Different sedentary behavior domains present distinct associations with eating-related indicators

doi:10.21203/rs.3.rs-3485992/v1

Background

Unhealthful dietary patterns have been consistently associated with low levels of physical activity (PA), but studies dedicated to sedentary behavior (SB) are scarce, especially in adults. The few studies that investigated the association between SB and dietary patterns focused mostly on specific types of SB, such as TV-watching or screen time. SB can be accumulated in distinct domains (i.e., work, transport, and leisure-time), thus, it is key to investigate in depth the impact that different domains of SB can have on eating-related indicators. We aimed to investigate the associations between different SB domains and eating-related indicators, in a sample of adults.

Methods

Cross-sectional data from students, teachers, and staff from a Portuguese University was collected in November/2021 through an online survey. Data analyses were performed using the IBM SPSS software (version 28.0) and included descriptive statistics, partial correlations, and group comparisons using one-way ANOVA. Daily average SB at work/study, transport, and in leisure-time were self-reported and eating-related indicators were measured with several items from the Mediterranean Diet Score. Specific eating-related behaviors reflecting a protective eating pattern (e.g., eating breakfast regularly), and eating behavior traits (e.g., external eating) were also assessed. Body mass index (BMI) was calculated as weight/height². The International Physical Activity Questionnaire/Short-Form was used to assess PA.

Results

The sample included 301 adults (60.1% women), with a mean age of 34.5 years. Leisure-time SB was inversely associated with adherence to the Mediterranean diet (r=-0.20; p < 0.001) and with a protective eating profile (r=-0.31; p < 0.001). Higher transport SB was also related to lower adherence to the Mediterranean diet (r=-0.20; p < 0.001) and to an unhealthier eating profile (r=-0.22; p < 0.001), but no associations were found for work-related SB (p > 0.05). These results persisted after the adjustment for BMI, age, sex, and self-reported PA.

Conclusions

These findings suggest that adults with higher levels of SB in leisure-time and transport domains tend to report less healthy eating-related behaviors, irrespective of BMI, age, sex, and PA level. This information may assist public health authorities in focusing their efforts and strategies on specific domains, thus promoting active forms of commuting, and reducing SB in the leisure setting.

Eating behavior

Eating patterns

Mediterranean diet

Physical activity

Sedentary behavior

Unhealthy diet and poor eating behaviors can contribute to lifestyle linked conditions such as obesity (1), cardiovascular diseases, type 2 diabetes (2), and some types of cancer (3). The mechanisms underlying the association between a higher physical activity (PA) level and a healthier diet are far from being settled (3), but it seems that PA is generally positively associated with a healthier diet and with improved eating choices (4, 5).

In young adults, a reduced preference for processed foods, red meat, fried foods, soft drinks, and snacking was observed in those exercising for longer durations, whereas those exercising at higher intensities presented a healthier eating pattern, including a higher intake of fruits, vegetables, and low-fat foods (6). In previously inactive adults, moderate-vigorous exercise has also been shown to reduce the reinforcing value of high energy dense foods compared to non-exercise controls (7). Structured and lifestyle PA have also been positively associated with several eating behavior traits, like higher cognitive restraint and eating self-efficacy, and lower eating disinhibition, emotional, and external eating (8, 9). Moderate-vigorous PA has also been associated with a higher attunement with body-related eating cues, especially with body-congruent food choices (10).

However, some contradictory findings regarding the association between PA and healthy eating have also been reported (11). This might be related with the interference of sedentary behavior (SB), which usually occupies the majority of our 24-hour day cycle. SB refers to any waking behavior characterized by an energy expenditure of less than 1.5 METs, in a seated/reclined position (12) and current evidence supports the notion that SB displaces light intensity PA (LIPA) (13). Excessive time in SB is independently associated with several non-communicable diseases (14, 15), and even with higher mortality (14, 16), thus, there is a possibility that SB may play a specific role in people’s diet and eating behaviors, that can somehow alter the association between PA and these important outcomes.

Regarding diet, a systematic review found that SB in children, adolescents, and adults was associated with elements of a less healthier diet, including lower fruit and vegetable consumption, higher consumption of energy-dense snacks, drinks, fast food, and a higher total energy intake (17). This seems especially observed in younger populations (18). Also, regardless of the use of objective or subjective measurements, older adults without a healthy diet appear to engage in a more inactive and sedentary lifestyle (19). Screen time has also shown negative correlations with the intake of breakfast, vegetables, and fruit (20). Furthermore, a review suggested that SB was consistently identified as an important correlate of dietary behavior (21) and recent evidence suggests that displacing SB with LIPA improves dietary quality in older females, and SB fragmentation seems valuable for various dietary outcomes (22). Although limited, there is some evidence suggesting that SB might also be impactful on eating behavior traits. One recent investigation found that decreased cognitive impulse control following after-school sedentary screen time may be associated with increased brain activation for high energy dense foods, possibly contributing to disinhibited eating (23).

SB incorporates several domains (e.g., work, leisure-time, and transport related SB), and to the authors best knowledge evidence on the associations between distinct SB domains and eating behaviors is scarce (24, 25). One investigation including participants from five urban regions in Europe found that domain-specific SBs were related to unhealthy dietary behaviors, except for transport related SB (24). Furthermore, the fact that almost none of the associations were moderated by age or gender suggests that these associations may hold across age and gender groups (24). A more recent investigation (25) found no relationship for high SB with dietary and lifestyle habits in adjusted analyses, however, breaks in SB at work were associated with high consumption of dairy products and cereals. Also, more breaks from SB at leisure-time were associated with higher consumption of fruits and vegetables, even among the ones with higher SB (25).

Though the available evidence supports the putative role of PA and, to a smaller extent, the detrimental role of SB on eating-related behaviors, prior research has focused mostly on isolated macronutrient categories within a diet, neglecting the potentially interactive effects of multiple diet components and related behaviors. No studies addressing the associations between SB domains and eating behavior traits were found. Indeed, behind one’s food choices and consumption, there are several behavior traits that explain the higher or lower capacity someone has, to self-regulate eating and one’s food choices in the current obesogenic food environment (26). Thus, it is important that PA and/or SB effects on eating indicators are addressed in a more comprehensive way.

In the present investigation and considering the limited evidence available, we aimed to investigate the cross-sectional associations between distinct SB domains (i.e., work, leisure-time, and transport) and several eating-related indicators (i.e., dietary intake, eating habits’ protective profile, and eating behavior traits) in adults. Additionally, we aim to explore if these relationships are independent of sex, age, body mass index (BMI), and PA levels.

Study design & Participants

The present study has a cross-sectional design and is based on self-reported data from 301 participants (students, teachers, and staff) from a Portuguese University. This project was approved by the Ethics Committee of Lusófona University, and all participants gave their consent prior to participation. Data was collected in November 2021 through an online survey (ILIND/F+/EI/02/2020). A survey was conducted using a Qualtrics™ link, disseminated to the entire community of Lusófona University, via institutional emails, social networks, and websites. All participants with complete data for the current analyses were considered. The survey took approximately 20 minutes.

Measures

Sociodemographic. The participants were asked about their age (years), height (m) and weight (kg). BMI was calculated using the ratio weight/height².

Sedentary Behaviors. For assessing SB domains, the participants were asked to indicate, “The time spent on a regular weekday and weekend day while: 1) seated working/studying without screen; 2) seated working/studying with screen; 3) seated watching TV/computer/tablet/phone for leisure; 4) seated doing a hobby like playing an instrument, playing cards, reading; 5) seated in transport (private or public)”. A daily weighted average of SB at work/study was calculated based on the sum of 1) and 2), considering the number of weekdays and weekend days. Daily transport-related SB was calculated as the weighted average of the number of weekdays and weekend days. A daily leisure-time SB was calculated by summing 3) and 4) and obtaining the weighted average of the number of weekdays and weekend days (27, 28).

Physical Activity. The International Physical Activity Questionnaire Short-Form (IPAQ-SF) was used to assess PA (29). The IPAQ questionnaire is formed of seven questions related to specific types of PA e.g., walking, moderate, and vigorous activities, in terms of weekly frequency and duration of each specific type of activity. The Total PA (min/week) was calculated by summing the marked minutes per week spent on walking, moderate, and vigorous activities (29).

Mediterranean Diet Adherence. The MedDietScore measures adherence to the Mediterranean Diet (30). Briefly, the MedDietScore is a composite 11-item index that examines the weekly intake of 9 food groups, namely non-refined cereals, fruits, vegetables (raw or cooked), legumes, potatoes, fish (and fisheries), meat and meat products, poultry, full fat dairy (including cheese), as well as olive oil use, and alcohol drinking, based on the principles of the traditional Mediterranean Diet. Each response given for an item of the index is assigned a score from 0 to 5 (never, rare, frequent, very frequent, weekly, and daily consumption) based on the rationale of the traditional dietary pattern (or on a reverse scale, i.e., from 5 to 0, for food groups presumed away from the Mediterranean Diet; a composite score was then calculated, with greater scores suggesting higher adherence to the Mediterranean Diet (30).

Eating-related indicators.

An eating protective profile was determined based on participants’ answers on three specific indicators of eating-related habits (i.e., eating breakfast regularly, not skipping meals, and low consumption of fast food). A composite score was calculated by averaging the scores on these three items, answered in a 4-point Likert scale ranging from never to six or more times/week (31).

Reliance on internal hunger and satiety cues, an indicator of an individual’s trust in his/her physiological signals and thus of a more intuitive eating style, was measured using two items ("I trust my body to tell me when to eat.” and “I trust my body to tell me how much to eat." (32).

Eating for non-emotional reasons, i.e., using food to satisfy physical hunger instead of to cope with emotional distress, was measured with a reverse-coded item (“I find myself eating when I’m feeling emotional (e.g., anxious, depressed, sad), even when I’m not physically hungry”; (32).

Cognitive eating restraint was assessed with three items “I usually make an effort to eat less than I want”; “When I have eaten too much, I eat less than usual the following days” (indicative of a flexible approach to eating restraint) and “The only way I can control what I eat is to follow a fixed plan and not depart from it” (reflecting a rigid approach of eating restraint) ) (33, 34). A total eating restraint score was calculated by averaging these three items.

Eating induced by external cues was assessed via two items: “Being with someone who is eating, often makes me want to also eat” and “When I smell appetizing food or see a delicious dish, I find it very difficult not to eat – even if I’ve just finished a meal” (33).

The power that food has over the individual was measured with two items: “I think I enjoy eating a lot more than most other people” and “It seems like I have food on my mind a lot” (35).

Uncontrolled eating was assessed with one item: “Sometimes when I start eating, I just can’t seem to stop” (33).

Participants answered all these items in a 5-point Likert scale ranging from Strongly Disagree (1) to Strongly Agree (5). Greater values signal greater levels on that eating indicator.

Statistical analyses

Data analyses were performed using the IBM SPSS software (version 28.0), including descriptive statistics, bivariate correlations, and partial correlations controlling for age, sex, BMI, and PA level. For the statistically significant correlations, linear regression models were performed using R, which are displayed in the Figures. Comparisons between sex, age terciles, and BMI categories (normal vs excess body weight) were performed by One-way ANOVA for all the included variables. A 5% significance level was adopted.

The final sample included 301 adults (60.1% women), with 64.8% students, 21.9% teachers, and 13.3% staff. Age ranged from 18 to 72 years, with a mean age of 34.5 years. Participants spent an average of 404.5 (188.6) min/day in SB at work/school, 212.9 (146.0) min/day in leisure-time SB, and 71.9 (58.5) min/day in transport-related SB.

As presented in Table 1, participants with higher BMI showed a higher consumption of red meat and poorer eating indicators (e.g., higher external eating, uncontrolled eating, and power of food). Older participants tend to display lower levels of PA but concurrently lower SB in all domains and more positive eating patterns (i.e., a protective eating profile with more breakfast consumption, less meal skipping and fast-food consumption; less uncontrolled or external eating; and a higher adherence to the Mediterranean diet with higher consumption of fruits, vegetables, legumes, fish, olive oil and less red meat). Finally, men tend to have more intuitive and external eating than women, and in terms of Mediterranean diet adherence some mixed results were found (i.e., higher consumption of non-refined cereals and legumes, but in opposition a higher consumption of red meat and alcohol, and a lower use of olive oil in cooking). No differences between men and women were found for any SB domain, but men presented significantly more PA than women.

*** Table 1 here ***

Correlations of SB domains and PA level with eating-related indicators are depicted in Table 2. Higher transport-related SB was associated with lower intake of fruits (r=-0.17; p=0.003) and vegetables (r=-0.18; p=0.001), lower use of olive oil (r=-0.16; p=0.005), lower breakfast consumption (r=-0.20; p<0.001), and higher intake of potatoes (r=0.12; p=0.042), higher fast-food consumption (r=0.12; p=0.042), which was translated into a lower adherence to the Mediterranean diet (r=-0.20; p<0.001) and a less protective eating profile (r=-0.22; p<0.001). A similar pattern of associations was observed for leisure-time SB. Negative associations were observed with fruits (r=-0.16; p=0.005), vegetables (r=-0.12; p=0.037), and fish consumption (r=-0.16; p=0.006), use of olive oil in cooking (r=-0.20; p<0.001), and breakfast consumption (r=-0.21; p<0.001); positive associations were found with fast-food consumption (r=0.27; p<0.001), red meat and products (r=0.12; p=0.040), and meal skipping (r=0.16; p=0.005). These results were also translated into a lower adherence to the Mediterranean diet (r=-0.20; p<0.001) and a less protective eating profile (r=-0.31; p<0.001). No significant associations were observed for the work-related domain. After adjusting for sex, BMI, and PA level (only for SB domains), similar results were obtained. After adjusting for age, several associations became non-significant for the SB leisure-time domain, namely with fruits, vegetables, fish, and red meat consumption.

Regarding eating behavior indicators, transport-related SB was negatively associated with cognitive eating restraint (r=-0.12; p=0.031). No other significant associations were found. Adjusting for age, the association between transport-related SB with eating for physical rather than emotional reasons became significant (r=0.13; p=0.022). When adjusting for BMI, cognitive eating restraint was no longer significant. When adjusting for PA, leisure-time SB became negatively associated with cognitive eating restraint (r=-0.11; p=0.05). Results were independent of sex.

Higher PA levels were positively associated with fruits (r=0.16; p=0.007), poultry (r=0.13; p=0.029), and alcoholic beverages consumption (r=0.14; p=0.016). After adjustments, associations with poultry were no longer significant. PA levels were not associated with any eating behavior indicator, but after controlling for age, a positive association with eating for physical rather than emotional reasons emerged (r=0.14; p=0.015).

*** Table 2 here ***

The results for the significant associations are additionally presented in Figures 1-3.

*** Figure 1 here ***

*** Figure 2 here ***

*** Figure 3 here ***

Research has explored the interplay between eating and movement behaviors, considering their potential interactional nature (i.e., changes in one domain may aggregate with changes in others) (36, 37), and that health-enhancing spill-over effects may arise when being more physically active and having a healthier eating pattern co-occur (10, 38), suggesting a pattern of protective versus deleterious associations (31). However, contrarily to the wealth of evidence concerning the protective role of overall PA in this regard, limited research is available on the relationship between SB specific domains and eating-related indicators. Thus, this study sought to analyze the associations between distinct SB related domains (i.e., work, leisure-time, and transport) and several eating-related markers (i.e., food intake, protective eating habits profile, and eating behavior indicators) in adults.

Overall, findings extend the ones from previous studies focusing on PA and reporting a reduced preference for processed foods, red meat, fried foods, soft drinks, and snacking in more physically active individuals (6). We found that SB, more specifically leisure-time SB, was consistently and negatively associated with fruits, vegetables, and fish consumption, use of olive oil, and breakfast consumption; and positively associated with fast-food consumption, red meat, and meal skipping, irrespective of sex, BMI category, and PA level. However, some of these results disappeared when adjusting for age, which means that these associations somehow vary with age. We could not find any significant associations between work-related SB and eating-related indicators; future investigations are needed to confirm these findings.

In line with previous evidence showing that lifestyle PA is positively associated with several eating behavior indicators (8, 9), results showed that leisure-time and transport related SB - but not work-related SB - also seem to be associated with these outcomes. Transport-related SB was negatively associated with cognitive eating restraint, a potential risk factor for the adoption of dysfunctional eating patterns (39, 40). Thus, these findings extend the ones from a systematic review that found SB to be associated with elements of a less healthier diet including lower fruit and vegetable consumption, higher consumption of energy-dense snacks, drinks, fast food, and a higher total energy intake (17). Still, according to our findings, not all SB domains appear to be related with a less healthy diet. Future studies must confirm these results.

One investigation including participants from five urban regions in Europe found that domain-specific SBs were related to unhealthy dietary behaviors, except for transport related SB (24). Our results do not align with these findings, as a higher transport-related SB was associated with a lower adherence to the Mediterranean diet (e.g., lower intake of fruits and vegetables, and lower use of olive oil), but also with a less protective eating profile, including lower breakfast consumption and higher fast-food consumption. Furthermore, after the adjustment for sex, BMI, and PA level, similar results were obtained, meaning that the associations were independent from these variables.

Regarding eating behavior indicators, transport-related SB was associated with lower cognitive eating restraint (i.e., tendency to restrict food intake to manage weight (41)), although these associations seem to depend on people’s BMI. One possible explanation for these findings might be related to the high percentage of our sample that commuted to and from work by using public transportation (approximately 45%). This type of transportation has been associated with greater steps and MVPA when compared to private transportation (42, 43). It might just be that those using public transportation feel less demanded to control their eating, especially if they are not overweight, due to their potential higher PA level. On the other hand, we can speculate that transport-related SB may boost screen time during this period, resulting in a decreased cognitive impulse control and subsequent increase in brain activation for high energy dense foods, thus possibly contributing to a disinhibited vs. restricted eating pattern (23). Screen time has also been associated with lower vegetables and fruits intake (20).

Work-related SB was not associated with any eating indicator, highlighting that the deleterious role of SB may not only be related to SB itself (physiologically), but with participants’ global health behavior profile. When considering other outcomes (e.g., mental health), this different pattern of associations by domain has also been observed. For example, leisure-time SB, characterized as mentally passive (e.g., watching TV), was found to increase the risk of depression in adults, while there was no harm associated with mentally active SB (e.g., using the computer at work/school) (44), therefore suggesting that not only the domain can play a role in the association between SB and mental health, as the specific type of SB within the domain may be of relevance. It may be the case that when leisure-time is not fulfilled with activities that nurture both body and mind, that may be a marker for risky health behaviors, such as a high sedentary time not occupied with any intentional activity triggering dysfunctional eating patterns and behaviors.

Indeed, excessive time in SB has been shown to be independently associated with several non-communicable diseases (14, 15), and even with higher mortality (14, 16). Hence, there is a possibility that SB may play a specific role in people’s diet and eating behaviors, that can somehow alter the association between PA and these important outcomes (e.g., by limiting PA’s health-related outcomes). Future research would do well to explore these interactions.

Despite our focus on SB, we also explored the associations between PA and eating-related indicators and confirmed previous findings showing that higher PA levels were positively associated with healthier dietary choices (45), therefore reinforcing PA’s role as a gateway behavior for healthier eating. Unexpectedly, PA levels were not associated with any eating behavior indicator in the overall sample, but after controlling for age, a positive association with eating for physical rather than emotional reasons emerged. This finding is in line with previous studies, reporting negative associations between PA and emotional eating (8, 9), further suggesting that age might be a relevant factor in this relationship. Post-hoc analysis, exploring these associations by age tercile, showed that PA was associated with higher intuitive eating (i.e., in response to physiological hunger and satiety signals and not driven by emotional reasons) only in the older participants (3rd tercile). In other words, age appears to be a protective factor against more emotional (less healthy) eating pattern.

In today’s obesogenic food environment, regulating eating behavior is very demanding, with homeostatic drives involved in appetite control being easily superseded by hedonic, reward-based drives, that encourage eating beyond physiological necessity (46). Also, emotional states have an important impact over one’s eating behavior, mainly by depleting his/her cognitive resources (47), which has in fact been associated with increased consumption of unhealthy, highly palatable foods (48). In this context, the quality of food choices is determined by food hedonics, and deliberately and successfully resisting external and internal eating cues requires the identification of factors that can facilitate one’s eating self-regulation. SB, as PA, could be such a factor.

This study’s data was cross-sectional, and therefore the direction of relationships analysed cannot be clarified. The study only aimed at exploring the interplay amongst the variables, thus, causality cannot be inferred. Another limitation concerns the use of a self-report survey to measure the variables under scrutiny, thus, potential bias, such as social desirability, cannot be excluded. In the domain of eating-related indicators, an adapted self-report set of questions was also created, following previous studies and what has been advocated to conduct studies pertaining to a behavioral epidemiology framework (31, 49). Furthermore, although device-based measures, to assess PA and SB, would be preferred due to their accuracy, these types of devices may not capture domain and context specific behaviors, essential for this study’s aims. Thus, according to best practices, the survey was based on widely used, validated methods to measure PA and SB. There was an effort in balancing the need for precision vs. feasibility.

The main analyses conducted in this study allowed us to explore the cross-sectional associations between distinct SB domains (i.e., work, leisure-time, and transport) and several eating-related markers (i.e., food intake, eating habits’ protective profile, and eating behavior indicators) in adults, while also exploring if these relationships are independent of sex, age, BMI, and PA levels. This was an underexplored research arena as very limited research addressing the associations between SB domains and eating is available, neglecting the potentially interactive effects of multiple diet components and related behaviors.

Taken together, our results highlight the importance of considering the complexity embedded in health behaviors. Sedentary behaviors should not only be considered in terms of their physiological impact. The potential health risk of this type of behavior lies on the behavioral health patterns involved and their psychological consequences. SB domains are, thus, of critical importance, also for eating behavior regulation, as demonstrated in other outcomes (e.g., mental health) (50). Health literacy and public health campaigns devoted to promoting more active and healthier lifestyles may benefit from these findings- targeting SB domains of higher risk (i.e., leisure-time, and transport to some extent).

Ethics approval and consent to participate

This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the Scientific Ethics Committee of the Lusófona University, Portugal. Written informed consent was obtained from all subjects/patients.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

This research received no external funding. The survey was funded by ILIND-Lusófona University Research Center (ILIND/F+/EI/02/2020).

Authors' contributions

Conceptualization - MNS, PBJ, EVC; formal analysis - PBJ, EVC, IS; investigation -MNS, PBJ, EVC, IS, ALP; methodology- PBJ, EVC, IS; project administration and supervision- MNS, PBJ, EVC; validation- ALP; writing—original draft; writing—review & editing – PBJ, EVC, MNS, IS, ALP, FJ. All authors have read and agreed to the published version of the manuscript.

Acknowledgments

The authors sincerely appreciate the time and efforts of all students, teachers, and staff from Lusófona University who participated in this study.

Seifu CN, Fahey PP, Hailemariam TG, Frost SA, Atlantis E. Dietary patterns associated with obesity outcomes in adults: an umbrella review of systematic reviews. Public Health Nutr. 2021;24(18):6390–414.
Medina-Remón A, Kirwan R, Lamuela-Raventós RM, Estruch R. Dietary patterns and the risk of obesity, type 2 diabetes mellitus, cardiovascular diseases, asthma, and neurodegenerative diseases. Crit Rev Food Sci Nutr. 2018;58(2):262–96.
Tabung FK, Brown LS, Fung TT. Dietary Patterns and Colorectal Cancer Risk: A Review of 17 Years of Evidence (2000–2016). Curr Colorectal Cancer Rep. 2017;13(6):440–54.
Moe ÅM, Sørbye SH, Hopstock LA, Carlsen MH, Løvsletten O, Ytterstad E. Identifying dietary patterns across age, educational level and physical activity level in a cross-sectional study: the Tromsø Study 2015–2016. BMC Nutr. 2022;8(1):102.
Beaulieu K, Hopkins M, Long C, Blundell J, Finlayson G. High Habitual Physical Activity Improves Acute Energy Compensation in Nonobese Adults. Med Sci Sports Exerc. 2017;49(11):2268–75.
Joo J, Williamson SA, Vazquez AI, Fernandez JR, Bray MS. The influence of 15-week exercise training on dietary patterns among young adults. Int J Obes. 2019;43(9):1681–90.
Panek LM, Jones KR, Temple JL. Short term aerobic exercise alters the reinforcing value of food in inactive adults. Appetite. 2014;81:320–9.
Carraca EV, Silva MN, Coutinho SR, Vieira PN, Minderico CS, Sardinha LB, et al. The association between physical activity and eating self-regulation in overweight and obese women. Obes Facts. 2013;6(6):493–506.
Andrade AM, Coutinho SR, Silva MN, Mata J, Vieira PN, Minderico CS, et al. The effect of physical activity on weight loss is mediated by eating self-regulation. Patient Educ Couns. 2010;79(3):320–6.
Carraca EV, Rodrigues B, Teixeira DS. A Motivational Pathway Linking Physical Activity to Body-Related Eating Cues. J Nutr Educ Behav. 2020;52(11):1001–7.
Ottevaere C, Huybrechts I, Béghin L, Cuenca-Garcia M, De Bourdeaudhuij I, Gottrand F, et al. Relationship between self-reported dietary intake and physical activity levels among adolescents: The HELENA study. Int J Behav Nutr Phys Activity. 2011;8(1):8.
Tremblay MS, Aubert S, Barnes JD, Saunders TJ, Carson V, Latimer-Cheung AE, et al. Sedentary Behavior Research Network (SBRN) - Terminology Consensus Project process and outcome. Int J Behav Nutr Phys Act. 2017;14(1):75.
Mansoubi M, Pearson N, Biddle SJ, Clemes S. The relationship between sedentary behaviour and physical activity in adults: a systematic review. Prev Med. 2014;69:28–35.
Patterson R, McNamara E, Tainio M, de Sa TH, Smith AD, Sharp SJ, et al. Sedentary behaviour and risk of all-cause, cardiovascular and cancer mortality, and incident type 2 diabetes: a systematic review and dose response meta-analysis. Eur J Epidemiol. 2018;33(9):811–29.
Wilmot EG, Edwardson CL, Achana FA, Davies MJ, Gorely T, Gray LJ, et al. Sedentary time in adults and the association with diabetes, cardiovascular disease and death: systematic review and meta-analysis. Diabetologia. 2012;55(11):2895–905.
Chau JY, Grunseit AC, Chey T, Stamatakis E, Brown WJ, Matthews CE, et al. Daily sitting time and all-cause mortality: a meta-analysis. PLoS ONE. 2013;8(11):e80000.
Pearson N, Biddle SJ. Sedentary behavior and dietary intake in children, adolescents, and adults. A systematic review. Am J Prev Med. 2011;41(2):178–88.
Hobbs M, Pearson N, Foster PJ, Biddle SJ. Sedentary behaviour and diet across the lifespan: an updated systematic review. Br J Sports Med. 2015;49(18):1179–88.
Hsueh MC, Rutherford R, Huang YH, Chang Chien HY, Chang CH, Park JH et al. Are Older Adults without a Healthy Diet Less Physically Active and More Sedentary? Nutrients. 2019;11(5).
Al-Hazzaa HM, Abahussain NA, Al-Sobayel HI, Qahwaji DM, Musaiger AO. Physical activity, sedentary behaviors and dietary habits among Saudi adolescents relative to age, gender and region. Int J Behav Nutr Phys Act. 2011;8:140.
Sleddens EF, Kroeze W, Kohl LF, Bolten LM, Velema E, Kaspers P, et al. Correlates of dietary behavior in adults: an umbrella review. Nutr Rev. 2015;73(8):477–99.
Grant D, Tomlinson D, Tsintzas K, Kolic P, Onambele-Pearson G. Displacing Sedentary Behaviour with Light Intensity Physical Activity Spontaneously Alters Habitual Macronutrient Intake and Enhances Dietary Quality in Older Females. Nutrients. 2020;12(8).
Efraim M, Kirwan CB, Muncy NM, Tucker LA, Kwon S, Bailey BW. Acute after-school screen time in children decreases impulse control and activation toward high-calorie food stimuli in brain regions related to reward and attention. Brain Imaging Behav. 2021;15(1):177–89.
Compernolle S, De Cocker K, Teixeira PJ, Oppert JM, Roda C, Mackenbach JD, et al. The associations between domain-specific sedentary behaviours and dietary habits in European adults: a cross-sectional analysis of the SPOTLIGHT survey. BMC Public Health. 2016;16(1):1057.
Delfino LD, Tebar WR, Gil FC, De Souza JM, Romanzini M, Fernandes RA, et al. Association of sedentary behaviour patterns with dietary and lifestyle habits among public school teachers: a cross-sectional study. BMJ open. 2020;10(1):e034322.
Giskes K, van Lenthe F, Avendano-Pabon M, Brug J. A systematic review of environmental factors and obesogenic dietary intakes among adults: are we getting closer to understanding obesogenic environments? Obes Rev. 2011;12(5):e95–e106.
Vizcaino M, Buman M, DesRoches CT, Wharton C. Reliability of a new measure to assess modern screen time in adults. BMC Public Health. 2019;19(1):1386.
Lynch BM, Friedenreich CM, Khandwala F, Liu A, Nicholas J, Csizmadi I. Development and testing of a past year measure of sedentary behavior: the SIT-Q. BMC Public Health. 2014;14:899.
Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35(8):1381–95.
Panagiotakos DB, Pitsavos C, Arvaniti F, Stefanadis C. Adherence to the Mediterranean food pattern predicts the prevalence of hypertension, hypercholesterolemia, diabetes and obesity, among healthy adults; the accuracy of the MedDietScore. Prev Med. 2007;44(4):335–40.
Silva MN, Gregorio MJ, Santos R, Marques A, Rodrigues B, Godinho C et al. Towards an In-Depth Understanding of Physical Activity and Eating Behaviours during COVID-19 Social Confinement: A Combined Approach from a Portuguese National Survey. Nutrients. 2021;13(8).
Tylka TL, Van Diest AM. The Intuitive Eating Scale-2: item refinement and psychometric evaluation with college women and men. J Couns Psychol. 2013;60(1):137–53.
Duarte PAS, Palmeira L, Pinto-Gouveia J. The Three-Factor Eating Questionnaire-R21: a confirmatory factor analysis in a Portuguese sample. Eat Weight Disord. 2020;25(1):247–56.
Meresko R, Rubin M, Shontz FC, Morrow WR. Rigidity of attitudes regarding personal habits and its ideological correlates. J Abnorm Social Psychol. 1954;49(1):89–93.
Lowe MR, Butryn ML, Didie ER, Annunziato RA, Thomas JG, Crerand CE, et al. The Power of Food Scale. A new measure of the psychological influence of the food environment. Appetite. 2009;53(1):114–8.
Silva AM, Judice PB, Carraca EV, King N, Teixeira PJ, Sardinha LB. What is the effect of diet and/or exercise interventions on behavioural compensation in non-exercise physical activity and related energy expenditure of free-living adults? A systematic review. Br J Nutr. 2018;119(12):1327–45.
Marks DF. A General Theory of Behaviour. first ed. London: SAGE Publications Ltd.; 2018.
Mata J, Silva MN, Vieira PN, Carraca EV, Andrade AM, Coutinho SR, et al. Motivational spill-over during weight control: increased self-determination and exercise intrinsic motivation predict eating self-regulation. Health Psychol. 2009;28(6):709–16.
Lowe MR. The effects of dieting on eating behavior: a three-factor model. Psychol Bull. 1993;114(1):100–21.
Lowe MR, Timko CA. Dieting: really harmful, merely ineffective or actually helpful? Br J Nutr. 2004;92(Suppl 1):19–22.
Herman CP, Polivy J. Anxiety, restraint, and eating behavior. J Abnorm Psychol. 1975;84(6):66–72.
Chaix B, Kestens Y, Duncan S, Merrien C, Thierry B, Pannier B, et al. Active transportation and public transportation use to achieve physical activity recommendations? A combined GPS, accelerometer, and mobility survey study. Int J Behav Nutr Phys Act. 2014;11:124.
Bista S, Debache I, Chaix B. Physical activity and sedentary behaviour related to transport activity assessed with multiple body-worn accelerometers: the RECORD MultiSensor Study. Public Health. 2020;189:144–52.
Huang Y, Li L, Gan Y, Wang C, Jiang H, Cao S, et al. Sedentary behaviors and risk of depression: a meta-analysis of prospective studies. Transl Psychiatry. 2020;10(1):26.
Blakely F, Dunnagan T, Haynes G, Moore S, Pelican S. Moderate physical activity and its relationship to select measures of a healthy diet. J Rural Health. 2004;20(2):160–5.
Finlayson G, King N, Blundell JE. Liking vs. wanting food: importance for human appetite control and weight regulation. Neurosci Biobehav Rev. 2007;31(7):987–1002.
Macht M. How emotions affect eating: a five-way model. Appetite. 2008;50(1):1–11.
Vohs KD, Heatherton TF. Self-regulatory failure: a resource-depletion approach. Psychol Sci. 2000;11(3):249–54.
Sallis JF, Adlakha D, Oyeyemi A, Salvo D. An international physical activity and public health research agenda to inform coronavirus disease-2019 policies and practices. J Sport Health Sci. 2020;9(4):328–34.
Wang X, Li Y, Fan H. The associations between screen time-based sedentary behavior and depression: a systematic review and meta-analysis. BMC Public Health. 2019;19(1):1524.

Table 1 Comparisons between BMI, age, and sex categories, for the main variables (mean ± SD).

Variables	BMI<25.0 (n=199)	BMI>25.0 (n=102)	Age 1^st tercile	Age 2^nd tercile	Age 3^rd tercile	Women (n=181)	Men (n=120)
Demographics
Age (years)^a	32.83 ± 14.02	37.84 ± 13.2 **	19.80 ± 1.4	31.01 ± 5.8	51.04 ± 6.8 ***	32.93 ± 13.3	36.78 ± 14.6 *
Height (m)	1.68 ± 0.08	1.70 ± 0.11 *	1.68 ± 0.10	1.69 ± 0.09	1.69 ± 0.09	1.64 ± 0.07	1.76 ± 0.06 ***
Body mass (kg)	61.7 ± 9.2	84.8 ± 16.7 ***	65.2 ± 14.8	69.1 ± 16.6	73.6 ± 16.5 **	63.2 ± 13.6	78.9 ± 15.7 ***
BMI (kg/m²)	21.8 ± 2.2	29.1 ± 4.9 ***	23.0 ± 4.2	24.0 ± 5.2	25.6 ± 4.6 ***	23.5 ± 4.7	25.4 ± 4.8 ***

Eating-related indicators (min-max)
*Mediterranean Diet Adherence*
Non-refined cereals (1-6)	2.51 ± 1.1	2.51 ± 1.1	2.51 ± 1.1	2.59 ± 1.1	2.43 ± 1.1	2.40 ± 1.0	2.67 ± 1.2 *
Potatoes (1-6)	2.89 ± 1.0	3.00 ± 1.2	3.02 ± 1.0	2.90 ± 1.0	2.88 ± 1.1	2.91 ± 1.1	2.97 ± 1.0
Fruits (1-6)	3.55 ± 1.3	3.52 ± 1.3	3.25 ± 1.2	3.47 ± 1.2	3.87 ± 1.3 **	3.50 ± 1.2	3.60 ± 1.3
Vegetables (raw or cooked) (1-6)	3.23 ± 1.2	3.30 ± 1.2	2.86 ± 1.1	3.30 ± 1.2	3.57 ± 1.2 ***	3.22 ± 1.3	3.32 ± 1.1
Legumes (1-6)	3.21 ± 1.2	3.35 ± 1.3	3.00 ± 1.2	3.43 ± 1.3	3.30 ± 1.1 *	3.09 ± 1.2	3.50 ± 1.2 *
Fish (and fisheries) (1-6)	3.28 ± 1.2	3.33 ± 1.1	3.13 ± 1.2	3.13 ± 1.2	3.63 ± 1.1 **	3.31 ± 1.2	3.28 ± 1.1
Red meat and products (1-6)	2.20 ± 1.2	2.54 ± 1.3 *	2.63 ± 1.4	2.21 ± 1.3	2.14 ± 1.3 *	2.06 ± 1.2	2.72 ± 1.3 ***
Poultry (1-6)	2.46 ± 1.3	2.58 ± 1.3	2.69 ± 1.4	2.49 ± 1.3	2.37 ± 1.2	2.41 ± 1.3	2.65 ± 1.4
Use of olive oil in cooking (1-6)	5.10 ± 1.2	4.99 ± 1.3	4.69 ± 1.4	5.08 ± 1.3	5.37 ± 0.9 ***	5.22 ± 1.2	4.81 ± 1.3 **
Alcoholic beverages (1-7)	1.51 ± 0.9	1.63 ± 1.0	1.57 ± 1.1	1.47 ± 0.9	1.61 ± 0.8	1.40 ± 0.7	1.78 ± 1.2 ***
*Eating-related profile*
Breakfast consumption (1-4)	3.52 ± 0.9	3.51 ± 0.8	3.40 ± 0.9	3.44 ± 0.9	3.71 ± 0.7 *	3.56 ± 0.8	3.46 ± 0.9
Meal skipping (1-4)	1.55 ± 0.8	1.57 ± 0.8	1.62 ± 0.9	1.66 ± 0.8	1.38 ± 0.6 *	1.56 ± 0.8	1.54 ± 0.8
Fast-food consumption (1-3)	1.59 ± 0.6	1.69 ± 0.6	1.86 ± 0.5	1.56 ± 0.6	1.48 ± 0.5 ***	1.63 ± 0.6	1.62 ± 0.6

Variables	BMI<25.0 (n=199)	BMI>25.0 (n=102)	Age 1^st tercile	Age 2^nd tercile	Age 3^rd tercile	Women (n=181)	Men (n=120)
Eating behavior indicators (min-max)
Reliance hunger/satiety cues (1-5)	3.21 ± 1.13	2.61 ± 1.1 ***	3.01 ± 1.2	3.14 ± 1.1	2.87 ± 1.2	3.00 ± 1.2	3.01 ± 1.1
Eating non-emotional reasons (1-5)	3.38 ± 1.5	2.66 ± 1.6 ***	3.00 ± 1.6	2.94 ± 1.5	3.45 ± 1.6 *	2.79 ± 1.5	3.66 ± 1.5 ***
External eating (1-5)	2.08 ± 1.0	2.43 ± 1.1 **	2.48 ± 1.1	2.13 ± 1.0	2.02 ± 1.0 **	2.31 ± 1.1	2.04 ± 1.0 *
Power of food (1-5)	1.88 ± 1.0	2.44 ± 1.2 ***	2.34 ± 1.2	1.97 ± 1.1	1.91 ± 1.1 *	2.11 ± 1.2	2.00 ± 1.1
Cognitive eating restraint (1-5)	2.28 ± 1.0	2.76 ± 0.9 ***	2.39 ± 1.1	2.39 ± 1.0	2.53 ± 1.0	2.45 ± 1.0	2.43 ± 1.0
Uncontrolled eating (1-5)	1.98 ± 1.2	2.45 ± 1.3 **	2.44 ± 1.4	2.11 ± 1.3	1.90 ± 1.2 *	2.24 ± 1.4	1.98 ± 1.1

Sedentary behavior domains & PA
Leisure-time (min/day)	213.9 ± 150.7	210.2 ± 137.5	290.5 ± 167.1	200.9 ± 141.8	155.3 ± 89.8 ***	212.1 ± 151.3	214.1 ± 138.1
Transport (min/day)	73.8 ± 57.5	68.5 ± 61.0	89.2 ± 64.9	67.8 ± 63.5	60.6 ± 42.1 **	67.9 ± 56.0	77.9 ± 61.9
Work (min/day)	412.5 ± 194.1	387.8 ± 177.7	439.3 ± 197.7	421.1 ± 218.6	356.8 ± 131.4 **	414.0 ± 196.8	390.1 ± 175.3
Total PA (min/week)	696.5 ± 620.4	604.6 ± 539.0	807.4 ± 717.8	671.0 ± 588.3	541.6 ± 441.1 **	612.8 ± 553.5	752.3 ± 648.3 *

* p < .05, ** p < .01, *** p < .001 for the One-way ANOVA comparisons between categories of BMI, age, and sex.

The means and SD for the eating-related indicators and behavior traits are in relation to the corresponding response Likert scales.

Table 2 Correlations of SB domains and Total PA with Eating-related Indicators and Eating Behavior Traits.

Variable	SB transport	SB work	SB leisure-time	PA level
Eating-related Indicators
*Mediterranean Diet Adherence*
Non-refined cereals	-0.10	-0.01	0.02	0.00
Potatoes	0.12*	0.03	0.10	0.06
Fruits	-0.17**	-0.02	-0.16**	0.16**
Vegetables	-0.18**	-0.03	-0.12*	0.07
Legumes	-0.07	0.01	-0.01	0.08
Fish (and fisheries)	-0.04	-0.01	-0.16**	-0.06
Red meat and products	0.10	0.09	0.12*	-0.01
Poultry	0.02	0.04	0.04	0.13*
Use of olive oil in cooking	-0.16**	-0.07	-0.20***	-0.09
Alcoholic beverages	0.03	-0.04	0.07	0.14*
MedDietScore	-0.20***	-0.05	-0.20***	0.00
*Eating-related profile*
Breakfast consumption	-0.20***	0.01	-0.21***	-0.06
Meal skipping	0.11	0.05	0.16**	0.05
Fast-food consumption	0.12*	-0.02	0.27***	-0.05
Protective eating profile	-0.22***	-0.02	-0.31***	-0.04
Eating Behavior Indicators
Reliance on hunger & satiety cues	-0.05	-0.03	0.04	0.01
Eating non-emotional reasons	0.10	-0.03	-0.05	0.11
External eating	-0.04	0.01	0.09	-0.02
Power of food	0.03	-0.02	0.07	-0.05
Cognitive eating restraint	-0.12*	0.02	-0.11	0.04
Uncontrolled eating	-0.06	0.02	0.04	-0.00

* p < .05, ** p < .01, *** p < .001

No competing interests reported.

Different sedentary behavior domains present distinct associations with eating-related indicators

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Materials and Methods

Study design & Participants

Measures

Statistical analyses

Results

Discussion

Declarations

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1