Lifestyle factors, dietary patterns, and social determinants of social and eating jetlag: A cross-sectional survey

ABSTRACT Social jetlag (SJL) and, more recently, eating jetlag have been linked with an increased risk of non-communicable diseases. Here we aim to investigate lifestyle factors (diet, eating behavior, smoking, perceived stress, time spent sedentary/day) and social determinants (education level, employment status, and place of residence) associated with SJL corrected for sleep duration (SJLsc) and eating jetlag. Self-declared data on age, gender, lifestyle, and eating behavior were collected online from March 2021 to February 2022 of 432 adults. Principal component analysis was used to extract three dietary patterns (Prudent, Western, and Risky). Prevalence of SJLsc was 35.2%, with no significant difference between men and women (p = 0.558). Adults with SJLsc had significantly larger eating jetlag (56.0 min vs 41.2 min, p = 0.001). Increasing SJLsc duration was associated with an increased adherence to a Risky dietary pattern (standardized β coefficient = .165, p = 0.012); increasing eating jetlag duration was associated with an increased adherence to a Western dietary pattern (standardized β coefficient = .127, p = 0.039) and a shorter sleep duration (standardized βcoefficient = −0.147, p = 0.011). Among social determinants analyzed, only being a student or employed was associated with eating jetlag (standardized β coefficient = 0.125, p = 0.044), while none displayed any relationship with SJLsc. Our survey provides evidence on a risky behavior among young persons with SJLsc and eating jetlag, characterized by a higher alcohol consumption, and a diet rich in processed meat and high-fat food, eating during nights, and shorter sleep duration with potential long-term negative health outcomes.


Introduction
Humans along with almost all living organisms display circadian pattern of activity independent of environmental cues, and disruption of these rhythms has been associated with negative health outcomes; increased risk of obesity, diabetes, cardiovascular diseases, cancer, and psychiatric disorders has been described (Baron and Reid 2014).Social jetlag (SJL) is a measure of circadian misalignment resulting from the chronic mismatch of the sleep/wake schedule during working days and free days (Roenneberg et al. 2019).Irregularity of sleep/wake schedule between weekdays and weekends leads to a misalignment between lifestyle behaviors and endogenous circadian system, and circadian misalignment predicts subsequent metabolic dysfunction (Makarem et al. 2021).Epidemiological data have linked SJL to an increased risk of obesity and other chronic noncommunicable diseases (Koopman et al. 2017;Roenneberg et al. 2012;Rutters et al. 2014).Also, worse academic performance, increased verbal and physical aggressivity, and depressive symptoms have been described in those with SJL as compared to those without (Haraszti et al. 2014;Qu et al. 2023;Randler and Vollmer 2013;Wittmann et al. 2006).
Recently, the regularity of meal timing has been included among the determinants of cardiometabolic health (St-Onge et al. 2017).Eating jetlag is a novel concept, similar to SJL, proposed by Zerón-Rugerio et al. (2019a) as a marker of variability in meal timing and it is used to assess regularity of meal timing during the workdays and free days.Food intake is a cue that entrains the peripheral clocks with the master clock and eating at an unusual time may determine internal misalignment of the central and peripheral clocks with adverse metabolic outcomes (St-Onge et al. 2017;Zarrinpar et al. 2016;Zerón-Rugerio et al. 2021).The mechanisms involved are largely unknown, but they may be related to circadian dysregulation (St-Onge et al. 2017).Data on the health consequence of eating jetlag are limited and come from two epidemiological studies that reported its association with worse cardiometabolic health -higher body mass index (BMI), increased adiposity, glycemic dysregulation, and increased blood pressure (Makarem et al. 2021;Zerón-Rugerio et al. 2019a).
The importance of lifestyle in health and disease has been evaluated in many studies and indubitable proofs linking lifestyle factors with chronic noncommunicable diseases (obesity, diabetes, cardiovascular diseases, and cancer), disability, and death are now available.Due to this relationship, tackling each individual lifestyle risk factor is an important element that should be considered in the prevention of these diseases (Centers for Disease Control and Prevention 1996; Office of the Surgeon General 2004; Policy and action for cancer prevention 2009; World Health Report 2002).Social determinants have recently gained attention in relation to health, and they are considered nonmedical factors that affect health outcomes (Social Determinants of Health at CDC 2022).These are not individual lifestyle risk factors, but economic and social conditions that shape the health status of a group of people and include five domains: education access and quality, health care quality, neighborhood and built environment, social context, and economic stability (Social Determinants of Health at CDC 2022).To our knowledge, nothing is known yet on lifestyle factors and social determinants associated with eating jetlag, and limited data is available for SJL (a higher probability of being a smoker, a poor dietary pattern, heavy episodic drinking) (Almoosawi et al. 2018;Borisenkov et al. 2019;Haynie et al. 2018).Lifestyle factors could act synergistically to circadian misalignment increasing the risk of cardiometabolic diseases or, together with social determinants, could be just the expression of a social context and certain lifestyle promoting circadian misalignment.Understanding lifestyle factors and social determinants associated with both SJL and eating jetlag could help identify groups at risk for these forms of circadian misalignment and inform risk prevention interventions aiming to control them.
Thus, the objective of this survey was to investigate the lifestyle factors, dietary patterns, and social determinants associated with SJL and eating jetlag in a cohort of Romanian adults.

Study design and participants
This was a cross-sectional survey performed in the Department of Diabetes and Nutrition Diseases of the Iuliu Hatieganu University of Medicine and Pharmacy Cluj-Napoca, Romania, between March 2021 and February 2022.Data was collected using a questionnaire distributed online, and participants were recruited by face-to-face meetings, advertisement on social media and by emails.All individuals aged 18 years or older were invited to complete the online questionnaire.Pregnancy or breastfeeding, jet lag resulting from more than two time zones flights within two weeks before study inclusion, working on shifts and the refusal to participate, all were the exclusion criteria chosen to reduce cofounding variables (Baron and Reid 2014;Won 2015).
The study was approved by the Ethics Committees of the Iuliu Hatieganu University of Medicine and Pharmacy Cluj-Napoca (No 15/18.01.2021).Before completing the questionnaire, all participants were informed about the purpose of data collection, their rights according to the applicable data privacy regulation and that by filling out the forms they consent to collecting and processing of their personal data.

Data collection, social determinants of health and lifestyle factors
Using Google Forms (Google LLC) we gathered online self-declared information on age, social determinants, medical history (previous diagnosis of ischemic heart disease, stroke, hypertension, diabetes, other chronic diseases), weight and height, and lifestyle factors.
Using Healthy People conceptual framework (Social Determinants of Health at CDC 2022) we collected data on three social determinants of health: place of residence (urban or rural), employment status (homemaker, unemployed, unable to work, retired, employed, student), and education (the highest education level attained -secondary school, professional education, high school, university).As lifestyle factors we collected information on smoking habits (current smoker/ex-smoker/never smoker), self-perceived stress, eating behavior, dietary intake, physical activity during leisure time (yes/no) and time spent being sedentary (hours/day), and sleep habits in the previous month.BMI was calculated as weight (kg)/height (m) 2 and a BMI ≥30 kg/m 2 was defined as obesity (World Health Organization 2000).

Eating behavior and dietary intake assessment
Eating behavior was assessed by using questions on the number of meals/day, the largest meal of the day (breakfast, lunch or dinner), breakfast consumption frequency (daily, most days, rarely), eating after 9:00 pm (yes/no), waking up to eat during night (night eating) (yes/no), eating while watching TV/playing computer/reading (yes/no) (Roman et al. 2015), meal schedule during working and rest days (weekdays and weekends).
Based on data on meal schedules during weekdays and weekends, we computed meal timing variability and eating jetlag as previously described by Zerón-Rugerio et al. (2019a).The variability in meal timing (breakfast, lunch and dinner jetlag, min) was calculated separately for breakfast, lunch, and dinner as the difference between mealtime on weekends and weekdays.Eating jetlag was computed as the absolute difference between eating midpoint time during weekends and weekdays and eating midpoint time was calculated using the formula [Timing of the last meal-Timing of the first meal]/2) + Timing of the first meal.
Food and beverage intake in the last 12 months was evaluated by an adapted version of the food frequency questionnaire from the Nurses' Health Study Questionnaire, which was previously translated in Romanian language and validated (Roman et al. 2016).The questionnaire evaluated the frequency of consumption of 59 food and beverage items, such as bread, cereals, pasta, eggs, meat, fish, milk and dairy products, sweets, snacks, cooked food, legumes, vegetables, fruits, fried and fast-food, and type of fat used for cooking.For each of them, the portion size was provided, and the participants were asked to choose the frequency of consumption of that portion size; the frequency ranged between "never" scored as 1 to "6 or more times per day", scored as 9. Exceptions were represented by the items "fried food consumption" and "fast-food consumption" for which potential scores ranged between 1 (less than once/week) and 4 (daily).For this analysis, based on previous literature reports (Atkins et al. 2016), the 59 food and beverage items were aggregated into 32 mutually exclusive food groups with similar nutrient profiles.Each food group was assigned a score calculated by summing the scores of individual food items and missing data was not imputed (Atkins et al. 2016).The list of the 32 food groups is provided in Table S1.

Sleep and circadian parameters
Data on bedtime, sleep-onset latency, wake-up time, and overall perceived sleep duration was collected separately for weekdays and weekends using questions from the Munich Chronotype Questionnaire (MCQT) (Roenneberg et al. 2003).Based on the information collected we assessed the sleep onset time (bedtime plus sleep onset latency), SJL corrected for sleep duration (SJLsc), weighted sleep duration and mid-sleep time during weekdays (MSWsc) and weekends (MSFsc) both corrected for the sleep duration (Jankowski 2017).As sleep duration has been associated with dietary intake (Theorell-Haglöw et al. 2020), to limit the influence of this cofounding factor we chose to use for SJL calculation the formula proposed by Jankowski (2017) which allows sleep debt correction instead of the original SJL formula proposed by Wittmann et al. (2006).
Weighted sleep duration was calculated as the weighted average of self-reported sleep duration for weekdays and weekends: [(reported weekday sleep duration × 5) + (reported weekend sleep duration × 2)]/7 (Reutrakul et al. 2013).SJLsc was calculated as the absolute difference between MSFsc and MSWsc.Based on SJLsc duration, subjects were categorized as having SJLsc (those with an SJLsc ≥1 h) and not having SJLsc (those with SJLsc <1 h).

Statistical analysis
Before beginning the study, assuming a SJLsc prevalence of 27%, a desired level of confidence of 98% and a relative precision of 5%, we estimated that at least 427 participants fulfilling the inclusion criteria and not meeting any exclusion criteria, and without missing data on sleep habits required to calculate SJLsc (main study objective) would be required.For the analysis provided here, we included only participants with complete demographic and sleep data.
Statistical analysis was performed using SPSS version 26.0 (Armonk, NY: IBM Corp).The quantitative variables with normal distribution were presented as mean ±SD whereas the quantitative variables with a nonnormal distribution were presented as median (quartile 1; quartile 3).The qualitative variables were presented as the number of observations and frequency.Continuous variables were compared between groups using the Mann-Whitney U test.Qualitative variables were compared between groups by Chi-squared test.Time variables were compared using the related-samples Wilcoxon Signed Rank test.
Despite globalization, food availability and consumption are influenced by numerous factors such as cultural, economic, and social ones (da Costa et al. 2022).The worldwide homogenization of diet occurs heterogeneously with a higher preservation of traditional foods in some regions.Thus, dietary patterns may vary among different populations (da Costa et al. 2022).To overcome this, we chose to establish the dietary patterns specific to the young Romanian sample enrolled instead of using previously identified dietary patterns in other populations.The assessment of dietary patterns was performed using an a posteriori approach with the principal component analysis with orthogonal varimax rotation for dimension reduction.Based on the scree plot analysis, the eigenvalue > 1 and the interpretability (Fransen et al. 2014), we retained three principal components (factors) that represented identified dietary patterns labelled based on food groups with a factor loading ≥ 0.4 (Costello and Osborne 2005) and according to a previous populationbased study performed in Romania (Roman et al. 2019).Principal component factors scores were calculated for each participant by regression method with Bartlett correction and could display negative values (low consumption) or positive values (high consumption of food groups characteristic to that dietary pattern).
The association of lifestyle factors, dietary patterns, and social determinants of health with SJLsc and eating jetlag (both measured in minutes and included as dependent continuous variable) was assessed using the linear regression analysis.Each regression model was mutually adjusted for other factors, age, and gender.

Study population
The online questionnaire was completed by 506 persons.Of these, 30 persons had exclusion criteria (one without any data completed, five pregnant or lactating women, 23 which worked in shifts, one person who travelled between time zones within the 2 weeks prior to survey completion) and nine did not provide sleep onset latency data or provided unrealistic data (e.g.4-7 h or 480 min).Also based on timestamp and data provided were identified 35 possible duplicates which were also removed.

Dietary patterns
Based on principal component analysis with Varimax rotation and Kaiser normalization we identified three factors (three dietary patterns) with eigenvalue > 1 that could be interpreted and which explained 34.3% of the diet variance.The Kaiser Meyer Olkin measure was 0.791 showing good sampling adequacy.Bartlett's test of sphericity had a p-value < .001rejecting the null hypothesis that the variables are not correlated and thus unsuitable for factor analysis (Supplementary Figure S1).Dietary pattern identified by Factor 1 explained 17.1% of the variance and was characterized by high intake of vegetables, legumes, fruits, whole wheat pasta, whole rice, cheese, milk and milk products, nuts, eggs, dark and milk chocolate/other sweets, biscuits, and snacks.Based on the food groups identified this pattern was labelled as Prudent pattern.Dietary pattern identified by Factor 2 was labelled Western pattern, explained 10.8% of the diet variance and was characterized by a high intake processed meat, potatoes, fried potatoes, fried food, fastfood, white wheat bread, biscuits and snacks, milk chocolate/other sweets, and sugar sweetened beverages.Dietary pattern identified by Factor 3 explained 6.5% of the diet variance, was labelled Risky pattern and was characterized by a high intake of processed meat, fish, high fat food, and alcoholic beverages (wine, beer, and spirits).Factor loadings for individual food groups are provided in Supplementary Table 2.
By gender, men had higher scores for Western and Risky dietary patterns as compared to women.No difference between genders was observed for Prudent pattern (Table 2).

Social jetlag, meal timing variability and eating jetlag
Prevalence of SJLsc (absolute duration ≥1 h) in the studied sample was 35.2%, with no statistically significant differences between men and women (33.1% vs. 36.0%,p = 0.558).Median SJLsc and eating jetlag were 30.0 min in the whole sample analyzed.Statistically significant variability between weekdays and weekends eating times was observed for breakfast and dinner, similar in both men and women (60 min and 30 min, respectively, p < .001;Table 3).

Lifestyle factors, dietary patterns, and social determinants of SJLsc
The association of lifestyle factors, dietary patterns, and social determinants with SJLsc was assessed using linear regression analysis adjusted for all other risk factors and for age and gender, overall and separately for men and women.All models had a low multicollinearity (collinearity statistics tolerance > 0.700 for all).In the whole sample the only variable independently associated with SJLsc was the Risky dietary pattern (standardized β coefficient = 0.165, p = 0.012).By gender, in women, SJLsc remained positively associated with the Risky dietary pattern (standardized β coefficient = 0.228, p = 0.002) and snacking during day (standardized β coefficient = 0.146, p = 0.049).In men, feeling stressed was inversely correlated with SJLsc (standardized βcoefficient = −0.326,p = 0.014), while waking up during night for eating was positively associated with SJLsc (standardized β coefficient = 0.387, p = 0.004) (Table 4).

Lifestyle factors, dietary patterns, and social determinants of eating jetlag
The association of eating jetlag with lifestyle factors, dietary patterns, and social determinants was also assessed using linear regression analysis adjusted for all other risk factors and for age and gender, overall and separately for men and women.In the whole sample, eating jetlag was positively associated with employment status (standardized β coefficient = 0.125, p = 0.044), Prudent and Western dietary patterns (standardized β coefficient = 0.169, p = 0.005 and standardized β coefficient = 0.127, p = 0.039, respectively) and negatively associated with weighted sleep duration (standardized βcoefficient = −0.147,p = 0.011).By gender, in women, the employment status, Prudent dietary pattern, and weighted sleep duration remained associated with eating jetlag (p < 0.05 for all).In men, eating breakfast, dinner as the largest meal of the day and waking up during night for eating were positively associated with eating jetlag, while eating three meals/day was inversely correlated with eating jetlag (p < .05for all) (Table 5).

Discussion
In this survey performed online, we found that SJLsc was highly prevalent among young and middle-aged adults in Romania, being present in 35% of the participants, and it frequently coexisted with eating jetlag.Both SJLsc and eating jetlag had a median duration in our sample of 30 minutes.Using data from this survey, we were also able to identify factors and dietary patterns suggestive for an unhealthy lifestyle and risky behavior which were associated with both SJLsc and eating jetlag with potential long term negative health outcomes.Overall, increasing SJLsc duration was associated with an increased adherence to a Risky dietary pattern and increasing eating jetlag duration was associated with an increased adherence to a Prudent or a Western dietary pattern and a shorter sleep duration.Among social determinants analyzed, only being employed or being a student were associated with eating jetlag, while none displayed any relationship with SJLsc.
a shorter sleep duration in women.In men, a longer eating jetlag was associated with an unhealthy eating behavior which included rarely having three meals/ day, eating breakfast daily/most of the times, having dinner as the most important meal of a day and waking up during night for eating.The association of SJL with a poor diet, a lower adherence to Mediterranean diet or to healthy Nordic diet was previously reported in studies assessing diet in persons with SJL from Finnish and Spanish populations (Suikki et al. 2021;Zerón-Rugerio et al. 2019b).Several studies performed in young adults from Finland, Spain, UK, Brazil, and Japan reported a lower intake of fruits, vegetables, berries, whole grains, beans as well as a higher intake of added sugars (Al Khatib et al. 2022;Bermingham et al. 2022Bermingham et al. ,2023;;Silva et al. 2016;Suikki et al. 2021;Yoshizaki and Togo 2021;Zerón-Rugerio et al. 2019b).In our study, for the Romanian sample we did not observe a low intake of fruits or vegetables with the Risky dietary pattern described in those with SJL, however this dietary pattern was characterized by a high intake of processed meat, high-fat food, and alcoholic beverages (wine, beer, and spirits).Similar results have been previously reported in other populations.A higher intake of dietary fat in general, and specifically of saturated fat, and cholesterol in individuals with SJL were reported in patients with obesity-related chronic diseases enrolled in a study performed in Brazil (Mota et al. 2019).Furthermore, in a cross-sectional study including 282 young inhabitants from the Komi Republic, participants with SJL consumed more meat, fats, oils, and alcohol than those without SJL (Borisenkov et al. 2019).Increased alcohol consumption has also been reported in night shift workers.Nightshift is an extreme form of misalignment between the central and peripheral clocks.Sleep loss was one of the mechanisms incriminated in binge and heavy alcohol consumption observed in these workers as a coping strategy to psychosocial stress associated with nightshifts (Richter et al. 2021) and this assumption could also be true for individuals with SJL.Additional mechanisms involved in the association of a diet rich in more palatable foods with high fat content, added sugars, processed meat and with a higher alcohol intake in subjects with SJL remain largely unknown.However, recent studies point toward emotional eating, loss of control of eating (Vrabec et al. 2022), increased self-rated appetite for energy-dense foods (Polugrudov et al. 2017;Rusu et al. 2021) and the involvement of the hedonic system as suggested by a higher activation of brain regions associated with reward in persons with SJL (Nechifor et al. 2020).Also, social factors and limited time to source healthy foods have been incriminated (Gupta et al. 2019).
While the association between SJL and poor lifestyle and diet habits is confirmed in many studies including our reported data here, limited information on eating jetlag's consequences is yet available.However, two studies, one cross-sectional and one longitudinal, reported a positive association of eating jetlag with BMI, adiposity, worse glycemic control, and increased blood pressure (Makarem et al. 2021;Zerón-Rugerio et al. 2019).In healthy young adults, eating patterns differ between weekdays and weekends (Gill and Panda 2015) and the difference is larger in those with SJL as compared to those without (Bodur et al. 2021).We also observed in our participants a larger eating jetlag, breakfast, and dinner jetlag in those with SJLsc as compared to those without.A novel finding of our study was the association of eating jetlag with a shorter sleep duration and a Western dietary pattern, characterized by high intake processed meat, potatoes, fried potatoes, fried food, fast-food, white wheat bread, biscuits and snacks, and food items with added sugar.Increased adherence to a Western dietary pattern in Romanian population, especially among young persons, has been reported previously (Roman et al. 2019), as shown by the increase of fast-food consumption by up to 70% of the inhabitants in urban regions (Pocol et al. 2015).Adherence to a Western dietary pattern has been associated with a higher risk of developing obesity, cardiovascular diseases, and cancer (Bahadoran et al. 2016).Thus, adherence to such diet of those with eating jetlag may explain, in addition to misalignment between central and peripheral clocks, the increased BMI reported previously in those with eating jetlag (Makarem et al. 2021;Zerón-Rugerio et al. 2019a).We also observed an association of eating jetlag with a Prudent dietary pattern, characterized by healthy food choices, in our sample and especially in women.This association may be explained by the high educational attainment of our participants which has been linked to healthy dietary choices in previous studies (Galobardes et al. 2001;Pomerleau et al. 1997;Tian et al. 1996;Turrell and Kavanagh 2006) and which may override the potential negative influence of eating jetlag in a highly-educated group.The relationship of education with dietary choices is complex and may be mediated by the knowledge of positive and negative influences of diet on health and how this knowledge is used (Johansson et al. 1999;Liberatos et al. 1998;Winkleby et al. 1992).
Social determinants have not been previously studied in association with SJL or eating jetlag.To better describe populations affected by SJLsc and eating jetlag we looked at education, employment status, and place of residence as proxies for education access, income, access to nutritious foods opportunities and literacy skills thus covering three of the five domains of social determinants of health (Priority Areas -Healthy People 2030).Employment status was the only social determinant associated with one of our variables of interest -i.e.eating jetlag was more frequent among students and those employed than in the unemployed, stay at home or retired group.Furthermore, being a student or employed was associated with eating jetlag only in women and not in men.This may reflect the time scarcity due to additional time spent by women in home and childcare in addition to paid work during weekdays.
Although this survey provides as novelty an analysis of lifestyle, dietary patterns and social determinants associated with SJL and eating jetlag, it also has several limitations.It was conducted online, and no recruitment strategy was put in place.Thus, all those who wanted to complete the questionnaire, had internet access and the skills to use an electronic device were enrolled.This approach did not ensure the representativity of the sample for all Romanian population as mainly young adults, students or employed, with higher education and women were enrolled.The imbalanced sample is characteristic to the online surveys and is due to both a selection bias and an interest bias (Andrade 2020); individuals not familiar to digital communication means, and those with a lower educational attainment were more likely not approached.Also, it is possible that individuals interested in the subject of sleep and eating habits, with poorer lifestyle habits were more likely to participate (Andrade 2020).Therefore, our online survey, as all online surveys, has a limited representativity for the general population.
In conclusion, our survey provides evidence on a risky behavior among young persons with SJLsc and eating jetlag characterized by a high alcohol consumption, and an unhealthy diet with high intake of processed meat, fast-food, food rich in fats and sweets, eating during night, and with shorter sleep duration, with potential long term negative health outcomes.Whether these are determinants of obesity and other chronic non-communicable diseases reported in those with variability in sleep and meal timing remains to be confirmed in prospective studies with a proper sampling strategy and population-based methods that will overcome the limitations described above for the online surveys.Meanwhile, the evaluation of social determinants in this investigation showing the association of eating jetlag with being a student or employed may represent a starting point to develop populational interventions targeting specific groups aiming to promote healthy sleep, diet and eating behaviors and moderate alcohol consumption.

Disclosure statement
CD reports honoraria for lectures and other support from Sanofi, AstraZeneca, Eli Lilly, Medtronic, and Novo Nordisk.MP reports honoraria for lectures from Alfa Wassermann and Sun Wave Pharma.CB reports honoraria for lectures, advisory board, and other support and fees from AstraZeneca, Novo Nordisk, Bayer, Boehringer Ingelheim, Eli Lilly, Medtronic, and Sanofi.GR reports honoraria for lectures, advisory board, expert testimony, and other support from Eli Lilly, Novo Nordisk, AstraZeneca, Boehringer Ingelheim, Medtronic, and Sanofi.AR declares support from Sanofi.The other authors declare no conflict of interest in relation to this work.

Table 1 .
Description of study sample according to gender.
n/N (%) = number (percentage) of participants in each category; BMI = body mass index; h = hour.

Table 2 .
Factor scores for identified dietary patterns according to gender.

Table 3 .
Difference in sleep onset, wake-up and meal timing between weekdays and weekends.

Table 5 .
Standardized β coefficient (p-value)for the association of lifestyle factors, dietary patterns, and social determinants with eating jetlag.Each model was adjusted for all other risk factors and for age and gender.

Table 4 .
Standardized β coefficient (p-value)for the association of lifestyle factors, dietary patterns, and social determinants with SJLsc.Each model was adjusted for all other risk factors and for age and gender.