"Doing well" : an initiative to improve depression care

• Main Author: Smith, Michael J.
• Published: University of Glasgow 2010
• Subjects: 610.7343; R Medicine (General)
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.559897

Aim: The aim of this thesis was to describe the service use, clinical outcomes and prescribing change associated with the implementation of a complex intervention designed to improve care for people with depression in a primary care setting. Background: Health systems have limited capacity to provide appropriate psychological and pharmacological treatments for people with depression. Although guidance on the treatment of depression in primary care in the UK was clarified by the National Institute for Clinical Excellence (NICE) in 2004, it is generally acknowledged that the current diagnostic classification of depression is not satisfactory. Antidepressant prescriptions have continued to rise in Scotland since the mid-1990s, even though there is no indication that the incidence or prevalence of depression is increasing. There is limited access to psychological therapies. Health services have not implemented consistent packages or systems of care in order to provide adequately for patient needs. Although the welfare of staff is critical to their therapeutic engagement with patients, this is rarely an explicit focus of health systems design. Method: This thesis describes an observational study examining the implementation of a complex intervention to improve depression care called “Doing Well”. The intervention was based in 14 General Practices in Renfrewshire, a mixed urban-rural area in Scotland. The catchment population for the study was 76,000 people. A small team of clinicians implemented a programme for people with low mood, depression and adjustment disorder, based in primary care. This programme incorporated a number of changes to standard mental health care, including the following: no “severity threshold” for referral to secondary care; the routine use of an objective measure of depression severity with continuous outcome monitoring; a paperless clinical record; prompt access to guided self-help; prompt “step-up” care to more formal psychological therapy or medical care if indicated; and careful attention to staff training and satisfaction. Findings: 1501 out of 1584 people referred to the programme met inclusion criteria and were included in the study. Three hundred and thirty-two people (22%) did not attend any appointment; 320 (21%) dropped out of treatment after at least one contact. One hundred and ninety-five people (13%) subsequently had their care transferred to other services (of which 43% were to secondary care mental health services), and 654 (44%) of patients completed treatment per protocol. There was good fidelity to the intended model of care, with patients in the “treatment complete” group receiving “brief interventions” of an average of five contacts. These contacts totalled 151 minutes over an average of 103 days of treatment. Referrals from GPs continued at a high and stable level throughout the period of the evaluation. Median waiting times of 15 days were satisfactory. The mean reduction in PHQ for patients completing treatment was 10.6 points, representing a reduction from baseline of 62%. Seventy-two percent of the treatment complete group showed a PHQ drop greater or equal than 50%, compared with seven percent in the “disengaged” and ten percent in the “transfer of care” groups. Doing Well received a lower than expected proportion of referrals from deprived areas, and there was a small negative association between clinical outcome and living in a more deprived area. Defined daily doses of antidepressants in the practices that had access to the Doing Well clinical intervention increased less rapidly (5.3% between the 12 months to June 2004 and the 12 months to June 2008) than in neighbouring areas or Scotland as a whole (15.8% over the same period). Gross ingredient costs of antidepressants in the Doing Well practices fell more substantially over this period (to 56% of baseline) than in Scotland as a whole (to 65% of baseline). Formulary compliance increased more rapidly in the Renfrewshire area than in a neighbouring area which used the same formulary, but had no contact with Doing Well. Conclusions: It was feasible to implement and sustain a system of care for depression that was consistent with NICE guidance, including the provision of some form of psychological therapy (including guided self help) for all who needed it. Access to the service was acceptable, and retention within the service compares favourably with equivalent studies in other parts of the UK. Clinical outcomes were satisfactory, but it was not possible to compare with outcomes in usual care in this observational study. Doing Well practices showed a reduction in the rate of rise of antidepressant use, although did not stop the rise altogether. The implications of this form of “stepped care” for depression for service development are discussed.