Debt Counselling for Depression in Primary Care: an adaptive randomised controlled pilot trial (DeCoDer study)

Gabbay, M. B. et al. (2017) Debt Counselling for Depression in Primary Care: an adaptive randomised controlled pilot trial (DeCoDer study). Health Technology Assessment, 21(35), (doi: 10.3310/hta21350) (PMID:28648148) (PMCID:PMC5502372)

Abstract

Background: Depression and debt are common in the UK. Debt Counselling for Depression in Primary Care: an adaptive randomised controlled pilot trial (DeCoDer) aimed to assess the clinical effectiveness and cost-effectiveness of the addition of a primary care debt counselling advice service to usual care for patients with depression and debt. However, the study was terminated early during the internal pilot trial phase because of recruitment delays. This report describes the rationale, methods and findings of the pilot study, and implications for future research. Objectives: The overarching aim of the internal pilot was to identify and resolve problems, thereby assessing the feasibility of the main trial. The specific objectives were to confirm methods for practice recruitment and the ability to recruit patients via the proposed approaches; to determine the acceptability of the study interventions and outcome measures; to assess contamination; to confirm the randomisation method for main trial and the level of participant attrition; and to check the robustness of data collection systems. Design: An adaptive, parallel, two-group multicentre randomised controlled pilot trial with a nested mixed-methods process and economic evaluation. Both individual- and cluster (general practice)-level were was used in the pilot phase to assign participants to intervention or control groups. Setting: General practices in England and Wales. Participants: Individuals were included who were aged ≥ 18 years, scored ≥ 14 on the Beck Depression Inventory II and self-identified as having debt worries. The main exclusion criteria were being actively suicidal or psychotic and/or severely depressed and unresponsive to treatment; having a severe addiction to alcohol/illicit drugs; being unable/unwilling to give written informed consent; currently participating in other research including follow-up phases; having received Citizens Advice Bureau (CAB) debt advice in the past year; and not wanting debt advice via a general practice. Interventions: The participants in the intervention group were given debt advice provided by the CAB and shared biopsychosocial assessment, in addition to treatment as usual (TAU) and two debt advice leaflets. The participants in the control group were given advice leaflets provided by the general practitioner and TAU only. Main outcome measures: (1) Outcomes of the pilot trial – the proportion of eligible patients who consented, the number of participants recruited compared with target, assessment of contamination, and assessment of patient satisfaction with intervention and outcome measures. (2) Participant outcomes – primary – Beck Depression Inventory II; secondary – psychological well-being, health and social care utilisation, service satisfaction, substance misuse, record of priority/non-priority debts, life events and difficulties, and explanatory measures. Outcomes were assessed at baseline (pre-randomisation) and at 4 months post randomisation. Other data sources – qualitative interviews were conducted with participants, clinicians and CAB advisors. Results: Of the 238 expressions of interest screened, 61 participants (26%) were recruited and randomised (32 in the intervention group and 29 in the control group). All participants provided baseline outcomes and 52 provided the primary outcome at 4 months’ follow-up (14.7% dropout). Seventeen participants allocated to the intervention saw a CAB advisor. Descriptive statistics are reported for participants with complete outcomes at baseline and 4 months’ follow-up. Our qualitative findings suggest that the relationship between debt and depression is complex, and the impact of each on the other is compounded by other psychological, social and contextual influences. Conclusions: As a result of low recruitment, this trial was terminated at the internal pilot phase and was too small for inferential statistical analysis. We recommend ways to reduce this risk when conducting complex trials among vulnerable populations recruited in community settings. These cover trial design, the design and delivery of interventions, recruitment strategies and support for sites.

• Item Type: Articles
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Taylor, Professor Rod
• Authors: Gabbay, M. B., Ring, A., Byng, R., Anderson, P., Taylor, R. S., Matthews, C., Harris, T., Berry, V., Byrne, P., Carter, E., Clarke, P., Cocking, L., Edwards, S., Emsley, R., Fornasiero, M., Frith, L., Harris, S., Huxley, P., Jones, S., Kinderman, P., King, M., Kosnes, L., Marshall, D., Mercer, D., May, C., Nolan, D., Phillips, C., Rawcliffe, T., Sardani, A. V., Shaw, E., Thompson, S., Vickery, J., Wainman, B., and Warner, M.
• College/School: College of Medical Veterinary and Life Sciences > School of Health & Wellbeing > MRC/CSO SPHSU
• Journal Name: Health Technology Assessment
• Publisher: NIHR Journals Library
• ISSN: 1366-5278
• ISSN (Online): 2046-4924
• Copyright Holders: Copyright © 2017 Queen’s Printer and Controller of HMSO
• First Published: First published in Health Technology Assessment 21(35)
• Publisher Policy: Reproduced in accordance with the copyright policy of the publisher