Clinical effectiveness and cost-effectiveness of cognitive behavioural therapy as an adjunct to pharmacotherapy for treatment-resistant depression in primary care: the CoBalT randomised controlled trial

Background: Only one-third of patients with depression respond fully to treatment with antidepressant medication. However, there is little robust evidence to guide the management of those whose symptoms are 'treatment resistant'. Objective: The CoBalT trial examined the clinical effectiveness and cost-effectiveness of cognitive behavioural therapy (CBT) as an adjunct to usual care (including pharmacotherapy) for primary care patients with treatment-resistant depression (TRD) compared with usual care alone. Design: Pragmatic, multicentre individually randomised controlled trial with follow-up at 3, 6, 9 and 12 months. A subset took part in a qualitative study investigating views and experiences of CBT, reasons for completing/not completing therapy, and usual care for TRD. Setting: General practices in Bristol, Exeter and Glasgow, and surrounding areas. Participants: Patients aged 18-75 years who had TRD [on antidepressants for >= 6 weeks, had adhered to medication, Beck Depression Inventory, 2nd version (BDI-II) score of >= 14 and fulfilled the International Classification of Diseases and Related Health Problems, Tenth edition criteria for depression]. Individuals were excluded who (1) had bipolar disorder/psychosis or major alcohol/substance abuse problems; (2) were unable to complete the questionnaires; or (3) were pregnant, as were those currently receiving CBT/other psychotherapy/secondary care for depression, or who had received CBT in the past 3 years. Interventions: Participants were randomised, using a computer-generated code, to usual care or CBT (12-18 sessions) in addition to usual care. Main outcome measures: The primary outcome was `response', defined as >= 50% reduction in depressive symptoms (BDI-II score) at 6 months compared with baseline. Secondary outcomes included BDI-II score as a continuous variable, remission of symptoms (BDI-II score of < 10), quality of life, anxiety and antidepressant use at 6 and 12 months. Data on health and social care use, personal costs, and time off work were collected at 6 and 12 months. Costs from these three perspectives were reported using a cost-consequence analysis. A cost-utility analysis compared health and social care costs with quality adjusted life-years. Results: A total of 469 patients were randomised (intervention: n = 234; usual care: n = 235), with 422 participants (90%) and 396 (84%) followed up at 6 and 12 months. Ninety-five participants (46.1%) in the intervention group met criteria for `response' at 6 months compared with 46 (21.6%) in the usual-care group {odds ratio [OR] 3.26 [95% confidence interval (CI) 2.10 to 5.06], p < 0.001}. In repeated measures analyses using data from 6 and 12 months, the OR for `response' was 2.89 (95% CI 2.03 to 4.10, p < 0.001) and for a secondary `remission' outcome (BDI-II score of < 10) 2.74 (95% CI 1.82 to 4.13, p < 0.001). The mean cost of CBT per participant was 910 pound, the incremental health and social care cost 850 pound, the incremental QALY gain 0.057 and incremental cost-effectiveness ratio 14,911 pound. Forty participants were interviewed. Patients described CBT as challenging but helping them to manage their depression; listed social, emotional and practical reasons for not completing treatment; and described usual care as mainly taking medication. Conclusions: Among patients who have not responded to antidepressants, augmenting usual care with CBT is effective in reducing depressive symptoms, and these effects, including outcomes reflecting remission, are maintained over 12 months. The intervention was cost-effective based on the National Institute for Health and Care Excellence threshold. Patients may experience CBT as difficult but effective. Further research should evaluate long-term effectiveness, as this would have major implications for the recommended treatment of depression.