Changes in affect from non-eating activity engagement as predictors of loss of control eating: an ecological momentary assessment study

PURPOSE: Loss of control eating (LOCE) is supported as a maladaptive regulation strategy for high negative affect (NA) and low positive affect (PA). Yet, little is known about the day-to-day activities outside of eating that may precipitate such changes in affect and impact risk for LOCE. The current ecological momentary assessment study sought to examine whether (1) low NA and high PA at a given survey would prospectively predict lower likelihood of a LOCE episode at the next survey, and (2) having recently engaged in various non-eating activities would be associated with changes in average NA and PA levels at the same survey. METHODS: Data collected from 66 participants with LOCE over 7–14 days were analyzed using generalized estimating equations. RESULTS: High NA but not low PA at a given survey prospectively predicted lower likelihood of a LOCE episode at the next survey. Having recently engaged in outdoor recreation and socializing was associated with lower than one’s usual NA and higher than one’s usual PA at the same survey, social media was associated with higher than one’s usual NA and lower than one’s usual PA, self-care was associated with higher than one’s usual PA, and napping was associated with lower than one’s usual PA. CONCLUSION: Individuals may benefit from increasing the frequency of daily non-eating activities that reduce NA, since it appears to be a more salient predictor of LOCE than PA. Further research is needed to elucidate this LOCE maintenance model, particularly in regard to the role of PA. LEVEL OF EVIDENCE: Level IV, multiple time series without intervention.