# Secular trends and determinants of ad libitum energy intake measured in a research setting from 1999-2020

## TL;DR

This study found that people consumed fewer calories over time in a controlled setting, even after accounting for factors like physical activity and body composition.

## Contribution

The study uses objective measures to show secular trends in ad libitum energy intake, adjusting for body composition and activity in a controlled environment.

## Key findings

- Ad libitum energy intake decreased over time, with significant reductions in total, protein, fat, and carbohydrate calories.
- Season had no significant effect on energy intake after adjusting for covariates.
- Body composition and spontaneous physical activity were positively associated with intake measures.

## Abstract

Historically, secular and seasonal trend analyses have been examined using self-report measures of intake. Rarely are objective measures and known determinants of dietary intake used in these analyses. Our objective was to quantify the seasonal and secular differences in an objective ad libitum intake paradigm while considering the contribution of determinants, such as fat-free mass (FFM) index and spontaneous physical activity (SPA) limited to the restricted space of a whole-room calorimeter.

For this study, recruitment of N = 292 healthy, diabetes free, adults occurred from 1999 to 2020. Assessment during their 10-day stay included body composition (by DXA), SPA (by an approximately 24-h stay in whole-room calorimetry), and ad libitum intake (by a vending machine for 3 days). This secondary analysis used general linear models (GLM) to investigate secular and seasonal differences while adjusting for sex, age, FFM index, FM (fat mass) index, SPA, and race/ethnicity.

FFM index and SPA were positively associated with all intake measures (p < 0.05). In all adjusted seasonal models, season did not affect intake. Adjusted secular trends models (kcals/year) demonstrated a decrease in total kcals (β = −55), intake as percent weight maintaining energy needs (β = −2), protein kcals (β = −10), fat kcals (β = −27), and carbohydrates kcals (β = −22) (all p < 0.05). After further adjustment for SPA, significance remained in all intake measures (p < 0.05). Secular trends in body composition revealed no changes in weight, BMI, and percent body fat (all p > 0.20).

Our results indicate that over time, ad libitum intake decreased in this controlled research setting and remained significant even after accounting for positive determinants of intake. A significant ad libitum decrease, coupled with no change in body composition, may highlight a participant bias toward calorie restriction in a controlled setting over time and deserves further investigation.