Abstract P125: EMA Data Reveal Associations Between Sleep and Self-Efficacy for Adhering to a Healthy Lifestyle Plan

Abstract

Background: We are conducting a 12-month behavioral weight loss study to provide the background for the study of relapse following intentional weight loss. Ecological momentary assessment (EMA) is being used to assess the triggers of relapse in real time and in the person’s own environment. Objective: The purpose of this abstract is to report on the associations between sleep and self-efficacy for sticking to or adhering to a healthy lifestyle. Methods: Using a smartphone for EMA data collection, participants were prompted daily for 6 months to complete a beginning of the day (BOD) survey that reported on 1) difficulty falling asleep, 2) number of hours slept, 3) number of awakenings, and 4) how well they slept the previous night. Using the BOD survey data, separate mixed linear models were carried out predicting confidence in sticking to a healthy diet from the four items to which participants responded. For each model, subject and subject by predictor interaction were included as random effects. Results: The sample for this analysis is comprised of 89 participants who have completed the first 6 mos. of the study. It is predominantly female (90.3%) and White (81.9%), 58.3% are married with 16.72±2.56 years of education and a mean BMI of 33.90±4.56 kg/m 2 . All four items on the BOD survey were significant predictors of self-efficacy for adhering to a healthy lifestyle: trouble sleeping (b = 0.0182, p 0.0068); hours slept (b = 0.0011, p < 0.0001); number awakenings (b = 0.0706, p 0.0002); and how well slept (b = 0.1117, p <0.0001). Self-efficacy increased with the number of hours slept and how well the person slept, and decreased with trouble sleeping and number of awakenings. Standard deviation (SD) components for intercept were generally larger than the SD components for error suggesting that there was more variability within subjects than between subjects. Moreover, SD components for the slopes of the regression lines were large relative to the slopes of the regression lines, suggesting that that there is considerable variation among subjects in the predictive equations. While on average confidence in maintaining a healthy lifestyle increases with how well subjects slept, a high SD component suggests that for some subjects, confidence in maintaining a healthy lifestyle is negatively related to how well they slept. Conclusions: These preliminary EMA data suggest that for most participants in a weight loss study, reduced quality and quantity of sleep negatively impacted their confidence in adhering to their established plan for a healthy lifestyle. However, some participants reported higher self-efficacy despite poor or reduced sleep the previous night, which might suggest the use of greater resolve and problem-solving to do well in the face of what might be a challenging day. These findings have implications for content and standard approaches to behavioral weight loss treatment.