Next Article in Journal
Lotus (Nelumbo nucifera Gaertn.) Leaf-Fermentation Supernatant Inhibits Adipogenesis in 3T3-L1 Preadipocytes and Suppresses Obesity in High-Fat Diet-Induced Obese Rats
Next Article in Special Issue
Unhealthy Food at Your Fingertips: Cross-Sectional Analysis of the Nutritional Quality of Restaurants and Takeaway Outlets on an Online Food Delivery Platform in New Zealand
Previous Article in Journal
Effects on Spirulina Supplementation on Immune Cells’ Parameters of Elite College Athletes
Previous Article in Special Issue
Predicting Perceived Problems in Self-Administered 24-Hour Dietary Recalls: A Quantitative Think-Aloud Study Comparing Automated Self-Assisted 24-Hour Dietary Assessment Tool (ASA24®) and INTAKE24© in University Students
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Nutrients
Volume 14
Issue 20
10.3390/nu14204349
nutrients-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Wearable Cameras Reveal Large Intra-Individual Variability in Timing of Eating among Young Adults

by
Leanne Wang
1,2,
Virginia Chan
1,2,
Margaret Allman-Farinelli
1,2,
Alyse Davies
2,
Lyndal Wellard-Cole
1,3 and
Anna Rangan
1,2,*
1
Discipline of Nutrition and Dietetics, Susan Wakil School of Nursing and Midwifery, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW 2006, Australia
2
Charles Perkins Centre, The University of Sydney, Sydney, NSW 2006, Australia
3
Cancer Prevention and Advocacy Division, Cancer Council NSW, Sydney, NSW 2011, Australia
*
Author to whom correspondence should be addressed.
Nutrients 2022, 14(20), 4349; https://doi.org/10.3390/nu14204349
Submission received: 20 September 2022 / Revised: 13 October 2022 / Accepted: 16 October 2022 / Published: 17 October 2022
(This article belongs to the Special Issue Nutrition in the Digital Age—Innovation and Trends in Dietary Patterns and Healthy Lifestyle)
Browse Figures
Versions Notes

Abstract

:
Studies have shown that young adults follow less structured eating patterns compared with older cohorts. This may have implications for dietary assessment methods which rely on memory and structured meal patterns. Our aim was to describe the intra-individual variation of eating times in young adults aged 18–30 years. Participants (n = 41) wore an Autographer camera that captured first-person perspective images every 30 s for three consecutive days. All images were timestamped and those showing food consumption were used to extract data such as the timing of the first and last eating occasions (EOs), number of EOs per day, and length of eating window. Intra-individual variability was calculated from these data using composite phase deviation (CPD) and coefficient of variation (CV). The number of individuals with high or very high variability was 28 and 18 for timing of first and last EOs, respectively (CPD > 1.70), and 27 and 17 for number of EOs and eating window, respectively (CV > 20%). In this sample of young adults, the lack of regularity in eating patterns should be considered when selecting a dietary assessment method.

Graphical Abstract

1. Introduction

Previous studies have shown that young adults eat at later times throughout the day, follow less structured eating patterns, and have a higher variability in energy intake when compared with older cohorts [1,2,3,4,5]. It is unclear whether this has an impact on dietary quality and health due to the heterogenous nature of the studies exploring this relationship in terms of methodology used and outcomes investigated as reported in previous reviews [6,7,8,9]. Common outcomes include meal frequency (number of meals per day), breakfast consumption or lack thereof [3,10,11,12,13,14], late-night eating [14,15,16,17,18,19], time-restricted eating such as intermittent fasting [8,20], and meal timing relative to sleep timing [21,22,23]. In recent years, there has been a greater interest in the intra-individual day-to-day variability of meal timing and its association with health outcomes [4,21,24,25,26,27,28,29,30,31,32,33].
Methods of data collection in meal timing variability studies are usually subjective. There is a heavy reliance on self-report and human recall such as dietary recalls [4,25,31,32,33,34], questionnaires [24,28,35,36,37,38], or user-initiated food image capture [5,21,27,30]. This, combined with the availability of feedback from these data collection methods, may influence or interfere with subjects’ food choices and behaviors [5]. Technologies such as wearable cameras present an opportunity to objectively monitor human food and beverage consumption. They can automatically capture time-stamped images of all activities throughout the day, including eating occasions, without requiring user input and can be analyzed for daily eating patterns [5,39,40].
In this study, we examine young adults’ eating patterns via time-stamped food images obtained from wearable cameras. Specifically, for our first aim, we sought to describe the intra-individual variation of eating times in young adults using four different metrics and categorize them as low, moderate, high, or very high variability. Our second aim was to investigate whether there was an association between the four metrics used to measure intra-individual variability and specific demographics such as gender, age, body mass index (BMI), and socioeconomic status (SES), as well as total energy intake across the duration of data collection.

2. Materials and Methods

2.1. Data Collection

This paper used data from the sub-study (n = 133) [41] of a larger cross-sectional MYMeals project (n = 1001) [42]. Participants in the sub-study were young adults (18–30 years) who wore an Autographer camera for three consecutive days. The wearable camera was worn on a lanyard around the neck and automatically captured images from a first-person perspective every 30 s. Participants were instructed to wear the camera for all waking hours over the three days while maintaining their usual daily activities. They were permitted to temporarily halt image capture or remove the camera when privacy was needed [43,44]. On the same three days, they also recorded food and beverage intake using a researcher-designed smartphone app named EaT and Track (The University of Sydney, Sydney, Australia) [45] and completed daily 24-h dietary recall interviews with research dietitians via the Automated Self-Administered 24-h recall Australia program (Deakin University, Melbourne, Australia) [46]. Recruitment methods have previously been described in the MYMeals study protocol [47]. In brief, participants had to be between the ages of 18–30 inclusive, consume foods or beverages prepared outside the home at least once a week, own a smartphone, and read and write English. Participants who were pregnant, lactating or had ever had an eating disorder were excluded. Participants completed a basic demographic questionnaire, providing information such as gender (male, female or prefer not to say), education status (primary school or less, secondary school, trade qualification/apprenticeship/diploma or university degree), employment status (full-time study, full-time work, part-time study/work, not working or studying) and residential postcode. Residential postcode was used to determine relative socio-economic advantage and disadvantage ranking within Australia using the Socio-Economic Indexes for Areas 2016 (high; top five deciles or low; bottom five deciles) [48]. Participants’ weight (kg) and height (cm) were self-reported and used to calculate BMI (underweight < 18.5, healthy weight 18.5–24.9, overweight 25–29.9, obese ≥ 30 kg/m2). Using self-reported weight and height to calculate BMI has been found to be sufficiently accurate [49]. Deidentified camera images were stored in the university’s research data store and demographic and anthropometric information were hosted and stored in the Research Electronic Data Capture (REDCap) data management system (Vanderbilt University, Nashville, TN, USA) [50]. This sub-study was approved by the Institutional Human Research Ethics Committee (2016/546) on the 15 July 2016.
All images captured by the Autographer camera were coded by an Accredited Practising Dietitian for the presence or absence of food and beverages and then matched with their 24-h recall data. This involved matching the time and date stamp of the wearable camera images with the self-reported times of the 24-h recall and annotating foods and beverages reported in the 24-h recall as: (i) reported by both methods, (ii) not reported in the 24-h recall, or (iii) not identified by the wearable camera (i.e., the camera may have been turned off). For entries labelled as not reported in the 24-h recall or not identified by the wearable camera, the omitted episode and associated food and beverage items were tabulated in Microsoft Excel. Two researchers checked all matching of data from the three sources (v.c. and a.d.) [40].

2.2. Inclusion Criteria

The 24-h recall was considered the ground truth method and foods that were reported in the 24-h recall but not identified by the camera were labelled as missing. If foods from an entire meal or snack were missing from the images, the eating occasion (EO), i.e., main meal or snack, would be labelled as missing. To ensure the inclusion of only high quality data, participants with no missing main meals and no more than three missing snacks over the course of the three days of data collection were included for analysis in this paper. Data from the EaT app and images of beverages were not used, as it was difficult to differentiate between nutritive and non-nutritive beverages, e.g., images of opaque drinking utensils.

2.3. Data Analysis

The times of EOs were collated using the time and date stamps available from the camera images. An EO was defined as the consumption of any food. In cases where there was more than one image captured of an EO, the timestamp of only the first image (i.e., the one with the earliest time) was used to represent its time of consumption. EO labels such as main meals or snacks were not used and were only labelled by time of day. Any EO that was recorded ≤ 15 min of a previous event was combined to form one EO as per previous studies [4,5]. Each day was defined as the 24-h period from 12:00 am to 11:59 pm on the same calendar day. No drinking occasions were included.
Eating pattern metrics such as the total number of EOs per day (after combining events within 15 min of each other), the clock time of the first and last EO of each day, the daily eating window, and the daily energy intake were extracted for all days. The daily eating window was defined as the duration between the start time of the first EO to the start time of the last EO of the same day. These metrics were stratified by weekdays and weekends and also compared between participants with healthy and overweight/obese BMIs.

2.3.1. Meal Timing Variability Metrics

Four metrics were applied to assess the stability of meal timing across the three days of data collection: Composite Phase Deviation of the first EO (CPD First), Composite Phase Deviation of the last EO (CPD Last), coefficient of variation of the daily number of EOs (CV No. of EOs), and coefficient of variation of the daily eating window (CV Eating Window).
The first two metrics both used Composite Phase Deviation (CPD) to assess the day-to-day stability of meal timing [21] for the first and last EO of each calendar day. CPD combines two components: (i) regularity, i.e., how different the timing of a particular meal is from the same meal of the previous day, and (ii) alignment, i.e., how different the timing of a particular meal is from the average time of that meal over the number of days of data collection [51]. The greater the CPD score, the greater the deviation in hours from a perfectly regular pattern of meal timing, i.e., the same time every day. CPD was originally developed to assess the stability of sleep timing but has recently been applied to social and eating events [21,52]. This is the formula for CPD:
ΔRegularityi(ΔDDi) = Meal timingi–1 − Meal timingi
ΔAlignmenti(ΔATi) = Average meal timing − Meal timingi
CPDi = √(ΔDDi2 + ΔATi2)
CPD = 1/N(∑Ni=1 CPDi)
Variables: DD stands for “day-to-day”, AT stands for “average timing”, i is any given day, and N is the total number of days.
The third and fourth metrics both used coefficient of variation (CV) to evaluate intra-individual variability. The formula for CV is: (standard deviation/mean) × 100. One measured the intra-individual day-to-day variability of the total number of EOs per day and the other measured the intra-individual day-to-day variability of the daily eating window.
The cut-offs used to categorize CPD and CV scores as low, moderate, high, and very high variability were based on hours or number of EOs (Table 1). For example, participants with low variability in CPD of the first EO varied in meal timing by less than two hours over the three days of data collection. An example of this is the consumption of the first meal or snack at 9:00 a.m. on the first day, 10:00 a.m. on the second day, and 10:45 a.m. on the third day. Participants with high variability in CV number of EOs differed by more than three EOs over the three days.

2.3.2. Statistics

Statistical analyses were conducted using SPSS software, v27.0 for Windows (IBM, Armonk, NY, USA). To test for differences in intra-individual meal timing variability metrics between demographic variables and eating pattern metrics between BMI categories (reduced to two categories—healthy weight versus overweight/obese), the Mann-Whitney U test was used. Spearman rank-order correlation coefficients were used to identify associations between energy intake and intra-individual variability metrics. A p-value of ≤0.05 was considered statistically significant.

3. Results

The MYMeals sub-study recruited 133 participants, of which 41 met the inclusion criteria for this paper. Participant characteristics including age, BMI, SES, education level, and employment status can be found in Table 2. A total of 577 EOs across 123 days were included for analysis. Of the 123 days, 72% (n = 88) were weekdays and 28% (n = 35) were weekend days.
The mean and range of eating patterns and intra-individual variability metrics are shown in Table 3. The average time of first and last EO were 10:18 a.m. and 8:06 p.m., respectively. The average number of EOs per day was 4.7, the average length eating window duration was 9.9 h, and the average daily energy intake was 8478 kJ. The intra-individual variability of the first EO was greater than the last EO as measured by CPD (2.9 versus 1.8), i.e., the time at which participants consumed their last EO over the three days was more consistent compared with the first EO.
Table 4 compares the eating pattern metrics of participants with a healthy versus an overweight/obese BMI. Although the window of time in which participants consumed food over the course of the day and the daily energy intake were slightly higher for people with overweight/obese BMIs, this was not statistically significant. All other eating pattern metrics also showed no statistically significant differences between BMI groups.
A comparison of the mean and range of eating patterns and intra-individual variability metrics between weekdays and weekends is shown in Table S1. Whilst the mean eating window was larger on weekdays than weekends (10.1 h versus 9 h), the mean daily energy intake was greater on weekends than weekdays (8875 kJ versus 8320 kJ).
Figure 1 is a visual representation of the three eating windows for each participant. The shortest observed eating window was 15 min, and the longest was 22 h 11 min. Ninety five percent of participants (n = 39) had at least one day out of the three days with an eating window of < 12 h. Fifty nine percent of participants (n = 24) had an eating window of < 12 h for all three days of data collection.
Table 5 compares the intra-individual variability metrics for each of the demographic variables studied. There were no significant associations for age, gender, and BMI, but individuals from a high SES (top five deciles) had a significantly greater variability in the timing of the first EO compared with individuals from a low SES (bottom five deciles) (CPD First 3.8 versus 1.9, p = 0.047).
Figure 2 shows the distribution of participants in the low, moderate, high, and very high variability categories for each meal timing stability metric. For the first EO (CPD First), 28 out of 41 participants had high to very high variability. This is similar (n = 27) to the variability for the number of EOs (CV No. of EOs). In contrast, variability was lower for the last EO (CPD Last) and the daily eating window (CV Eating Window) with high to very high variability in 18 and 17 participants, respectively. There were no participants who fell into the low category for all four metrics simultaneously, whereas six participants had all four metrics in the high or very high variability category.
Total energy intake across three days was not significantly associated with any of the metrics used to measure day-to-day variability of meal timing (Table S2). The average daily energy intake (kJ) for participants in the low, moderate, high, and very high variability categories for each metric is shown in Figure 2.

4. Discussion

Overall, our results showed that the timing of the first EO and the daily number of EOs were highly variable from day-to-day. It was common for the timing of the first EO to vary by more than three hours across days and for the number of EOs per day to vary by more than two EOs. In contrast, the timing of the last EO and daily eating window were relatively more stable with more participants in the low to moderate variability groups. Most of the participants in these groups had less than three hours of variation for the timing of the last EO and less than four hours of variation for the daily eating window. These results add to previous studies which have found that young adults tend to eat more frequently and erratically throughout waking hours [4,5].

4.1. Comparing Our Methods and Results with Previous Studies

Studies that measure the intra-individual variability of eating timing use subjective methods of data collection and have inconsistent definitions for meal time variability or regularity. These methods include using metrics such as CPD, CV, SD, and mean meal shift [21,25,27,30,31,32,33], calculating the proportion of daily energy intake per hourly bin [4,5,26,27,29,30,32,34], using self-reported questionnaires [24,28,36,37,38], measuring eating jetlag between weekends and weekdays [28,31], and assessing meal frequency [53,54,55]. The metrics that we developed for our paper (CPD and CV) were adapted from methods used by McHill et al. [21] and Popp et al. [30] but adjusted for three days of meal timing data. Previously, epidemiology papers have established cut-offs for CPD and CV to categorize scores as low, moderate, or high variability for sleep timing [56,57,58]. However, these cut-offs were based on levels of sleep variability that were associated with increased health risks and were not appropriate for categorizing meal timing variability [58]. It is important to establish standardized cut-offs for eating patterns to effectively compare results between dietary studies.
McHill et al. evaluated intra-individual variability of food intake in healthy young adults across two timescales—a daily timescale (seven days) using CPD and a monthly timescale (seven days per month for three months) using intra-class correlation. They found that the CPD score for the last caloric event was the highest (i.e., least consistent across days) relative to the first caloric event and caloric midpoint. This contrasts with our results, where we showed that CPD Last was lower than CPD First. While no association was seen between BMI and meal timing variability in our study, McHill et al. found that non-lean individuals had a higher stability in meal timing when measured between months [21]. Popp et al. measured intra-individual variability in a population of adults who were overweight or obese using CV No. of EOs and the CV of the timing of the first and last EO over two or more days. Their results indicated that, compared to weekdays, the timing of the first EO was later but more consistent and that there were fewer EOs and a shorter eating window on weekends. This is consistent with our results which also saw a later first EO and shorter eating window on weekends but a similar number of EOs across weekdays and weekends. Popp et al. also found no significant differences between genders for all metrics, aligning with our findings [30].

4.2. Implications of Meal Timing Variability on Dietary Quality and Health

Most epidemiological findings appear to suggest that following eating patterns with a large variability in meal timing has negative implications on dietary quality and/or cardiometabolic health [27,28,29,30,31,32,33,34,37,38,54]. Data collected in the 1994–2004 National Health and Nutrition Examination Survey (NHANES) and 2005–2007 German National Nutrition Survey II (NVS II) showed that a regular meal pattern with greater energy intake earlier in the day scores higher on the Healthy Eating Index [29,34]. This provides physiological benefits such as a reduced prevalence of metabolic syndrome, reduced waist circumference, improved HDL cholesterol, and reduced γ-glutamyl transferase concentrations in Swedish men and women aged 60 years or above [37]. A prospective study also showed an association between irregular eating at 16 years of age and the development of metabolic syndrome at the age of 43 [38], although this was explained by concurrent unhealthy lifestyles such as smoking, alcohol, and low levels of physical activity.
Makarem et al., Meth et al., and Zhao et al. used SD to measure intra-individual variability in a range of populations including women aged 20–64 years with cardiometabolic risk factors [31,33], 70-year-old men [25], and adults with overweight or obese BMIs [27]. All three studies found that a high day-to-day variability was associated with adverse effects on health such as an increased blood pressure, worse glycemic control [31], elevated high-sensitivity C-reactive protein [33], higher fatal cancer risk [25], and increased adiposity [27,31]. These results are also supported by Fleischer et al. for healthy adults without obesity [32]. Eating jetlag, defined as weekday-weekend differences in meal timing, has been measured in two studies [28,31]. Zerón-Rugerio et al. showed that an eating jetlag of 3.5 h or more was significantly associated with an increased BMI in young adults aged 18–25 years [28], and this was supported by Makarem et al. who found similar associations both cross-sectionally and longitudinally in women with cardiometabolic risk factors [31].
Given the negative health effects of eating irregularly, three randomized crossover trials have examined the effects of a consistent versus an inconsistent number of meals per day on the cardiometabolic health of lean [54,59] and overweight or obese women [53]. A regular meal pattern where participants were provided six meals a day for 14 days had a beneficial impact on peak insulin and fasting total and LDL-cholesterol levels when compared to an irregular meal pattern varying from three to nine meals a day for the same duration of time [53,54]. Although only one study out of all of the above looked specifically at meal timing variability in young adults [28] overall, they all show that following eating patterns with large variability has negative implications with regard to various health outcomes.
In contrast, McHill et al. found that young adults who are overweight or obese have a higher stability in meal timing across months [21], which contradicts the findings of previous studies [28,34]. The authors explained that this may have been due to the lack of consistency in the definition of meal timing stability [21]. Two large studies found no statistically significant association between meal time regularity and cardiometabolic outcomes such as metabolic syndrome [36] and BMI [24]. One of these studies was conducted in a sample of 5337 Korean men aged 30 years and above [36], whereas the other was conducted in 1175 healthy UK adults aged 19–64 years [24].
The mechanisms behind the impact of meal timing variability on cardiometabolic health are yet to be fully understood due to the limited number of observational and intervention studies in humans [60]. However, it has been hypothesized that the timing and regularity of meals play an important role in entraining circadian rhythms [31]. Although the master clock in the brain is regulated by exposure to light, meal timing is a strong cue for peripheral clocks [61,62], which are present in nearly all organ systems [63,64]. These systems are involved in nutritionally-related metabolic processes such as glycolysis and gluconeogenesis, cholesterol and lipid metabolism, oxidative phosphorylation and detoxification [64,65]. Studies have shown that an irregular eating pattern, defined as an inconsistent number of EOs from day to day [59] or a change in the time of meal consumption [66], could lead to circadian misalignment [66], and has been linked with lower energy expenditure, greater hunger ratings, and lower fullness rating, resulting in a positive energy balance which potentially increases the risk of developing obesity and cardiometabolic diseases [59]. Animal studies have exposed genes that link circadian rhythms with metabolic regulation such as the circadian locomotor output cycles kaput gene in mutant mice. Mice lacking this gene were hyperphagic, obese, and rapidly developed metabolic syndrome [67]. The relationship between food intake and circadian rhythms is reciprocal—circadian rhythms drive changes in metabolic pathways and changes in metabolic pathways alter molecular components of circadian rhythms [68,69].
Having a high variability in eating patterns also has implications for dietary assessment. Traditional dietary assessment methods that rely on human memory to initiate the recording of food intake or recall foods consumed such as food diaries and 24-h recalls are high burden and subject to memory bias [70,71]. Important information such as foods consumed at breakfast may be omitted as its timing may vary considerably from day-to-day. Increasing the frequency of data collection throughout the day such as progressive 24-h recalls [72] or collecting dietary information near real-time may be warranted for this cohort. This can be achieved with the use of time-triggered ecological momentary assessment (EMA) to capture food intake and contextual data at pre-determined times personalized to subjects’ usual pattern of food intake [73,74]. Although not yet commercially available nor feasible for use in practice settings [75], wearable sensors may also be useful for capturing real-time eating behavior [76]. Sensors embedded in smartwatches can be used to detect eating-related behaviors such as hand-to-mouth movements to deliver timely prompts via a connected device such as a smartphone to remind users to record their dietary intake, minimizing inaccurate dietary recalls because of memory decay [77,78].

4.3. Strengths and Limitations

A key strength of our study is the use of wearable cameras. To the best of our knowledge, our paper is the first to use automated wearable cameras to objectively measure the intra-individual variability of meal timing in young adults. Previous studies have either used methods of data collection that are subjective [4,24,28,31,34,35,55] and require user-initiation [21,30] or examined meal timing in terms of when meals and snacks are consumed rather than its consistency or variability across days [79].
Our study had several limitations, the most notable being the small sample size. Our results are likely not generalizable to the larger population, and one example of this is the disparity between the proportion of our sample who had overweight and obese BMIs and the young adult Australian population (32% versus 46%) [80]. Another limitation is the exclusion of beverages from our analysis despite beverages contributing up to one quarter of total daily energy intake [81,82]. Including nutritive beverages may change the intra-individual variability of consumption timing. The data that we collected may not be representative of participants’ usual intake, as they only wore the camera for three days, may have altered their eating behavior on the days the camera was worn, or chosen to wear the camera on days that were more convenient such as at home or away from social settings. Week-long cross-sectional studies are likely needed to capture habitual eating behaviors [21].

5. Conclusions

Our results contribute to the literature on the timing of eating patterns in young adults and highlight the urgent need to develop more objective methods of dietary assessment that rely less on human recall. Methods that are based on data capture around the traditional structure of meal times are not appropriate for this cohort. However, to date, some of our innovations in the use of technology via apps and websites for dietary data collection have still maintained this structure. While wearable cameras such as those used in this paper are an objective means of recording the holistic food and beverage consumption process, their widescale use in the general population is not feasible. Signal-contingent EMA delivered on a personalized schedule to capture near real-time dietary and contextual data or event-contingent EMA triggered by the detection of eating-related behaviors via wearable sensors are some methods that may be appropriate for capturing dietary data with high variability.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/nu14204349/s1, Table S1: Mean and range of eating pattern metrics across weekdays and weekends; Table S2: Associations between day-to-day variability and total energy intake (kJ) across three days.

Author Contributions

Conceptualization, A.R. and M.A.-F.; methodology, A.R., V.C., A.D. and L.W.-C.; data curation, V.C., A.D. and L.W.-C.; formal analysis, L.W.; visualization, A.R. and M.A.-F.; writing—original draft preparation, L.W.; writing—review and editing, A.R. and M.A.-F.; supervision, A.R and M.A.-F.; project administration, M.A.-F. All authors have read and agreed to the published version of the manuscript.

Funding

This research analysis received no external funding. However, the main study that collected the data was funded by an Australian Research Council Linkage grant LP150100831 with Cancer Council NSW.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Human Research Ethics Committee of the University of Sydney (2016/546) on the 15 July 2016.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding author subject to ethical approval.

Acknowledgments

The authors thank all of the investigators of the original MYMeals study. The authors also acknowledge the technical assistance of Alex Shaw of the Sydney Informatics Hub, a Core Research Facility of the University of Sydney.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

  1. Dashti, H.S.; Scheer, F.A.J.L.; Saxena, R.; Garaulet, M. Timing of Food Intake: Identifying Contributing Factors to Design Effective Interventions. Adv. Nutr. 2019, 10, 606–620. [Google Scholar] [CrossRef] [PubMed]
  2. Winkler, G.; Döring, A.; Keil, U. Meal patterns in middle-aged men in Southern Germany: Results from the MONICA Augsburg dietary survey 1984/85. Appetite 1999, 32, 33–37. [Google Scholar] [PubMed] [Green Version]
  3. Kant, A.K.; Graubard, B.I. 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, 661–670. [Google Scholar] [PubMed] [Green Version]
  4. Leech, R.M.; Worsley, A.; Timperio, A.; McNaughton, S.A. Temporal eating patterns: A latent class analysis approach. Int. J. Behav. Nutr. Phys. Act. 2017, 14, 3. [Google Scholar] [CrossRef] [Green Version]
  5. Gill, S.; Panda, S. A Smartphone App Reveals Erratic Diurnal Eating Patterns in Humans that Can Be Modulated for Health Benefits. Cell Metab. 2015, 22, 789–798. [Google Scholar]
  6. Paoli, A.; Tinsley, G.; Bianco, A.; Moro, T. The Influence of Meal Frequency and Timing on Health in Humans: The Role of Fasting. Nutrients 2019, 11, 719. [Google Scholar]
  7. St-Onge, M.-P.; Ard, J.; Baskin, M.L.; Chiuve, S.E.; Johnson, H.M.; 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, e96–e121. [Google Scholar] [CrossRef]
  8. Templeman, I.; Gonzalez, J.T.; Thompson, D.; Betts, J.A. The role of intermittent fasting and meal timing in weight management and metabolic health. Proc. Nutr. Soc. 2020, 79, 76–87. [Google Scholar] [CrossRef] [Green Version]
  9. Maugeri, A.; Vinciguerra, M. The Effects of Meal Timing and Frequency, Caloric Restriction, and Fasting on Cardiovascular Health: An Overview. J. Lipid Atheroscler. 2020, 9, 140–152. [Google Scholar]
  10. Ballon, A.; Neuenschwander, M.; Schlesinger, S. Breakfast Skipping Is Associated with Increased Risk of Type 2 Diabetes among Adults: A Systematic Review and Meta-Analysis of Prospective Cohort Studies. J. Nutr. 2018, 149, 106–113. [Google Scholar]
  11. 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, 1–8. [Google Scholar] [CrossRef]
  12. Wicherski, J.; Schlesinger, S.; Fischer, F. Association between Breakfast Skipping and Body Weight—A Systematic Review and Meta-Analysis of Observational Longitudinal Studies. Nutrients 2021, 13, 272. [Google Scholar] [CrossRef]
  13. Bonnet, J.P.; Cardel, M.I.; Cellini, J.; Hu, F.B.; Guasch-Ferré, M. Breakfast Skipping, Body Composition, and Cardiometabolic Risk: A Systematic Review and Meta-Analysis of Randomized Trials. Obesity 2020, 28, 1098–1109. [Google Scholar] [CrossRef]
  14. Kutsuma, A.; Nakajima, K.; Suwa, K. Potential Association between Breakfast Skipping and Concomitant Late-Night-Dinner Eating with Metabolic Syndrome and Proteinuria in the Japanese Population. Scientifica 2014, 2014, 253581. [Google Scholar] [CrossRef]
  15. Berg, C.; Lappas, G.; Wolk, A.; Strandhagen, E.; Torén, K.; Rosengren, A.; Thelle, D.; Lissner, L. Eating patterns and portion size associated with obesity in a Swedish population. Appetite 2009, 52, 21–26. [Google Scholar]
  16. Marinac, C.R.; Sears, D.D.; Natarajan, L.; Gallo, L.C.; Breen, C.I.; Patterson, R.E. Frequency and Circadian Timing of Eating May Influence Biomarkers of Inflammation and Insulin Resistance Associated with Breast Cancer Risk. PLoS ONE 2015, 10, e0136240. [Google Scholar] [CrossRef] [Green Version]
  17. Cahill, L.E.; Chiuve, S.E.; Mekary, R.A.; Jensen, M.K.; Flint, A.J.; Hu, F.B.; Rimm, E.B. Prospective study of breakfast eating and incident coronary heart disease in a cohort of male US health professionals. Circulation 2013, 128, 337–343. [Google Scholar]
  18. Tsuchida, Y.; Hata, S.; Sone, Y. Effects of a late supper on digestion and the absorption of dietary carbohydrates in the following morning. J. Physiol. Anthropol. 2013, 32, 9. [Google Scholar] [CrossRef] [Green Version]
  19. Pelletier, J.E.; Laska, M.N. Balancing healthy meals and busy lives: Associations between work, school, and family responsibilities and perceived time constraints among young adults. J. Nutr. Educ. Behav. 2012, 44, 481–489. [Google Scholar]
  20. Welton, S.; Minty, R.; O’Driscoll, T.; Willms, H.; Poirier, D.; Madden, S.; Kelly, L. Intermittent fasting and weight loss: Systematic review. Can. Fam. Phys. 2020, 66, 117–125. [Google Scholar]
  21. McHill, A.W.; Hilditch, C.J.; Fischer, D.; Czeisler, C.A.; Garaulet, M.; Scheer, F.A.J.L.; Klerman, E.B. Stability of the timing of food intake at daily and monthly timescales in young adults. Sci. Rep. 2020, 10, 20849. [Google Scholar] [CrossRef]
  22. Culnan, E.; Reid, K.J.; Zee, P.C.; Crowley, S.J.; Baron, K.G. Meal timing relative to DLMO: Associations with BMI and body fat. Sleep Health 2021, 7, 339–344. [Google Scholar]
  23. Reid, K.J.; Baron, K.G.; Zee, P.C. Meal timing influences daily caloric intake in healthy adults. Nutr. Res. 2014, 34, 930–935. [Google Scholar] [CrossRef] [Green Version]
  24. Zimmerman, A.R.; Johnson, L.; Brunstrom, J.M. Assessing “chaotic eating” using self-report and the UK Adult National Diet and Nutrition Survey: No association between BMI and variability in meal or snack timings. Physiol. Behav. 2018, 192, 64–71. [Google Scholar] [CrossRef] [Green Version]
  25. Meth, E.M.S.; van Egmond, L.T.; Moulin, T.C.; Cedernaes, J.; Rosqvist, F.; Benedict, C. Association of Daily Eating Duration and Day-To-Day Variability in the Timing of Eating With Fatal Cancer Risk in Older Men. Front. Nutr. 2022, 9, 889926. [Google Scholar]
  26. Gupta, N.J.; Kumar, V.; Panda, S. A camera-phone based study reveals erratic eating pattern and disrupted daily eating-fasting cycle among adults in India. PLoS ONE 2017, 12, e0172852. [Google Scholar] [CrossRef] [Green Version]
  27. Zhao, L.; Teong, X.T.; Liu, K.; Liu, B.; Melaku, Y.A.; Vincent, A.; Manoogian, E.; Panda, S.; Wittert, G.A.; Hutchinson, A.; et al. Eating architecture in adults at increased risk of type 2 diabetes: Associations with body fat and glycemic control. Br. J. Nutr. 2021, 128, 324–333. [Google Scholar] [CrossRef]
  28. Zerón-Rugerio, M.F.; Hernáez, Á.; Porras-Loaiza, A.P.; Cambras, T.; Izquierdo-Pulido, M. Eating Jet Lag: A Marker of the Variability in Meal Timing and Its Association with Body Mass Index. Nutrients 2019, 11, 2980. [Google Scholar] [CrossRef] [Green Version]
  29. Wittig, F.; Hummel, E.; Wenzler, G.; Heuer, T. Energy and macronutrient intake over the course of the day of German adults: A DEDIPAC-study. Appetite 2017, 114, 125–136. [Google Scholar] [CrossRef]
  30. Popp, C.J.; Curran, M.; Wang, C.; Prasad, M.; Fine, K.; Gee, A.; Nair, N.; Perdomo, K.; Chen, S.; Hu, L.; et al. Temporal Eating Patterns and Eating Windows among Adults with Overweight or Obesity. Nutrients 2021, 13, 4485. [Google Scholar] [PubMed]
  31. Makarem, N.; Sears, D.D.; St-Onge, M.P.; Zuraikat, F.M.; Gallo, L.C.; Talavera, G.A.; Castaneda, S.F.; Lai, Y.; Aggarwal, B. Variability in Daily Eating Patterns and Eating Jetlag Are Associated With Worsened Cardiometabolic Risk Profiles in the American Heart Association Go Red for Women Strategically Focused Research Network. J. Am. Heart Assoc. 2021, 10, e022024. [Google Scholar] [CrossRef] [PubMed]
  32. Fleischer, J.G.; Das, S.K.; Bhapkar, M.; Manoogian, E.N.C.; Panda, S. Associations between the timing of eating and weight-loss in calorically restricted healthy adults: Findings from the CALERIE study. Exp. Gerontol. 2022, 165, 111837. [Google Scholar] [CrossRef] [PubMed]
  33. Makarem, N.; Zuraikat, F.M.; Caceres, B.; Sears, D.D.; St-Onge, M.P.; Lai, Y.; Aggarwal, B. Variable Eating Patterns: A Potential Novel Risk Factor for Systemic Inflammation in Women. Ann. Behav. Med. 2022, kaac042. [Google Scholar] [CrossRef]
  34. Eicher-Miller, H.A.; Khanna, N.; Boushey, C.J.; Gelfand, S.B.; Delp, E.J. Temporal Dietary Patterns Derived among the Adult Participants of the National Health and Nutrition Examination Survey 1999–2004 Are Associated with Diet Quality. J. Acad. Nutr. Diet. 2016, 116, 283–291. [Google Scholar] [CrossRef] [Green Version]
  35. Mesas, A.E.; Muñoz-Pareja, M.; López-García, E.; Rodríguez-Artalejo, F. Selected eating behaviours and excess body weight: A systematic review. Obes. Rev. 2012, 13, 106–135. [Google Scholar] [CrossRef]
  36. Shin, A.; Lim, S.Y.; Sung, J.; Shin, H.R.; Kim, J. Dietary intake, eating habits, and metabolic syndrome in Korean men. J. Am. Diet. Assoc. 2009, 109, 633–640. [Google Scholar] [CrossRef]
  37. Sierra-Johnson, J.; Undén, A.L.; Linestrand, M.; Rosell, M.; Sjogren, P.; Kolak, M.; De Faire, U.; Fisher, R.M.; Hellénius, M.L. Eating meals irregularly: A novel environmental risk factor for the metabolic syndrome. Obesity 2008, 16, 1302–1307. [Google Scholar] [CrossRef]
  38. Wennberg, M.; Gustafsson, P.E.; Wennberg, P.; Hammarström, A. Irregular eating of meals in adolescence and the metabolic syndrome in adulthood: Results from a 27-year prospective cohort. Public Health Nutr. 2016, 19, 667–673. [Google Scholar] [CrossRef] [Green Version]
  39. König, L.M.; Van Emmenis, M.; Nurmi, J.; Kassavou, A.; Sutton, S. Characteristics of smartphone-based dietary assessment tools: A systematic review. Health Psychol. Rev. 2021, 1–25. [Google Scholar] [CrossRef]
  40. Chan, V.; Davies, A.; Wellard-Cole, L.; Lu, S.; Ng, H.; Tsoi, L.; Tiscia, A.; Signal, L.; Rangan, A.; Gemming, L.; et al. Using Wearable Cameras to Assess Foods and Beverages Omitted in 24 Hour Dietary Recalls and a Text Entry Food Record App. Nutrients 2021, 13, 1806. [Google Scholar] [CrossRef]
  41. Chan, V.; Wellard-Cole, L.; Davies, A.; Watson, W.; Hughes, C.; Chapman, K.; Signal, L.; Ni Mhurchu, C.; Wang, L.; D’Souza, D. The association of social and food preparation location context with the quality of meals and snacks consumed by young adults: Findings from the MYMeals wearable camera study. Eur. J. Nutr. 2022, 61, 3407–3422. [Google Scholar] [CrossRef] [PubMed]
  42. Wellard-Cole, L.; Davies, A.; Chen, J.; Jung, J.; Bente, K.B.; Kay, J.; Watson, W.L.; Hughes, C.; Rangan, A.; Yacef, K. The Contribution of Foods Prepared Outside the Home to the Diets of 18- to 30-Year-Old Australians: The MYMeals Study. Nutrients 2021, 13, 1761. [Google Scholar] [CrossRef] [PubMed]
  43. Davies, A.; Allman-Farinelli, M.; Owen, K.; Signal, L.; Hosking, C.; Wang, L.; Bauman, A. Feasibility Study Comparing Physical Activity Classifications from Accelerometers with Wearable Camera Data. Int. J. Environ. Res. Public Health 2020, 17, 9323. [Google Scholar] [CrossRef] [PubMed]
  44. Davies, A.; Chan, V.; Bauman, A.; Signal, L.; Hosking, C.; Gemming, L.; Allman-Farinelli, M. Using wearable cameras to monitor eating and drinking behaviours during transport journeys. Eur. J. Nutr. 2021, 60, 1875–1885. [Google Scholar] [CrossRef]
  45. Wellard-Cole, L.; Potter, M.; Jung, J.J.; Chen, J.; Kay, J.; Allman-Farinelli, M. A Tool to Measure Young Adults’ Food Intake: Design and Development of an Australian Database of Foods for the Eat and Track Smartphone App. JMIR Mhealth Uhealth 2018, 6, e12136. [Google Scholar] [CrossRef]
  46. Park, Y.; Dodd, K.W.; Kipnis, V.; Thompson, F.E.; Potischman, N.; Schoeller, D.A.; Baer, D.J.; Midthune, D.; Troiano, R.P.; Bowles, H.; et al. Comparison of self-reported dietary intakes from the Automated Self-Administered 24-h recall, 4-d food records, and food-frequency questionnaires against recovery biomarkers. Am. J. Clin. Nutr. 2018, 107, 80–93. [Google Scholar] [CrossRef] [Green Version]
  47. Wellard-Cole, L.; Jung, J.; Kay, J.; Rangan, A.; Chapman, K.; Watson, W.L.; Hughes, C.; Ni Mhurchu, C.; Bauman, A.; Gemming, L. Examining the Frequency and Contribution of Foods Eaten Away from Home in the Diets of 18- to 30-Year-Old Australians Using Smartphone Dietary Assessment (MYMeals): Protocol for a Cross-Sectional Study. JMIR Res. Protoc. 2018, 7, e24. [Google Scholar] [CrossRef] [Green Version]
  48. Australian Bureau of Statistics. Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA). 2016. Available online: http://www.abs.gov.au/ausstats/[email protected]/mf/2033.0.55.001 (accessed on 19 February 2022).
  49. Davies, A.; Wellard-Cole, A.; Rangan, A.; Allman-Farinelli, M. Validity of self-reported weight and height for BMI classification: A cross-sectional study among young adults. Nutrition 2020, 71, 110622. [Google Scholar] [CrossRef]
  50. Harris, P.A.; Taylor, R.; Thielke, R.; Payne, J.; Gonzalez, N.; Conde, J.G. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J. Biomed. Inform. 2009, 42, 377–381. [Google Scholar] [CrossRef] [Green Version]
  51. Fischer, D.; Vetter, C.; Roenneberg, T. A novel method to visualise and quantify circadian misalignment. Sci. Rep. 2016, 6, 38601. [Google Scholar] [CrossRef]
  52. Fischer, D.; McHill, A.W.; Sano, A.; Picard, R.W.; Barger, L.K.; Czeisler, C.A.; Klerman, E.B.; Phillips, A.J.K. Irregular sleep and event schedules are associated with poorer self-reported well-being in US college students. Sleep 2020, 43, zsz300. [Google Scholar] [CrossRef]
  53. Farshchi, H.R.; Taylor, M.A.; Macdonald, I.A. Beneficial metabolic effects of regular meal frequency on dietary thermogenesis, insulin sensitivity, and fasting lipid profiles in healthy obese women. Am. J. Clin. Nutr. 2005, 81, 16–24. [Google Scholar] [CrossRef] [Green Version]
  54. Farshchi, H.R.; Taylor, M.A.; Macdonald, I.A. Regular meal frequency creates more appropriate insulin sensitivity and lipid profiles compared with irregular meal frequency in healthy lean women. Eur. J. Clin. Nutr. 2004, 58, 1071–1077. [Google Scholar] [CrossRef]
  55. Zeballos, E.; Chelius, C. The effects of grazing on daily caloric intake and dietary quality. Int. J. Behav. Nutr. Phys. Act. 2021, 18, 163. [Google Scholar] [CrossRef]
  56. Fischer, D.; Klerman, E.B.; Phillips, A.J.K. Measuring sleep regularity: Theoretical properties and practical usage of existing metrics. Sleep 2021, 44, zsab103. [Google Scholar] [CrossRef]
  57. Claudino, J.G.; Gabbet, T.J.; de Sá Souza, H.; Simim, M.; Fowler, P.; de Alcantara Borba, D.; Melo, M.; Bottino, A.; Loturco, I.; D’Almeida, V.; et al. Which parameters to use for sleep quality monitoring in team sport athletes? A systematic review and meta-analysis. BMJ Open Sport Exerc. Med. 2019, 5, e000475. [Google Scholar] [CrossRef] [Green Version]
  58. Huang, T.; Mariani, S.; Redline, S. Sleep Irregularity and Risk of Cardiovascular Events: The Multi-Ethnic Study of Atherosclerosis. J. Am. Coll. Cardiol. 2020, 75, 991–999. [Google Scholar]
  59. Alhussain, M.H.; Macdonald, I.A.; Taylor, M.A. Irregular meal-pattern effects on energy expenditure, metabolism, and appetite regulation: A randomized controlled trial in healthy normal-weight women. Am. J. Clin. Nutr. 2016, 104, 21–32. [Google Scholar] [CrossRef] [Green Version]
  60. Pot, G.K.; Almoosawi, S.; Stephen, A.M. Meal irregularity and cardiometabolic consequences: Results from observational and intervention studies. Proc. Nutr. Soc. 2016, 75, 475–486. [Google Scholar] [CrossRef] [Green Version]
  61. Drapeau, V.; Gallant, A.R. Homeostatic and Circadian Control of Food Intake: Clinical Strategies to Prevent Overconsumption. Curr. Obes. Rep. 2013, 2, 93–103. [Google Scholar]
  62. Stokkan, K.A.; Yamazaki, S.; Tei, H.; Sakaki, Y.; Menaker, M. Entrainment of the circadian clock in the liver by feeding. Science 2001, 291, 490–493. [Google Scholar] [CrossRef] [Green Version]
  63. Tahara, Y.; Shibata, S. Chronobiology and nutrition. Neuroscience 2013, 253, 78–88. [Google Scholar] [CrossRef]
  64. Green, C.B.; Takahashi, J.S.; Bass, J. The meter of metabolism. Cell 2008, 134, 728–742. [Google Scholar] [CrossRef] [Green Version]
  65. Johnston, J.D. Physiological links between circadian rhythms, metabolism and nutrition. Exp. Physiol. 2014, 99, 1133–1137. [Google Scholar] [CrossRef] [Green Version]
  66. Wehrens, S.M.T.; Christou, S.; Isherwood, C.; Middleton, B.; Gibbs, M.A.; Archer, S.N.; Skene, D.J.; Johnston, J.D. Meal Timing Regulates the Human Circadian System. Curr. Biol. 2017, 27, 1768–1775.e3. [Google Scholar] [CrossRef] [Green Version]
  67. Turek, F.W.; Joshu, C.; Kohsaka, A.; Lin, E.; Ivanova, G.; McDearmon, E.; Laposky, A.; Losee-Olson, S.; Easton, A.; Jensen, D.R.; et al. Obesity and metabolic syndrome in circadian Clock mutant mice. Science 2005, 308, 1043–1045. [Google Scholar] [CrossRef] [Green Version]
  68. Johnston, J.D. Physiological responses to food intake throughout the day. Nutr. Res. Rev. 2014, 27, 107–118. [Google Scholar] [CrossRef] [Green Version]
  69. Guan, D.; Lazar, M.A. Interconnections between circadian clocks and metabolism. J. Clin. Investig. 2021, 131, e148278. [Google Scholar] [CrossRef]
  70. Livingstone, M.B.; Robson, P.J.; Wallace, J.M. Issues in dietary intake assessment of children and adolescents. Br. J. Nutr. 2004, 92 (Suppl. 2), S213–S222. [Google Scholar] [CrossRef] [Green Version]
  71. Westerterp, K.R.; Goris, A.H. Validity of the assessment of dietary intake: Problems of misreporting. Curr. Opin. Clin. Nutr. Metab. Care 2002, 5, 489–493. [Google Scholar] [CrossRef]
  72. Osadchiy, T.; Poliakov, I.; Olivier, P.; Rowland, M.; Foster, E. Progressive 24-Hour Recall: Usability Study of Short Retention Intervals in Web-Based Dietary Assessment Surveys. J. Med. Internet Res. 2020, 22, e13266. [Google Scholar] [CrossRef] [PubMed]
  73. Battaglia, B.; Lee, L.; Jia, S.S.; Partridge, S.R.; Allman-Farinelli, M. The Use of Mobile-Based Ecological Momentary Assessment (mEMA) Methodology to Assess Dietary Intake, Food Consumption Behaviours and Context in Young People: A Systematic Review. Healthcare 2022, 10, 1329. [Google Scholar] [CrossRef] [PubMed]
  74. Schembre, S.M.; Liao, Y.; O-Connor, S.G.; Hingle, M.D.; Shen, S.E.; Hamoy, K.G.; Huh, J.; Dunton, G.F.; Weiss, R.; Thomson, C.A.; et al. Mobile Ecological Momentary Diet Assessment Methods for Behavioral Research: Systematic Review. JMIR Mhealth Uhealth 2018, 6, e11170. [Google Scholar] [CrossRef] [Green Version]
  75. Wang, L.; Allman-Farinelli, M.; Yang, J.A.; Taylor, J.C.; Gemming, L.; Hekler, E.; Rangan, A. Enhancing Nutrition Care Through Real-Time, Sensor-Based Capture of Eating Occasions: A Scoping Review. Front. Nutr. 2022, 9, 852984. [Google Scholar] [CrossRef] [PubMed]
  76. Crochiere, R.J.; Zhang, F.Z.; Juarascio, A.S.; Goldstein, S.P.; Thomas, J.G.; Forman, E.M. Comparing ecological momentary assessment to sensor-based approaches in predicting dietary lapse. Transl. Behav. Med. 2021, 11, 2099–2109. [Google Scholar] [CrossRef] [PubMed]
  77. Thompson, F.E.; Subar, A.F.; Loria, C.M.; Reedy, J.L.; Baranowski, T. Need for technological innovation in dietary assessment. J. Am. Diet. Assoc. 2010, 110, 48–51. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  78. Bell, B.M.; Alam, R.; Mondol, A.S.; Ma, M.; Emi, I.A.; Preum, S.M.; de la Haye, K.; Stankovic, J.A.; Lach, J.; Spruijt-Metz, D. Validity and Feasibility of the Monitoring and Modeling Family Eating Dynamics System to Automatically Detect In-field Family Eating Behavior: Observational Study. JMIR Mhealth Uhealth 2022, 10, e30211. [Google Scholar] [CrossRef] [PubMed]
  79. Gage, R.; Girling-Butcher, M.; Joe, E.; Smith, M.; Ni Mhurchu, C.; McKerchar, C.; Puloka, V.; McLean, R.; Signal, L. The Frequency and Context of Snacking among Children: An Objective Analysis Using Wearable Cameras. Nutrients 2020, 13, 103. [Google Scholar] [CrossRef]
  80. Australian Bureau of Statistics. National Health Survey: First Results. 2018. Available online: https://www.abs.gov.au/statistics/health/health-conditions-and-risks/national-health-survey-first-results/latest-release (accessed on 31 August 2022).
  81. Riley, M.D.; Hendrie, G.A.; Baird, D.L. Drink Choice is Important: Beverages Make a Substantial Contribution to Energy, Sugar, Calcium and Vitamin C Intake among Australians. Nutrients 2019, 11, 1389. [Google Scholar] [CrossRef] [Green Version]
  82. Malisova, O.; Bountziouka, V.; Zampelas, A.; Kapsokefalou, M. Evaluation of Drinks Contribution to Energy Intake in Summer and Winter. Nutrients 2015, 7, 3724–3738. [Google Scholar] [CrossRef]
Nutrients 14 04349 g001 550
Figure 1. Individual differences in eating window (time from the first eating occasion of the day until the last eating occasion of the day). Each color represents a separate participant. Each column represents one day. Solid columns represent weekdays and unfilled columns represent weekend days.
Figure 1. Individual differences in eating window (time from the first eating occasion of the day until the last eating occasion of the day). Each color represents a separate participant. Each column represents one day. Solid columns represent weekdays and unfilled columns represent weekend days.
Nutrients 14 04349 g001
Nutrients 14 04349 g002 550
Figure 2. Stacked bar chart showing the number of participants that fall into the low, moderate, high, and very high intra-individual variability categories for each metric. Numbers within each category represent the average daily energy intake (kJ) for participants in that category. CPD First: Composite Phase Deviation of the first eating occasion for each participant over the three days of data collection. CPD Last: Composite Phase Deviation of the last eating occasion for each participant over the three days of data collection. CV No. of EOs: Coefficient of variation for the total daily number of eating occasions for each participant over the three days of data collection. CV Eating Window: Coefficient of variation for the daily eating window for each participant over the three days of data collection.
Figure 2. Stacked bar chart showing the number of participants that fall into the low, moderate, high, and very high intra-individual variability categories for each metric. Numbers within each category represent the average daily energy intake (kJ) for participants in that category. CPD First: Composite Phase Deviation of the first eating occasion for each participant over the three days of data collection. CPD Last: Composite Phase Deviation of the last eating occasion for each participant over the three days of data collection. CV No. of EOs: Coefficient of variation for the total daily number of eating occasions for each participant over the three days of data collection. CV Eating Window: Coefficient of variation for the daily eating window for each participant over the three days of data collection.
Nutrients 14 04349 g002
Table
Table 1. Cut-offs used to categorize meal timing variability metric scores as low, moderate, high, and very high variability and the equivalent difference in number of hours or eating occasions (EOs) over three days.
Table 1. Cut-offs used to categorize meal timing variability metric scores as low, moderate, high, and very high variability and the equivalent difference in number of hours or eating occasions (EOs) over three days.
CPD (First and Last EO)CV No. of EOsCV Eating Window
Cut-offHoursCut-offEOsCut-offHours
Variability
Low<1.15<2 h<10%<1 EO<10%<2 h
Moderate1.15–1.702–3 h11–20%1–2 EOs11–20%2–4 h
High1.71–2.403–4 h21–30%2–3 EOs21–30%4–5 h
Very high>2.40> 4 h>30%>3 EOs>30%>5 h
Table
Table 2. Participant characteristics and differences between male and female participants.
Table 2. Participant characteristics and differences between male and female participants.
All Participants (n = 41)Female (n = 27)Male (n = 14)
Age (years)
18–2425187
25–301697
Body Mass Index (BMI, kg/m2)
<18.5000
≥18.5 < 2528208
≥25 < 30734
≥30642
Socio-economic status (SES)
Higher22139
Lower19145
Highest education attained
Secondary school or less15105
Trade or diploma770
University degree19109
Employment/study
Full-time study281810
Full-time work541
Part-time study/work532
Not studying or working321
Table
Table 3. Mean and range of eating pattern and intra-individual variability metrics.
Table 3. Mean and range of eating pattern and intra-individual variability metrics.
All Days (n = 123)
MeanMinMax
Eating pattern metrics
Time of first EO (hh:mm)10:1800:4619:52
Time of last EO (hh:mm)20:0612:4323:38
No. of EOs per day4.71.09.0
Daily eating window (h)9.80.322.2
Daily energy intake (kJ)847876022,879
Intra-individual variability metrics
CPD First (h)2.90.316.3
CPD Last (h)1.80.25.8
CV No. of EOs (%)28.30.078.1
CV Eating Window (%)25.61.6106.0
EO, eating occasion; CPD, Composite Phase Deviation; CV, coefficient of variation. CPD First: CPD (i.e., average deviation in hours from a perfectly regular pattern of meal timing) of the first EO for each participant over the three days of data collection. CPD Last: CPD of the last EO for each participant over the three days of data collection. CV No. of EOs: CV for the total daily number of EOs for each participant over the three days of data collection. CV Eating Window: CV for the daily eating window (the duration of time between the first and last EOs of the same calendar day) for each participant over the three days of data collection.
Table
Table 4. Associations between eating pattern metrics and Body Mass Index (BMI, kg/m2).
Table 4. Associations between eating pattern metrics and Body Mass Index (BMI, kg/m2).
Mean ± SDp-Value
Body Mass Index (BMI)<25 kg/m2 (n = 28)≥25 kg/m2 (n = 13)
Time of first EO (hh:mm)10.58 ± 2.1149.777 ± 1.0210.338
Time of last EO (hh:mm)20.00 ± 1.42820.41 ± 1.1160.589
No. of EOs per day4.726 ± 1.2474.615 ± 0.9890.709
Daily eating window (h)9.419 ± 2.46310.63 ± 1.4230.195
Daily energy intake (kJ)8169 ± 22589143 ± 24870.311
Table
Table 5. Associations between day-to-day variability and demographic variables.
Table 5. Associations between day-to-day variability and demographic variables.
Mean ± SDp-Value
GenderMale (n = 14)Female (n = 27)
CPD First (h)4.128 ± 4.6612.295 ± 1.7720.176
CPD Last (h)1.555 ± 0.8052.009 ± 1.3750.263
CV No. of EOs (%)22.96 ± 9.90931.00 ± 17.5380.122
CV Eating Window (%)24.82 ± 22.5925.98 ± 24.430.883
Body Mass Index (BMI)<25 kg/m2 (n = 28)≥25 kg/m2 (n = 13)
CPD First (h)2.466 ± 1.7843.903 ± 4.9220.325
CPD Last (h)1.843 ± 1.3101.879 ± 1.0450.932
CV No. of EOs (%)30.54 ± 16.7523.32 ± 12.360.174
CV Eating Window (%)24.68 ± 24.8727.53 ± 21.190.723
Socioeconomic Status (SES)Top five deciles (n = 22)Bottom five deciles (n = 19)
CPD First (h)3.786 ± 4.0121.920 ± 1.0910.047 *
CPD Last (h)1.890 ± 1.3271.813 ± 1.1150.843
CV No. of EOs (%)27.58 ± 14.9729.03 ± 16.900.773
CV Eating Window (%)28.07 ± 24.8622.70 ± 22.230.474
* Significance at p ≤ 0.05
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Wang, L.; Chan, V.; Allman-Farinelli, M.; Davies, A.; Wellard-Cole, L.; Rangan, A. Wearable Cameras Reveal Large Intra-Individual Variability in Timing of Eating among Young Adults. Nutrients 2022, 14, 4349. https://doi.org/10.3390/nu14204349

AMA Style

Wang L, Chan V, Allman-Farinelli M, Davies A, Wellard-Cole L, Rangan A. Wearable Cameras Reveal Large Intra-Individual Variability in Timing of Eating among Young Adults. Nutrients. 2022; 14(20):4349. https://doi.org/10.3390/nu14204349

Chicago/Turabian Style

Wang, Leanne, Virginia Chan, Margaret Allman-Farinelli, Alyse Davies, Lyndal Wellard-Cole, and Anna Rangan. 2022. "Wearable Cameras Reveal Large Intra-Individual Variability in Timing of Eating among Young Adults" Nutrients 14, no. 20: 4349. https://doi.org/10.3390/nu14204349

APA Style

Wang, L., Chan, V., Allman-Farinelli, M., Davies, A., Wellard-Cole, L., & Rangan, A. (2022). Wearable Cameras Reveal Large Intra-Individual Variability in Timing of Eating among Young Adults. Nutrients, 14(20), 4349. https://doi.org/10.3390/nu14204349

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop