Randomised controlled trial of a web-based self-management intervention for people with type 2 diabetes (HeLP-Diabetes).

Conference: 
Talk Code: 
3A.3
Presenter: 
Elizabeth Murray
Co-authors: 
E Murray, K Pal, C Dack, K Modrow, J Ross, F Stevenson, M Sweeting, S Parrott, J Li, S Michie, L Yardley, A Farmer, M Barnard.
Author institutions: 
University College London, Universities of Bath, York, Cambridge, Southampton, Oxford, and Whittington Hospital NHS Trust

Problem

Type 2 Diabetes Mellitus (T2DM) is an NHS priority, affecting 6% of the population, absorbing 10% of the NHS budget, and causing significant morbidity and mortality. Self-management education improves outcomes and GPs are incentivised through the Quality and Outcomes Framework to refer patients with T2DM for self-management education at diagnosis. However, uptake remains low (5.6% in 2014-15). Almost all structured education in the UK is group-based, which may not suit people who work, have family or other caring commitments, or simply do not like groups. Moreover, patient needs vary with time and illness progression, and a single education session at diagnosis is unlikely to meet these evolving needs. Web-based self-management support may help increase uptake and be affordable and deliverable at scale, but its effectiveness is unclear.

Approach

AimTo determine the effectiveness and cost-effectiveness of the HeLP-Diabetes self-management programme.MethodsDesign and setting: Individually randomised controlled trial in English primary care. Participants: Adults with T2DM.Intervention: HeLP-Diabetes, an evidence-based, theoretically informed web-based self-management programme developed with substantial user input. It addresses medical, emotional and role management, including behaviour change, decision, and peer support, with interactive tailored tools for self-assessment and self-monitoring. Comparator: simple information website. Outcomes: Joint primary outcomes were glycated haemoglobin (HbA1c) and diabetes-related distress (PAID). Secondary outcomes were clinical (BP, BMI, lipids); emotional (anxiety, depression); cognitive (self-efficacy, satisfaction); quality of life (EQ-5D) and economic (costs, health service use). Data were collected at baseline (prior to randomisation), and 3 and 12 months post-randomisation, with 12 months the primary end-point. Sample size and analysis: Allowing for 15% attrition, randomising 345 patients provided 90% power at 5% significance to detect a 0.25% difference in HbA1c and 4 point difference in PAID (both small effect sizes). The groups will be compared as randomised (intention to treat). Health economic analysis follows NICE guidance and determines the incremental cost-effectiveness ratio.

Findings

374 patients recruited from 20 practices were randomised. Baseline characteristics included mean age 64.8 years (SD 9.3), 69% (n = 258) male, 80.4% (n = 300) White British. Baseline mean HbA1c was 7.3% (SD 1.4) and PAID 19 (SD 18.6). Retention at 3 months was 86% for PAID and 82% for HbA1c and at 12 months was 90% for PAID and 84% for HbA1c (83% for both primary outcomes). 12 month follow-up completed on 15.2.16; data base lock is March 2016 and the results of the analyses will be available by May.

Consequences

Clinical Commissioning Groups are urgently seeking ways to increase uptake of diabetes self-management education, and the Five Year Forward View emphasises the need to embrace technology. The results of this trial will be of interest to policy makers, clinical commissioners, health care professionals and patients.

Submitted by: 
Elizabeth Murray
Funding acknowledgement: 
This presentation presents independent research funded by the National Institute for Health Research (NIHR) under its Programme Grants for Applied Research Programme (Grant Reference Number RP-PG-0609-10135). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health. Additional authors whose names did not fit in the author box: O O’Donnell, G Alkhaldi, D Patterson, B Fisher, C May.