Collaborative mental health care in the NHS has small but meaningful benefits for people with depression

Collaborative care, that places a care manager in primary care to deliver treatment and coordinate care between GPs and specialists, improved recovery of people with moderate to severe depression. Primary responsibility for prescribing remained with the GP. This large trial found that improvements were modest, but were similar to those found in an evaluation of the Improving Access to Psychological Therapies programme.

The collaborative care model was cost-effective too. Collaborative care cost on average £272.50 per participant and had a cost per quality-adjusted life year (QALY) of £14,248, which is within NHS willingness-to-pay thresholds. Because depression is so common, the implementation of a collaborative care model is likely to require a considerable investment of resources. It would also be necessary to get GPs on board – the researchers found it was difficult for GPs to engage with the collaborative care framework.

Why was this study needed?

Depression affects about 6% of men and 10% of women each year, and is the reason for two-thirds of all suicides. It is frequently long-lasting, and about 50% of people who have suffered one depressive episode will experience additional episodes. GPs are often the clinician caring for people with depression, and although effective treatments for depression exist, there is often very limited support for GPs or time available to provide non-pharmacological therapies.

The collaborative care model introduces a care manager (see Definitions tab) into the primary care system to help deliver care and liaise between GPs and specialists. The model was mostly developed in the USA and has been shown to be effective for depression management there. However, there are differences between the primary care systems in the US and UK – there is little UK evidence on its clinical- and cost-effectiveness. NICE issued a research recommendation that chronic disease management programmes for depression, such as collaborative care, should be tested in large-scale multicentre trials. The intervention tested here was tested for acceptability as part of the design of this trial.

What did this study do?

This cluster randomised controlled trial allocated 51 UK primary care practices to give usual care or usual care plus collaborative care for 14 weeks to 581 people with moderate to severe depression. Usual care included antidepressant therapy and/or referral to specialist mental health care as deemed necessary by the GP. Collaborative care consisted of a care manager providing a face-to-face session and six to 12 telephone contacts with the participant, plus regular liaison with the GP. Details of the collaborative care intervention can be found in the definitions tab.

The average age of participants was 44.8 years and 72% were women; 44% were in full- or part-time employment. Just over half of the participants had moderately severe depression, with a further 30% having severe depression. At the start of the trial, 83% had been prescribed antidepressant drugs by their GP. Almost two-thirds had a long-standing physical illness.

The primary outcome was depression scored on the PHQ-9 score. The researchers also wanted to ensure that the results could be easily interpreted from a clinical perspective. They calculated rates of ‘recovery’ as the proportion of participants with a PHQ-9 score of 9 or less at the end of the trial.

Due of the nature of the intervention, it was only possible to blind research workers to the allocation and not the participants, care managers or GPs.

What did it find?

  • After four months, more people receiving collaborative care had recovered (47%) compared to those in the usual group (35%) (odds ratio 1.67, 95% confidence interval [CI] 1.22 to 2.29). This gave a number needed to treat of 8.4.
  • Depression scores on a 28 point scale (PHQ-9) were 1.33 lower for collaborative care participants after four months (95% CI ‑0.35 to ‑2.31). Both groups reduced from moderately severe depression to moderate depression on this scale; the collaborative group reduced from 17.4 to 11.1 while usual care group reduced from 18.1 to 12.7. After 12 months the collaborative care group score was 1.36 points lower (95% CI ‑0.07 to ‑2.64), but there was no difference after three years.
  • Participants receiving collaborative care reported being significantly more satisfied with treatment than those in usual care.
  • Although 98% of participants in the trial had a secondary diagnosis of anxiety, there was no difference between groups on anxiety outcomes at any time-point.
  • Collaborative care cost on average £272.50 per participant (care manager costs at £232 and clinical supervision costs of £40.50). Health and social care service use costs (excluding the collaborative care intervention) were £1,571 and £1,614 for usual care and collaborative care participants respectively.
  • Collaborative care had an estimated mean cost per QALY of £14,248, which is within NHS willingness-to-pay thresholds. There was a 58% probability that collaborative care is cost-effective compared to usual care at a willingness to pay of £20,000 per QALY, and 65% at a willingness to pay of £30,000 per QALY.

What does current guidance say on this issue?

NICE do not make any recommendation around collaborative care for people with depression only.

The 2009 NICE guideline on depression in adults recommends considering collaborative care (in a stepped care model) for people with moderate to severe depression when their depression is in the context of a chronic physical health problem with associated functional impairment.

What are the implications?

Although the mean quality-adjusted life year gain is modest, it is comparable to that recently reported in the evaluation of the IAPT service. It seems likely that IAPT may consider implementing collaborative care into their service model.

The main component of the collaborative care intervention that predicted better outcomes for patients was the amount of behavioural activation undertaken—the more the better. This was despite the fact that there was considerable variation in how behavioural activation was both explained and used by care managers. Further research could examine behavioural activation and the other different components of collaborative care in isolation.

The researchers found that GPs’ understanding of collaborative care, compounded by organisational barriers, hindered their engagement in the intervention. While case managers and supervisors regarded collaborative care as coherent, implementation of collaborative care was made difficult by GPs’ lack of engagement. All participants identified the potential of a collaborative approach. NIHR is funding other work in this area including, for example, a trial of case management in primary care for people with depression and other physical health problems.


Citation and Funding

Richards DA, Bower P, Chew-Graham C, et al. Clinical effectiveness and cost-effectiveness of collaborative care for depression in UK primary care (CADET): a cluster randomised controlled trial. Health Technol Assess. 2016;20(14):1-192.

This project was funded by the Medical Research Council (MRC) (G0701013) and managed by the National Institute for Health Research (NIHR) on behalf of the MRC–NIHR partnership.



NICE. Depression in adults: recognition and management. CG90. London: National Institute for Health and Care Excellence; 2009.

NICE. Common mental health problems: identification and pathways to care. CG123. London: National Institute for Health and Care Excellence; 2011.

Mukuria C, Brazier J, Barkham M, et al. Cost-effectiveness of an improving access to psychological therapies service. Br J Psychiatry. 2013;202(3):220–7.

Ross S, Curry N, Goodwin N. Case management: What it is and how it can best be implemented. London: The King’s Fund; 2011.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

Care managers were existing NHS mental health workers who undertook care management alongside their usual NHS caseload in the community. They were trained over five days and supervised by psychiatrists and psychological therapists from the trial team. The sessions they provided included symptom assessment, solving any antidepressant issues, support of self-guided “behavioural activation” (a brief low-intensity psychosocial treatment for depression), and advice on relapse prevention.

The Patient Health Questionnaire (PHQ-9) is a brief self-report tool which is used for screening purposes, to aid in the diagnosis of depression and for monitoring the severity. It consists of nine questions about how frequently the person has had problems in the previous week with symptoms such as lack of interest, poor sleep or concentration, change in appetite and suicidal thoughts. Scores indicate the following levels of depression: 5 to 9 (mild), 10 to 14 (moderate), 15 to 19 (moderately severe) and 20 to 28 (severe).

People with a PHQ-9 score of less than 9 were considered “recovered”.


Expert commentary

This well-designed study confirms the potential of collaborative care to confer a small but meaningful, and cost-effective, improvement in the clinical effectiveness of the primary care management of depression. Superiority over usual care was demonstrated at four and 12-month follow-up but not at 36 months. Interestingly, the effect was fully mediated by behavioural activation; medication adherence did not mediate outcome. There remain questions regarding mechanism and the optimal configuration of collaborative care but, as delivered in this study, it is worthy of implementation as it will improve outcomes for this often extremely disabling condition.

Professor Jonathan I Bisson, Director, Health and Care Research Wales