Furthering Scientific Inquiry for Weight Loss Maintenance: Assessing the Psychological Processes Impacted by a Low intensity Technology-Assisted Intervention (NULevel Trial)

Background NULevel was a randomized control trial to evaluate a technology-assisted weight loss maintenance (WLM) program in the UK. The program included: (a) a face-to-face goal-setting session; (b) an internet platform, a pedometer, and wirelessly connected scales to monitor and report diet, physical activity, and weight, and; (c) regular automated feedback delivered by mobile phone, tailored to participants’ progress. Components were designed to target psychological processes linked to weight-related behavior. Though intervention participants showed increased physical activity, there was no difference in WLM between the intervention and control groups after 12 months (Sniehotta FF, Evans EH, Sainsbury K, et al. Behavioural intervention for weight loss maintenance versus standard weight advice in adults with obesity: A randomized controlled trial in the UK (NULevelTrial). PLoS Med. 2019; 16(5):e1002793. doi:10.1371/journal.pmed.1002793). It is unclear whether the program failed to alter targeted psychological processes, or whether changes in these processes failed to influence WLM. Purpose We evaluate whether the program influenced 16 prespecified psychological processes (e.g., self-efficacy and automaticity toward diet and physical activity), and whether these processes (at 6 months) were associated with successful WLM (at 12 months). Methods 288 adults who had previously lost weight were randomized to the intervention or control groups.The control group received wireless scales and standard advice via newsletters. Assessments occurred in person at 0, 6, and 12 months. Results The intervention significantly altered 10 of the 16 psychological processes, compared with the control group. However, few processes were associated with WLM, leading to no significant indirect effects of the intervention via the processes on WLM. Conclusions Changes in targeted processes were insufficient to support WLM. Future efforts may more closely examine the sequence of effects between processes, behavior, and WLM.