Salley: Accuracy of a Bite-Count Based Calorie Estimate Compared to Human Estimates With and Without Calorie Information Available

When: Monday, April 15th, 2013. 8:30 am

Where: Brackett 419

Title: Accuracy of a Bite-Count Based Calorie Estimate Compared to Human Estimates With and Without Calorie Information Available

Abstract:

Obesity is an increasing health problem in the US, associated with such dangerous health risks as heart disease and diabetes. Self-monitoring in the form of calorie counting is a critical aspect of successful weight loss. However, calorie estimations are subject to several perceptual and cognitive biases, and there are limited tools available to assist these estimations. The present study seeks to assess the accuracy of participants’ estimations of the calorie content of meals in the presence or absence of calorie information, and to compare their accuracy with calorie estimations based on bite count. Data were analyzed for 87 participants from a study in which participants were allowed to select from a wide variety of meals in a cafeteria setting, which they consumed while wearing a device designed to count bites of food. They were asked to estimate the number of calories they consumed either with or without calorie information available. True calorie intake and a calorie intake estimation based on bite count were calculated for each participant. A 2×2 Mixed-Design ANOVA revealed a significant main effect for estimation method (F(1, 83) = 14.381, p < .001), a marginally significant effect for the presence of calorie information (F(1, 83) = 3.835, p = .054), and a significant interaction between estimation method and the presence of calorie information (F(1, 83) = 6.384, p < .05). Post-hoc tests revealed that errors in human calorie estimations were significantly improved by the presence of calorie information (t(45.89) = -2.731 p < .01). Calorie estimations based on bite count were significantly more accurate than human estimates without the aid of calorie information (t(32) = -3.578, p < .005), but there was no significant difference between estimations based on bite count and human estimates with the aid of calorie information (t(52) = -1.116, p = .270). The results suggest that bite count may be a valid proxy for calorie intake.

Similar Posts (auto-generated):