Chrononutrition Patterns in People Who Attempted Weight Loss in the Past Year: A Descriptive Analysis of the National Health and Nutrition Examination Survey (NHANES) 2017–2020 Pre-Pandemic

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article is excellently written. Furthermore, as a reviewer, I have no additional comments. My only question is why the p-values for the Chrononutrition Profile are not displayed in Tables 4 and 5?

 

Author Response

Comment 1: This article is excellently written. Furthermore, as a reviewer, I have no additional comments. My only question is why the p-values for the Chrononutrition Profile are not displayed in Tables 4 and 5?

Response 1: Thank you for your comments. The chrononutrition profile is treated as a categorical variable. We revised tables 4, 5, and 6 to include the corresponding p-values from the chi-square test.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you very much for the opportunity to review the manuscript entitled “Chrononutrition Patterns in People Who Attempted Weight 2 Loss in the Past Year: A Descriptive Analysis of the National 3 Health and Nutrition Examination Survey (NHANES) 2017-4 2020 Pre-Pandemic”.

This study by Kim et al. examines chrononutrition patterns in a nationally representative sample of U.S. adults who attempted weight loss through dietary modifications. Using NHANES 2017–2020 data, the authors analyzed meal timing, frequency, and caloric distribution in relation to weight change and adiposity. Key findings suggest that individuals who gained weight tended to have a longer delay between waking and their first meal, consumed fewer calories later in the day, and ate less frequently. Additionally, participants with obesity had a shorter eating window but also delayed their first meal and ate less frequently. These results highlight the potential role of chrononutrition in personalized weight management strategies.

Overall, the study addresses an interesting and timely topic. The manuscript is generally very well written and clearly structured throughout. However, I have a few questions and some comments regarding ways to further improve the clarity and impact of the paper.

Please find my specific comments below.

1) The introduction is well-structured and logically presents the rationale for the study. However, certain claims could be strengthened by specifying key findings from cited literature. For example:

- line 37-40: Consider briefly mentioning the efficacy of these interventions based on prior research.

- line 46-48: This is a strong statement, but it would be useful to briefly explain how these clocks influence metabolism (e.g., regulating insulin secretion, lipid metabolism).

- line 50-52: Consider adding a reference to a specific study that demonstrated the metabolic consequences of late-night eating.

- line 68-71: The study used BMI for subgroup analysis to assess variations in dietary patterns and meal timing strategies. However, BMI may not fully capture differences in body composition, particularly in distinguishing between fat mass and lean mass. Given that body fat percentage is a more precise indicator of metabolic risk and obesity-related health outcomes, its inclusion would strengthen the analysis. The authors should discuss the limitations of using BMI for subgroup analysis and , if possible, consider incorporating or referencing body fat percentage as a more relevant measure of adiposity.

2) Methods

- line 76: The phrasing "collected prior to the COVID-19 pandemic" could be misleading since NHANES 2020 was disrupted by the pandemic. Were data from 2020 fully available or only partial? This should be clarified.

- Diet assessment: The methodology for defining eating episodes is well explained, but it is unclear how meal skipping was determined (e.g., was an absence of an eating episode within a certain timeframe classified as skipping?).

- Diet assessment: Were there any limitations in dietary recall accuracy? Self-reporting may introduce several sources of bias, including recall bias, underreporting or social desirability. A brief mention of potential biases and accuracy of the assessment method would strengthen the section.

- Weight loss attempt/adiposity: See previous comment on BMI. The use of BMI as a primary measure of adiposity is a limitation, as BMI does not distinguish between fat mass and lean mass. A brief discussion of why body fat percentage was not used (or why BMI was preferred) would be valuable.

- Covariates: The categorization of physical activity follows WHO guidelines, but did the study also consider sedentary time? Including sedentary behavior could provide additional context for weight loss and chrononutrition patterns. Moreover, recent research shows that even lower physical activity levels than the standard recommendation of 150 min moderate or 75 min vigorous exercise per week can improve cardiometabolic health. Thus, the sharp cutoff used in this study to classify participants as physically active (≥150 min moderate or ≥75 min vigorous per week) may overlook individuals who engage in lower but still beneficial levels of physical activity. It would be valuable for the authors to discuss this limitation.

3) Results

- Effect sizes should be included in addition to p-values where possible to show the strength of observed differences.

- The classification of eating profiles (Early Eating, Later Eating, etc.) is useful, but the rationale for defining these categories based on specific cutoff points is unclear: was this data-driven (e.g., clustering analysis) or based on prior research?

-Physical Activity: Since physical activity and total calorie intake differ between groups, were these factors adjusted for in the analyses? And, the extended eating window profile is linked to weight gain, but is this association independent of total energy intake and physical activity?

4) Discussion

- Some comparisons, such as the difference in total caloric intake between weight loss attempters and non-attempters (1,966 vs. 2,066 kcal), appear numerically small. The authors should discuss (based on references) whether this difference is clinically meaningful or simply statistically significant.

- The authors emphasize that the present study builds on prior research (Farsijani et al.) but they do not clearly explain how it advances the field beyond introducing new chrononutrition profiles. I suggest that the authors should highlight specific new insights gained from this study that were not present in earlier research.

- The discussion links weight gain and obesity to delayed first meals, fewer eating episodes, and extended eating windows, but it does not sufficiently address potential reverse causality (i.e., do these patterns cause weight gain, or do people with obesity develop these eating habits?). The authors should discuss some alternative explanations, such as stress, sleep patterns, and work schedules. Moreover, the discussion should briefly mention existing studies on time-restricted eating and intermittent fasting and clarify whether they have demonstrated clear benefits for weight loss.

- Socioeconomic factors and lifestyle constraints (e.g., shift work, caregiving responsibilities) are briefly mentioned but not deeply explored. I suggest that the authors should explicitly state that these variables could influence both eating patterns and weight status, making it difficult to isolate the effect of meal timing alone.

Author Response

Comment 1:  The introduction is well-structured and logically presents the rationale for the study. However, certain claims could be strengthened by specifying key findings from cited literature. For example:

 

- line 37-40: Consider briefly mentioning the efficacy of these interventions based on prior research.

Reponse 1: Thank you for your comment. We revised the introduction to summarize key findings from priors regarding the efficacy of diet intervention. The revised section now reads:

"Common dietary recommendations include caloric restriction, macronutrient adjustments, and portion control, while physical activity guidelines promote increased aerobic exercise and resistance training to enhance metabolic health. Both strategies have been shown to result in 5–10% weight loss and significant improvements in insulin sensitivity, blood pressure, and lipid profiles.”

 

Reference:

Jensen MD, Ryan DH, Apovian CM, Ard JD, Comuzzie AG, Donato KA, Hu FB, Hubbard VS, Jakicic JM, Kushner RF, Loria CM, Millen BE, Nonas CA, Pi-Sunyer FX, Stevens J, Stevens VJ, Wadden TA, Wolfe BM, Yanovski SZ; American College of Cardiology/American Heart Association Task Force on Practice Guidelines; Obesity Society. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and The Obesity Society. J Am Coll Cardiol. 2014 Jul 1;63(25 Pt B):2985-3023. doi: 10.1016/j.jacc.2013.11.004. Epub 2013 Nov 12. Erratum in: J Am Coll Cardiol. 2014 Jul 1;63(25 Pt B):3029-3030. PMID: 24239920.

 

Garber CE, Blissmer B, Deschenes MR, Franklin BA, Lamonte MJ, Lee IM, Nieman DC, Swain DP; American College of Sports Medicine. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc. 2011 Jul;43(7):1334-59. doi: 10.1249/MSS.0b013e318213fefb. PMID: 21694556.

 

Response 2: line 46-48: This is a strong statement, but it would be useful to briefly explain how these clocks influence metabolism (e.g., regulating insulin secretion, lipid metabolism).

 

Response 2: We have revised the following sentence as follows:

“The central and peripheral circadian clocks regulate key metabolic processes, such as insulin secretion, glucose metabolism, lipid oxidation, appetite signaling, and energy balance, all of which align with the body’s daily rhythms to support metabolic health and influence weight gain and loss.”

 

Comment 3: line 50-52: Consider adding a reference to a specific study that demonstrated the metabolic consequences of late-night eating.

 

Response 3: We have added the reference and revised the sentence as follows:

“For example, studies suggest that eating earlier in the day, maintaining consistent eating timing, and limiting late-night energy intake are associated with greater metabolic efficiency and improved weight management outcomes, while late-night eating late-night eating has been associated with impaired glucose tolerance, reduced fat oxidation and mobilization, and increased BMI.”

 

Reference:

Gu C, Brereton N, Schweitzer A, Cotter M, Duan D, Børsheim E, Wolfe RR, Pham LV, Polotsky VY, Jun JC. Metabolic Effects of Late Dinner in Healthy Volunteers-A Randomized Crossover Clinical Trial. J Clin Endocrinol Metab. 2020 Aug 1;105(8):2789–802. doi: 10.1210/clinem/dgaa354. PMID: 32525525; PMCID: PMC7337187.

Yoshida, J., Eguchi, E., Nagaoka, K. et al. Association of night eating habits with metabolic syndrome and its components: a longitudinal study. BMC Public Health 18, 1366 (2018).

 

Comment 4: line 68-71: The study used BMI for subgroup analysis to assess variations in dietary patterns and meal timing strategies. However, BMI may not fully capture differences in body composition, particularly in distinguishing between fat mass and lean mass. Given that body fat percentage is a more precise indicator of metabolic risk and obesity-related health outcomes, its inclusion would strengthen the analysis. The authors should discuss the limitations of using BMI for subgroup analysis and , if possible, consider incorporating or referencing body fat percentage as a more relevant measure of adiposity.

 

Response 4: Thanks for your thoughtful feedback. To address this limitation, we expanded our analysis to include abdominal obesity, measured by waist circumference, as a more specific indicator of central adiposity and cardiometabolic risk. We agree that BMI alone doesn't distinguish between fat and lean mass. While we used BMI because it provides a clinically standardized and widely recognized cutoff point for underweight, overweight, and obesity that facilitates comparisons across studies, we recognize that direct measures of body fat would improve precision. Unfortunately, NHANES did not collect body fat percentage data for all participants. We've acknowledged this limitation in the discussion section:

“Lastly, obesity in this study was classified using BMI cutoffs. While BMI provides clinically standardized and widely accepted thresholds for underweight, overweight, and obesity that facilitates comparisons across studies, it does not differentiate between fat mass and lean body mass. To address this limitation, we also incorporated abdominal obesity defined by waist circumference as a more specific indicator of central adiposity and cardiometabolic risk.”

 

Comment 5: line 76: The phrasing "collected prior to the COVID-19 pandemic" could be misleading since NHANES 2020 was disrupted by the pandemic. Were data from 2020 fully available or only partial? This should be clarified.

 

Response 5: We revised the sentence to clarify that the 2019–March 2020 NHANES data were collected prior to the disruption caused by the COVID-19 pandemic and were combined with the 2017–2018 data. The sentence now reads:

“This cross-sectional study utilized the most recent available NHANES data from two survey cycles: the complete 2017-2018 cycle and partial data from the 2019-2020 cycle (data collected through March 2020), prior to data collection being disrupted by the COVID-19 pandemic.”

 

Comment 6: Diet assessment: The methodology for defining eating episodes is well explained, but it is unclear how meal skipping was determined (e.g., was an absence of an eating episode within a certain timeframe classified as skipping?).

 Response : We have revised our definition to emphasize that meal skipping explicitly was determined based on the absence of a participant-defined eating occasion. The revised sentence now reads:

"Eating occasion skipping was defined as the absence of a self-defined eating occasion (e.g., breakfast, lunch, dinner, or snack) based on how participants labeled each eating episode during the 24-hour dietary recalls."

 

Comment 7: Diet assessment: Were there any limitations in dietary recall accuracy? Self-reporting may introduce several sources of bias, including recall bias, underreporting or social desirability. A brief mention of potential biases and accuracy of the assessment method would strengthen the section.

 

Response 7: Thank you for this important point. We acknowledge that dietary data collected via 24-hour dietary recall methods can be subject to potential biases. We have revised the discussion section to discuss this potential limitation. We also emphasize that this method is valid despite the limitations. Specifically, previous research has extensively validated the USDA Automated Pass Method, used in NHANES, to demonstrate that it provides valid and reliable estimates of group-level total energy and nutrient intake. We have added the following sentence to the manuscript to address this point:

“The 24-hour dietary recall method may also be subject to recall and social desirability bias. However, it has been extensively validated in prior studies using the USDA Automated Multiple Pass Method and provides reliable estimates of group-level energy and nutrient intake [49, 50].”

 

Reference:

Blanton CA, Moshfegh AJ, Baer DJ, Kretsch MJ. The USDA Automated Multiple-Pass Method accurately estimates group total energy and nutrient intake. J Nutr 2006 Oct; 136(10):2594-9. https://pubag.nal.usda.gov/catalog/10039

Moshfegh AJ, Rhodes DG, Baer DJ, Murayi T, Clemens JC, Rumpler WV, Paul DR, Sebastian RS, Kuczynski KC, Ingwersen LA, Staples RC, Cleveland LC. The USDA Automated Multiple-Pass Method reduces bias in the collec-tion of energy intakes. Am J Clin Nutr 2008; 88:324-332. https://pubag.nal.usda.gov/catalog/21951

 

 

Comment 8: Weight loss attempt/adiposity: See previous comment on BMI. The use of BMI as a primary measure of adiposity is a limitation, as BMI does not distinguish between fat mass and lean mass. A brief discussion of why body fat percentage was not used (or why BMI was preferred) would be valuable.

 

Response 8: Thank you for reiterating this important point. As noted in our response to the related comment above, we used BMI because it has widespread clinical and research application and recognized cutoffs for categorizing underweight, overweight, and obesity. However, to address BMI's limitation in distinguishing between fat and lean mass, we additionally included abdominal obesity using waist circumference.

 

Comment 9: Covariates: The categorization of physical activity follows WHO guidelines, but did the study also consider sedentary time? Including sedentary behavior could provide additional context for weight loss and chrononutrition patterns. Moreover, recent research shows that even lower physical activity levels than the standard recommendation of 150 min moderate or 75 min vigorous exercise per week can improve cardiometabolic health. Thus, the sharp cutoff used in this study to classify participants as physically active (≥150 min moderate or ≥75 min vigorous per week) may overlook individuals who engage in lower but still beneficial levels of physical activity. It would be valuable for the authors to discuss this limitation.

 

Response 9: We have added sedentary time (minutes/day) to Table 1 to provide additional context. We also acknowledged in the limitations section that physical activity levels below the WHO recommendations may still confer health benefits, as reflected in the following sentence: “Although we classified participants as physically active based on WHO recommendations (≥150 minutes of moderate or ≥75 minutes of vigorous activity per week), lower levels of physical activity may still provide health benefits. Therefore, our use of this binary cutoff may have underestimated the beneficial effects of sub-threshold activity levels.”

 

Comment 10: Effect sizes should be included in addition to p-values where possible to show the strength of observed differences.

 

Response 10: Thank you for pointing this out. We have added the effect sizes in addition to the p-values throughout the Results section.

 

Comment 11: The classification of eating profiles (Early Eating, Later Eating, etc.) is useful, but the rationale for defining these categories based on specific cutoff points is unclear: was this data-driven (e.g., clustering analysis) or based on prior research?

 

Response 11: Thank you for your comment. To clarify, we used Latent Profile Analysis, a data-drive approach to identifying subgroups based on shared patterns in the data rather than applying predefined cutoff points. We believe that this is a strength of methods. We revised the statistical analysis section to highlight this approach as a strength of our study. The additional sentence reads:

 

“Since standardized thresholds for classifying chrononutrition patterns do not exist, we used Latent Profile Analysis (LPA) to derive data-driven subgroups based on shared eating timing, eating frequency, and caloric distribution patterns. This approach allowed for the identification of conceptually meaningful profiles grounded in the observed data, rather than relying on arbitrary or predefined cut points—an important strength of using LPA in this context.”

 

Comment 12: Physical Activity: Since physical activity and total calorie intake differ between groups, were these factors adjusted for in the analyses? And, the extended eating window profile is linked to weight gain, but is this association independent of total energy intake and physical activity?

 

Response 12: We appreciate your thoughtful comment. To clarify, our primary goal was to provide a descriptive snapshot of chrononutrition patterns across different obesity and weight change statuses, rather than to infer causal relationships or explain observed group differences. As such, we did not adjust for physical activity or total energy intake in the analysis. Instead, our intention was to identify patterns and generate hypotheses for future analytic studies that can more rigorously evaluate whether specific chrononutrition profiles are independently associated with metabolic outcomes, accounting for behavioral and lifestyle factors such as diet and physical activity..

 

Comment 13:  Some comparisons, such as the difference in total caloric intake between weight loss attempters and non-attempters (1,966 vs. 2,066 kcal), appear numerically small. The authors should discuss (based on references) whether this difference is clinically meaningful or simply statistically significant.

 

Response 13: We have added the following sentence and references to highlight that relatively small behavioral changes, such as reducing daily caloric intake by 100 kcal, may be sufficient to produce clinically meaningful effects in preventing weight gain for most individuals:

“Although the observed difference in total caloric intake between weight loss attempters and non-attempters was modest (approximately 100 kcal), prior research suggests that even small daily reductions in energy intake can be clinically meaningful for weight management and the prevention of gradual weight gain over time [35, 36].”

 

Reference:

Hill JO, Wyatt HR, Peters JC. The Importance of Energy Balance. Eur Endocrinol. 2013 Aug;9(2):111-115. doi: 10.17925/EE.2013.09.02.111. Epub 2013 Aug 23. PMID: 29922364; PMCID: PMC6003580.

Chen L, Appel LJ, Loria C, Lin PH, Champagne CM, Elmer PJ, Ard JD, Mitchell D, Batch BC, Svetkey LP, Caballero B. Reduction in consumption of sugar-sweetened beverages is associated with weight loss: the PREMIER trial. Am J Clin Nutr. 2009 May;89(5):1299-306. doi: 10.3945/ajcn.2008.27240. Epub 2009 Apr 1. PMID: 19339405; PMCID: PMC2676995.

 

Comment 14: The authors emphasize that the present study builds on prior research (Farsijani et al.) but they do not clearly explain how it advances the field beyond introducing new chrononutrition profiles. I suggest that the authors should highlight specific new insights gained from this study that were not present in earlier research.

 

Response 14: We revised the discussion section to articulate better how this study builds on prior research. Specifically, prior studies, including Farsijani et. al. described patterns of individual chrononutrition indices. We looked at patterns to identify chrononutrition profiles:

“Taking into account the intervals between waketime, bedtime, and eating timing, weight loss attempters were grouped into profiles that appeared most frequently in the following order: Typical Eating, Extended Eating Window, Early Eating, and Later Eating. Typical Eating, the most prevalent profile, showed a relatively even distribution across eating timing. Extended Eating Window, the second most common, was marked by an early start to eating after waking and a late end to eating before bed, resulting in the longest eating duration. Early Eating consumed half of their daily calories during the earlier hours of the day and had a long fasting period before bedtime. Later Eating, the least represented group, showed a noticeably delayed eating onset and tended to consume most of their calories later in the day. These profiles represent prototypical chrononutrition patterns among weight loss attempters in the U.S. and provide a framework for identifying key behavioral features, offering valuable insights that can inform more personalized and targeted weight management strategies.”

 

Comment 15: The discussion links weight gain and obesity to delayed first meals, fewer eating episodes, and extended eating windows, but it does not sufficiently address potential reverse causality (i.e., do these patterns cause weight gain, or do people with obesity develop these eating habits?). The authors should discuss some alternative explanations, such as stress, sleep patterns, and work schedules. Moreover, the discussion should briefly mention existing studies on time-restricted eating and intermittent fasting and clarify whether they have demonstrated clear benefits for weight loss.

 

Response 16: Thank you for your suggestions. We agree that the cross-sectional nature of the study limits causal inference and have revised the discussion to explicitly address the possibility of reverse causality. We have added the following sentences to the Discussion:

“It is also possible that such patterns arise as a result of other factors or reflect consequences of excess weight gain rather than causes. For instance, individuals experiencing higher levels of stress may skip or delay eating and eat irregularly due to disrupted routines or emotional coping mechanisms. Similarly, poor sleep patterns—such as short sleep duration, late sleep timing, or sleep disturbances—can alter appetite-regulating hormones and shift eating behaviors toward later or more erratic patterns. In addition, varying work schedules, including night shifts or long work hours, may limit opportunities for regular eating and lead to delayed or reduced eating frequency. These alternative explanations could also be considered when interpreting the observed associations between eating patterns and metabolic status, as the relationship may be bidirectional.”

 

We have also briefly summarized key findings on time-restricted eating and intermittent fasting from previous studies as follows:

“While the evidence is still evolving, several randomized controlled trials and observational studies have reported that TRF and intermittent fasting may support weight loss and improve metabolic markers, particularly when eating windows are aligned with the body’s natural circadian rhythm [40, 41, 46].”

 

Comment 17: Socioeconomic factors and lifestyle constraints (e.g., shift work, caregiving responsibilities) are briefly mentioned but not deeply explored. I suggest that the authors should explicitly state that these variables could influence both eating patterns and weight status, making it difficult to isolate the effect of meal timing alone.

 

Response 17: We have added the following sentences to the Discussion to provide a more detailed explanation of the potential confounding effects of socioeconomic factors and lifestyle constraints on the relationship between chrononutrition and weight status:

“These contextual factors may influence both eating timing and weight status, making it challenging to disentangle the independent effect of chrononutrition from other social determinants of health. For instance, individuals with nonstandard work hours or caregiving demands may have limited flexibility in eating timing, potentially leading to delayed or irregular eating patterns, and these socioeconomic and lifestyle constraints may also be associated with obesity. Despite these limitations, the observed chrononutrition patterns offer valuable insights into real-world eating behaviors across diverse groups. These findings provide a useful foundation for generating hypotheses and guiding the design of future, more targeted interventions. Future studies should incorporate more detailed measurements of these variables to better account for their potential confounding effects.”

Reviewer 3 Report

Comments and Suggestions for Authors

The article presents an interesting issue related to nutrition in accordance with the biological clock. The authors searched existing results and delved into information that was not there. The manuscript shows an interesting perspective on a weight loss diet from the perspective not of calorie balance or macronutrients but of behaviors related to the frequency of meal intake. The literature used is adequate and up-to-date. However, the entire article is well written and the results are appropriately presented.

I recommend it for publication with minor changes. Minor suggestions are presented below.

Table 1 is described in a completely different chapter than it was placed. Tables are placed immediately after quoting its numbering and description in the text. It is worth sticking to this principle throughout the manuscript.

The concept of Health Eating Index appears in the tables. It is nowhere explained what it refers to and what the result associated with it means.
Bolding in line 278 is not necessary.

Author Response

Comment 1: Table 1 is described in a completely different chapter than it was placed. Tables are placed immediately after quoting its numbering and description in the text. It is worth sticking to this principle throughout the manuscript.

Response 1: We have moved each table to appear immediately after the section where it is quoted.

Comment 2: The concept of Health Eating Index appears in the tables. It is nowhere explained what it refers to and what the result associated with it means.

Response 2: We have added a brief explanation of the Healthy Eating Index at the end of the Diet Assessment section: "we evaluated the participants’ overall diet quality using the Healthy Eating Index 2015 (HEI-2015), which ranges from 0 to 100, with higher scores reflecting better overall diet quality". In Tables 2, 4, 5, and 6, we presented the HEI-2015 score alongside total kcal to provide a broader picture of both the quantity and quality of diet, in addition to chrononutrition patterns. Since HEI-2015 was not our main exposure of interest, we did not elaborate on it further in the main text.

Comment 3: Bolding in line 278 is not necessary.

Response 3: Thank you for your comment. We have removed the bold formatting accordingly.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have adequately addressed all of my comments and the quality of the paper has further improved through the revision.

I congratulate the authors and recommend the manuscript for publication in Dietetics.