Many champion wearables as devices that will revolutionise 21st century medicine. However, this technology has often failed to provide new insights for healthcare professionals and patients. Regarding behaviour change specifically, current research paints a mixed picture when demonstrating their effectiveness. While some patients claim that these devices are beneficial, recent clinical trials targeting weight loss observed that patients provided with a wearable tracker lost significantly less weight than patients provided with lifestyle support alone. Additional research also highlights failures in the design and implementation of similar devices. Nevertheless, these outcomes remain a key cornerstone in the literature because they emphasise the importance of understanding errors in order to capture the ideal functioning of any new device and/or intervention. Therefore, our white paper has two complementary aims that will hopefully help generate discussion and realise the potential of future technologies.

First, we identify patterns and document the key reasons why wearables and other mobile technologies often fail to change behaviour. The design of any digital device or intervention that aims to improve health and well-being combines elements of engineering, computer science and social science. Therefore, our paper critically considers aspects of (a) device design, (b) theoretical contributions and (c) the individual experience. A clear understanding of these issues can be derived not only from the latest research, but also by looking backwards at early attempts to self-tracking within health and social care (e.g. telehealth). This, in turn, provides new insights concerning how the next wave of wearable technology can be better designed and implemented – increasing the chances of success within specific target environments (e.g., schools, workplaces) and populations (e.g., adolescents, elderly adults).

Second, our recommendations provide guidance on how future research designs and outcome measures may need to be adapted in the future. For example, the vast majority of current investigations fail to adopt measures that are sensitive enough when it comes to capturing progress and accurately quantifying key outcomes. While research continues to rely on outcome measures used in traditional behavioural interventions (e.g., lifestyle modification), these are often not appropriate when feedback relating to physical activity is available 24/7. The absence of recording human-computer interactions is particularly evident when most wearable technology can easily collect this information (e.g., opening an application) alongside health related (e.g., physical activity) behaviours.