Supporting people with type 2 diabetes in effective use of their medicine through a system comprising mobile health technology integrated with clinical care

Background Type 2 diabetes affects 2.9 million people in the UK. Treatments to reduce risks of complications from type 2 diabetes are effective if taken as intended. Concerns about medicines and difficulties in taking them regularly, whether intentional or unintentional, are common. Some services to help are available, but evidence of their effectiveness and cost-effectiveness is weak. These services may not be right for everyone, and are not designed for repeated use. Not everyone has access to this extra help. Importance In addition to the burden of preventable death and disability, the cost of non-adherence with diabetes treatments has been estimated at £100 million a year in avoidable treatment costs. Understanding and improving this situation could make a major contribution to health and NHS costs. Need for Research Systematic reviews do not provide strong evidence to support the wider use of current approaches to developing interventions to support adherence. Brief messages delivered at a wide-scale and low cost via digital health systems added to usual care have been shown to be effective in improving health for some conditions and are a promising approach to the problem. Past and current research Systematic reviews of text-messaging to support adherence to treatment and mobile health interventions in diabetes identify some effective interventions. There are few trials testing the impact of brief messaging in type 2 diabetes; they do not have systematically developed interventions and are at risk of bias. Recent trials of SMS text-messaging for cardiovascular risk prevention and blood pressure lowering have shown clinically relevant changes in outcomes compared with usual care. Research Plans We will ask patients, clinicians and experts in behaviour change what sort of messages are likely to be relevant and acceptable for people starting and continuing taking a diabetes medicine. We will examine predictors of non-adherence in a retrospective primary-care cohort using routinely collected electronic health record data. We will explore, using qualitative methods, the added value of tailoring messages based on personal preferences, the type of medication prescribed, duration of treatment and timing of medication collection. We will carry out a feasibility study to inform decisions about the final design of the intervention and the trial to evaluate it. We will test the intervention in a randomised clinical trial and linked health economic study to estimate the extent of benefit and costs. We aim to understand how the intervention might work and be implemented by interviews with patients and health care staff, and using quantitative measures based on psychological constructs. Research team Members of this research team have previously worked together to develop new interventions and test them in trials funded by Oxford NIHR Biomedical Research Centre and others. These include interventions to help people take their medicine more regularly and mobile-phone based interventions for people with diabetes. The research team will work in partnership with Patient Groups, NHS Trusts, Research Networks, and the Health and Social Care Information Centre to deliver this research. Research Environment Applicants are based at leading academic institutions with strong training programmes for early-career clinical researchers and methodologists. The lead Department has a silver Athena SWAN award. There are strong links with patient involvement groups to ensure a strong PPI reference group for the programme, well developed clinical research networks to support recruitment, and good links with an Academic Health Sciences Consortium specifically interested in facilitating digital health care projects. Outputs This research will provide evidence about how to give support to patients, in addition to their usual care, using a range of messages targeting a wide range of the factors that can contribute to stoppin