How can we promote effective engagement with web-based support for weight management?

Talk Code: 

The problem

There is a need for cost-effective weight management interventions that primary care can deliver to reduce the morbidity caused by obesity. Automated web-based interventions might provide a solution, but engagement with these interventions is often limited, resulting in small effect sizes. We need a better understanding of how to enhance effective usage of web-based support for weight management using the most efficient combinations of digital and human support.

The approach

We developed the Positive Online Weight Management (POWeR) intervention, using the LifeGuide software which automatically records time spent and data entered on every page of the intervention. We evaluated POWeR with varying levels of nurse support in a feasibility primary care trial (n = 179) and a full primary care trial (n = 834). We also evaluated POWeR without nurse support rolled out by public health teams in the workplace (n = 942). This paper compares engagement with POWeR across these settings, analysed by means of a novel tool for visualising usage. We examine usage and outcomes with different levels of support, and investigate how usage predicts outcome, in order to inform the design and implementation of web-based support for weight management in the future. To date, online data collection has been completed for all trials, the visualisation analyses have been completed for the feasibility and workplace trials and are commencing for the full trial.


Analyses from the primary care feasibility trial revealed that usage was similar with and without nurse support, but outcomes were better with modest nurse support. Users engaged most with weekly goal and weight review (i.e. receiving weekly personalised feedback based on their progress relative to their weight and eating goals) and accessing weight management tools such as eating plans and food and steps diaries. Users were less likely to then go on to access in-depth sessions providing CBT-based behavioural advice on a variety of topics, and usage of these sessions was unrelated to weight outcomes. Analyses from the workplace trial indicated that usage of an app giving mobile access to these tools was associated with greater usage of the website sessions.


Analyses completed to date provide valuable information for web-based intervention design, suggesting that the majority of users may require and engage best with easy access to brief personalised feedback and self-regulation tools support rather than in-depth behaviour management advice – although in our full trial we are testing whether users can be prompted to make greater use of in-depth advice. Final analyses will compare outcomes with and without nurse support in the full trial, and examine usage predictors of outcomes in that trial.


  • Lucy Yardley, University of Bournemouth, Bournemouth, UK
  • Leanne Morrison, University of Bournemouth, Bournemouth, UK
  • Emily Smith, University of Bournemouth, Bournemouth, UK
  • Emily Arden-Close, University of Oxford, Oxford, UK
  • Mary Steele, University of Bournemouth, Bournemouth, UK
  • Katherine Bradbury, University of Bournemouth, Bournemouth, UK
  • Richard Hobbs
  • Paul Little, University of Bournemouth, Bournemouth, UK