clinical obesity
doi: 10.1111/cob.12001
Integrating user perspectives into the development of a web-based weight management intervention L. Yardley1, S. Williams2, K. Bradbury1, G. Garip1, S. Renouf1, L. Ware1, H. Dorling1, E. Smith1 and P. Little3 What is already known about this subject
What this study adds
• Without sustained human support, patients can find it difficult to maintain
their
weight
independently
using
web-based
• In order to be credible and motivating to people with long-term weight
weight
problems, interventions may need to offer novel approaches to weight
• Development of complex interventions of this kind may benefit from
• Interventions that employ prescriptive weight loss techniques may be
drawing on insights into user perspectives derived from inductive
difficult to follow long-term if they entail too much conscious effort and
management programmes.
management.
qualitative research.
disruption of lifestyle. • A more flexible and autonomous approach to changing eating habits is attractive to potential intervention users but could be difficult to implement successfully.
1
Faculty of Human and Life Sciences,
University of Southampton, Southampton, UK, 2
School of Design, Engineering and
Computing, Bournemouth University, Bournemouth, UK, 3Faculty of Medicine, University of Southampton, Southampton, UK
Received 29 April 2012; revised 30 August 2012; accepted 12 October 2012
Address for correspondence: Professor L Yardley, Academic Unit of Psychology, Faculty of Human and Life Sciences, University of Southampton, Highfield, Southampton SO17 1BG, UK. E-mail:
[email protected]
Summary The objective of this study was to adapt the design of our weight management intervention to the needs, expectations and capabilities of potential users. In study 1, we interviewed 25 people about their experiences of weight management. The findings of these interviews were combined with findings from existing theory and research in a process of ‘intervention planning’ that informed the design of the intervention. Study 2 comprised in-depth think-aloud studies with a further 16 people interested in using a web-based intervention to manage their weight, in order to elicit reactions to the intervention techniques and materials. In study 1, overly intrusive and restrictive aspects of eating self-regulation were commonly cited reasons for failure to maintain weight management long-term. We therefore designed an intervention with a more flexible approach to autonomous selfregulation. This approach was broadly welcomed in study 2, but there were indications that some participants might have difficulty effectively implementing self-regulation techniques independently. A flexible and autonomous approach to changing eating habits is attractive to potential intervention users but may be difficult for some users to implement successfully. Keywords: Health communication, overweight, qualitative research, weight reduction programme.
Introduction Internet-based support for weight management has the potential to provide a low cost means of promoting weight reduction, but there is currently mixed evidence concerning the effectiveness of this type of intervention. Existing 132
research suggests that without human input, most automated interactive web interventions achieve only modest weight loss at best, but the combination of web-based support with some degree of human contact may help people to lose weight and maintain weight loss (1–7). The aim of the study described in this paper was to inform the
© 2013 The Authors Clinical Obesity © 2013 International Association for the Study of Obesity. clinical obesity 2, 132–141
clinical obesity
User perspectives on weight management L. Yardley et al. 133
Table 1 POWeR intervention philosophy Characteristics designed to foster credibility • Distinctive – containing new, surprising and interesting content, e.g. ‘POWeR tools’ (self-regulation techniques) • Explicitly evidence based, presenting scientific rationale for recommendations and proof of their effectiveness • Non-commercial, developed by named team of medical and behaviour change experts, linked to NHS, supported by nurse Characteristics designed to encourage adherence and long-term maintenance • Emphasis on building autonomous motivation, non-prescriptive approach (e.g. no forbidden foods, choice of eating plans and goals) • Focus on creating lifestyle-compatible long-term habits rather than conscious self-regulation through calorie counting, diary keeping • Self-efficacy and positive affect promoted by encouraging and rewarding achievable goals, modelling overcoming barriers using engaging stories
development of a web-based weight management intervention that could be integrated within primary care support for overweight and obese patients. There has been increasing emphasis in recent years on the need to develop complex interventions promoting healthrelated behaviour change more systematically, using theory and evidence to inform the selection of behaviour change methods employed by the intervention (8–10). Drawing on theory and existing research allows intervention developers to learn from previous experience, while systematic development ensures that the full range of relevant and important determinants of behaviour are identified and appropriately addressed. In addition, undertaking a comprehensive description of the intervention elements and their rationale can help other intervention developers to identify and incorporate the effective ingredients of successful interventions. Systematic, theory-based intervention development can draw on deductive and inductive, quantitative and qualitative methods (11,12). However, to date, most papers on systematic intervention development have focused principally on the use of deductive methods, whereby theory is used to choose intervention techniques (13–15) and quantitative evidence is relied on for the identification of important behavioural determinants and the evaluation of the effects of intervention elements on determinants and behaviour (16,17). Although formal and informal qualitative research also often informs successful intervention development (18,19), there has been less explicit recognition and discussion of the vital role that inductive and qualitative methods can play in systematic intervention development. This paper sought to redress this imbalance by considering and illustrating how inductive, qualitative methods can contribute not only to intervention development but also to our understanding of the perspectives and experiences of those attempting to lose weight and how these may impact on the effectiveness of weight management interventions. Here we report an in-depth, iterative, qualitative approach, fundamental to the development of our webbased weight management intervention (Positive Online Weight Reduction; POWeR). This paper highlights the
significance of emerging themes that seemed most important to understanding how users may view and respond to weight loss interventions.
Overview of the intervention development process We describe our approach to intervention development as ‘intervention planning’ to signal our commitment to drawing explicitly and systematically on theory and evidence, while distinguishing our approach from more prescriptive procedures such as ‘intervention mapping’ (13). The initial phase of intervention planning was based on the findings of study 1, an exhaustive (unpublished) review of existing research into the effectiveness of digital weight management programmes, a systematic review and meta-ethnography of qualitative studies of weight management experiences (20), a scoping review of components of weight management programmes delivered faceto-face and consultations with other experts in health behaviour change. We also drew on the experience of team members and expert consultants, who had complementary expertise in health psychology, cognitive– behavioural therapy, health promotion, nutrition and medicine, and experience implementing a variety of health behaviour change programmes. Initially, we drew up a set of principles, which would define the philosophy of the POWeR programme (Table 1). We then identified and incorporated components of successful interventions that were consistent with our philosophy. Study 2 was used to gain an in-depth appreciation of how users related to the resulting intervention content and format, allowing us to modify them as necessary. Further qualitative research (not reported here) was then carried out to check how users experienced the modified intervention after using it independently over several weeks. Finally, to facilitate comparison with other interventions we made our intervention materials available to other researchers and we documented the design and content of the intervention (see Supporting Information Appendix S1), recording how we implemented each theory-based technique we employed using standardized terminology (21).
© 2013 The Authors Clinical Obesity © 2013 International Association for the Study of Obesity. clinical obesity 2, 132–141
clinical obesity
134 User perspectives on weight management L. Yardley et al.
Table 2 Study 1 participant characteristics
Study 1
Characteristics
Men (n = 7)
Women (n = 18)
Materials and methods Aim, design and procedure The aim of study 1 was to ensure that we understood the needs and perspectives of members of our target population, and the context in which they would be engaging with our intervention. Specifically, we wanted to understand the general populations’ perceptions of weight management and weight management interventions. Ethical approval for both studies was obtained from the Research Governance Office at the University of Southampton, and all participants gave written informed consent prior to interview. For study 1, we interviewed a diverse sample of 25 men and women of different ages. We asked about their previous experiences of weight management, and then presented them with a weight management booklet from a previous study to elicit opinions of messages and advice that might be included in our intervention. We purposively sampled people at a healthy body weight as well as people with excess weight, to permit comparison of successful and unsuccessful experiences. Posters advertising the study were posted around the university campus and emails advertising the study were circulated among university staff to reach a non-student population. To be eligible for the study, participants had to be concerned about their weight, over the age of 18 and fluent in English. Participants received £6 for participating in an audiotaped interview lasting between 45 and 60 min (median = 51 min), carried out between May and June 2010 at the University of Southampton by GG. Data analysis Interviews were transcribed and analysed using an inductive thematic analysis approach (Braun & Clarke, 2006), informed by techniques from grounded theory (22), including line-by-line open coding grounded in the data and constant comparison of all instances of codes. Deviant case analysis was employed to ensure that perspectives that diverged from dominant trends were not overlooked (23), but all themes presented in this paper are based on the accounts of multiple participants. GG and LY immersed themselves in the data by reading the transcripts (a full coding and description of all the data was undertaken at this time and is available from the authors on request). A topic that emerged that seemed to have particular relevance to the intervention development was ‘people’s previous experiences with dieting’, and so for the purpose of the analysis presented here, all data relating to this theme was identified by GG. Further coding of this text into four sub-themes was undertaken by LY, and interrater agreement of all the final coded text relating to these themes was undertaken between LY, GG and KB.
Age 18–25 26–40 40–57 Nationality British Chinese European Work status Employed Student Level of education
