Clinical and cost-effectiveness of remote-delivered, online lifestyle therapy versus psychotherapy for reducing depression: results from the CALM non-inferiority, randomised trial
Adrienne O'Neil,Joahna Perez,Lauren M Young,Tayla John,Megan Turner,Dean Saunders,Sophie Mahoney,Marita Bryan,Deborah N Ashtree,Felice N Jacka,Courtney Bruscella,Megan Pilon,Mohammadreza Mohebbi,Megan Teychenne,Simon Rosenbaum,Rachelle Opie,Meghan Hockey,Lucija Peric,Samantha De Araugo,Khyati Banker,India Davids,Monica Tembo,Jessica A Davis,Jerry Lai,Tetyana Rocks,Melissa O'Shea,Niamh L Mundell,Grace McKeon,Murat Yucel,Pilvikki Absetz,Vincent Versace,Sam Manger,Mark Morgan,Anna Chapman,Craig Bennett,Jane Speight,Michael Berk,Steve Moylan,Lara Radovic,Mary Lou Chatterton
DOI: https://doi.org/10.1016/j.lanwpc.2024.101142
2024-07-31
Abstract:Background: We conducted the first non-inferiority, randomised controlled trial to determine whether lifestyle therapy is non-inferior to psychotherapy with respect to mental health outcomes and costs when delivered via online videoconferencing. Methods: An individually randomised, group treatment design with computer-generated block randomisation was used. Between May 2021-April 2022, 182 adults with a Distress Questionnaire-5 score = ≥8 (indicative depression) were recruited from a tertiary mental health service in regional Victoria, Australia and surrounds. Participants were assigned to six 90-min sessions over 8-weeks using group-based, online videoconferencing comprising: (1) lifestyle therapy (targeting nutrition, physical activity) with a dietitian and exercise physiologist (n = 91) or (2) psychotherapy (Cognitive Behavioural Therapy) with psychologists (n = 91). The primary outcome was Patient Health Questionnaire-9 (PHQ-9) depression at 8-weeks (non-inferiority margin ≤2) using Generalised Estimating Equations (GEE). Cost-minimisation analysis estimated the mean difference in total costs from health sector and societal perspectives. Outcomes were assessed by blinded research assistants using Computer Assisted Telephone Interviews. Results are presented per-protocol (PP) and Intention to Treat (ITT) using beta coefficients with 95% Confidence Intervals (CIs). Findings: The sample was 80% women (mean: 45-years [SD:13.4], mean PHQ-9:10.5 [SD:5.7]. An average 4.2 of 6 sessions were completed, with complete data for n = 132. Over 8-weeks, depression reduced in both arms (PP: Lifestyle (n = 70) mean difference:-3.97, 95% CIs:-5.10, -2.84; and Psychotherapy (n = 62): mean difference:-3.74, 95% CIs:-5.12, -2.37; ITT: Lifestyle (n = 91) mean difference:-4.42, 95% CIs: -4.59, -4.25; Psychotherapy (n = 91) mean difference:-3.82, 95% CIs:-4.05, -3.69) with evidence of non-inferiority (PP GEE β:-0.59; 95% CIs:-1.87, 0.70, n = 132; ITT GEE β:-0.49, 95% CIs:-1.73, 0.75, n = 182). Three serious adverse events were recorded. While lifestyle therapy was delivered at lower cost, there were no differences in total costs (health sector adjusted mean difference: PP AUD$156 [95% CIs -$182, $611, ITT AUD$190 [95% CIs -$155, $651] ]; societal adjusted mean difference: PP AUD$350 [95% CIs:-$222, $1152] ITT AUD$ 408 [95% CIs -$139, $1157]. Interpretation: Remote-delivered lifestyle therapy was non-inferior to psychotherapy with respect to clinical and cost outcomes. If replicated in a fully powered RCT, this approach could increase access to allied health professionals who, with adequate training and guidelines, can deliver mental healthcare at comparable cost to psychologists. Funding: This trial was funded by the Australian Medical Research Future Fund (GA133346) under its Covid-19 Mental Health Research Grant Scheme.