Effectiveness of an integrated telehealth service for patients with depression: a pragmatic randomised controlled trial of a complex intervention

Background Many countries are exploring the potential of telehealth interventions manage the rising number of people with chronic conditions. However, evidence of effectiveness of telehealth is equivocal. Based on an evidence-based conceptual framework, we developed an integrated telehealth service (the ‘Healthlines Service’) for chronic conditions and assessed it’s effectiveness in patients with depression. Methods Pragmatic, multi-centre, randomised controlled trial comparing the Healthlines Depression Service plus usual care with usual care alone. The Healthlines Service comprised regular telephone calls from non-clinical, trained health advisors following standardised scripts generated by interactive software. Advisors supported participants to use online resources (including computerised cognitive-behavioural therapy) and sought to encourage healthier lifestyles, optimise medication and improve treatment adherence. We recruited participants from 43 general practices in three areas of England. Eligible participants required access to the internet and email, a PHQ9 score ≥ 10, and a confirmed diagnosis of depression. Participants were individually allocated in 1:1 ratio using an automated randomisation system, stratified by site, minimised by practice and PHQ-9 score. Participants were aware of their allocation but outcomes were analysed masked. The primary outcome was the proportion of participants responding to the intervention four months after randomisation (defined as PHQ9 <10 and reduction in PHQ-9 of ≥ 5 points), with further follow-up at eight and 12 months. Primary analysis was based on intention-to-treat without imputation. Trial registration: Current Controlled Trials (ISRCTN 14172341). Findings Between 24th July 2012 and 31st July 2013 we recruited 609 participants, randomly allocating 307 to the Healthlines Service and 302 to usual care. 525 (86%) provided primary outcome data. Response to treatment at four months was higher in the intervention arm (