Late eating and shortened fasting are associated with higher ultra-processed food intake across all age groups: a population-based study
Late eating and shortened fasting are associated with higher ultra-processed food intake across all age groups: a population-based study
PURPOSE: Global dietary patterns are increasingly driven by ultra-processed foods-cheap, highly palatable, and ready-to-eat options. Exploring time-related eating patterns and its association with ultra-processed foods could help in intervention efforts, but knowledge on this topic is still limited. This study assessed the association of time-related eating patterns with unprocessed/minimally processed and ultra-processed food consumption across different life stages.
METHODS: Two 24-hour food recalls from a nationally representative sample in Brazil (Brazilian Household Budget Survey, POF, 2017-2018; n = 46,164) were used to estimate tertiles of first and last intake times, eating midpoint, caloric midpoint time, and night fasting (independent variables). All consumed foods were classified according to the Nova classification system, and the outcomes of interest were consumption of unprocessed/minimally processed and ultra-processed foods. Multiple linear regression models were performed for all individuals and stratified for each age group: adolescents (10-19 years, n = 8,469), adults (20-59 years, n = 29,332), and older individuals (≥ 60 years, n = 8,322).
RESULTS: The later tertile of first food intake time, last food intake time, caloric midpoint, and eating midpoint were positively associated with consumption of ultra-processed foods (β = 3.69, 95%CI = 3.04, 4.34; β = 1.89, 95%CI = 1.32, 2.47; β = 5.20, 95%CI = 4.60, 5.81; β = 3.10, 95%CI = 2.49, 3.71, respectively) and negatively associated with consumption of unprocessed or minimally processed foods (β=-2.79, 95%CI=-3.37; -2.22; β=-1.65, 95%CI=-2.24, -1.05; β=-3.94, 95%CI=-4.44, -3.44; β=- 2.35, 95%CI=-2.93, -1.78, respectively) compared to the first "earlier" tertile (reference). An inverse association was found for night fasting (β=-1.74, 95%CI=-2.28, -1.22 and β = 1.52, 95%CI = 0.98, 2.06 for ultra-processed and unprocessed/minimally processed foods, respectively). These associations were consistent across all age groups.
CONCLUSION: Chrononutrition patterns characterized by late intake timing and shortened overnight fasting were associated with higher consumption of ultra-processed foods and lower intake of unprocessed/minimally processed foods across all age groups.
Humans, Adult, Adolescent, Male, Female, Middle Aged, Fasting, Child, Young Adult, Fast Foods/statistics & numerical data, Brazil, Feeding Behavior, Diet/statistics & numerical data, Energy Intake, Age Factors, Time Factors, Aged, Cross-Sectional Studies, Diet Surveys, Eating, Food, Processed
Crispim, Cibele A.
4595d83e-1415-4d73-8aa4-84f65e20643a
Azeredo, Catarina M.
6e98ab23-36d1-47ac-8ce3-66e06832234f
Rinaldi, Ana E.M.
e36a4dc8-3cf7-4dc3-8778-28569402d4a7
Alves, Alexessander Couto
87b9179e-abde-4ca5-abfc-4b7c5ac8b03b
Skene, Debra J.
b16a25b9-8fe8-4176-83ac-7934fee8beb4
Moreno, Claudia R.C.
9a76bd06-a176-4946-8c4d-99d1bea67951
21 March 2025
Crispim, Cibele A.
4595d83e-1415-4d73-8aa4-84f65e20643a
Azeredo, Catarina M.
6e98ab23-36d1-47ac-8ce3-66e06832234f
Rinaldi, Ana E.M.
e36a4dc8-3cf7-4dc3-8778-28569402d4a7
Alves, Alexessander Couto
87b9179e-abde-4ca5-abfc-4b7c5ac8b03b
Skene, Debra J.
b16a25b9-8fe8-4176-83ac-7934fee8beb4
Moreno, Claudia R.C.
9a76bd06-a176-4946-8c4d-99d1bea67951
Crispim, Cibele A., Azeredo, Catarina M., Rinaldi, Ana E.M., Alves, Alexessander Couto, Skene, Debra J. and Moreno, Claudia R.C.
(2025)
Late eating and shortened fasting are associated with higher ultra-processed food intake across all age groups: a population-based study.
European Journal of Nutrition, 64 (3), [134].
(doi:10.1007/s00394-025-03633-w).
Abstract
PURPOSE: Global dietary patterns are increasingly driven by ultra-processed foods-cheap, highly palatable, and ready-to-eat options. Exploring time-related eating patterns and its association with ultra-processed foods could help in intervention efforts, but knowledge on this topic is still limited. This study assessed the association of time-related eating patterns with unprocessed/minimally processed and ultra-processed food consumption across different life stages.
METHODS: Two 24-hour food recalls from a nationally representative sample in Brazil (Brazilian Household Budget Survey, POF, 2017-2018; n = 46,164) were used to estimate tertiles of first and last intake times, eating midpoint, caloric midpoint time, and night fasting (independent variables). All consumed foods were classified according to the Nova classification system, and the outcomes of interest were consumption of unprocessed/minimally processed and ultra-processed foods. Multiple linear regression models were performed for all individuals and stratified for each age group: adolescents (10-19 years, n = 8,469), adults (20-59 years, n = 29,332), and older individuals (≥ 60 years, n = 8,322).
RESULTS: The later tertile of first food intake time, last food intake time, caloric midpoint, and eating midpoint were positively associated with consumption of ultra-processed foods (β = 3.69, 95%CI = 3.04, 4.34; β = 1.89, 95%CI = 1.32, 2.47; β = 5.20, 95%CI = 4.60, 5.81; β = 3.10, 95%CI = 2.49, 3.71, respectively) and negatively associated with consumption of unprocessed or minimally processed foods (β=-2.79, 95%CI=-3.37; -2.22; β=-1.65, 95%CI=-2.24, -1.05; β=-3.94, 95%CI=-4.44, -3.44; β=- 2.35, 95%CI=-2.93, -1.78, respectively) compared to the first "earlier" tertile (reference). An inverse association was found for night fasting (β=-1.74, 95%CI=-2.28, -1.22 and β = 1.52, 95%CI = 0.98, 2.06 for ultra-processed and unprocessed/minimally processed foods, respectively). These associations were consistent across all age groups.
CONCLUSION: Chrononutrition patterns characterized by late intake timing and shortened overnight fasting were associated with higher consumption of ultra-processed foods and lower intake of unprocessed/minimally processed foods across all age groups.
This record has no associated files available for download.
More information
Accepted/In Press date: 21 February 2025
e-pub ahead of print date: 21 March 2025
Published date: 21 March 2025
Additional Information:
© 2025. Springer-Verlag GmbH Germany, part of Springer Nature.
Keywords:
Humans, Adult, Adolescent, Male, Female, Middle Aged, Fasting, Child, Young Adult, Fast Foods/statistics & numerical data, Brazil, Feeding Behavior, Diet/statistics & numerical data, Energy Intake, Age Factors, Time Factors, Aged, Cross-Sectional Studies, Diet Surveys, Eating, Food, Processed
Identifiers
Local EPrints ID: 509401
URI: http://eprints.soton.ac.uk/id/eprint/509401
ISSN: 1436-6207
PURE UUID: 370710c8-799b-447a-9e12-be29c8654813
Catalogue record
Date deposited: 20 Feb 2026 17:38
Last modified: 21 Feb 2026 03:20
Export record
Altmetrics
Contributors
Author:
Cibele A. Crispim
Author:
Catarina M. Azeredo
Author:
Ana E.M. Rinaldi
Author:
Alexessander Couto Alves
Author:
Debra J. Skene
Author:
Claudia R.C. Moreno
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics