INTRODUCTIONDisparities in fruit and vegetable consumption exist by occupation, with blue-collar workers consuming fewer servings than white-collar workers (1). Among blue-collar workers, disparities also exist in terms of chronic disease risk (2). For example, construction laborers and motor freight workers have a higher risk for developing chronic diseases than the general adult male population (3–5). While the strength of evidence that consuming fruits and vegetables reduces cancer risk has weakened with recent studies, (6–9) the United States (US) government recommends consuming at least 3 ½ cups of fruits and vegetables per day to lower risk of cardiovascular disease, stroke, type 2 diabetes and lower the risk of developing obesity, which can indirectly affect cancer risk (9–17). Thus disparities in consumption faced by blue-collar workers may contribute to occupation-related gaps in morbidity and mortality.
Workers in some blue-collar industries, such as construction and transportation, may experience common job-related barriers to consuming the recommended amounts of fruit and vegetables. For example, construction laborers switch job sites frequently and may work long hours of overtime. Motor freight workers, including truck drivers and dockworkers, are under the “obligatory vigilance” of extended driving and pressure to deliver goods on time (18). In addition, the lack of healthful food options at construction sites and truck stops are structural barriers that may make eating fruits and vegetables challenging for these mobile blue-collar workers (19, 20). One study found that while long-haul truck drivers in the Midwest of the US placed a high value on healthful food choices, due to large-vehicle parking restrictions, they were limited by the food options that truck-stops and fast food places offered (20). Because of the higher risk for chronic diseases experienced by these workers, it is crucial to understand how factors on and off the job may influence dietary patterns. Doing so may inform interventions seeking to enhance fruit and vegetable intake.
Some studies have looked at individual factors related to fruit and vegetable consumption in these two working populations, such as sleep adequacy in truck drivers and the role of work-family spillover in construction laborers (21, 22) However, the reasons these workers don’t eat the recommended daily amount of fruits and vegetables are mutlifacted and need to be examined at multiple levels within a comprehensive framework. Additionally, instead of looking at each occupational group individually, there may be benefits to comparing the same factors in different groups. First, it identifies how different occupations may operate through the same social contextual factors to influence fruit and vegetable consumption. Second, identifying common factors can facilitate the development of interventions that can be used across multiple groups, including blue-collar workers in other mobile industries, thus optimizing intervention effects.
The main purpose of this study is to compare and contrast correlates of fruit and vegetable consumption in two blue-collar populations: construction laborers and motor freight workers. Both groups completed similar baseline surveys as a part of two different intervention studies. The Social Contextual Model of Health Behavior Change (SCM) was used as a framework to develop the surveys, and in this study, to organize potential predictors of fruit and vegetable consumption according to their level of influence (Figure 1).
RESULTSCharacteristics of the study samplesThe study samples differed on several characteristics, as would be expected with the different study populations. The incomes of the two groups differed substantially with more than half of laborers earning $50,000 or less, while two-thirds of the motor freight workers made over $60,000 a year. Laborers were also more diverse by race/ethnicity with 64% being White and 19% Hispanic/Latino as compared to 83% of motor freight workers being White and only 3% of them being Hispanic/Latino. Laborers were also, on average, almost 10 years younger than motor freight workers.
For both groups, more than a quarter of the samples had some college education or post-high-school training; the mean fruit and vegetable consumption for both groups was also below the recommend number of servings per day (Table 1).
Bivariate analysis between social contextual factors and fruit and vegetable consumptionWe tested the bivariate association between each factor shown in Table 1 and fruit and vegetable consumption, controlling for trucking terminal in the Gear Up for Health sample. We also tested the association between response mode (mail verses phone) and fruit and vegetable intake in the Tools for Health sample (data not shown). Because this association was not statistically significant (p=0.56), we combined the data from both response modes in our analyses.
For both motor freight workers and construction laborers, fruit and vegetable consumption was significantly associated with the same seven factors (Table 2). Agreeing with the statements, “Eating fast food is often my only choice when I am working;” “Because of my work, I don’t have enough time to eat right;” “I often eat junk food because I am tired and stressed from work,” and being White were associated with lower daily intake of fruits and vegetables, with larger effects observed in the motor freight workers. Fruit and vegetable consumption was also positively associated with income and agreeing with the statement “Because of my work, it is especially important that I eat right.” The only difference between the two samples was the direction of the effect for “responsibility for shopping/cooking”: motor freight workers who were responsible for most or all cooking had a −0.29 lower mean servings of fruit and vegetables than those who did not. Among laborers, those who reported responsibility for cooking had a mean consumption of 0.38 servings higher than those who did not.
Multivariate analysis between social contextual factors and fruit and vegetable consumptionWe used the same multivariable model in each group by selecting the variables that were statistically significant in either group. The seven predictors included in each multivariable model were: (1) “Eating fast food is often my only choice when I am working”; (2)”Because of my work, I don’t have time to eat right”; (3) “I often eat junk food because I am tired and stressed from work”; and (4) “Because of my work it is especially important that I eat right,” (5) Responsibility for cooking or shopping for people they live with; (6) Race/ethnicity; and (7) Income. Because responsibility for cooking was significant in both bivariate samples but in opposite directions, and was then not significant in the multivariable models when other variables were controlled for, we believed this may have been a spurious association and therefore did not include it in our final models.
Table 3 shows our final multivariable models. For both construction laborers and motor freight workers, fruit and vegetable consumption was positively associated with income – as income increased, so did fruit and vegetable intake after controlling for other variables in the models. For both groups, agreeing with the statement, “Because of my work it is especially important that I eat right,” was positively associated with consumption as compared to those who disagreed with this statement, holding other variables constant (0.51 for laborers and 0.64 for motor freight workers). Fruit and vegetable consumption was also negatively associated with being White, as compared to other racial/ethnic groups.
We also found that construction laborers who agreed with the statement, “I often eat junk food because I am tired and stressed from work,” had lower intake of fruits and vegetables, compared to those who disagreed with this statement. For motor freight workers, agreement with the statements, “Eating fast food is often my only choice when I am working,” and “Because of my work, I don’t have time to eat right,” was negatively associated with fruit and vegetable intake, compared to those who disagreed with these statements.
DISCUSSIONThe purpose of this study was to compare and contrast correlates of fruit and vegetable consumption in construction laborers and motor freight workers. In these two samples of workers, we found three factors that were shared between the groups and others that were unique to each one.
When other variables were controlled for, greater fruit and vegetable consumption was associated with higher income and agreeing with the statement, “Because of my work, it is especially important that I eat right” for both construction laborers and motor freight workers. Across all income categories for construction laborers, we found a positive gradient in consumption - as income increased, so did fruit and vegetable intake - which is consistent with the literature (35, 36). However, we observed a somewhat different pattern for motor freight workers: while the group making $40,000–$59,000 had lower consumption as compared to the reference group (>$80,000) was anticipated, it was even less than the group making $40,000–$59,000, which was not expected. Additionally, both construction laborers and motor freight workers who agreed that it was especially important to eat right because of their work, consumed more fruits and vegetables compared to those not agreeing with this statement. As part of the formative research conducted in the Tools for Health and Gear Up for Health studies, focus group participants articulated a ‘strong concern about being physically fit for work’ and wanting to ‘fuel their bodies’ to be better able to do their jobs. This corroborates with qualitative research conducted by Sabinksy and colleagues with Danish, unskilled male workers who expressed that “becoming more effective and a greater asset to one’s workplace,” were key motivators for improved nutrition and weight loss (37). Thus it appears believing it is important to eat a healthful diet in order to be effective in the workplace has the potential to shape blue-collar workers’ fruit and vegetable intake.
Finally, construction laborers and motor freight workers who were White had a lower consumption of fruits and vegetables compared to non-white responders. This finding supports other studies which have found that White men on average consume fewer servings of fruits and vegetables compared to other racial/ethnic groups (38, 39).
Other perceptions related to each group of workers reflect the specific nature of the job experiences of these two occupations. In terms of unique factors, construction laborers who agreed with the statement, “I often eat junk food because I am tired and stressed from work” had lower fruit and vegetable intake as compared to those who disagreed with this statement. Given this finding, we would have expected job strain also to be significantly associated with lower intake but it was not in this study. It is conceivable our measure of job strain did not capture elements of strain or stress that were pertinent to the job experiences of construction laborers. For example, it is possible that continually changing worksites - sometimes from week-to-week or even day-to-day – which may involve long commutes could be a source of stress, which is something our measure would not capture (23). While some studies have suggested that workers decrease their intakes of fruits and vegetables in response to job strain, others have found that there was either a weak or no association between the two (33, 40–42). More research is needed to better understand this finding.
For motor freight workers, lower fruit and vegetable intake was significantly associated with agreeing with the statements: “Because of my work, I don’t have enough time to eat right,” and “Eating fast food is often my only choice when I am working.” In the context of their work, these findings were expected. Truck drivers often spend long days on the road and are under time pressures to deliver their loads. Their job experiences involve irregular shifts and mealtimes, and anxiety about traffic, schedules and economic pressures (4, 43–45). Because Federal law restricts the number of hours they can drive, to maximize them, they may continuously snack and eat one large meal at the end of their shift (20). Due to large-vehicle parking constraints, they are confined to eating at truck-stop restaurants and fast food outlets, which may have limited selections of fresh produce (19, 20, 25, 46, 47). Our findings that motor freight workers believed they did not have enough time to eat properly and eating fast food was often their only choice at work are consistent with other research on truck drivers conducted in the US, Europe and Brazil (19, 20, 46–48).
This study had many strengths, most notably that this was amongst the first that examined similarities and differences between social contextual factors and fruit and vegetable consumption in two groups of male, blue-collar workers. By using similar questions and measures, we harmonized the variables used in this study to the best comparability. In addition, we had representation of workers from trucking terminals in four states along the Eastern seaboard of the US and representation from LIUNA’s national membership of construction laborers.
There are also limitations to this study. The construction laborer and motor freight worker data were collected at different time periods and consequently the effects of the modifying factors are not constant over time; the income categories were also not adjusted for inflation or converted to a standard. We used different survey and administration techniques (telephone and mail verses completing the survey on-site). In some cases, there were different response categories, which we addressed in our methods by trying to make them as similar as possible. Our study did not control for the form in which in which the fruits and vegetables were consumed. We also had a low response rate (44%) in the construction laborers background survey, which limits the ability to generalize our findings beyond construction laborers who would respond to a similar type of survey. As others have noted, this in part reflects the challenges of conducting telephone surveys at a time when many calls are blocked or screened as well as for a population that is mobile by profession (23, 49). All data were self-reported. Furthermore, the data are cross-sectional precluding conclusions about temporality or causation.
Nonetheless, this study has important implications for practice and research to improve fruit and vegetable consumption among blue-collar workers. For example, because most construction laborers (89%) and motor freight workers (86%) believed it was especially important to eat right because of work, future interventions could take advantage of this strong belief and provide health messages related to “the importance of healthy eating to workplace effectiveness” through different channels. In addition, management can demonstrate its commitment to healthful eating through organizational changes, such as increasing the availability of fruits and vegetables at worksites. For example, an intervention in bus garages in the US Midwest increased the sale of healthful foods and beverages sold in vending machines by lowering prices and increasing the availability of these items (50). In California, low-income workers increased both fruit and vegetable intake when fresh fruit deliveries were made to their worksites (51) Furthermore, a Swedish intervention improved the eating habits of lorry drivers by educating restaurant staff in healthful food preparation and using truck stop staff as proxy health promoters for customers (52). These types of interventions hold promise for motor freight workers and construction laborers by modifying the offerings at truck stops, trucking terminals, and on food trucks near construction sites.
An interesting finding from this study was that individual approaches to coping with job experiences (i.e. eating in response to workplace stress) were significantly associated with fruit and vegetable intake, but not the organizational factors themselves, such as hours worked or job strain. The nature of these occupations, in which there are relatively loose ties to the work environment, may contribute to the week associations of organizational factors to these dietary factors. This underscores the need for future research to examine the mechanisms through which these experiences may influence dietary behavior. For example, researchers using similar questions could probe “why” workers don’t have time to eat right and “what” about the workplace makes them tired and stressed. In this way, specific features of the work context can be targeted in order to improve worker health.
In conclusion, this study highlights the importance of examining associations between the same set of social contextual factors and fruit and vegetable consumption. It illustrates how work experiences may be related to consumption in different ways and facilitates the development of interventions that can be used across blue-collar occupational groups. Given that construction and motor freight are large and growing industries (53), efforts to help these workers eat a healthful diet may close occupation-related health disparity gaps.