Abstract 10130: Relationship Between Perceived and Measured Diet Quality Improvements in a Randomized Weight Loss Trial

Abstract

Background: Cross-sectional evidence suggests individuals overestimate their consumption of healthful foods. However, how perceived diet quality (PDQ) relates to diet quality calculated from standard methods is unclear. Examining study participants' misperceptions of either diet quality, and/or improvements in diet quality, may enhance our understanding of factors that influence the effectiveness of dietary interventions. Methods: The Healthy Eating Index (HEI-2015) was calculated pre- and post-intervention from dietary recalls in this randomized weight loss trial. Participant self-perception of pre-intervention and post-intervention diet quality was rated on a scale of 0 to 100 at the 12-month post-intervention time point. Agreement between perceived and calculated diet quality was examined using Bland-Altman method. Good agreement between measures was considered scores within 6 points. Results: Participants with complete HEI and PDQ data (n=116) was similar to the total sample and mostly female (79%), white (84%), and with median (q1, q3) age of 51.5 (35.5, 58.5) years. A quarter of participants had good agreement between HEI and PDQ scores at 12 months with differences between scores ranging from -44.8 to 29.9 points. Most of the disagreement arose from PDQ scores being higher (mean [std] 67.6 [18.5]) than HEI scores (mean [std] 56.4 [13.6]). Only 12% of participants had good agreement between change in HEI and change in PDQ scores with differences ranging from -68.4 to 39.6 points. Again, disagreement mainly arose from higher perceived improvement in diet quality (mean=18.6 points) compared to improvement in HEI scores (mean=1.3 points). Discussion: The diet quality of weight loss-seeking adults is less than ideal with little evidence of improvement; however, the majority of participants appear to perceive their diet quality and improvements in diet quality as better than measured. Future studies might consider the effect of better aligning participants’ perceptions with reality.