Stepped Care for Depression: A Systematic Review and Feasibility Study

Abstract

Background. Stepped care is widely implemented as a means to organise depression treatment. However, it is unclear how this system and the system it was designed to replace – long-term intensive psychotherapy for all – compare. Aim. To further the development and evaluation of stepped care. Specifically, assess the clinical effectiveness of stepped care and prepare for a fully-powered evaluation of stepped care vs. high-intensity psychotherapy alone for depressed adults. Design. A systematic review and mixed methods feasibility study encompassing a pilot randomised controlled trial and semi-structured interviews. Results of the systematic review. Fourteen randomised controlled trials involving 4580 participants were included. Relative to controls, there was significantly greater improvement in depression for adults treated with stepped care (d=0.34 at six months; 95% CI 0.20 to 0.48). The quality of included studies was good and there was little evidence of publication bias. All comparisons were with usual care. Results of the feasibility study. 66 patients were recruited to the pilot trial. The recruitment rate was 2.9% and follow-up data was obtained from 90.9% of participants. A third of stepped care patients stepped up to high-intensity therapy. Patients improved in both groups: the mean reduction in depressive symptoms was 13.4 in the stepped care group and 13.6 in the high-intensity therapy alone group. Recruitment methods were appropriate to patients and therapists but only somewhat appropriate to IAPT staff. Although the stepped care intervention was broadly acceptable to therapists, patient experience varied and some patients who demonstrated a low level of self-efficacy declined any therapy or dropped out of treatment. Conclusions. The effectiveness of stepped care compared with long-term intensive psychological therapy for all has not yet been established. A fully-powered trial of stepped care vs. high-intensity therapy alone is feasible although pilot trial methods and procedures should be modified to improve recruitment and acceptability.