Exploring trait- and state-based dietary restraint using ecological momentary assessment

Researchers often use trait-based questionnaires to characterise dietary restraint, but it is unclear how this relates to real-life appetite and eating behaviour. Advances in technology have facilitated the development of ecological momentary assessment (EMA), an approach which captures everyday experiences in real-time. The aim of this study was to use our novel EMA tool (Momentary appetite capture; MAC) to investigate appetite and prospective portion-size selection throughout the day in 'trait-based' restrained eaters, and to explore whether 'state-based' dietary restraint can be measured using MAC and how this relates to BMI. Ninety-four participants (mean BMI = 22.4, 78 females) received seven text message MAC prompts per day for two consecutive weekdays. Each MAC measured current appetite, portion-size selection, and state-based dietary restraint. For each appetite measure, and each participant, we calculated a daily area-under-the-curve and daily range. For state-based restraint, the percentage of occasions that participants reported not eating a food they were tempted by but trying to avoid was calculated. We found that trait-based dietary restraint (TFEQ-18) was negatively correlated with daily area under the curve for hunger (r(83) = -0.25, p = .022), and prospective portion-size selection (r(83) = -0.27, p = .013), as well as daily range for prospective portion-size selection (r(93) = -0.24, p = .021). Exploratory analysis revealed a negative relationship (ρ = -0.40, p = .003, N = 53) between our MAC state-based measure of dietary restraint and BMI. These findings highlight the utility of MAC to measure fluctuations in appetite and state-based dietary restraint throughout the day. This study also demonstrates the potential to apply this work to understand how real-world dietary restraint impacts BMI.