JMIR Publications

ABSTRACT

Background:

Digital approaches to weight management have potential to produce cost-effective and scalable weight management solutions.

Objective:

The core components of effective weight management interventions encourage self-regulation of energy balance behaviors which could be enhanced using aspects of emotion regulation.

Methods:

The NoHoW trial studied 1627 participants in a 2×2 factorial, randomised, single blind, controlled trial. The trial evaluated a digital toolkit for weight management subsequent to initial ≥5% weight loss in the prior 12 months (weight loss maintenance (WLM)) involving: (i) self-regulation and motivation (M), (ii) contextual behavioural elements of emotion regulation (E), or (iii) these intervention components in combination (M+E), compared to an active control (generic toolkit content, regular self-weighing, and wearable tracking device (Fitbit®) use). We hypothesized that the combined intervention would be more effective for WLM compared to the self-regulatory or emotion regulatory interventions alone, which individually would also be more effective than the active control group. The primary outcome was weight change from baseline and secondary outcomes were health biomarkers, physical activity, sleep, dietary intake and self-reported psychological processes at 12 months from baseline. Linear models evaluated weight change from 0 -12 months using (i) intention to treat (ITT), (ii) participants who completed 12 months (completers), and (iii) participants who had measurements at 12 months and completed ≥80% of toolkit modules (per protocol). Explanatory terms were recruitment centre, gender, M intervention, E intervention and two-way interactions between these terms. Age group, BMI group and pre-trial weight loss were included as additional covariates. Predefined subgroups analyses were carried out for men and women separately.

Results:

For the ITT (76% retention at 12 months) and the completer populations, the M arm, the E arm and the E+M arms did not significantly affect weight outcomes compared to the active control. Analyses of gender differences showed a small but clinically negligible effect of the self-regulation/motivation arm on weight in men only, for the ITT, per protocol and completer populations. There were no significant effects of the interventions on biomarkers, body composition or health outcomes. Regardless of imputation method for missing weight data, due to participant drop-out there was a tendency for pre-trial weight loss to be regained on average.

Conclusions:

The current study highlights some possible design aspects of the digital behaviour change intervention that could be improved to enhance intervention effectiveness. Clinical Trial: ISRCTN88405328.