Trends in diet quality and related sociodemographic, health, and occupational characteristics among workers in Spain: results from three consecutive national health surveys (2006–2017)

nutrients Article Trends in Diet Quality and Related Sociodemographic, Health, and Occupational Characteristics among Workers in Spain: Results from Three Consecutive National Health Surveys (2006–2017) Silvia Portero de la Cruz 1and Jesús Cebrino 2,*   Citation: Portero de la Cruz, S.; Cebrino, J. Trends in Diet Quality and Related Sociodemographic, Health, and Occupational Characteristics among Workers in Spain: Results from Three Consecutive National Health Surveys (2006–2017). Nutrients 2021,13, 522. https://doi.org/ 10.3390/nu13020522 Academic Editors: Javier Sanz-Valero and Elena Ronda Received: 27 December 2020 Accepted: 2 February 2021 Published: 5 February 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1Department of Nursing, Pharmacology and Physiotherapy, Faculty of Medicine and Nursing, University of Córdoba, Avda. Menéndez Pidal S/N, 14071 Córdoba, Spain; [email protected] 2Department of Preventive Medicine and Public Health, Faculty of Medicine, University of Seville, Avda. Doctor Fedriani S/N, 41009 Seville, Spain *Correspondence: [email protected]; Tel.: +34-954-551-771 Abstract: Poor dietary practices are commonly reported in working populations from different economic sectors, resulting in increased absenteeism and a decrease in productivity. The aims of this study were to describe the frequency of food consumption and diet quality in workers aged ≥ 16 years from 2006 to 2017 in Spain and to evaluate the factors associated with diet quality. A nationwide cross-sectional study was carried out among workers using data from the Spanish National Health Surveys in 2006 (n= 11,068), 2011 (n= 7497) and 2017 (n= 8890). Sociodemographic, occupational, and health-related variables were used as well as diet quality data. A multiple linear regression was performed to determine the characteristics related to overall diet quality. The percentage of workers who consumed vegetables, at most, once or twice per week decreased from 2006 to 2017 (p< 0.001). A lower diet quality score was related to the consumption of tobacco and alcohol and being aged ≥ 25 years old, while a higher diet quality score was linked to being a woman, having Spanish nationality, receiving optimal perceived social support, being physically active in one’s main occupation, doing leisure-time physical activity, and the type of contract. Keywords: diet; food consumption; occupational health; Spain; trends 1. Introduction Worldwide, having a poor diet is a major preventable cause of adverse health and many non-communicable diseases, such as diabetes, heart disease, stroke, and cancer [ 1 – 3 ]. Having an unhealthy diet is now generally considered a leading risk factor for disease and death [ 4 ]. Spain is a country with dietary recommendations based on the Mediterranean diet [ 5 ]. Nevertheless, the incidence of chronic conditions, such as cardiovascular, oncological, metabolic, and musculoskeletal diseases has risen in Spain over the last few decades, as has happened worldwide [ 6 ]. Although the Spanish dietary guidelines [ 7 ] promote healthy dietary patterns, a high percentage of individuals do not meet the existing dietary guidelines in terms of consumption of fruit and vegetables (94.6% consume less than the recommended five daily servings), cereals (84.2% consume less than four servings/day), whole grains (83.5% consume less than half a serving/day), and dairy (66.4% consume less than two servings/day) and consume higher quantities of meat than recommended (59.0% of subjects have more than one serving/day) [ 8 ]. Nowadays, the current dietary pattern is moving away from the traditional Mediterranean diet pattern towards increased consumption of energy-dense food characterized by poor nutritional quality [9]. Diet is associated with individual, lifestyle, social, economic, and geographical factors, amongst others [ 10 – 13 ], which can produce a social gradient in diet quality that contributes to health inequalities [ 14 ]. For instance, diet quality is positively related to increased age, Nutrients 2021,13, 522. https://doi.org/10.3390/nu13020522 https://www.mdpi.com/journal/nutrients Nutrients 2021,13, 522 2 of 16 higher educational level, engaging in physical activity, multivitamin use, and not living alone, and is negatively related to obesity and smoking [15–18]. Similar to the general population, risk factors for chronic conditions are prevalent in working individuals, specifically among those with low-income occupations [ 19 ]. Around the world, poor dietary practices have been reported in employed people from different economic sectors [ 20 – 22 ], resulting in absenteeism and productivity loss, which generate substantial costs for societies and employers [ 23 , 24 ]. To date, a number of workplace conditions, including work stress, long working hours, shift work, and the unavailability of foods to allow a healthy diet, have been reported as contributing factors to poor dietary practices [ 25 , 26 ]. However, the findings are varied and have been reported inconsistently, which may be explained, in part, by the limitations related to synthesizing both qualitative and quantitative studies. Given that a large number of adults spend lengthy periods at work, workplaces play an important role in providing an opportunity to inform workers about healthy eating [19]. To the best of our knowledge, this study is the first research study conducted in a large, representative working population that shows the evolution over time of eating habits and diet quality and also analyzes sociodemographic, occupational, and health-related factors as a prior step to developing and implementing sound healthy diet-promoting policies and practices in workplaces. This could play a key role in achieving the aims stated in Sustainable Development Goal 3.4 to reduce premature mortality from non-communicable diseases by one-third by 2030 [ 27 ]. The aims of the current study were to describe the frequency of food consumption and diet quality in workers aged ≥ 16 years from 2006 to 2017 in Spain and to evaluate the factors associated with diet quality. 2. Materials and Methods 2.1. Study Design, Data Source and Study Population A nationwide, cross-sectional study was carried out. The sources of information were the Spanish National Health Survey (SNHS) 2006 [ 28 ], the SNHS 2011 [ 29 ], and the SNHS 2017 [ 30 ], all conducted by the Ministry of Health, Consumer Affairs and Social Welfare (MHCASW) and the National Institute of Statistics (NIS). The sampling framework consisted of non-institutionalized Spanish individuals. It used a three-stage sampling design. The first stage units were the census section, the second stage units were the households, and the third stage units were the individuals. The participants were informed about the survey through an informative letter from the MHCASW, detailing the purposes of the survey, the voluntary and anonymous nature of participation, and visit of a duly authorized interviewer. The study population was restricted to individuals aged 16 to 64 years old who reported being in employment at the time that the surveys were conducted. The sample originally consisted of 31,501 subjects (SNSH 2006: n= 13,064; SNHS 2011: n= 8653; SNHS 2017: n= 9784). We excluded 4046 (12.84%) individuals who did not respond or refused to answer the interview questions (SNHS 2006: n= 1996; SNHS 2011: n= 1156; SNHS 2017: n= 894 ). Those individuals excluded were not different from the rest of workers. The total sample included 27,455 workers: 11,068 in SNHS 2006; 7497 in SNHS 2011; and 8890 in SNHS 2017. 2.2. Variables Diet quality was the dependent variable, and it was examined through the Spanish Health Eating Index (SHEI) [ 31 ], which is a modified version of the North American Healthy Eating Index [ 32 ]. The SHEI is based on 10 food groups (cereals, vegetables, fruit, dairy, meat, legumes, cold meets, sweets, soft drinks and variety of the diet) classified into five categories (never or hardly ever, <1 time per week, 1–2 times per week, ≥ 3 times per week, but not daily, and daily), according to the frequency of food consumption proposed by the recommendations of the Spanish Society of Community Nutrition (SSCN) [ 33 ]. Each food group varies from 0 to 10 points, as reported by Supplementary Table S1. The total score for Nutrients 2021,13, 522 3 of 16 the SHEI was evaluated by summing the frequency of consumption of food groups. The higher the score, the higher the degree of compliance with the recommendations of SSCN. The overall SHEI score was categorized into three groups, taking into account the cut-offs previously described in the literature [ 32 ]: >80 points, good diet quality; 51–80 points, improvable diet quality; and <51 points, poor diet quality. The independent variables were as follows: Sociodemographic characteristics: survey year (2006, 2011, 2017); gender (woman, man); age (16–24 years, 25–44 years, 45–64 years); educational level (without studies, primary, secondary or professional training, university); marital status (single, married, widowed, separated/divorced); citizenship (Spanish, foreigner); and town size (<10,000 inhabitants, 10,000–100,000 inhabitants, >100,000 inhabitants). Health-related factors: self-reported height and weight, which were used to calculate body mass index (BMI), which was classified as underweight (BMI below 18.50 kg/m 2 ), normal weight (BMI of 18.50 to 24.99 kg/m 2 ), overweight (BMI of 25.00 to 29.99 kg/m 2 ), or obese (BMI of 30 kg/m 2 or higher) [ 34 ], and common mental disorders, which were evaluated using the General Health Questionnaire (GHQ-12) [ 35 ], adapted to the Spanish language [ 36 , 37 ]. The questionnaire was composed of 12 items with four response options on a Likert scale from zero (more than usual) to three (much less than usual). The bimodal response scale was used. For this, if one of the first two options were answered, a score of zero was given and if one of the last two were answered, a score of one was given (0-0-1-1) [ 38 ]. The total score of the GHQ-12 was calculated by summing the 12 items (minimum score: 0 points and maximum: 12 points). The final score was categorized using a cut-off at ≥ 3 points as the absence of common mental disorders (<3 points) and presence of common mental disorders ( ≥ 3 points). Other variables in the study related to health were ongoing tobacco use (yes, no); alcohol consumption in the last 12 months (yes, no); physical activity in the main occupation (yes: walking, carrying some weight, frequently moving/carrying out tasks that require great physical effort, no: sitting down for most of the day/standing up for most of the day, without moving much or making an effort); and leisure-time physical activity (yes, no). Perceived personal social support: this was evaluated using the Duke-UNC-11 questionnaire [ 39 ], which was validated in Spain [ 40 , 41 ]. The questionnaire consists of 11 items evaluated using a scale of five possible responses, ranging from one (Much less than I would like) to five (As much as I would like). The total score is calculated as the sum of scores for all items. Social support varies from 11 to 55 points. The level of social support was categorized into optimal support (≥33 points) or sub-optimal support (≤32 points). Occupational variables: the company activity in which the adult works (primary economic activity, secondary economic activity, tertiary economic activity). This activity was categorized according to the National Classification of Economic Activities 1993 [ 42 ] and 2009 [ 43 ]. Other occupational variables used in the study were type of contract (without contract, self-employed, temporary contract, ongoing contract, civil servant or permanent contract); type of working day (regular: split working day and continuous morning shift; irregular: continuous afternoon shift, continuous night shift, shift work, reduced working day and irregular working day); work-related stress (evaluated by a Likert scale ranging from one “not at all stressful” to seven “very stressful”, categorizing as: none or low work-related stress (1 and 2 points), moderate work-related stress (3–5 points) and high work-related stress (6 and 7 points)); and job satisfaction (evaluated by a Likert-type response format ranging from one “not satisfactory” to seven “very satisfying”, grouped as no or low job satisfaction (1 and 2 points), moderate job satisfaction (3–5 points) and high job satisfaction (6 and 7)). 2.3. Procedure and Ethical Aspects Anonymized data, which are available to the public, were downloaded from the NIS and MHCASW websites [ 29 – 31 ]. According to Spanish laws, Ethics Committee approval Nutrients 2021,13, 522 4 of 16 is not necessary if secondary data are being used. The Supplementary File contains the research data. 2.4. Statistical Analysis The workers are described through the counts and percentages for categorial variables and arithmetic means and standard deviations (SDs) for quantitative variables. For the bivariate analysis, the 2-means Student’s t-test was employed, while for the analysis of three or more means, the ANOVA repeated means test was applied. Linear regression models were performed to recognize significant trends in the frequency of food consumption from 2006 to 2017 using the regression coefficient and the coefficient of determination (R 2 ). In addition, a multiple linear regression (forward stepwise selection) was made using diet quality as the result variable. Previously, variables relevant to the final model were selected from the univariate analysis (variables with p< 0.15 were removed from the model). All possible interactions were assessed. Variables with p> 0.05 were assessed as potential confounding factors and were considered as such if the percentage of change in the coefficient was >20%. To assess goodness of fit, the adjusted coefficient of determination (R 2 ), the F statistic and the normality of the residues were analyzed. Values of p< 0.05 were considered significant. For statistical analysis, the IBM SPSS Statistics version 25 (IBM Corp, Armonk, NY, USA), licensed to the University of Cordoba (Spain) was used. 3. Results 3.1. Sociodemographic, Occupational and Health-Related Variables A total of 27,455 workers constituted the total sample, including 47.54% (n= 13,052) women and 52.46% (n= 14,403) men with a mean age of 42.50 years (SD ± 10.15). Most of them were married (60.08%), lived in towns with >100,000 inhabitants (41.30%), had completed secondary-level study (57.29%), were Spanish (91.70%), were of normal weight (48.58%), were non-smokers at the time of study (66.22%), had consumed alcohol in the past twelve months (68.47%), were not physically active at work (77.27%) but were active during their leisure time (60.27%), had no common mental disorders (85.37%), and received normal social support (97.33%). Regarding occupational variables, most of the participants had a current contract (51.81%), had an irregular working day (59.94%), belonged to the tertiary sector of the economy (71.38%), had moderate levels of perceived stress (58.82%), and had a high level of job satisfaction (51.00%). 3.2. Food Consumption and Diet Quality Most of the participants reported daily consumption of bread or grains (85.69%), fresh fruit (63.41%), and milk, cheese, or yoghurt (86.19%). In addition, 30.88% consumed sweet foods daily (Figure 1). Regarding diet quality, this needed to be improved in 70.89% of the subjects. From 2006 to 2017, the percentage of workers who consumed leafy greens, salads, and vegetables, at most, once or twice per week decreased (2006: 23.70%, 2011: 17.28%, 2017: 11.43% (B = − 0.78, R 2 = 0.95, p< 0.001)). On the other hand, there was an increase in the consumption of milk, cheese, or yoghurt ≥ 3 times per week but not in daily consumption (2006: 4.18%, 2011: 7.08%, 2017: 9.79% (B = 10.15, R 2 = 0.99, p= 0.04)). In addition, there was an increase in workers who consumed soft drinks ≥ 3 times per week but not in daily consumption (2006: 9.59%, 2011: 9.94%, 2017: 10.48% (B = 1.53, R 2 = 0.98, p= 0.04)) ( Table 1 ). Nutrients 2021,13, 522 5 of 16 Nutrients 2021, 13, x FOR PEER REVIEW 5 of 16 Figure 1. Frequency of food consumption in workers aged 16–64 years (n = 27,455) (2006–2017). From 2006 to 2017, the percentage of workers who consumed leafy greens, salads, and vegetables, at most, once or twice per week decreased (2006: 23.70%, 2011: 17.28%, 2017: 11.43% (B = −0.78, R2 = 0.95, p < 0.001)). On the other hand, there was an increase in the consumption of milk, cheese, or yoghurt ≥3 times per week but not in daily consumption (2006: 4.18%, 2011: 7.08%, 2017: 9.79% (B = 10.15, R2 = 0.99, p = 0.04)). In addition, there was an increase in workers who consumed soft drinks ≥3 times per week but not in daily consumption (2006: 9.59%, 2011: 9.94%, 2017: 10.48% (B = 1.53, R2 = 0.98, p = 0.04)) (Table 1). Table 1. Frequency of food consumption in workers aged 16–64 years (N = 27,455) (2006–2017). Variables 2006 2011 2017 Β R2 p-Value n = 11,068 (%) n = 7497 (%) n = 8890 (%) Frequency of consumption of bread or grains Never or hardly ever/<1 time per week/1–2 times per week 751 (6.79%) 632 (8.43%) 631 (7.10%) 1.59 0.45 0.52 ≥3 times per week1 526 (4.75%) 629 (8.39%) 761 (8.56%) 6.68 0.74 0.34 Daily 9791 (88.46%) 6236 (83.18%) 7498 (84.34%) -8.01 0.49 0.5 Frequency of consumption of leafy greens, salads and vegetables Never or hardly ever/<1 time per week/1–2 times per week 2622 (23.70%) 1296 (17.28%) 1017 (11.43%) −0.78 0.95 <0.001 ≥3 times per week1 3806 (34.39%) 2611 (34.83%) 4217 (47.44%) 24.08 0.82 0.28 Daily 4640 (41.92%) 3590 (47.89%) 3656 (41.12%) -2.6 0.03 0.9 Frequency of fresh fruit (excluding juices) consumption Never or hardly ever/<1 time per week/1–2 times per week 2191 (19.80%) 1448 (19.31%) 1267 (14.25%) −1.07 0.83 0.27 ≥3 times per week1 1648 (14.89%) 1472 (19.63%) 2020 (22.72%) 13.99 0.97 0.11 Daily 7229 (65.31%) 4577 (61.05%) 56.03 (63.03%) -4.44 0.24 0.67 Figure 1. Frequency of food consumption in workers aged 16–64 years (n= 27,455) (2006–2017). Table 1. Frequency of food consumption in workers aged 16–64 years (N= 27,455) (2006–2017). Variables 2006 2011 2017 BR2p-Value n= 11,068 (%) n= 7497 (%) n= 8890 (%) Frequency of consumption of bread or grains Never or hardly ever/<1 time per week/1–2 times per week 751 (6.79%) 632 (8.43%) 631 (7.10%) 1.59 0.45 0.52 ≥3 times per week 1526 (4.75%) 629 (8.39%) 761 (8.56%) 6.68 0.74 0.34 Daily 9791 (88.46%) 6236 (83.18%) 7498 (84.34%) −8.01 0.49 0.5 Frequency of consumption of leafy greens, salads and vegetables Never or hardly ever/<1 time per week/1–2 times per week 2622 (23.70%) 1296 (17.28%) 1017 (11.43%) −0.78 0.95 <0.001 ≥3 times per week 13806 (34.39%) 2611 (34.83%) 4217 (47.44%) 24.08 0.82 0.28 Daily 4640 (41.92%) 3590 (47.89%) 3656 (41.12%) −2.6 0.03 0.9 Frequency of fresh fruit (excluding juices) consumption Never or hardly ever/<1 time per week/1–2 times per week 2191 (19.80%) 1448 (19.31%) 1267 (14.25%) −1.07 0.83 0.27 ≥3 times per week 11648 (14.89%) 1472 (19.63%) 2020 (22.72%) 13.99 0.97 0.11 Daily 7229 (65.31%) 4577 (61.05%) 56.03 (63.03%) −4.44 0.24 0.67 Frequency of consumption of dairy products (milk, cheese, yoghurt) Never or hardly ever/<1 time per week/1–2 times per week 596 (5.38%) 565 (7.54%) 766 (8.62%) 1.16 0.95 0.15 ≥3 times per week 1463 (4.18%) 531 (7.08%) 870 (9.79%) 10.15 0.99 0.04 Daily 10,009 (90.43%) 6401 (85.38%) 7254 (81.60%) − 16.89 0.98 0.09 Nutrients 2021,13, 522 6 of 16 Table 1. Cont. Variables 2006 2011 2017 BR2p-Value n= 11,068 (%) n= 7497 (%) n= 8890 (%) Frequency of meat (chicken, beef, pork, lamb, etc.) consumption Never or hardly ever/<1 time per week/1–2 times per week 2651 (23.95%) 2276 (30.36%) 2304 (25.92%) 0.36 0.52 0.49 ≥3 times per week 16317 (57.07%) 4444 (59.28%) 5682 (63.91%) 12.00 0.98 0.09 Daily 2100 (18.97%) 777 (10.36%) 904 (10.17%) − 15.71 0.72 0.35 Frequency of legume consumption Never or hardly ever/<1 time per week/1–2 times per week 8225 (74.31%) 5848 (78.00%) 6761 (76.05%) 1.27 0.18 0.72 ≥3 times per week 12500 (22.59%) 1563 (20.85%) 2069 (23.27%) 1.25 0.1 0.79 Daily 343 (3.10%) 86 (1.15%) 60 (0.67%) −4.37 0.86 0.25 Frequency of consumption of cold meats and cuts Never or hardly ever/<1 time per week/1–2 times per week 6240 (56.38%) 4523 (60.33%) 4354 (48.98%) -0.51 0.92 0.18 ≥3 times per week 12575 (23.27%) 1793 (23.92%) 3183 (35.80%) 23.19 0.83 0.27 Daily 2253 (20.36%) 1181 (15.75%) 1353 (15.22%) -9.31 0.78 0.31 Frequency of consumption of sweets (biscuits, pastries, jams, cereals with sugar, sweets, etc.) Never or hardly ever/<1 time per week/1–2 times per week 5730 (51.77%) 4141 (55.24%) 4493 (50.54%) −0.86 0.24 0.67 ≥3 times per week 11414 (12.78%) 1111 (14.82%) 2087 (23.48%) 19.71 0.92 0.19 Daily 3924 (35.45%) 2245 (29.95%) 2310 (25.98%) − 17.49 0.98 0.09 Frequency of consumption of soft drinks with sugar Never or hardly ever/<1 time per week/1–2 times per week 8125 (73.41%) 5835 (77.83%) 7110 (79.98%) 0.31 0.66 0.4 ≥3 times per week 11061 (9.59%) 745 (9.94%) 932 (10.48%) 1.53 0.98 0.04 Daily 1882 (17.00%) 917 (12.23%) 848 (9.54%) − 13.61 0.96 0.13 Diet quality Poor diet quality 520 (4.70%) 143 (1.91%) 204 (2.29%) −4.25 0.58 0.45 Improvable diet quality 8008 (72.35%) 4616 (61.57%) 6800 (76.49%) 8.27 0.1 0.79 Good diet quality 2540 (22.95%) 2738 (36.52%) 1886 (21.21%) −5.03 0.02 0.9 B = Estimated parameter; R2= Adjusted coefficient of determination; 1But not daily. 3.3. Association between Sociodemographic, Occupational, and Health-Related Variables and Diet Quality The bivariate analysis revealed no relationship between a poor diet quality and marital status, size of town of residence, BMI, common mental disorders, or work-related stress. In the adjusted multivariate analysis, workers who were over 25 years old, had a temporary contract, and consumed tobacco and alcohol had lower mean diet quality scores. In contrast, there were positive significant relationships between diet quality and being female, being Spanish, being physically active in one’s main occupation, doing physical activity during leisure time, perceiving optimal social support, being self-employed, and being a civil servant or having a permanent contract or being a non-contracted worker (Table 2). Nutrients 2021,13, 522 7 of 16 Table 2. Univariate and adjusted linear regression models predicting diet quality scores in workers aged 16–64 years (n= 27,455) (2006–2017). Variables Mean (SD) Univariate Analysis Adjusted Analysis Bβp-Value B βp-Value Gender Man 71.48 (11.02) Reference Reference Reference Reference Woman 74.57 (10.65) 3.09 0.14 <0.001 2.96 0.14 <0.001 Age group (years) 16–24 64.05 (12.21) Reference Reference Reference Reference 25–44 56.84 (10.94) −7.21 −0.33 <0.001 −6.27 −0.29 <0.001 45–64 52.02 (9.77) −12.03 −0.54 <0.001 −10.71 −0.48 <0.001 Educational level No education 72.82 (11.30) Reference Reference Primary 72.37 (11.32) −0.45 −0.01 0.35 Secondary or professional training 72.19 (11.15) −0.63 −0.03 0.16 University 74.86 (10.04) 2.04 0.08 <0.001 Marital status Single 70.24 (11.61) Reference Reference Married 74.07 (10.43) 3.83 0.17 0.07 Widowed 76.62 (9.89) 6.38 0.07 0.15 Separated/Divorced 73.91 (10.63) 3.67 0.09 0.23 Citizenship Foreigner 69.82 (11.15) Reference Reference Reference Reference Spanish 70.06 (10.89) 0.24 0.09 <0.001 1.56 0.04 <0.001 Town size <10,000 inhabitants 73.05 (10.69) Reference Reference 10,000–100,000 inhabitants 71.77 (11.00) −1.28 −0.11 0.09 >100,000 inhabitants 73.36 (11.05) 0.31 0.01 0.19 Body Mass Index Normal weight 69.35 (11.85) Reference Reference Underweight 65.71 (11.05) −3.64 −0.04 0.59 Overweight 69.28 (10.67) −0.07 −0.003 0.83 Obese 69.31 (11.15) −0.04 −0.001 0.38 Ongoing tobacco use No 74.18 (10.46) Reference Reference Reference Reference Yes 70.51 (11.48) −3.67 −0.16 <0.001 −2.62 −0.11 <0.001 Alcohol consumption in the last12 months No 73.56 (11.17) Reference Reference Reference Reference Yes 72.67 (10.84) −0.89 −0.04 <0.001 −5.17 −0.02 <0.001 Physical activity as main activity No 70.95 (11.26) Reference Reference Reference Reference Yes 73.54 (10.79) 2.59 0.09 <0.001 1.17 0.05 <0.001 Leisure-time physical activity No 71.23 (11.25) Reference Reference Reference Reference Yes 74.07 (10.60) 2.84 0.13 <0.001 2.39 0.11 <0.001 Perceived social support Sub-optimal social support 70.36 (11.94) Reference Reference Reference Reference Optimal social support 73.02 (10.91) 2.66 0.04 <0.001 2.03 0.03 <0.001 Nutrients 2021,13, 522 8 of 16 Table 2. Cont. Variables Mean (SD) Univariate Analysis Adjusted Analysis Bβp-Value B βp-Value Common mental disorders Absence of common mental disorders 72.71 (11.71) Reference Reference Presence of common mental disorders 72.99 (10.82) 0.28 0.009 0.14 Sector of economy Tertiary 73.59 (10.84) Reference Reference Secondary 72.15 (10.63) −1.44 −0.6 <0.001 Primary 74.55 (11.13) 0.96 0.04 <0.01 Type of contract Ongoing contract 72.90 (10.84) Reference Reference Reference Reference Without contract 76.84 (10.67) 3.94 0.06 <0.001 1.12 0.02 0.01 Self-employed 75.09 (9.80) 2.19 0.08 <0.001 0.76 0.03 <0.01 Temporary contract 67.13 (11.45) −5.77 −0.2 <0.001 −2.3 −0.08 <0.001 Civil servant or permanent contract 76.03 (12.10) 3.13 0.14 <0.001 1.35 0.06 <0.001 Type of working day Irregular 72.65 (11.08) Reference Reference Regular 73.38 (10.74) 0.73 0.03 <0.001 Work-related stress None or low 73.05 (11.27) Reference Reference Moderate 73.02 (10.79) −0.03 −0.001 0.87 High 72.70 (11.12) −0.35 −0.01 0.1 Job satisfaction None or low 71.40 (11.73) Reference Reference Moderate 72.36 (10.80) 0.96 0.04 0.36 High 73.59 (10.96) 2.19 0.1 <0.01 SD = Standard Deviation; B = Estimated parameter; β = Beta coefficient; Adjusted coefficient of determination (R 2 ) = 13.50%, F = 30.61, p= < 0.001. 4. Discussion The main results of this study suggest that the frequency of food consumption did not vary among workers from 2006 to 2017, except for the consumption of milk, cheese and yoghurt, leafy green, salads, vegetables, and soft drinks. Similarly, the overall quality of the workers’ diets did not vary over the years of the study. On the other hand, the quality of diet was related to gender, age, nationality, type of contract, consumption of tobacco and alcohol, physical activity, and social support. Spanish dietary recommendations are based on the traditional Mediterranean diet [ 44 ]. In our study, most of the respondents from the Spanish working population consumed bread (85.69%), fruit (63.41%), vegetables (43.29%), and milk (86.19%) daily. In addition, 61.64% of the workers in our study consumed legumes once or twice per week, while the frequency of consumption was lower for meat (23.49%). Although the Mediterranean dietary pattern is also dominant in countries such as Croatia and Portugal [ 45 , 46 ], a study carried out in workers from Croatia [ 47 ] showed higher bread consumption (91%) and lower consumption of fruit (51.20%) and vegetables (33.60%) than in our study. In Portugal, the frequency of consumption of fruit and vegetables was found to be higher (53% and 76%, respectively), but it was lower for milk (74.60%) and meat (17%) compared with the results of our study [ 48 ]. Paradoxically, and despite the fact that the Nordic diet has many similarities regarding the use of fruit and vegetables in the Mediterranean diet [ 49 ], Hemiö et al. [50] reported a lower intake of fruit and vegetables among Finnish workers (27.97% and 22.97%, respectively) than in the workers in the current study. While the Brazilian diet is based on the consumption of high-fat foods, salt, and sugars [ 51 ], the Nutrients 2021,13, 522 9 of 16 frequency of legume consumption in Brazilian workers was found to be similar to that found in our study (62.60%) [ 52 ]. In addition, the findings of the present study corroborate the results of a previous study, indicating that one-third of workers consumed sweet foods daily [ 53 ]. It is important to consider that a comparison between countries needs to be considered carefully, owing the use of different dietary guidelines and survey methods to assess the frequency of food consumption. In the current study, diet needed to be improved in 70.89% of the workers. Diet quality is determined by a multitude of individual, social, and organizational factors. A number of studies have reported several barriers to maintaining a healthy diet for workers including long working hours, shift work and high workloads, low staffing levels, moderate levels of self-efficacy, and a lack of nutritional knowledge [ 25 , 54 ]. In contrast, the availability of healthy food options, the enjoyment of healthy behaviors, and social support have been reported as facilitators of eating healthy [55]. The percentage of workers who consumed leafy greens, salads, and vegetables, at most, once or twice per week decreased from 2006 to 2017, and those who consumed milk, cheese, or yogurt ≥ 3 times per week, but not daily, increased in our study. Workplaces are potentially promising environments for health promotion, due to the fact that workers spend one-third of their time there [ 56 ]. In Spain, health promotion strategies in the workplace are a relatively recent phenomenon. The Spanish Strategy on Occupational Safety and Health (SSOSH) 2007–2012 [ 57 ] and 2015–2020 [ 58 ] aimed to promote the improvement of health and safety conditions at work through companies adopting health promotion plans, which include activities aimed at following a healthy diet, preventing a sedentary lifestyle or avoiding unsuitable work postures [ 57 , 58 ]. In contrast, in the present study, the percentage of workers who consumed soft drinks at least once per week—but not daily—increased from 2006 to 2017. A study conducted in Nepal reported that the main barriers to healthy eating, as informed by the working population, are the unavailability of healthy food, the lack of sufficient human resources to prepare more healthy food options, and the greater cost of healthy foods [59]. When analyzing the predictors of diet quality in our study, women were found to have an increased diet quality score compared with men. Similar results have been obtained in other studies [ 15 , 60 , 61 ], and this could be explained by the fact that women are more health-conscious than men, and they may be more worried about the quality of their foods as well as having different job roles or shorter working hours, which may allow female workers to have more control over dietary choices [ 60 , 62 , 63 ]. However, in another study, no associations were found between those variables [ 64 ]. Although older age has been related to higher diet quality [ 65 ], in the current study, older workers had lower mean diet quality scores. Possible reasons for this finding could be that appetite and sensory abilities diminish with age [ 66 ]. Moreover, medical conditions, the use of drugs, and dental changes, affecting mouth movement and chewing ability, usually increase with age and could affect eating behavior, food preparation, and food intake [67]. Our results also corroborated previous findings [ 14 , 68 ], indicating that, in the univariate analysis, workers with a higher level of education had an increased overall diet quality score compared with employees with no education. This is probably due to people with a higher level of education having greater knowledge about nutrition, better cooking skills, or the ability to follow prevention messages. Regarding nationality, we found that Spanish employees’ dietary quality was better than that of foreign workers. This finding is consistent with another study conducted in Canada [ 69 ]. Immigrant individuals generally show lower rates of healthy behavior than the majority of non-immigrant people. Factors contributing to suboptimal eating behavior among immigrant groups include abnormally high social, economic, and environmental vulnerability due to the impacts of acculturation, low health literacy, and other migration-specific effects on health [ 70 ]. Moreover, it has also been suggested that the community-based participatory research approach has led to sustained improvements in the dietary quality of the foreign population [70]. In line with our results, most studies have shown that social support is usually associated with better Nutrients 2021,13, 522 16 of 16 90. Kong, J.-S.; Min, K.-B.; Min, J.-Y. Temporary workers’ skipping of meals and eating alone in South Korea: The Korean National Health and Nutrition Examination Survey for 2013–2016. Int. J. Environ. Res. Public Health 2019,16, 2319. [CrossRef] 91. Vahid Shahidi, F.; De Moortel, D.; Muntaner, C.; Davis, O.; Siddiqi, A. Do flexicurity policies protect workers from the adverse health consequences of temporary employment? A cross-national comparative analysis. SSM. Popul. Health 2016 ,2, 674–682. [CrossRef] 92. Kwon, K.; Park, J.B.; Lee, K.-J.; Cho, Y.-S. Association between employment status and self-rated health: Korean working conditions survey. Ann. Occup. Environ. Med. 2016,28, 43. [CrossRef] [PubMed] 93. Quinlan, M. The Effects of Non-Standard Forms of Employment on Worker Health and Safety, 1st ed.; International Labour Office: Geneva, Switzerland, 2015. 94. Tanaka, R.; Tsuji, M.; Tsuchiya, T.; Kawamoto, T. Association between work-related factors and diet: A review of the literature. Workplace Health Saf. 2019,67, 137–145. [CrossRef] 95. Julià, M.; Belvis, F.; Vives, A.; Tarafa, G.; Benach, J. Informal employees in the European Union: Working conditions, employment precariousness and health. J. Public Health 2019,41, e141–e151. [CrossRef] 96. Instituto Nacional de Estadística (INE). Distribución Porcentual de Los Activos por Sector Económico y Provincia. Available online: https://www.ine.es/jaxiT3/Datos.htm?t=3994#!tabs-tabla (accessed on 20 November 2020). 97. Sorensen, G.; Stoddard, A.; Hammond, S.K.; Hebert, J.R.; Avrunin, J.S.; Ockene, J.K. Double jeopardy: Workplace hazards and behavioral risks for craftspersons and laborers. Am. J. Health Promot. 1996,10, 355–363. [CrossRef] 98. Qaim, M.; Sibhatu, K.T. On the Link between Production Diversity and Dietary Quality in Smallholder Farm Households. In Hidden Hunger: Strategies to Improve Nutrition Quality; Biesalski, H.K., Birner, R., Eds.; Karger Publishers: Basel, Switzerland, 2018; Volume 118, pp. 102–111. 99. Vieira Souza, R.; Aguiar Sarmento, R.; Carnevale de Almeida, J.; Canuto, R. The effect of shift work on eating habits: A systematic review. Scand. J. Work Environ. Health 2019,45, 7–21. [CrossRef] 100. Proper, K.I.; van de Langenberg, D.; Rodenburg, W.; Vermeulen, R.C.H.; van der Beek, A.J.; van Steeg, H.; van Kerkhof, L.W.M. The relationship between shift work and metabolic risk factors: A systematic review of longitudinal studies. Am. J. Prev. Med. 2016,50, e147–e157. [CrossRef] 101. Sun, M.; Feng, W.; Wang, F.; Li, P.; Li, Z.; Li, M.; Tse, G.; Vlaanderen, J.; Vermeulen, R.; Tse, L.A. Meta-analysis on shift work and risks of specific obesity types. Obes. Rev. 2018,19, 28–40. [CrossRef] 102. Okumus, B.; Chaulagain, S.; Giritlioglu, I. Examining the impacts of job stress and job satisfaction on hotel employees’ eating behavior. J. Hosp. Mark. Manag. 2019,28, 558–575. [CrossRef] 103. Grant, N.; Wardle, J.; Steptoe, A. The relationship between life satisfaction and health behavior: A cross-cultural analysis of young adults. Int. J. Behav. Med. 2009,16, 259–268. [CrossRef] [PubMed] 104. Bongers, P.; de Graaff, A.; Jansen, A. “Emotional” does not even start to cover it: Generalization of overeating in emotional eaters. Appetite 2016,96, 611–616. [CrossRef] [PubMed] 105. Cardi, V.; Leppanen, J.; Treasure, J. The effects of negative and positive mood induction on eating behaviour: A meta-analysis of laboratory studies in the healthy population and eating and weight disorders. Neurosci. Biobehav. Rev. 2015 ,57, 299–309. [CrossRef] 106. Hann, C.S.; Rock, C.L.; King, I.; Drewnowski, A. Validation of de Healthy Eating Index with use of plasma biomarkers in a clinical sample of women. Am. J. Clin. Nutr. 2001,74, 479–486. [CrossRef] 107. Weinstein, S.J.; Vogt, T.M.; Gerrior, S.A. Healthy Eating Index scores are associated with blood nutrient concentrations in the Third National Health and Nutrition Examination Survey. J. Am. Diet. Assoc. 2004,104, 576–584. [CrossRef]