The Effects of a Commitment Device on Health Outcomes: Reputational Commitment and Weight Loss in an Online Experiment

Introduction

Obesity remains a top public health priority, with 27% of adults in England and 40% of adults in the US now obese (Fuller, Mindell, & Prior, 2015; Hales, Carroll, Fryar, & Ogden, 2017). Public health discourse widely applies the term ‘epidemic’ in describing obesity trends (Butland et al., 2007, p. 17). Medical treatments are advancing, policy interventions increasingly target the ‘obesogenic environment’, and effective prevention strategies are known to combine an active lifestyle with a diet; but for any of these approaches to deliver sustained weight management, individual behaviour change is fundamental.

Like other preventative health behaviours, maintaining a healthy exercise and dietary regimen involves intertemporal choices: tradeoffs between satisfying our current desires and achieving our longer term objectives (Rogers, Milkman, & Volpp, 2014). Left to manage ourselves, too often we privilege the near-term payoff over future wellbeing (O’ Donoghue & Rabin, 1999). Such ‘present bias’ can explain the failure to follow up on good intentions with concrete action (Liu, Wisdom, Roberto, Liu, & Ubel, 2014), and has been linked empirically to weight management issues (Fan & Jin, 2013). A potential solution is pre-commitment strategies, or commitment devices, that bind our future selves to certain behaviours (Strotz, 1955; Thaler & Shefrin, 1981). Such strategies have been shown to promote smoking cessation (Gine, Karlan, & Zinman, 2010; Halpern et al., 2015), exercise (Prestwich et al., 2012; Royer, Stehr, & Sydnor, 2015) and a switching to safer water sources (Inauen, Tobias, & Mosler, 2014).

Where commitment devices have been tested as weight management aids they have shown some promising results.¹ Financial commitment devices in the form of a deposit contract stake money on achieving a weight loss outcome have been shown to have positive short-term effects (Volpp et al., 2008). An alternative form of commitment device relies on reputational rather than monetary stakes, and these too have been shown to promote weight loss: Nyer and Dellande (2010) report improved weight loss when goals are posted to a public gym noticeboard; Prestwich et al (2012) report greater weight loss when exercise is planned with a partner rather than as a solo activity. However, other studies have shown more mixed results from planning and commitment strategies (Chapman, Campbell, & Wilson, 2015; Verhoeven, Adriaanse, Ridder, Vet, & Fennis, 2013). Questions remain around the impact of reputational commitment devices.

What is striking is that none of these studies are set in the context of digital interventions, defined as programmes that promote health via a digital platform such as a website or smartphone (Alkhaldi, Hamilton, & Murray, 2016, p. 1). Recent studies highlight the considerable potential of such tools to improve health, and their increasing application to a wide range of health issues including weight management (Alkhaldi et al., 2016; Murray et al., 2016). A gap exists however in our understanding of how commitment devices may promote weight loss attempts supported by digital interventions.

This paper contributes to the scholarly debate with a novel test of how a reputational commitment device affects weight outcomes in a digital health setting. A randomised control trial introduced an additional reputational commitment for some online users of a weight management service. The comparison group comprised fee-paying clients. A second group of participants were asked to nominate an independent referee (a ‘coach’) who could verify their progress, thereby adding a reputational commitment element to their weight loss endeavours. A third group were offered a refund on their monthly subscription fee.