Abstract(resumo):

Our study aimed to explore factors influencing the timing of food intake, including physiological, behavioral, personal preferences, cultural, and environmental dynamics. Using data from 5,260 Brazilian adults collected through a virtual, exploratory, population-based survey (Sonar-Brazil/2023-24), Lasso regression identified the determinants of the timing of the first and last food intake, lunch, eating midpoint, and weekly breakfast skipping. Random Forest analysis was used to assess the feature importance for the eating midpoint. Chronotype emerged as the primary predictor across all outcomes: as chronotype shifted towards eveningness, the frequency of breakfast skipping increased and food intake timings delayed. While depression, obesity, evening screen time, smoking, caffeine consumption, having dinner as the largest meal, short sleep duration, and working remotely were associated with a later eating midpoint, age, female sex, exercising, evening alcohol consumption, diet quality, working in-office, and higher education were associated with an earlier eating midpoint. Insights into modifiable factors influencing eating timing can inform public health interventions aimed at promoting healthier eating habits aligned with the circadian system, tailored to diverse regional and individual contexts.

Keywords(palavra-chave):

meal timing, circadian rhythm, chrononutrition, food intake

Content(conteúdo):

1. Introduction
In recent decades, advancements in the field of chrononutrition have underscored the pivotal role of meal timing ('when to eat'), complementing established principles regarding dietary composition ('what to eat') and caloric intake ('how much to eat') for human metabolism1-9.
For instance, eating later in terms of both clock hour and relative circadian time, as well as consuming a larger proportion of daily calories later in the day rather than earlier, may disrupt the synchronization of the internal circadian clock 1-3,5,8. This disruption is associated with adverse metabolic outcomes, including obesity, diabetes, and cardiovascular diseases4,8-11.
In modern society, skipping breakfast and delaying the eating midpoint have become more common due to longer work hours, increased evening activities, and widespread digital device use. Additionally, circadian preferences shaped by chronotypes influence meal timings, along with cultural, behavioral, and physiological factors.4,7.
In Brazil, a country with diverse latitudes, a study observed a tendency for earlier eating times in low-latitude regions, though it did not explore the reasons for these regional differences12. Data from the same study also demonstrated positive associations between the timing of the first and last food intake, as well as the eating midpoint, and BMI13.
Similar associations were observed in the population-based virtual study Sonar-Brazil, designed to explore the chronobiological aspects of sleep and eating habits among Brazilian adults. Early findings revealed significant correlations: skipping breakfast and delaying lunch14, longer intervals between waking and the first food intake, a later eating midpoint15, timing of the largest meal 16, and dining closer to the sleep midpoint17, all linked to higher BMI values.
These findings, consistent with numerous studies globally, highlight the crucial role of meal timing in strategies to promote healthy eating and prevent or manage obesity and related chronic diseases3,4,7,18.
However, developing effective interventions requires understanding the biological, social, and particularly modifiable factors that influence food intake timings across various geographical, cultural, and economic contexts. This understanding is vital for assessing the long-term sustainability of adjusted eating schedules in natural and uncontrolled environments. While most studies have focused on Europe, North America, and Asia, our research aims to comprehensively investigate the determinants of food intake timing across Brazil, examining physiological, behavioral, personal, cultural, and environmental factors.
2. Methods
2.1. Participants and Methods
This cross-sectional study used data from the third cycle of the Sonar-Brazil Survey (2023-2024), conducted from May 2023 to May 2024. The survey is virtual, exploratory, and population-based, focusing on chronobiological aspects of sleep, food, and nutrition among Brazilian adults.
The inclusion criteria for Sonar-Brazil were: 1. Brazilian nationality, 2. Residence in Brazil, 3. Age between 18 and 65, 4. Not pregnant. The survey link was emailed to public and private higher education institutions registered with the Brazilian Ministry of Education, regional councils for nutritionists, dentists, and doctors; health units; media; sleep and nutrition congresses; and scientific events. It was also widely promoted on social media and websites, requesting sharing among eligible contacts. Dedicated social media pages were established to enhance the survey's visibility nationwide.
A total of 5,580 individuals completed the Sonar-Brazil questionnaire. Sixty responses were excluded: 39 did not meet the age criteria (under 18 or over 65), 12 had missing or inconsistent weight and height data, and 9 provided inconsistent data on daily routines. Thus, the final sample included 5,520 Brazilian adults. Night and shift workers were also excluded due to their distinct circadian rhythms and irregular sleep patterns (n=260). Consequently, our analysis included 5,260 Brazilian adults.
Brazil is divided into five macro-regions—North, Northeast, Center-West, Southeast, and South—each with distinct climate, economic, and cultural characteristics. The 2022 IBGE Demographic Census reported Brazil's total population as 207,750,291, with regional distributions as follows: North (8.6%), Northeast (26.7%), Southeast (42.0%), South (14.8%), and Midwest (7.9%). Our final sample (n=5,260) mirrors this demographic: North (9.1%), Northeast (27.5%), Southeast (41.2%), South (14.3%), and Midwest (7.9%).
2.2. Data Collection
Participants completed an online structured questionnaire after agreeing to the informed consent form. The questionnaire covered sociodemographic variables, lifestyle, and health factors, with a focus on sleep and dietary habits. Responses were automatically stored in spreadsheets compatible with Microsoft Office Excel and later exported to Stata 18 for analysis. All data collection procedures were conducted in accordance with the principles outlined in the Declaration of Helsinki and received approval from the Research Ethics Committee.
2.3. Dependent Variables: Eating timing and Breakfast Skipping
Using the Chrononutrition Profile Questionnaire, translated and validated in Brazilian Portuguese20,21, participants reported their daily habits over the past month, focusing on typical weekdays and weekends. They provided their usual wake times, bedtimes, and times for the lunch, first and last eating event. Responses were recorded in 30-minute intervals, and the questionnaire specified that the first and last eating events included any food or drink with calories.
From the collected responses, we evaluated the following chrononutrition behaviors as dependent variables: timing of the first eating event (the clock time of participants' first food and/or drink), lunch timing, timing of the last eating event (the clock time of participants' last food and/or drink), and the eating midpoint. The eating midpoint was calculated as follows: eating midpoint = [(last eating time-first eating time)/2]+first eating time.
To calculate the weekly average for each variable, we applied the following formula: [(5×weekdays value) + (2×weekend value)]/7. Participants were also asked to specify on which days of the week they typically have breakfast, and we assessed the weekly frequency of breakfast skipping as a dependent variable.
Each dependent variable was also categorized into tertiles.
2.4. Focal independent Variables
2.4.1. Physiological Factors
Physiological factors were assessed through inquiries about participants' age, sex, and whether they had been diagnosed with depression. Additionally, Body Mass Index (BMI) was calculated based on self-reported weight and height using the formula [weight (kg)/height (m)²]. Individuals with a BMI greater than 29.9 kg/m² were classified as having obesity22.
2.4.2. Behavioral and Personal Preferences
Participants reported their engagement in physical exercise, specifying the timing: morning (from 5am to 11am), afternoon (from 11am to 6pm), and evening (from 6pm to 10pm). Additionally, participants were queried about their smoking habits and screen time (cell phones, TV, computers, notebooks, and tablets) before bed.
Food consumption was assessed using a food frequency questionnaire that covered 17 food and beverage categories. Participants reported their typical consumption days from Monday to Sunday, specifying both the frequency and timing of daily consumption (5-11am, 12-5pm, 6-9pm, >9pm). Each food group received positive scores for healthy choices (e.g., fresh fruits, vegetables, dairy) and negative scores for unhealthy choices (e.g., sweets, fried snacks, sugary drinks). The resulting total score, ranging from 19 to 77, reflected diet quality and was categorized into tertiles: low (19-45), intermediate (46-54), and good (55-77).
Weekly alcohol consumption after 6pm and daily intake of caffeine sources (coffee, cola-based soft drinks, black tea, guarana powder) were also recorded. Participants identified their largest meal of the day as breakfast, lunch, dinner, none, or other, indicating the meal where they consumed the highest calorie intake.
Sleep duration (h) was calculated by subtracting wake time from bedtime, and weekly averages were computed using the formula [(5×weekdays value) + (2×weekend value)]/7. Participants were categorized as short sleepers (<7h/night) based on recommendations23.
We adopted the “midpoint of sleep on free days corrected for sleep extension on free days (MSFsc)” as an indicator of chronotype, which is proposed to clean the chronotype of the confounder sleep debt. For participants whose sleep duration on free days was longer than workdays, the midpoint was calculated as follows: [bedtime on free days + (sleep duration on free days/2)]. For participants whose sleep duration on free days was shorter than workdays, due to the sleep debt accumulated over the workweek, the corrected midpoint of sleep was applied, and calculated as follows: [bedtime on free days+(weekly average sleep duration/2)]. Participants were also classified into quartiles based on their MSFsc values. Values below the 25th percentile were categorized as morning chronotype, above the 75th as evening chronotype, and those between these percentiles as intermediate chronotype24.
2.4.3. Cultural and Environment Factors
Information on work setup (not working, remote, or in-person), educational attainment, marital status (single, married, stable union, widow), and city and state of residence was collected. Based on this data, participants were categorized into Brazil's five political and geographical regions: Northeast, North, Midwest, Southeast, and South.
2.5. Statistical Analysis
We assessed participant characteristics across tertiles of eating midpoint (Table 1) and differences in eating timing outcomes across Brazil's regions (Supplemental Table A.1) using ANOVA for continuous variables and the chi-square test for categorical variables. Significant differences among groups were further explored using Bonferroni correction.
To identify predictors of eating behaviors, we employed a series of Lasso regression analyses (Least Absolute Shrinkage and Selection Operator) for each dependent variable: timing of the first eating event, lunch, last eating event, eating midpoint, and weekly frequency of breakfast skipping (Table 2).
The hypothesized determinants analyzed were drawn from existing literature4,7: 1.Physiological: age (years), sex (male vs. female), depression (percent), obesity (percent). 2.Behavioral and Personal Preferences: morning (5am to 10am) and evening (6pm to 11pm) exercise (percent), evening (after 6pm) screen time before bed (minutes), tobacco smoking (percent), evening (after 6pm) alcohol consumption (days/week), caffeine intake (frequency/day), diet quality (score), largest meal being dinner/supper (percent), chronotype (MSFsc), short (<7hours/night) sleep duration (percent). 3.Cultural and Environment: work setup (not working as reference, remote vs. in-office work), education (graduate vs. not graduated), marital status (single/widowed vs. married/stable union) and Brazil’s macro-region (Northeast as reference).
Lasso regression was chosen for its dual capability of variable selection and regularization, crucial for mitigating overfitting risks, particularly in studies with numerous potential predictors. This method effectively reduces less influential variables' coefficients to zero, promoting a sparse model that retains only variables with significant coefficients. Within Lasso regression, we applied a cross-validation selection method to iteratively adjust coefficients, minimizing prediction error and optimizing model performance. The resulting models provided coefficients for selected variables, identifying the factors most strongly associated with each outcome by considering the collective impact of multiple predictors.
Additionally, to pinpoint the primary predictors of eating midpoint, we conducted a feature importance analysis using the Random Forest algorithm. Implementing the scikit-learn library, we constructed the model to evaluate the relative importance of each variable in predicting the outcome (Figure 1).
Based on the results of preliminary analyses, restricted cubic splines were employed to illustrate the shape of the association of age, chronotype, diet quality, and screen time with eating midpoint (Figure 2).

3. Results
A total of 5,260 adults (mean age 36 years, range 18 to 65; 70% women) participated in the study, with an average eating midpoint of 14.69 (±1.23) hours. Significant differences in participant characteristics were observed across tertiles of eating midpoint (Tertile 1: 13.45 ± 0.49, Tertile 2: 14.56 ± 0.27, and Tertile 3: 16.05 ± 0.90), as detailed in Table 1.
Regarding physiological factors, Tertile 3 (the latest group) was younger, with nearly half of the participants aged 18-29 years. This group also exhibited delayed timings for all eating events, as well as a higher frequency of breakfast skipping compared to Tertiles 1 and 2. Additionally, Tertile 3 showed a higher prevalence of obesity (38.32% vs. 31.07% in Tertile 1 and 30.61% in Tertile 2) and depression (43.28% vs. 26.47% in Tertile 1 and 30.25% in Tertile 2) (all p<0.001) (Table 1).
In terms of behavioral and personal preferences, participants in Tertile 3 had the lowest diet quality score, along with longer screen time before bedtime, higher daily caffeine intake, and later chronotype, wake time, and bedtime compared to those in Tertiles 1 and 2. Additionally, the highest percentages of short sleepers (39%), smokers (51%), and non-exercisers (42%) were found in Tertile 3. Weekly alcohol consumption was also notably higher in Tertiles 2 and 3 compared to Tertile 1 (all p<0.001) (Table 1).
Among all participants, lunch was the largest meal, followed by dinner. Moreover, more than half of those who reported dinner as their largest meal belonged to the latest eating midpoint group (56% in Tertile 3), while the majority of those who indicated breakfast as their largest meal were in the earliest group (45% in Tertile 1) (all p<0.001) (Table 1).
Work setup also showed significant differences: a higher proportion of the non-working group ate later, with 44% in Tertile 3; remote workers were mostly in Tertile 2 (36%); and in-office workers (37.15%) were in Tertile 1, reflecting earlier eating patterns (p<0.001) (Table 1).
A detailed comparison by Brazil's regions is available in Supplemental Table A.1, which highlights significant differences. The eating midpoint was statistically earlier in the Northeast (14.58) compared to the Southeast (14.73) and South (14.81). Participants in the North and Northeast started eating earlier (8.08 and 8.09, respectively), with statistically significant differences compared to the Southeast (8.27) and South (8.43). Lunch timing ranged from 12.45 in the Midwest to 12.68 in the Southeast, and the timing of the last food intake ranged from 20.96 in the Midwest to 21.33 in the North. Lunch consistently constituted the largest meal of the day, with dinner surpassing breakfast in all regions.
The determinants of eating timing and breakfast skipping frequency were identified using Lasso linear models. Table 2 presents the variables with non-zero coefficients, selected through cross-validation, highlighting their association with the outcomes and the direction of the relationship, as indicated by the sign of the beta (positive or negative).
Regarding the physiological variables, older age was associated with a lower frequency of breakfast skipping and earlier timings for first eating, last eating, and eating midpoint, while it was associated with a later lunch. Females were associated with a lower frequency of breakfast skipping, earlier timings for last eating and eating midpoint, but later first eating and lunch. Depression was positively associated with all outcomes, indicating a higher frequency of breakfast skipping and later eating events. Obesity was associated with a higher frequency of breakfast skipping, delayed first eating and eating midpoint, but earlier last eating (Table 2).
Among the behavioral and personal preference variables, later chronotype, smoking, and individuals who identified dinner as the largest meal were positively associated with all outcomes, indicating more frequent breakfast skipping and delays in all eating events. Morning exercise was associated with a lower frequency of breakfast skipping and earlier timings for all meals. Similarly, evening exercisers showed less frequent breakfast skipping and earlier first eating and eating midpoint, but were also associated with a delayed last eating event. By increasing screen time before bedtime, the frequency of breakfast skipping also increased, along with delays in both first eating and eating midpoint (Table 2).
Evening alcohol consumption was associated with a lower frequency of breakfast skipping and earlier eating timings, except for the last eating timing. Higher caffeine consumption was associated with less breakfast skipping and earlier first eating times, while being positively linked to delays in both the last eating time and eating midpoint. Better diet quality was associated to a reduced frequency of breakfast skipping and earlier timings for first eating, last eating, and eating midpoint, but showed no association with lunch timing. Short sleepers were associated with a higher frequency of breakfast skipping and an earlier first eating time, but also with delays in lunch, last eating, and eating midpoint (Table 2).
Finally, regarding cultural and environmental determinants, in addition to confirming the variations across regions previously observed, work setup - using the 'not working' group as a reference - was associated with a lower frequency of breakfast skipping for both remote and in-office workers. However, remote work was linked to delayed first eating, lunch time, and eating midpoint, while in-office work was associated with earlier first eating and eating midpoint. Being graduated was associated with a lower frequency of breakfast skipping, alongside earlier first eating, lunch, and eating midpoint. Being in a stable union was a determinant only for earlier first eating, with no association observed for other variables. Lastly, during the winter and autumn seasons, both the frequency and timing of breakfast skipping were reduced, while the timing of the last eating event and eating midpoint were delayed (Table 2).
We employed the Random Forest algorithm to assess the importance of variables predicting the eating midpoint. Figure 1 displays specific feature importance on the x-axis, with longer bars indicating greater significance. The y-axis labels categorize input names, highlighted in red for physiological features, purple for behavioral and personal preference features, and green for cultural and environmental features.
Chronotype emerged as the most influential feature (0.474), followed by age (0.089) and diet quality (0.088). Screen exposure before bed (0.060) ranked fourth, followed by daily caffeine consumption (0.051) and evening alcohol consumption (0.033). Region (0.032) and season (0.026) followed closely, alongside short sleep duration (0.020), work setup (0.019), and having dinner as the largest meal (0.015). All other feature importances were ? 0.01. Depression ranked lowest in importance on the list (Figure 1).
Based on these feature importance results, we utilized Restricted Cubic Splines (Figure 2) to visualize the dose-response relationships between age, chronotype, diet quality, and the eating midpoint. Our analysis indicates that the earliest eating midpoint occurs in the absence of screen time before bedtime.

4. Discussion
This groundbreaking Brazilian study explores the intricate interplay of physiological, behavioral, personal preferences, cultural, and environmental factors that influence eating timing. By thoroughly investigating these dimensions, our findings not only align with previous research but also reveal novel nuances in these relationships within the Brazilian context.
4.1. Physiological Factors
Age was identified as the second most influential factor for eating midpoint in the Random Forest analysis, showing a negative association with both breakfast skipping and eating timings. This may reflect both lifestyle changes and age-related shifts in circadian rhythms, with younger individuals, particularly young adults, exhibiting later chronotypes that naturally shift toward morningness as they age4.
Among modifiable physiological conditions, our study revealed that both depression and obesity are associated with delayed first food intake, a later eating midpoint, and increased frequency of breakfast skipping14,25. Similar findings were observed in population-based cohort studies, indicating that individuals with overweight are more prone to skip breakfast and consume more food in the evening4,26. Although the causal role of food intake timing on mental health among general population awaits further exploration, research suggests a potential link between late-night eating and increased anxiety and depressive symptoms. Thus, our findings may pave the way for future studies to clarify the direction of this association and inform interventions aimed at addressing these conditions27.
4.2. Behavioral and Personal Preference
Chronotype emerged as the most influential predictor of eating midpoint, as expected given its central role in regulating circadian rhythms and daily activities, including meal timing. Individuals with an evening chronotype tend to eat later than those with a morning chronotype, consistent with previous studies15,28-30. This natural alignment with endogenous circadian rhythms highlights the need for greater flexibility in work and study schedules to accommodate different chronotypes. Additionally, practices such as early morning exposure to natural light can help realign biological rhythms, promoting healthier sleep and eating patterns.
Following chronotype, diet quality, screen use before bed, daily caffeine consumption, evening alcohol consumption, short sleep duration, and having dinner as the largest meal emerged in descending order as the most influential behavioral factors for eating midpoint in the Random Forest analysis.
Diet quality, ranked third, suggests that individuals with healthier diets are more likely to follow structured eating routines, which influence meal timing. Higher diet quality, characterized by increased consumption of fruits, vegetables, and minimally processed foods, was associated with earlier timings for the first, last, and midpoint of eating in our study. This implies that a healthier diet may promote better metabolic regulation through mechanisms such as enhanced satiety from nutrient-dense foods, improved regulation of hunger hormones, and a reduced likelihood of late-night snacking31,32.
The earliest eating midpoint was observed in individuals with no evening screen exposure, whereas longer screen time was associated with delayed first food intake and increased frequency of breakfast skipping. Because circadian clocks rely on light as the primary zeitgeber responsible for synchronizing endogenous biological rhythms to the 24-hour light-dark cycles, artificial light at night may contribute to circadian misalignment, sleep-related problems, and later eating habits. Research on nocturnal rodents has shown that light at night (LAN) increases relative daytime (rest phase) food consumption (56% of food intake with LAN compared to 37% without LAN) and leads to significant weight gain33. Moreover, excessive screen time, especially engaging in social media scrolling, watching TV shows, or viewing YouTube on smartphones, can lead to bedtime procrastination, which reduces sleep duration. To compensate for lost sleep, individuals may extend morning sleep, potentially leading to skipping breakfast and prolonging morning fasting to accommodate work or study schedules34.
Higher daily caffeine intake and evening alcohol consumption were, respectively, associated with delayed and advanced eating midpoints. While the evidence is limited and requires further investigation4, the short-term stimulating effects of caffeine and alcohol on appetite may contribute to shifts in meal timings7,32,33. Caffeine, acting as an adenosine receptor antagonist, can reduce sleep pressure during wakefulness and delay sleep onset, potentially promoting nighttime eating even when consumed earlier in the day34,35. Research also indicates that nighttime caffeine consumption, particularly when consumed three hours before bedtime, can delay the circadian rhythm of melatonin by approximately 40 minutes, akin to exposure to bright light at night36.
The association of having dinner as the largest meal with both a higher frequency of breakfast skipping and later eating patterns in our study suggests an interconnectedness between eating events4,40. Indeed, it has been demonstrated that overeating at night might suppress morning appetite, contributing to both breakfast skipping and a later eating midpoint41.
While contributing less in the Random Forest analysis, morning exercise also showed an inverse association with all eating events and breakfast skipping. In contrast, individuals who exercised in the evening exhibited a delay in the timing of their last food intake, but the onset and eating midpoint occurred earlier compared to those who did not exercise. Indeed, chronobiological research suggests that exercise influences circadian rhythms and neuroendocrine systems that regulate appetite and metabolism. Moreover, exercise can alter dietary habits, promoting preferences for healthier foods and reducing compensatory behaviors such as emotional eating, thereby influencing daily meal timings43.
4.3. Cultural and Environment Factors
Results confirmed significant regional and seasonal differences in food intake timings across Brazil12. Regions at lower latitudes, like the Northeast and North, began eating earlier, with the Northeast having the earliest eating midpoint compared to higher latitude regions like the Southeast and South. Participants in winter and autumn had a later eating midpoint than those in summer and spring.
The observed influence of remote work on delaying the timing of the first meal, lunchtime, and eating midpoint reflects the current post-COVID-19 pandemic work environment, which has fostered this new work arrangement with greater flexibility in both work and meal schedules44,45.
4.4. Strengths and Limitations
This study's strengths are rooted in its use of the initial national survey focusing on chrononutrition behaviors, involving a substantial sample from Brazil's population. Previous research has mainly targeted high-income countries, revealing significant diversity in eating behaviors—timing and distribution—across global regions, influenced by cultural norms, seasonal factors, and local work routines46. In this context, our study is both unique and innovative, offering unprecedented insights into regional differences in a country of continental proportions, and underscoring the need for public health strategies tailored to address these disparities. Moreover, to ensure data authenticity reflecting real-life habits, participants completed the questionnaire based on their routines from the previous month. The survey differentiated between weekdays (workdays) and weekends (free days) and calculated weighted aggregate scores to accurately represent weekly patterns.
However, our study faced limitations, including the use of self-reported questionnaires susceptible to underreporting or misreporting. Nevertheless, it's noteworthy that the assessment of chrononutrition behaviors relied on validated questions deemed suitable for online, community-based samples47. Another limitation was the absence of data on energy and nutrient intake, which would have allowed for an objective assessment of energy distribution throughout the meals. However, a previous study20 found moderate to strong correlations between aspects of chrononutrition preferences and behaviors assessed using the Chrononutrition Profile Questionnaire and the 24-hour Food Recall Questionnaire. Furthermore, although our study addressed different dimensions of physiological, behavioral, cultural, and environmental variables, including education as a proxy for socioeconomic status, it is possible that confounding factors, such as income or comorbidities, were not fully measured, which could have influenced the findings. Lastly, due to the cross-sectional nature of our data, we cannot establish causality between predictors and eating outcomes.
5. Conclusions
This pioneering study highlights the intricate interplay of physiological, behavioral, and environmental factors in shaping eating habits, with chronotype emerging as the primary predictor. Additionally, diet quality, evening screen time, daily caffeine consumption were the most influential features for eating midpoint.
Furthermore, our study underscores the significant impact of various modifiable factors on the timing of food intake, underscoring its importance in shaping public health strategies aimed at promoting eating patterns synchronized with the circadian system. While chrononutrition research has primarily focused on timing as predictors of health and circadian outcomes, understanding the determinants of eating behaviors is crucial for effective interventions. This approach enables the development of targeted, proactive, and sustainable strategies essential in combating the obesity epidemic and associated metabolic diseases.

Acknowledgements
We extend our gratitude to all Brazilians who participated in the Sonar-Brazil survey and to the universities, institutions, and professionals whose invaluable contributions helped disseminate this research nationwide.

References:
1. Garaulet M, Gómez-Abellán P, Alburquerque-Béjar JJ, Lee YC, Ordovás JM, Scheer FA. Timing of food intake predicts weight loss effectiveness. Int J Obes 2013; 37(4):604-611.
2. Fong M, Caterson ID, Madigan CD. Are large dinners associated with excess weight, and does eating a smaller dinner achieve greater weight loss? A systematic review and meta-analysis. Br J Nutr 2017; 118(8):616-628.
3. Ahluwalia MK. Chrononutrition—when we eat is of the essence in tackling obesity. Nutrients 2022; 14(23):5080.
4. Dashti HS, Scheer FA, Saxena R, Garaulet M. Timing of food intake: identifying contributing factors to design effective interventions. Adv Nutr 2019; 10(4):606-620.
5. Manoogian EN, Chaix A, Panda S. When to eat: the importance of eating patterns in health and disease. J Biol Rhythms 2019; 34(6):579-581.
6. Mode WJ, Small C, Johns S, Slater T, Hough J, Varley I, James RM, James LJ, Clayton DJ. The acute effects of morning and evening fasting on energy intake, appetite and glycaemic control in young, healthy males. Appetite 2021; [Online].
7. Peters B, Vahlhaus J, Pivovarova-Ramich O. Meal timing and its role in obesity and associated diseases. Front Endocrinol 2024; 15:1359772.
8. Wehrens SM, Christou S, Isherwood C, Middleton B, Gibbs MA, Archer SN, Skene DJ, Johnston JD. Meal timing regulates the human circadian system. Curr Biol 2017; 27(12):1768-1775.
9. Zerón-Rugerio MF, Longo-Silva G, Hernáez Á, Ortega-Regules AE, Cambras T, Izquierdo-Pulido M. The elapsed time between dinner and the midpoint of sleep is associated with adiposity in young women. Nutrients 2020; 12(2):410.
10. BaHammam AS, Pirzada A. Timing matters: the interplay between early mealtime, circadian rhythms, gene expression, circadian hormones, and metabolism—a narrative review. Clocks Sleep 2023; 5(3):507-535.
11. St-Onge MP, Ard J, Baskin ML, Chiuve SE, Johnson HM, Kris-Etherton P, Varady K. Meal timing and frequency: implications for cardiovascular disease prevention: a scientific statement from the American Heart Association. Circulation 2017; 135(9).
12. Santos JS, Skene DJ, Crispim CA, Moreno CRDC. Seasonal and regional differences in eating times in a representative sample of the Brazilian population. Nutrients 2023; 15(18):4019.
13. Crispim CA, Rinaldi AE, Azeredo CM, Skene DJ, Moreno CR. Is time of eating associated with BMI and obesity? A population-based study. Eur J Nutr 2024; 63(2):527-537.
14. Longo-Silva G, de Oliveira PMB, Pedrosa AKP, da Silva JR, Bernardes RS, de Menezes RCE, de Menezes Marinho P. Breakfast skipping and timing of lunch and dinner: Relationship with BMI and obesity. Obes Res Clin Pract 2022; 16(6):507-513.
15. Longo-Silva G, Serenini R, Pedrosa AKP, Lima MO, Soares LL, Melo JS, Menezes RCE. Chrononutrition Behaviors and Their Association with Body Weight: differences between chronotypes. Endocrinol Diabetes Nutr. In press 2024.
16. Longo-Silva G, de Oliveira Lima M, Pedrosa AKP, Serenini R, de Menezes Marinho P, de Menezes RCE. Association of largest meal timing and eating frequency with body mass index and obesity. Clin Nutr ESPEN 2024; 60:179-186.
17. Pedrosa AKP, de Oliveira Lima M, de Oliveira PMB, Serenini R, de Menezes RCE, Longo-Silva G. Circadian dinner timing and BMI among adults in a Brazilian national survey. Obes Med 2023; 44:100526.
18. Franzago M, Alessandrelli E, Notarangelo S, Stuppia L, Vitacolonna E. Chrono-nutrition: circadian rhythm and personalized nutrition. Int J Mol Sci 2023; 24(3):2571.
19. IBGE. Instituto Brasileiro de Geografia e Estatística. Population Census 2022. [cited 2024 Jul 15]. Available from: https://www.ibge.gov.br/en/statistics/social/labor/22836-2022-census-3.html.
20. Lira NDCC, de Araújo SM, de Medeiros ACQ, de Souza JC. Translation, adaptation and validation of the Chrononutrition Profile–Questionnaire (CP-Q) in Brazilian Portuguese. Chronobiol Int 2023; 40(4):473-482.
21. Veronda AC, Allison KC, Crosby RD, Irish LA. Development, validation and reliability of the Chrononutrition Profile-Questionnaire. Chronobiol Int 2020; 37(3):375-394.
22. World Health Organization. Obesity and Overweight. Geneva, Switzerland; 2017.
23. Ohayon M, Wickwire EM, Hirshkowitz M, Albert SM, Avidan A, Daly FJ, Dauvilliers Y, Ferri R, Fung C, Gozal D, Hazen N, Krystal A, Lichstein K, Mallampalli M, Plazzi G, Rawding R, Scheer F, Somers V, Vitiello MV. National Sleep Foundation's sleep quality recommendations: first report. Sleep Health 2017; 3(1):6-19.
24. Roenneberg T, Pilz LK, Zerbini G, Winnebeck EC. Chronotype and social jetlag: a (self-) critical review. Biology 2019; 8(3):54.
25. Ma X, Chen Q, Pu Y, Guo M, Jiang Z, Huang W, Long Y, Xu Y. Skipping breakfast is associated with overweight and obesity: A systematic review and meta-analysis. Obes Res Clin Pract 2020; 14(1):1-8.
26. Kant AK, Graubard BI. Within-person comparison of eating behaviors, time of eating, and dietary intake on days with and without breakfast: NHANES 2005–2010. Am J Clin Nutr 2015; 102(3):661-670.
27. Qian J, Vujovic N, Nguyen H, Rahman N, Heng SW, Amira S, Scheer FAJL, Chellappa SL. Daytime eating prevents mood vulnerability in night work. Proc Natl Acad Sci U S A 2022; 119(38).
28. Almoosawi S, Vingeliene S, Gachon F, Voortman T, Palla L, Johnston JD, Van Dam RM, Darimont C, Karagounis LG. Chronotype: implications for epidemiologic studies on chrono-nutrition and cardiometabolic health. Adv Nutr 2019; 10(1):30-42.
29. Muñoz JG, Gallego MG, Soler ID, Ortega MB, Cáceres CM, Morante JH. Effect of a chronotype-adjusted diet on weight loss effectiveness: A randomized clinical trial. Clin Nutr 2020; 39(4):1041-1048.
30. Phoi YY, Rogers M, Bonham MP, Dorrian J, Coates AM. A scoping review of chronotype and temporal patterns of eating of adults: tools used, findings, and future directions. Nutr Res Rev 2022; 35(1):112-135.
31. Li B, Tang X, Le G. Dietary Habits and Metabolic Health. Nutrients 2023; 15(18):3975.
32. Wang P, Tan Q, Zhao Y, Zhao J, Zhang Y, Shi D. Night eating in timing, frequency, and food quality and risks of all-cause, cancer, and diabetes mortality: findings from national health and nutrition examination survey. Nutr Diabetes 2024; 14(1):5.
33. Salgado-Delgado R, Angeles-Castellanos M, Saderi N, Buijs RM, Escobar C. Food intake during the normal activity phase prevents obesity and circadian desynchrony in a rat model of night work. Endocrinology 2010; 151(3):1019-1029.
34. Hill VM, Rebar AL, Ferguson SA, Shriane AE, Vincent GE. Go to bed! A systematic review and meta-analysis of bedtime procrastination correlates and sleep outcomes. Sleep Med Rev 2022 Dec;66:101697.
35. Caton SJ, Ball M, Ahern A, Hetherington MM. Dose-dependent effects of alcohol on appetite and food intake. Physiol Behav 2004; 81(1):51-58.
36. Panek-Shirley LM, DeNysschen C, O’Brien E, Temple JL. Caffeine transiently affects food intake at breakfast. J Acad Nutr Diet 2018; 118(10):1832-1843.
37. Clark I, Landolt HP. Coffee, caffeine, and sleep: A systematic review of epidemiological studies and randomized controlled trials. Sleep Med Rev 2017; 31:70-78.
38. Landolt HP, Werth E, Borbély AA, Dijk DJ. Caffeine intake (200 mg) in the morning affects human sleep and EEG power spectra at night. Brain Res 1995; 675(1-2):67-74.
39. Burke TM, Markwald RR, McHill AW, Chinoy ED, Snider JA, Bessman SC, Wright KP Jr. Effects of caffeine on the human circadian clock in vivo and in vitro. Sci Transl Med 2015; 7(305):305ra146-305ra146.
40. de Castro JM. Macronutrient relationships with meal patterns and mood in the spontaneous feeding behavior of humans. Physiol Behav 1987; 39(5):561-569.
41. Mekary RA, Giovannucci E, Willett WC, van Dam RM, Hu FB. Eating patterns and type 2 diabetes risk in men: breakfast omission, eating frequency, and snacking. Am J Clin Nutr 2012; 95(5):1182-1189.
42. Okada C, Imano H, Muraki I, Yamada K, Iso H. The association of having a late dinner or bedtime snack and skipping breakfast with overweight in Japanese women. J Obes 2019; 2019(1):2439571.
43. Shen B, Ma C, Wu G, Liu H, Chen L, Yang G. Effects of exercise on circadian rhythms in humans. Front Pharmacol 2023; 14:1282357.
44. Bazzani A, Marantonio S, Andreozzi G, Lorenzoni V, Bruno S, Cruz-Sanabria F, d'Ascanio P, Turchetti G, Faraguna U. Late chronotypes, late mealtimes. Chrononutrition and sleep habits during the COVID-19 lockdown in Italy. Appetite 2022; 172:105951.
45. Restrepo BJ, Zeballos E. The effect of working from home on major time allocations with a focus on food-related activities. Rev Econ Household 2020; 18(4):1165-1187.
46. Almoosawi S, Vingeliene S, Karagounis LG, Pot GK. Chrono-nutrition: a review of current evidence from observational studies on global trends in time-of-day of energy intake and its association with obesity. Proc Nutr Soc 2016; 75(4):487-500.
47. Veronda AC, Irish LA. Evaluation of the Chrononutrition profile–questionnaire in an online community sample of adults. Eat Behav 2022; 45:101633.