Abstract
Background
Evidence on the association between shift workers’ dietary pattern, particularly in the Middle Eastern populations, remains limited. This study aimed to investigate dietary habits among male shift workers compared to non–shift workers in Iran.
Methods
A cross-sectional analysis was conducted using baseline data (2015–2017) of employed male workers enrolled in the Shahedieh cohort study in Yazd, Iran. Dietary intake was assessed using a validated food frequency questionnaire. Shift working status, demographic, socioeconomic, and lifestyle data were collected through structured interviews. Mean daily intake of processed and ultraprocessed food categories was compared between shift workers and non–shift workers. Models were adjusted for covariates. Associations were also analyzed according to the job categories and age.
Results
Among 3,158 participants, 530 (16.8%) were shift workers. Shift workers reported significantly higher intakes of high-fat dairy products (46.3 vs. 39.2 g/day), snacks (1.50 vs. 1.07 g/day), soft drinks (121 vs. 98.2 g/day), and total processed foods (212 vs. 180 g/day) than non–shift workers. Shift workers had lower pizza consumption and different meal frequency patterns compared to non–shift workers. The strong associations observed among driver shift-workers, especially for soft drinks and total processed food intake. Age modified the association of several processed and ultraprocessed food categories.
Conclusion
The findings suggest a less healthy dietary profile in shift workers than in non–shift workers in Iran. These findings highlight the need for targeted nutritional interventions and policies to support healthy eating behavior among shift workers.
Keywords: Circadian disruption, Iranian workers, Nutritional behavior, Occupational health, Ultraprocessed food intake
1. Introduction
Shift work is a common feature of modern employment, encompassing approximately 25% of the workforce in the developed and developing countries [1]. This nonstandard work schedule disrupts normal circadian rhythms [2,3], creating a mismatch between internal biological rhythms and external behavioral patterns. This mismatch extends to eating behavior and can alter the secretion of appetite-regulating hormones such as leptin and ghrelin [4], potentially increasing hunger and preference for energy-dense foods during nonstandard work hours. These behaviors have been linked to a range of adverse health outcomes [5,6]. Epidemiologic studies consistently show that shift workers are at an elevated risk for cardiometabolic disorders [5]. Shift workers have greater odds of metabolic syndrome [7,8]. Additionally, long-term shift work is associated with a 17% higher incidence of cardiovascular events and a 26% higher risk of coronary heart disease than nonshift work [9].
Beyond direct physiological effects, shift work can also adversely affect health behaviors [10,11], including diet and nutrition [5,12]. Irregular work hours often lead to irregular eating patterns [13,14]. Shift workers frequently skip traditional meals, eat at unconventional times (e.g., late at night), and rely more on readily available, calorie-dense foods [5,13,14]. A systematic review concluded that shift workers tend to consume more “unhealthy” foods, such as those high in saturated fat and sugar, and to drink more sugar-sweetened beverages than day workers, partly due to limited healthy options during night shifts and disrupted hunger cues [15]. Such dietary habit changes can describe some of the elevated risk of disease observed in shift workers [16].
Although several studies have examined the dietary habits of shift workers, findings remain somewhat inconsistent and context dependent. Some investigations report mixed results or no clear pattern. For example, while shift workers often eat more at night and consume more snacks, total daily energy intake does not consistently differ from that of day workers [15]. A cross-sectional study found no statistically significant differences in overall caloric intake or macronutrient distribution between shift workers and non–shift workers [17]. A systematic review highlighted that the majority of available studies on the topic had relatively high risk of bias and often did not adequately control for confounders [15].
Another gap in the literature is the under-representation of certain regions and populations. There are relatively sparse data from the Middle East and other low- and middle-income regions. For example, among the available systematic reviews examining eating behaviors and dietary intake in shift workers [15,18,19], few comparable studies have been conducted in Middle Eastern countries [[20], [21], [22]], and most of the existing evidence originates from North American and European populations. Given the documented link between circadian disruption and appetite regulation and the change in healthy eating behavior during night shifts, we hypothesized that shift workers would be associated with higher consumption of processed and ultraprocessed foods than non–shift workers. The present study aimed to explore the dietary habits regarding consumption of processed and ultraprocessed food items in shift workers using a large Iranian population dataset form the first phase of the Shahedieh cohort study (SCS, 2015–2020) in Yazd, Iran, a prospective population-based cohort of approximately 10,000 adults aged 35–70 years [23,24].
2. Methods
2.1. Study design and population
This cross-sectional analysis was conducted using data from the first phase of the SCS (2015–2017), part of the larger PERSIAN Cohort initiative in Iran [23,24]. The SCS is a prospective, population-based study designed to investigate noncommunicable diseases and their risk factors among residents aged 35 to 70 years in Yazd, Iran. Data collection began in May 2015 and continued until October 2021. The cohort enrolled participants from a broad range of occupations and industries (e.g., healthcare, manufacturing, and service sectors). Inclusion criteria for the SCS were (1) age 35–70 years; (2) residence in Shahedieh for at least one year; and (3) Iranian nationality. From 11,314 eligible residents who were invited to the study, finally, 9978 accepted to participate in the study (88.2% participation rate). Although the original cohort comprised 8,291 individuals, we limited our analyses on (1) male sex and (2) participants with current employment. From 8,291 participants, 3705 had a job. Women were also excluded due to their very low representation among shift workers (9 of 547 employed women, 1.6%). Finally, for the present analysis, data from 3,158 male employed participants were included. The study has b protocol was approved by the Ethics Committee of the School of public health of Shahid Sadoughi University of Medical Sciences (IR.SSU.SPH.REC.1403.098).
2.2. Data collection and variables
Data were obtained from clinical and paraclinical evaluation and comprehensive questionnaires filled by face-to-face interviews, covering domains such as demographic characteristics, general health information, socioeconomic status, occupational history, work schedule, nutrition and dieatery habits, sleep, and other lifestyle factors. These data were originally collected through in-person interviews and systematically recorded in the cohort electronic database. Individual-level sociodemographic variables were age at enrollment (continuous), gender (men vs. women), education (illiterate or primary school education, secondary school education, or university degree), marital status (married vs. single/divorced/widow), and job-holding situation (employed vs. unemployed, including housewives or retired participants). Data on sleep habits were obtained using a modified and validated Persian version of the Pittsburgh Sleep Quality Index questionnaire. We calculated habitual sleep efficiency by dividing the sleep duration by the total time in bed (sleep latency + sleep duration) [25]. Data on several comorbidities including body mass index (BMI, in kg/m2), type 2 diabetes, and depression were collected. BMI was calculated by dividing weight in kilograms per squared of height in meters and categorized into two groups of underweight or normal weight (BMI ≤ 25 vs. overweight or obese; BMI > 25). Self-reported type 2 diabetes was determined based on a yes or no answer to the question “Do you have been medically diagnosed with type 2 diabetes?” Data on self-reported depression were collected based on a yes or no answer to the question “Do you have been medically diagnosed with depression?” The wealth index, as an individual-level socioeconomic status indicator, was derived by principal component analysis of infrastructure facilities, housing condition, possession of different assets, and education level [26].
Dietary intakes were assessed using the validated semiquantitative food frequency questionnaire (FFQ) developed for the PERSIAN Cohort Study [27]. This FFQ was designed to assess usual dietary intake over the one year preceding the interview, rather than short-term intake over a limited number of days. It includes approximately 113 standard food items (small variation in items according to regions to compensate regional dishes), grouped into major food categories, with additional locally consumed items. The FFQ was administered by trained nutritionists following a standardized protocol described elsewhere [[27], [28], [29]]. Participants were asked to report the frequency of consumption of each food item (on a daily, weekly, monthly, or yearly base) as well as the usual portion size consumed per occasion, based on predefined standard portion sizes. To enhance the accuracy of portion size estimation, interviewers used household measures (e.g., cups, plates, and spoons), food models, and a picture booklet illustrating standard portion sizes. Interviewers recorded all responses directly in to a study web platform. Then, reported frequencies were converted into daily intake values and multiplied by the corresponding portion-size weights to estimate average intake in grams per day for each food item. Food items were subsequently aggregated into food groups for analysis.
Nutrient and energy values for food items were derived primarily from the United States Department of Agriculture MyPlate food groups [28]. For Iranian foods not directly available in the United States Department of Agriculture categories, equivalent food items were identified based on their major ingredients and macronutrient composition or weighted averages of constituent ingredients were used, following standard procedures applied in the PERSIAN Cohort Study. Local food items were subsequently harmonized with standard FFQ food items prior to analysis. Dietary data processing and aggregation into food groups were performed according to the predefined PERSIAN Cohort food grouping scheme [28].
In this study, our analyses focused on the consumption of processed and ultraprocessed food categories, including processed meat products, margarine and solid fats, high-fat dairy products, refined grains (e.g., white bread and refined rice), pizza (used as a proxy for fast foods), salty snacks (chips, crackers, etc.), mayonnaise, desserts and sweets, condiments, and sugar-sweetened soft drinks. These food groups were selected because they are typically energy dense and nutrient poor and have been consistently linked to adverse metabolic and cardiometabolic outcomes in previous studies [30,31].
Data on dietary habits were obtained using a structured, interviewer-administered questionnaire designed to capture details on common food consumption patterns and eating behaviors. Participants were asked about the frequency of their consumption of specific food categories, including fried foods and grilled foods, and the use of salt and herbal drinks. In addition, they reported their usual daily meal frequency, categorized as fewer than three times per day, three times, four times, five to six times, or more than six times per day. Salt-related behaviors were assessed using a self-reported dietary habit questionnaire developed for the Iranian population. Addition of salt to prepared food was recorded using the response options no, sometimes, and yes. Food saltiness preference was assessed by asking participants to report their usual taste preference as not salty, normal, or salty, without further quantitative specification.
Shift workers were categorized into five occupational groups (drivers, healthcare workers, industrial workers, nonindustrial workers, and service workers) based on their self-reported job titles and work characteristics. Industrial workers were defined as those engaged in manufacturing, construction, or other production-related large industries. Nonindustrial workers include small and medium enterprise workers and administrative, clerical, and office-based employees whose work was primarily sedentary and did not involve industrial production processes. We used this classification to capture occupational contexts that may influence dietary behaviors among shift workers, accounting for variations in physical demands, work environment, food accessibility, and shift scheduling patterns across different job sectors.
2.3. Statistical analysis
The association between mean daily intake (grams/day) of ultraprocessed food groups was examined by shift work status. To further explore dietary patterns, participants were categorized into tertiles based on total ultraprocessed food consumption. Differences across these tertiles were analyzed to assess trends in association between demographic characteristics and ultraprocessed food group intake. To examine potential heterogeneity in the associations between shift work and processed and ultraprocessed food consumption across job categories, subgroup analyses were conducted by stratifying the study population according to the five predefined job groups. Linear regression models were fitted to evaluate the association between shift work and continuous ultraprocessed food group intake. Both crude models (including shift work status as the sole predictor) and adjusted models (model 2: adjusted for age, BMI, and socioeconomic status; model 3: additionally adjusted for marital status and sleep quality) were performed. Results from the models were reported as regression coefficients (β) with 95% confidence intervals (CIs) and p values. To assess effect modification, an interaction term between age at interview (continuous) and shift work status (yes/no) was included in each model. Regression coefficients (β) and 95% CIs were reported. The interaction models were adjusted for BMI, socioeconomic status, sleep quality, and family status. All analyses were conducted using R software (version 4.2.2, R Foundation for Statistical Computing, Vienna, Austria), utilizing packages such as tidyverse, broom, gtsummary, Table 1, and flextable for data management, analysis, and tabulation.
Table 1.
Demographic characteristics of working population enrolled in Shahedieh cohort study
| Covariate | Non–shift workers (n = 2628) | Shift workers (n = 530) | Overall (n = 3158) | p |
|---|---|---|---|---|
| Age at interview | ||||
| Mean (SD) | 46.2 (8.42) | 43.2 (7.13) | 45.7 (8.29) | p < 0.001 |
| Median [min, max] | 45.0 [35.0, 70.0] | 41.0 [35.0, 70.0] | 44.0 [35.0, 70.0] | |
| Education | ||||
| Illiterate/elementary | 158 (6.0%) | 20 (3.8%) | 178 (5.6%) | p < 0.001 |
| Guidance | 1115 (42.4%) | 235 (44.3%) | 1350 (42.7%) | |
| High school | 688 (26.2%) | 186 (35.1%) | 874 (27.7%) | |
| University degree | 544 (20.7%) | 81 (15.3%) | 625 (19.8%) | |
| Marital status | ||||
| Married | 2587 (98.4%) | 522 (98.5%) | 3109 (98.4%) | 1 |
| Other | 41 (1.6%) | 8 (1.5%) | 49 (1.6%) | |
| Smoking status | ||||
| No | 1472 (56.0%) | 270 (50.9%) | 1742 (55.2%) | 0.146 |
| Sometimes | 702 (26.7%) | 158 (29.8%) | 860 (27.2%) | |
| Regular | 392 (14.9%) | 84 (15.8%) | 476 (15.1%) | |
| Body mass index | ||||
| Mean (SD) | 27.3 (4.31) | 27.0 (4.44) | 27.2 (4.33) | 0.301 |
| Median [Min, Max] | 27.1 [15.5, 52.5] | 26.9 [15.2, 43.2] | 27.0 [15.2, 52.5] | |
| Wealth score | ||||
| Mean (SD) | 0.406 (0.684) | 0.269 (0.584) | 0.383 (0.670) | p < 0.001 |
| Median [min, max] | 0.389 [-1.94, 2.81] | 0.257 [-1.91, 1.94] | 0.330 [-1.94, 2.81] | |
| Depression | ||||
| No | 2403 (91.4%) | 479 (90.4%) | 2882 (91.3%) | 0.894 |
| Yes | 163 (6.2%) | 34 (6.4%) | 197 (6.2%) | |
| Type 2 diabetes | ||||
| No | 2280 (86.8%) | 469 (88.5%) | 2749 (87.0%) | 0.101 |
| Yes | 286 (10.9%) | 44 (8.3%) | 330 (10.4%) | |
| Hypertension | ||||
| No | 2231 (84.9%) | 468 (88.3%) | 2699 (85.5%) | 0.009 |
| Yes | 335 (12.7%) | 45 (8.5%) | 380 (12.0%) | |
| Habitual sleep deficiency | ||||
| Mean (SD) | 97.0 (6.62) | 97.1 (6.42) | 97.0 (6.59) | 0.716 |
| Median [min, max] | 100 [37.5, 100] | 100 [40.0, 100] | 100 [37.5, 100] | |
SD, standard deviation.
3. Results
A total of 3,158 male participants were included in the final analysis, of whom 530 (16.8%) were classified as shift workers and 2,628 (83.2%) as non–shift workers. The mean (standard deviation) age of shift workers was 43.2 (7.13) years, significantly lower than that of non–shift workers (46.2 (8.42) years; p < 0.001). The proportion of participants with a university degree was lower among shift workers than among non–shift workers (15.3% vs. 20.7%, p < 0.001). No significant differences were observed in smoking status or BMI between the two groups. However, shift workers had a lower mean wealth score than non–shift workers (0.269 vs. 0.406; p < 0.001) (Table 1). Those in the highest tertile of ultraprocessed food consumption had a lower mean age (43.9 [7.75] years), education, and smoking prevalance than those in the lowest tertile (Table S1).
Regarding dietary habits, shift workers had significantly higher mean intakes of high-fat dairy products (46.3 [42.1] vs. 39.2 [31.9] g/day; p < 0.001), snacks (1.50 [4.56] vs. 1.07 [3.53] g/day; p = 0.017), soft drinks (121 [180] vs. 98.2 (184) g/day; p = 0.010), and total processed foods (212 [227] vs. 180 [221] g/day; p = 0.003) than non–shift workers (Table 2). The correlation analyses showed significant positive associations among most processed food items. Pizza consumption was most strongly correlated with snacks (r = 0.39), followed by processed meat (r = 0.23), high-fat dairy products (r = 0.22), and mayonnaise (r = 0.18). Soft drinks showed the highest correlation with the overall processed food score (r = 0.91), followed by sweet desserts (r = 0.35) and high-fat dairy products (r = 0.29) (Figure S1).
Table 2.
Dietary habits according to shift working status in Shahedieh cohort study male shift workers
| Food items | Non–shift workers (n = 2566) | Shift workers (n = 513) | Overall (n = 3079) | p |
|---|---|---|---|---|
| Processed meat | ||||
| Mean (SD) | 1.20 (4.20) | 1.50 (4.71) | 1.25 (4.29) | 0.155 |
| Median [min, max] | 0 [0, 103] | 0 [0, 68.4] | 0 [0, 103] | |
| Margarine | ||||
| Mean (SD) | 0.0820 (0.807) | 0.0763 (0.598) | 0.0811 (0.776) | 0.88 |
| Median [min, max] | 0 [0, 25.0] | 0 [0, 8.55] | 0 [0, 25.0] | |
| High fat dairy products | ||||
| Mean (SD) | 39.2 (31.9) | 46.3 (42.1) | 40.3 (33.8) | p < 0.001 |
| Median [min, max] | 33.0 [0, 488] | 35.3 [0, 507] | 33.5 [0, 507] | |
| Refined grains | ||||
| Mean (SD) | 562 (395) | 582 (358) | 565 (389) | 0.282 |
| Median [min, max] | 474 [7.53, 8410] | 494 [51.9, 2210] | 478 [7.53, 8410] | |
| Pizza | ||||
| Mean (SD) | 7.50 (14.3) | 6.29 (8.91) | 7.30 (13.6) | 0.067 |
| Median [min, max] | 3.29 [0, 375] | 3.29 [0, 64.1] | 3.29 [0, 375] | |
| Snacks | ||||
| Mean (SD) | 1.07 (3.53) | 1.50 (4.56) | 1.14 (3.72) | 0.017 |
| Median [min, max] | 0 [0, 77.0] | 0.196 [0, 57.5] | 0.0497 [0, 77.0] | |
| Mayonnaise | ||||
| Mean (SD) | 1.79 (4.14) | 1.66 (3.27) | 1.77 (4.01) | 0.513 |
| Median [min, max] | 0.493 [0, 75.0] | 0.493 [0, 45.0] | 0.493 [0, 75.0] | |
| Sweet desserts | ||||
| Mean (SD) | 18.3 (19.5) | 19.9 (19.4) | 18.6 (19.5) | 0.191 |
| Median [min, max] | 13.1 [0, 238] | 13.3 [0.844, 131] | 13.1 [0, 238] | |
| Condiment | ||||
| Mean (SD) | 5.41 (6.86) | 5.24 (5.99) | 5.38 (6.73) | 0.604 |
| Median [min, max] | 3.38 [0, 108] | 3.35 [0, 60.8] | 3.38 [0, 108] | |
| Soft drink | ||||
| Mean (SD) | 98.2 (184) | 121 (180) | 102 (184) | 0.010 |
| Median [min, max] | 37.8 [0, 2760] | 60.5 [0, 1910] | 40.3 [0, 2760] | |
| Sum of processed foods | ||||
| Mean (SD) | 180 (221) | 212 (227) | 185 (223) | 0.003 |
| Median [min, max] | 114 [0, 3060] | 154 [0, 2220] | 119 [0, 3060] | |
SD, standard deviation.
The proportion of participants who reported consuming food without added salt (“no salty”) was significantly lower among shift workers than among non–shift workers (25.5% vs. 32.4%, p = 0.011) (Table 3). Conversely, a higher proportion of shift workers reported using normal (62.6% vs. 58.4%) or salty (9.9% vs. 8.3%) food, though these differences were not statistically significant individually. Regarding fried food consumption, most participants in both groups reported consuming it one to three times per week (58.7% among shift workers and 56.3% among non–shift workers), and no significant overall difference was found (p = 0.231). In terms of salt use, a higher proportion of shift workers reported regular use of salt (22.2% vs. 17.7%, p = 0.040), while a lower proportion reported no salt use (65.7% vs. 70.2%). Finally, meal frequency differed significantly between groups (p = 0.018). A higher proportion of shift workers reported eating three main meals per day (19.7% vs. 15.7%), while a lower proportion consumed four meals per day (40.9% vs. 47.4%).
Table 3.
Dietary habits according to shift working status in Shahedieh cohort study among male shift workers∗
| Food habit | Non–shift workers (n = 2566) | Shift workers (n = 513) | Overall (n = 3079) | p |
|---|---|---|---|---|
| Food saltiness preference | ||||
| No salty | 831 (32.4%) | 131 (25.5%) | 962 (31.2%) | 0.011 |
| Normal | 1498 (58.4%) | 321 (62.6%) | 1819 (59.1%) | |
| Salty | 213 (8.3%) | 51 (9.9%) | 264 (8.6%) | |
| Use of herbal drinks | ||||
| No | 366 (14.3%) | 80 (15.6%) | 446 (14.5%) | 0.421 |
| Yes | 2176 (84.8%) | 423 (82.5%) | 2599 (84.4%) | |
| Fried food | ||||
| Never | 18 (0.7%) | 3 (0.6%) | 21 (0.7%) | 0.231 |
| Less than one time a month | 76 (3.0%) | 10 (1.9%) | 86 (2.8%) | |
| 1–3 times a month | 702 (27.4%) | 120 (23.4%) | 822 (26.7%) | |
| 1–3 times a week | 1444 (56.3%) | 301 (58.7%) | 1745 (56.7%) | |
| Daily | 302 (11.8%) | 69 (13.5%) | 371 (12.0%) | |
| Grilled food | ||||
| Never | 35 (1.4%) | 4 (0.8%) | 39 (1.3%) | 0.153 |
| Less than one time a month | 649 (25.3%) | 127 (24.8%) | 776 (25.2%) | |
| 1–3 times a month | 1329 (51.8%) | 262 (51.1%) | 1591 (51.7%) | |
| 1–3 times a week | 528 (20.6%) | 108 (21.1%) | 636 (20.7%) | |
| Daily | 1 (0.0%) | 2 (0.4%) | 3 (0.1%) | |
| Addition of salt to prepared meals | ||||
| Yes | 454 (17.7%) | 114 (22.2%) | 568 (18.4%) | 0.040 |
| Sometimes | 286 (11.1%) | 52 (10.1%) | 338 (11.0%) | |
| No | 1802 (70.2%) | 337 (65.7%) | 2139 (69.5%) | |
| Eating intervals | ||||
| Three times (breakfast, lunch, dinner) | 402 (15.7%) | 101 (19.7%) | 503 (16.3%) | 0.018 |
| Four times (breakfast, lunch, dinner, one middle time) | 1217 (47.4%) | 210 (40.9%) | 1427 (46.3%) | |
| Five-six times (breakfast, lunch, dinner, two-three middle time) | 760 (29.6%) | 155 (30.2%) | 915 (29.7%) | |
| More than 6 times | 83 (3.2%) | 13 (2.5%) | 96 (3.1%) | |
| Less than three | 80 (3.1%) | 24 (4.7%) | 104 (3.4%) | |
Missing observations are not reported.
Linear regression models showed that shift work was significantly associated with higher consumption of high-fat dairy products (adjusted β: 5.21; 95% CI: 1.97, 8.46; p = 0.002), pizza (adjusted β: −2.20; 95% CI: −3.49, −0.90; p = 0.001), soft drinks (adjusted β: 18.31; 95% CI: 0.55, 36.06; p = 0.043), and total processed foods (adjusted β: 24.01; 95% CI: 2.63, 45.38; p = 0.028). No significant association was observed for other food items after adjustment (Table 4). Additional adjustment of models with marital status and sleep quality (based on habitual sleep efficiency) did not change the estimates.
Table 4.
Results of multiple linear regression for association between shift working and use of processed food items
| Processed food item | Model 1 β (95% CI) |
p model 1 | Model 2 β (95% CI) |
p model 2 | Model 3 β (95% CI) |
p model 3 |
|---|---|---|---|---|---|---|
| Condiment | −0.171 (−0.816, 0.475) | 0.604 | −0.155 (−0.812, 0.501) | 0.643 | −0.168 (−0.819, 0.484) | 0.614 |
| High fat dairy products | 7.083 (3.848, 10.318) | 0.000 | 5.214 (1.969, 8.459) | 0.002 | 5.121 (1.899, 8.342) | 0.002 |
| Margarine | −0.006 (−0.08, 0.069) | 0.880 | −0.031 (−0.106, 0.045) | 0.426 | −0.032 (−0.107, 0.042) | 0.397 |
| Mayonnaise | −0.128 (−0.512, 0.256) | 0.513 | −0.315 (−0.701, 0.072) | 0.110 | −0.311 (−0.697, 0.075) | 0.114 |
| Pizza | −1.214 (−2.512, 0.085) | 0.067 | −2.195 (−3.493, −0.897) | 0.001 | −2.155 (−3.454, −0.857) | 0.001 |
| Processed meat | 0.297 (−0.113, 0.707) | 0.155 | 0.089 (−0.326, 0.504) | 0.675 | 0.085 (−0.327, 0.497) | 0.686 |
| Refined grains | 20.465 (−16.816, 57.745) | 0.282 | −8.527 (−46.081, 29.028) | 0.656 | −1.538 (−38.955, 35.88) | 0.936 |
| Snacks | 0.432 (0.076, 0.787) | 0.017 | 0.114 (−0.233, 0.46) | 0.520 | 0.163 (−0.181, 0.508) | 0.353 |
| Soft drink | 22.996 (5.39, 40.602) | 0.010 | 18.305 (0.548, 36.062) | 0.043 | 19.235 (1.508, 36.962) | 0.033 |
| Sweet desserts | 1.508 (−0.755, 3.772) | 0.191 | 1.112 (−1.176, 3.401) | 0.341 | 0.938 (−1.32, 3.197) | 0.415 |
| Sum of processed foods | 31.783 (10.708, 52.858) | 0.003 | 24.009 (2.634, 45.384) | 0.028 | 24.351 (3.005, 45.697) | 0.025 |
Model 1: Crude model.
Model 2: Adjusted for age, body mass index, and socioeconomic status.
Model 3: Model 2 + additionally adjusted for marital status and sleep quality.
CI, confidence interval.
A heterogeneity was observed in the associations between shift work and processed and ultraprocessed food consumption across occupational groups. For drivers, shift work was significantly associated with higher consumption of high-fat dairy products (β = 8.44; 95% CI: 2.67, 14.21) and soft drinks (β = 98.03; 95% CI: 64.97, 131.09), resulting in an elevated total processed and ultraprocessed food intake (β = 110.63; 95% CI: 71.10, 150.15). Healthcare workers showed significant positive associations with high-fat dairy products (β = 20.01; 95% CI: 3.36, 36.66) and mayonnaise (β = 4.23; 95% CI: 2.03, 6.43). For the industrial workers group, shift work was positively associated with high-fat dairy consumption (β = 4.92; 95% CI: 0.88, 8.96) but inversely associated with pizza consumption (β = −2.39; 95% CI: −4.15, −0.63), with no significant association observed for total processed and ultraprocessed food intake. Service workers (n= 30) and nonindustrial workers showed no statistically significant associations between shift work and any individual or total processed and ultraprocessed food item consumption (Table S2).
In multivariable-adjusted linear regression models, age was inversely associated with most food items intake. Significant modification by age was observed for high-fat dairy product consumption (p for interaction = 0.033), soft drink intake (p for interaction = 0.006), and total processed food intake (p for interaction = 0.013). No significant interactions were observed for the remaining dietary components (Table S3).
4. Discussion
In this study on Iranian male workers, those engaged in shift work showed different dietary intake and habits compared to non–shift workers. Male shift workers reported significantly higher consumption of high-fat dairy products, sugar-sweetened beverages (soft drinks), sweet snacks, and total ultraprocessed foods. These findings suggest that shift workers may rely more on energy-dense, processed items during their meals and breaks. However, we observed that pizza intake was lower in shift workers than in daytime workers. This contradictory result might reflect differences in meal patterns or food availability. The results were heterogeneous according to age and job groups.
Our findings align with a growing body of evidence indicating that shift workers tend to have poorer diet quality and different eating patterns compared to daytime workers. Several studies have reported that shift-working populations consume more energy-dense, high-fat foods and sugary drinks. A study of industrial employees in Bulgaria reported that shift workers were significantly more likely to consume fast foods, fried snacks (e.g., chips), processed meats, white bread, and margarine, whereas day workers ate more fruits and vegetables [32]. This shift-worker tendency toward convenient, high-fat choices and lower-produce intake appears to be a consistent theme across populations. These international findings reinforce that the dietary changes we observed in Iranian workers reflect a global occupational health challenge rather than a culture-specific phenomenon. A review by Gupta et al. (2019) examined the factors affecting shift workers’ eating behavior and concluded that work schedule is an important factor in food choices, with shift workers across multiple industries reporting increased reliance on convenient, energy-dense options [33]. Another systematic review and meta-analysis also found that rotational shift workers demonstrate consistently altered dietary patterns, contributing to elevated metabolic risk [34]. In a study on nurses, it has been shown that despite lower overall energy intake, nurses on evening and night shifts had significantly higher energy intake from snacking, which was associated with elevated saturated fat consumption [35]. This pattern of increased snacking has been consistently observed across diverse populations.
Prior research has shown that shift workers often consume more snacks and sugar-sweetened beverages to sustain energy during odd hours [36]. Wolska et al. (2022) found that Polish healthcare shift workers had nearly twice the odds of adhering to a “meat/fats/alcohol” dietary pattern (characterized by high fat and calorie content) and were less likely to follow a “pro-healthy” diet pattern, compared to daytime staff [37]. Their shift workers also had higher odds of exceeding 35% of energy from fat, underscoring the shift-associated preference for fatty foods [37]. In our study, increased intake of soft drinks and snacks among shift workers is in accordance with the reports of heightened snacking and sugary beverage consumption in off-hour workers [35].
It is worth mentioning that not all dietary differences point in the unhealthful direction; our study noted lower pizza intake in shift workers. While pizza is an ultraprocessed food, this result might reflect shift workers skipping traditional meal items or having limited access to certain foods during night shifts. Prior studies have shown that shift workers often skip formal meals (like family dinners) and instead graze on snacks at irregular times [15,35]. Some have even found shift workers eating fewer overall calories on night shifts due to disrupted routines [38]. In a French study of emergency personnel, night-shift workers consumed approximately 200 kcal less than day workers but still had poorer nutrient quality, suggesting that any reduction in meal foods may be offset by suboptimal snack choices [38]. Thus, our finding of lower pizza intake might indicate that shift workers forego certain large meals, potentially only to replace them with quick snacks.
The observed heterogeneity in associations between shift work and processed and ultraprocessed food consumption across occupational subgroups suggests that the relationship between shift work and dietary behavior is not uniform but rather depends on occupation-specific contextual factors. The strong associations observed among drivers, especially for soft drinks and total processed food intake, may reflect the unique demands of driving occupations, including prolonged sedentary periods, limited access to healthy food options during shifts, long-distance driving, and reliance on convenience foods and caffeinated beverages to maintain alertness during extended working hours. The elevated consumption of high-fat dairy products among healthcare workers, alongside their positive association with mayonnaise intake, may be attributable to the food environment characteristic of healthcare settings, where cafeterias and vending machines often provide readily accessible but nutritionally suboptimal options during irregular shift hours. Conversely, the absence of significant associations among service and nonindustrial workers, with point estimates suggesting even lower processed and ultraprocessed food consumption, indicates that certain occupational contexts may buffer against or not exacerbate the dietary disruptions typically attributed to shift work. These divergent patterns highlight the importance of considering occupational heterogeneity when examining the health consequences of shift work as pooled estimates may fade meaningful subgroup differences.
In our study, age modified the association between shift work and several processed food items. More specifically, the association between shift work and high-fat dairy consumption weakened with increasing age, whereas the associations with soft drink intake and total processed food consumption became stronger at older ages. Although age was generally inversely associated with most dietary intakes (including ultraprocessed foods) [39], these findings suggest that age should be considered not only as a confounder but also as an important modifier when evaluating the dietary impacts of shift work.
Several pathways could explain our findings. Circadian biology, sleep deprivation, workplace food environments, and psychosocial stressors all can push shift workers’ diets toward processed and ultraprocessed food options. Human appetite and metabolism are governed by the body’s internal clock, and working at night or rotating schedules can misalign hormonal patterns. Studies show that people undergoing circadian disruption experience reduced leptin and elevated ghrelin (the hormones of satiety and hunger, respectively), as well as impaired insulin sensitivity [[40], [41], [42]]. This hormonal milieu can increase cravings for calorie-dense foods and sweets at night. Experimental sleep restriction has been shown to boost snacking on high-sugar, high-fat foods [33]. Shift workers often suffer from chronic sleep debt and fatigue, which likely drives them to seek quick energy fixes via snacks and caffeinated sugary drinks. However, in our study, no significant difference was found in habitual sleep deficiency between shift and non–shift workers. This finding should be interpreted with caution. Available evidence indicates a bidirectional relationship between sleep quality and dietary behaviors [43], whereby insufficient or poor-quality sleep is associated with irregular eating patterns, increased snacking, greater intake of energy-dense and processed foods, and dysregulation of appetite-related hormones [44]. Conversely, dietary behaviors such as late-night eating and high consumption of ultraprocessed foods have been shown to negatively influence sleep quality [45]. Therefore, the lack of observed differences in habitual sleep deficiency in our study does not diminish the potential role of sleep as an important factor influencing dietary behaviors among shift workers. More detailed and longitudinal assessments of sleep habits are needed to better understand these complex inter-relationships.
Shift workers often have limited access to freshly prepared or healthier food, especially during overnight shifts. Cafeterias or restaurants may be closed at night, leaving vending machines and fast food as the convenient options [33]. It has been documented that night-shift staff members frequently rely on vending-machine snacks or bring shelf-stable processed foods from home due to lack of other options. Additionally, irregular or skipped meal times are common. For example, emergency healthcare workers on night shifts may go many hours without eating a proper meal [38]. When they do eat, it may be during a brief break, making grab-and-go processed foods more likely than balanced meals. Social factors can contribute as well: shift workers often lack regular family meal times, and peer influence during night shifts can normalize snacking. Some workers report that stress and time pressure lead them to “eat on the run” or indulge in comfort foods during shifts [46].
A key strength of this study is its large sample size, community-based design in a less presented non-Western setting. We used a detailed FFQ tailored to capture different food items including processed food consumption, which enabled us to document specific food-group differences (e.g., high-fat dairy, pizza, and snacks) rather than just macronutrient totals. While several previous studies on shift workers assessed diet more crudely or focused only on nutrients [15], our study by examining ultraprocessed food categories adds nuance to understanding shift work dietary patterns. Additionally, to our knowledge, this is among the very few reports from the Middle East region to quantify the impact of shift work on diet, which broadens the evidence base beyond Western and East Asian populations. Despite these strengths, several limitations need to be acknowledged. First, the analysis is cross-sectional, and we cannot infer causality or determine whether the poor dietary habits were a pre-existing trait or developed after commencing shift work. Second, diet was assessed by a self-report FFQ, which is prone to recall bias and under-reporting, especially for unhealthy foods. Salt-related variables were measured using simple self-reported categorical items without quantitative assessment; therefore, they should be interpreted as rough indicators of salt-related behavior rather than precise measures of salt intake. Third, our focus on male workers (and a relatively homogeneous population in a regional cohort) may limit generalizability of our findings. Dietary patterns in female shift workers or in other cultural contexts might be different [47]. Fourth, we did not have detailed data on shift schedules, and the definition of “shift worker” in our study did not distinguish between potentially diverse schedules together. Different shift types can have varied impacts on eating (e.g., rotating shifts might disturb routines more than permanent night shifts) [36]. Fifth, unmeasured confounders such as income or job stress could contribute to different results and need to be considered in future studies. Lastly, lack of detailed information on the sources and locations of food consumption among shift workers is another limitation. This limits our study ability to link dietary patterns to food availability during different shifts. We propose that future studies collect contextual data on eating locations, food sources, and workplace food environments to better understand barriers to healthy eating among shift workers.
5. Conclusion
Our study findings are in line with international evidence and support several recommendations consistent with global occupational health frameworks. Workplace canteens and vending facilities operating during nonstandard hours should be restructured to prioritize healthier options over ultraprocessed snacks and sugar-sweetened beverages. Additionally, workplace nutrition education programs should be integrated into occupational health services, with specific attention to the timing of food intake in relation to circadian rhythms. Shift schedules should allow adequate meal breaks as time constraints have been shown to drive reliance on convenient but less healthy food choices across diverse populations. Future occupational health policies should recognize poor diet as a workplace hazard comparable to other occupational exposures, warranting systematic intervention rather than individual-level advice alone. From a public health perspective, our findings suggest that dietary interventions targeting shift workers should be tailored to specific occupational groups rather than adopting a uniform approach, with particular attention warranted for drivers given their substantially elevated processed and ultraprocessed food and soft drink consumption. These recommendations are applicable not only to our study population but also to shift-working populations globally, given the consistency of dietary patterns observed across international studies. Longitudinal studies are needed to confirm causation and track how dietary habits change after transitioning into shift schedules. Moreover, investigating female shift workers and various industries would be valuable to see if similar processed-food patterns emerge; gender and occupational differences could inform tailored solutions.
CRediT authorship contribution statement
Akram Heydarizadeh: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Sayyed S. Khayyatzade: Writing – review & editing, Writing – original draft, Methodology, Conceptualization. Reyhane Sefidkar: Software, Methodology. Amir H. Mehrparvar: Writing – review & editing, Writing – original draft, Conceptualization. Mohammad J.Z. Sakhvidi: Writing – review & editing, Writing – original draft, Visualization, Supervision, Software, Project administration, Methodology, Formal analysis, Conceptualization.
Statement on the Use of AI Tools
The artificial intelligence tools (ChatGPT) have only been used for language editing in this manuscript.
Conflicts of interest
The study has been funded by Shahid Sadoughi University of Medical Sciences. The funder had no role in hypothesis generation, data pre-/processing, analysis and reporting. The authors declare no conflict of interest.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.shaw.2026.02.009.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
References
- 1.Reinganum M.I., Thomas J. StatPearls [Internet] StatPearls Publishing; 2024. Shift work hazards. [PubMed] [Google Scholar]
- 2.Li X, He Y, Wang D, Momeni MR. Chronobiological disruptions: unravelling the interplay of shift work, circadian rhythms, and vascular health in the context of stroke risk. Clin Exp Med. 2024;25(1):6. doi: 10.1007/s10238-024-01514-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Boivin D.B., Boudreau P., Kosmadopoulos A. Disturbance of the circadian system in shift work and its health impact. J Biol Rhythms. 2022;37(1):3–28. doi: 10.1177/07487304211064218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Crispim CA, Waterhouse J, Dâmaso AR, Zimberg IZ, Padilha HG, Oyama LM, Tufik S, de Mello MT. Hormonal appetite control is altered by shift work: a preliminary study. Metabolism. 2011;60(12):1726–1735. doi: 10.1016/j.metabol.2011.04.014. [DOI] [PubMed] [Google Scholar]
- 5.Hemmer A, Mareschal J, Dibner C, Pralong JA, Dorribo V, Perrig S, Genton L, Pichard C, Collet TH. The effects of shift work on cardio-metabolic diseases and eating patterns. Nutrients. 2021;13(11):4178. doi: 10.3390/nu13114178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wu QJ, Sun H, Wen ZY, Zhang M, Wang HY, He XH, Jiang YT, Zhao YH. Shift work and health outcomes: an umbrella review of systematic reviews and meta-analyses of epidemiological studies. J Clin Sleep Med. 2022;18(2):653–662. doi: 10.5664/jcsm.9642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Khosravipour M, Khanlari P, Khazaie S, Khosravipour H, Khazaie H. A systematic review and meta-analysis of the association between shift work and metabolic syndrome: the roles of sleep, gender, and type of shift work. Sleep Med Rev. 2021;57:101427. doi: 10.1016/j.smrv.2021.101427. [DOI] [PubMed] [Google Scholar]
- 8.Sooriyaarachchi P, Jayawardena R, Pavey T, King NA. Shift work and the risk for metabolic syndrome among healthcare workers: a systematic review and meta-analysis. Obesity Rev. 2022;23(10) doi: 10.1111/obr.13489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Torquati L, Mielke GI, Brown WJ, Kolbe-Alexander T. Shift work and the risk of cardiovascular disease. A systematic review and meta-analysis including dose–response relationship. Scand J Work Environ Health. 2018;44(3):229–238. doi: 10.5271/sjweh.3700. [DOI] [PubMed] [Google Scholar]
- 10.Bae MJ, Song YM, Shin JY, Choi BY, Keum JH, Lee EA. The association between shift work and health behavior: findings from the Korean National Health and Nutrition Examination Survey. Korean J Fam Med. 2017;38(2):86. doi: 10.4082/kjfm.2017.38.2.86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Proper KI, Jaarsma E, Robroek SJ, Schram JL, Boshuizen H, Picavet HS, Verschuren WM, van Oostrom SH. The mediating role of unhealthy behavior in the relationship between shift work and perceived health. BMC Public Health. 2021;21:1–9. doi: 10.1186/s12889-021-11350-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lowden A, Moreno C, Holmbäck U, Lennernäs M, Tucker P. Eating and shift work—effects on habits, metabolism, and performance. Scand J Work Environ Health. 2010:150–162. doi: 10.5271/sjweh.2898. [DOI] [PubMed] [Google Scholar]
- 13.Lin TT, Park CG, Kapella MC, Martyn-Nemeth P, Tussing-Humphreys L, Rospenda KM, Zenk SN. Shift work domains and their interactions with empty calorie food/beverage consumption: a 14-day intensive longitudinal study. Int J Nurs Stud. 2023;141:104490. doi: 10.1016/j.ijnurstu.2023.104490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Manoogian EN, Zadourian A, Lo HC, Gutierrez NR, Shoghi A, Rosander A, Pazargadi A, Ormiston CK, Wang X, Sui J, Hou Z. Feasibility of time-restricted eating and impacts on cardiometabolic health in 24-h shift workers: the Healthy Heroes randomized control trial. Cell Metab. 2022;34(10):1442–1456. e7. doi: 10.1016/j.cmet.2022.08.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Souza RV, Sarmento RA, de Almeida JC, Canuto R. The effect of shift work on eating habits: a systematic review. Scand J Work Environ Health. 2019;45(1):7–21. doi: 10.5271/sjweh.3759. [DOI] [PubMed] [Google Scholar]
- 16.Loef B, van Der Beek AJ, Hulsegge G, van Baarle D, Proper KI. The mediating role of sleep, physical activity, and diet in the association between shift work and respiratory infections. Scand J Work Environ Health. 2020;46(5):516. doi: 10.5271/sjweh.3896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mohit M., Mousavinezhad H., Mohsenpour M.A., Golestaneh M.A., Yousefinejad A. The relationship between dietary intake, obesity and shift working on employees of Pasargad Petrochemical Company in Asaluyeh. Iran. J Nutr Food Secur. 2023;8(2):276–282. [Google Scholar]
- 18.Pepłońska B., Nowak P., Trafalska E. The association between night shift work and nutrition patterns among nurses: a literature review. Medycyna Pracy. 2019;70 doi: 10.13075/mp.5893.00816. [DOI] [PubMed] [Google Scholar]
- 19.Phoi Y.Y., Keogh J.B. Dietary interventions for night shift workers: a literature review. Nutrients. 2019;11(10):2276. doi: 10.3390/nu11102276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Almajwal A.M. Stress, shift duty, and eating behavior among nurses in Central Saudi Arabia. Saudi Med J. 2016;37(2):191. doi: 10.15537/smj.2016.2.13060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hakim S.A., Eldin W.S., Mohsen A. Dietary behavior and its relation with lifestyle, rotating work shifts and job satisfaction among nurses of Ain Shams university hospitals. Egypt J Community Med. 2016;34(2):75–86. [Google Scholar]
- 22.Naghashpour M, Amani R, Nematpour S, Haghighizadeh MH. Dietary, anthropometric, biochemical and psychiatric indices in shift work nurses. Food Nutr Sci. 2013;4(12):1239–1246. [Google Scholar]
- 23.Eghtesad S, Mohammadi Z, Shayanrad A, Faramarzi E, Joukar F, Hamzeh B, Farjam M, Sakhvidi MJ, Miri-Monjar M, Moosazadeh M, Hakimi H. The PERSIAN cohort: providing the evidence needed for healthcare reform. Arch Iran Med. 2017;20(11):691–695. [PubMed] [Google Scholar]
- 24.Poustchi H., Eghtesad S., Kamangar F., Etemadi A., Keshtkar A.A., Hekmatdoost A., Mohammadi Z., Mahmoudi Z., Shayanrad A., Roozafzai F., Sheikh M. Prospective epidemiological research studies in Iran (the PERSIAN Cohort Study): rationale, objectives, and design. Am J Epidemiol. 2018;187(4):647–655. doi: 10.1093/aje/kwx314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Beck SL, Berger AM, Barsevick AM, Wong B, Stewart KA, Dudley WN. Sleep quality after initial chemotherapy for breast cancer. Support Care Cancer. 2010;18(6):679–689. doi: 10.1007/s00520-009-0662-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Cheraghian B, Hashemi SJ, Hosseini SA, Poustchi H, Rahimi Z, Sarvandian S, Malehi AS, Alipour M, Eghtesad S, Fatahiasl J, Bayat A. Cohort profile: the Hoveyzeh Cohort Study (HCS): a prospective population-based study on non-communicable diseases in an Arab community of Southwest Iran. Med J Islam Repub Iran. 2020;34:141. doi: 10.34171/mjiri.34.141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Eghtesad S, Hekmatdoost A, Faramarzi E, Homayounfar R, Sharafkhah M, Hakimi H, Dehghani A, Moosazadeh M, Mortazavi Z, Pasdar Y, Poustchi H. Validity and reproducibility of a food frequency questionnaire assessing food group intake in the PERSIAN Cohort Study. Front Nutr. 2023;10:1059870. doi: 10.3389/fnut.2023.1059870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Eghtesad S, Masoudi S, Sharafkhah M, Rashidkhani B, Esmaeili-Nadimi A, Najafi F, Faramarzi E, Homayounfar R, Ebrahimnejad P, Ansari-Moghaddam A, Mirjalili M. Validity and reproducibility of the PERSIAN Cohort food frequency questionnaire: assessment of major dietary patterns. Nutr J. 2024;23(1):35. doi: 10.1186/s12937-024-00938-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Eghtesad S, Sharafkhah M, Hekmatdoost A, Ostadrahimi A, Farjam M, Vakilian A, Pasdar Y, Kheradmand M, Shahraki-Sanavi F, Mehrparvar AH, Shayanrad A. Relative validity and reproducibility of a semi-quantitative food frequency questionnaire to assess nutrient intakes of PERSIAN cohort participants: comparisons to 24-hour dietary recalls and selected biomarkers. Arch Iran Med. 2025;28(9):485. doi: 10.34172/aim.34573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Silva Meneguelli T, Viana Hinkelmann J, Hermsdorff HH, Zulet MÁ, Martínez JA, Bressan J. Food consumption by degree of processing and cardiometabolic risk: a systematic review. Int J Food Sci Nutr. 2020;71(6):678–692. doi: 10.1080/09637486.2020.1725961. [DOI] [PubMed] [Google Scholar]
- 31.Juul F, Deierlein AL, Vaidean G, Quatromoni PA, Parekh N. Ultra-processed foods and cardiometabolic health outcomes: from evidence to practice. Curr Atheroscler Rep. 2022;24(11):849–860. doi: 10.1007/s11883-022-01061-3. [DOI] [PubMed] [Google Scholar]
- 32.Todorova MA, Yaneva AH, Bakova DR, Harizanova SN. Investigating the types of eating behavior among shift workers in the machine-building industry. Folia Medica. 2024;66(5):699–706. doi: 10.3897/folmed.66.e132829. [DOI] [PubMed] [Google Scholar]
- 33.Gupta CC, Coates AM, Dorrian J, Banks S. The factors influencing the eating behaviour of shiftworkers: what, when, where and why. Ind Health. 2019;57(4):419–453. doi: 10.2486/indhealth.2018-0147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Clark AB, Coates AM, Davidson ZE, Bonham MP. Dietary patterns under the influence of rotational shift work schedules: a systematic review and meta-analysis. Adv Nutr. 2023;14(2):295–316. doi: 10.1016/j.advnut.2023.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lin TT, Guo YL, Gordon CJ, Chen YC, Wu HC, Cayanan E, Ouyang CM, Shiao JS. Snacking among shiftwork nurses related to non-optimal dietary intake. J Adv Nurs. 2022;78(11):3629–3640. doi: 10.1111/jan.15253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Hulsegge G, Boer JM, Van Der Beek AJ, Verschuren WM, Sluijs I, Vermeulen R, Proper KI. Shift workers have a similar diet quality but higher energy intake than day workers. Scand J Work Environ Health. 2016:459–468. doi: 10.5271/sjweh.3593. [DOI] [PubMed] [Google Scholar]
- 37.Wolska A, Stasiewicz B, Kaźmierczak-Siedlecka K, Ziętek M, Solek-Pastuszka J, Drozd A, Palma J, Stachowska E. Unhealthy food choices among healthcare shift workers: a cross-sectional study. Nutrients. 2022;14(20):4327. doi: 10.3390/nu14204327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Bouillon-Minois J.B., Thivel D., Croizier C., Ajebo É., Cambier S., Boudet G., Adeyemi O.J., Ugbolue U.C., Bagheri R., Vallet G.T., Schmidt J. The negative impact of night shifts on diet in emergency healthcare workers. Nutrients. 2022;14(4):829. doi: 10.3390/nu14040829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Romeiro AC, Brito FD, Dos Santos DM, Da Costa ES, Curioni CC, Adegboye AR. Diet quality and consumption of ultra-processed foods according to age groups in Brazil: insights from the National Dietary Survey 2017–2018. Br J Nutr. 2025:1–12. doi: 10.1017/S000711452510411X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Oosterman J.E., Wopereis S., Kalsbeek A. The circadian clock, shift work, and tissue-specific insulin resistance. Endocrinology. 2020;161(12):bqaa180. doi: 10.1210/endocr/bqaa180. [DOI] [PubMed] [Google Scholar]
- 41.Schiavo-Cardozo D, Lima MM, Pareja JC, Geloneze B. Appetite-regulating hormones from the upper gut: disrupted control of xenin and ghrelin in night workers. Clin Endocrinol. 2013;79(6):807–811. doi: 10.1111/cen.12114. [DOI] [PubMed] [Google Scholar]
- 42.Söylemez S, Sivri AB, Şimşek E, Polat B, Çakır B. Melatonin, leptin, and ghrelin levels in nurses working night shifts. J Surg Med 2019;3(1):22–26.
- 43.Sun M, Scherffius A, Xu B, Zhang X, Li H, Wang X, Qiu Y, Chen QW, Wang D. Bidirectional associations between insomnia symptoms and eating disorders: a two-wave longitudinal study among Chinese college students. Eat Behav. 2025;56:101943. doi: 10.1016/j.eatbeh.2025.101943. [DOI] [PubMed] [Google Scholar]
- 44.Barreto EM, Almeida AT, Neta AC, Ferreira FE. Influence of sleep time on the consumption of ultra-processed foods in adolescents in a capital of Brazil: a longitudinal study. Nutrients. 2024;17(1):22. doi: 10.3390/nu17010022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Ekici EM, Çelik ÖM, Göbel P, Güzelalp AH. Association between ultra-processed food intake, night eating behavior, and sleep quality: a cross-sectional study from Türkiye. Health Qual Life Outcomes. 2025;23(1):103. doi: 10.1186/s12955-025-02415-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Gifkins J., Johnston A., Loudoun R. The impact of shift work on eating patterns and self-care strategies utilised by experienced and inexperienced nurses. Chronobiol Int. 2018;35(6):811–820. doi: 10.1080/07420528.2018.1466790. [DOI] [PubMed] [Google Scholar]
- 47.Hemiö K, Puttonen S, Viitasalo K, Härmä M, Peltonen M, Lindström J. Food and nutrient intake among workers with different shift systems. Occup Environ Med. 2015;72(7):513–520. doi: 10.1136/oemed-2014-102624. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.