TY - JOUR
T1 - Self-Reported Eating-Occasion Frequency and Timing are Reproducible and Relatively Valid in the American Cancer Society Cancer Prevention Study-3 Diet Assessment Substudy
AU - Hartman, Terryl J.
AU - Masters, Matthew
AU - Flanders, W. Dana
AU - Wang, Ying
AU - Li, Mengyi
AU - Mitchell, Diane C.
AU - Guinter, Mark
AU - Patel, Alpa V.
AU - McCullough, Marjorie L.
N1 - Publisher Copyright:
© 2022 American Society for Nutrition.
PY - 2022/12
Y1 - 2022/12
N2 - Background: Accurate assessment of eating-occasion behaviors, such as timing, frequency, and consumption intervals, is important for evaluating associations with obesity and other chronic diseases. Objectives: The main objective of this study was to assess the relative validity of a 24-hour grid approach to assess eating-occasion timing and frequency in comparison to data derived from repeated 24-hour dietary recalls (DRs). A second objective was to assess the 1-year test-retest reproducibility of the 24-hour grid. Methods: Between 2015 and 2016, 626 participants in the Cancer Prevention Study-3 (CPS-3) Diet Assessment Substudy (mean age, 52 years; age range, 31–70 years; 64% female; 64% non-Hispanic white, 22% non-Hispanic black, 14% Hispanic) completed 2 grids and up to 6 unannounced, telephone, interviewer-administered DRs over 1 year. Spearman correlations (ρ; 95% CIs) were calculated to assess reproducibility between the repeated eating-occasion grid–derived variables (e.g., numbers of snacks and meals per day, timing of eating occasions) and to assess relative validity by comparing the meal grid and DR-derived summary data separately for weekdays and weekend days. Results: Reproducibility correlations for eating-occasion variables derived from the eating-occasion grids completed 1 year apart were ≥0.5 for the majority of variables analyzed for both weekdays and weekend days, including numbers of snacks and meals per day and timing of the first and last eating occasions of the day. Relative validity was highest among weekday variables and was ≥0.5 for the majority of variables, with correlations ranging from ρ values of 0.32 (number of meals per day) to 0.68 (hour of the first eating occasion). Conclusions: These findings suggest the eating-occasion grid used in CPS-3 has good reproducibility over 1 year and yields estimates comparable to those from a more detailed method of assessment of eating timing and frequency.
AB - Background: Accurate assessment of eating-occasion behaviors, such as timing, frequency, and consumption intervals, is important for evaluating associations with obesity and other chronic diseases. Objectives: The main objective of this study was to assess the relative validity of a 24-hour grid approach to assess eating-occasion timing and frequency in comparison to data derived from repeated 24-hour dietary recalls (DRs). A second objective was to assess the 1-year test-retest reproducibility of the 24-hour grid. Methods: Between 2015 and 2016, 626 participants in the Cancer Prevention Study-3 (CPS-3) Diet Assessment Substudy (mean age, 52 years; age range, 31–70 years; 64% female; 64% non-Hispanic white, 22% non-Hispanic black, 14% Hispanic) completed 2 grids and up to 6 unannounced, telephone, interviewer-administered DRs over 1 year. Spearman correlations (ρ; 95% CIs) were calculated to assess reproducibility between the repeated eating-occasion grid–derived variables (e.g., numbers of snacks and meals per day, timing of eating occasions) and to assess relative validity by comparing the meal grid and DR-derived summary data separately for weekdays and weekend days. Results: Reproducibility correlations for eating-occasion variables derived from the eating-occasion grids completed 1 year apart were ≥0.5 for the majority of variables analyzed for both weekdays and weekend days, including numbers of snacks and meals per day and timing of the first and last eating occasions of the day. Relative validity was highest among weekday variables and was ≥0.5 for the majority of variables, with correlations ranging from ρ values of 0.32 (number of meals per day) to 0.68 (hour of the first eating occasion). Conclusions: These findings suggest the eating-occasion grid used in CPS-3 has good reproducibility over 1 year and yields estimates comparable to those from a more detailed method of assessment of eating timing and frequency.
UR - http://www.scopus.com/inward/record.url?scp=85146364304&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85146364304&partnerID=8YFLogxK
U2 - 10.1093/jn/nxac206
DO - 10.1093/jn/nxac206
M3 - Article
C2 - 36055776
AN - SCOPUS:85146364304
SN - 0022-3166
VL - 152
SP - 2827
EP - 2836
JO - Journal of Nutrition
JF - Journal of Nutrition
IS - 12
ER -