Randomised controlled trial of non-directive counselling, cognitive-behaviour therapy and usual general practitioner care in the management of depression as well as mixed anxiety and depression in primary care.

Abstract

OBJECTIVES: The aim of this study was to determine both the clinical and cost-effectiveness of usual general practitioner (GP) care compared with two types of brief psychological therapy (non-directive counselling and cognitive-behaviour therapy) in the management of depression as well as mixed anxiety and depression in the primary care setting. DESIGN: The design was principally a pragmatic randomised controlled trial, but was accompanied by two additional allocation methods allowing patient preference: the option of a specific choice of treatment (preference allocation) and the option to be randomised between the psychological therapies only. Of the 464 patients allocated to the three treatments, 197 were randomised between the three treatments, 137 chose a specific treatment, and 130 were randomised between the psychological therapies only. The patients underwent follow-up assessments at 4 and 12 months. SETTING: The study was conducted in 24 general practices in Greater Manchester and London. SUBJECTS: A total of 464 eligible patients, aged 18 years and over, were referred by 73 GPs and allocated to one of the psychological therapies or usual GP care for depressive symptoms. INTERVENTIONS: The interventions consisted of brief psychological therapy (12 sessions maximum) or usual GP care. Non-directive counselling was provided by counsellors who were qualified for accreditation by the British Association for Counselling. Cognitive-behaviour therapy was provided by clinical psychologists who were qualified for accreditation by the British Association for Behavioural and Cognitive Psychotherapies. Usual GP care included discussions with patients and the prescription of medication, but GPs were asked to refrain from referring patients for psychological intervention for at least 4 months. Most therapy sessions took place on a weekly basis in the general practices. By the 12-month follow-up, GP care in some cases did include referral to mental healthcare specialists. MAIN OUTCOME MEASURES: The clinical outcomes included depressive symptoms, general psychiatric symptoms, social function and patient satisfaction. The economic outcomes included direct and indirect costs and quality of life. Assessments were carried out at baseline during face-to-face interviews as well as at 4 and 12 months in person or by post. RESULTS: At 4 months, both psychological therapies had reduced depressive symptoms to a significantly greater extent than usual GP care. Patients in the psychological therapy groups exhibited mean scores on the Beck Depression Inventory that were 4-5 points lower than the mean score of patients in the usual GP care group, a difference that was also clinically significant. These differences did not generalize to other measures of outcome. There was no significant difference in outcome between the two psychological therapies when they were compared directly using all 260 patients randomised to a psychological therapy by either randomised allocation method. At 12 months, the patients in all three groups had improved to the same extent. The lack of a significant difference between the treatment groups at this point resulted from greater improvement of the patients in the GP care group between the 4- and 12-month follow-ups. At 4 months, patients in both psychological therapy groups were more satisfied with their treatment than those in the usual GP care group. However, by 12 months, patients who had received non-directive counselling were more satisfied than those in either of the other two groups. There were few differences in the baseline characteristics of patients who were randomised or expressed a treatment preference, and no differences in outcome between these patients. Similar outcomes were found for patients who chose either psychological therapy. Again, there were no significant differences between the two groups at 4 or 12 months. Patients who chose counselling were more satisfied with treatment than those who chose c