figure 2, the result of the interaction effect suggested that the relationship between consumers’ food safety knowledge and fresh-cut produce handling practices was stronger for consumers with high levels of risk perceptions toward fresh-cut produce, compared with those with low risk perceptions.