Abstract

Eating rate, defined as the consumption of food per unit of time, has been associated with energy intake and obesity. However, eating rate is difficult to measure, necessitating most studies to rely on self-reported eating rate (SRER). The wrist-worn Eat Less Move More (ELMM) device has been validated to count bites by detection of a wrist-roll motion specific to eating, but has not yet been used to measure seconds between bites, as a proxy to eating rate, in free-living settings. We aimed to examine ELMM-assessed eating rate in free-living settings compared to SRER. This was a secondary analysis of the initial three days of seconds between bites data from an 8-week weight loss intervention. Participants (n=37; 62.2% female; age 36.5 ± 16.1 years; BMI 31.2 ± 3.5 kg/m²) ranked their SRER and tracked their eating in free-living settings by turning the ELMM on and off at start/end of meals. Based on baseline data, five SRER categories were collapsed into three groups: very slow/slow (n=5), medium (n=12), and fast/very fast (n=16). One-way ANOVA examined SRER group differences in average seconds between bites.There was a significant difference in seconds between bites among SRER groups (SBB, M ± SD: slow=25.5 ± 4.3, medium=21.0 ± 3.4, and fast=20.8 ± 3.0). Tukey tests showed lower average seconds between bites, indicating faster eating, in the fast SRER group compared to the slow SRER group. Lower average seconds between bites was seen in the medium SRER group compared to the slow SRER group, but there was no difference between the medium and fast SRER groups.The ELMM assessed average seconds between bites during eating occasions, and seconds between bites corresponded to SRER. Future work should assess eating rate with ELMM in larger samples and in other populations.

Keywords: Eating rate; Speed of eating; Self-monitoring; Obesity; Wearable device; Wrist-worn device; Seconds between bites