Project objectives

To better manage the high prevalence of depression in the community, many services seek to provide simple, accessible and effective forms of psychological therapy (so called ‘low intensity’ interventions) to the majority of depressed patients in the first instance, with more intensive and costly treatments retained for patients who do not benefit from ‘low intensity’ interventions. However, despite large numbers of patients receiving such interventions, there is a lack of evidence to assist clinical decision-making about which patients should receive ‘low intensity’ interventions. We assessed whether initial severity of depression was a key determinant of the benefit that depressed patients derive from ‘low-intensity’ interventions.

Changes to project objectives

Research Plan and Methodology:

The original plan was that 3 months of the time would be spent analysing routine data sets, to complement the main individual patient data analysis. However, given changes in the project team with the loss of the PI and difficulties in getting timely access to data, the planned analyses could not be completed, and the analysis focussed on the  individual patient data analysis

Brief summary

Methods

Individual patient data meta-analysis, using 16 datasets comparing ‘low intensity’ interventions with usual care, from primary care and community settings, and including 2470 patients with depression. ‘Low intensity’ interventions for depression (such as guided self-help using written materials and limited professional support, and internet delivered interventions). Main outcomes were depression outcomes, and the effect of initial depression severity on the effects of ‘low intensity’ interventions

Results

Although referred for ‘low intensity’ interventions, baseline depression severity highlighted that many patients had significant symptoms. We found a significant interaction between baseline severity and treatment effect (coefficient -0.1, 95% CI -0.19 to -0.002), suggesting that patients who are more severely depressed at baseline demonstrate larger treatment effects than those who are less severely depressed. However, the magnitude of the interaction (equivalent to an additional drop of around 1 point on the Beck Depression Scale for a one standard deviation increase in initial severity) was small and may not be clinically significant.

Conclusions and impact

The data suggests that patients with more severe depression at baseline still show at least as good clinical benefit from ‘low intensity’ interventions as less severely ill patients and could be offered these treatments as part of a stepped care model. The data can inform the further development of the NICE guidelines for depression.

Plain English summary

Significant variation in response to treatments, and failure to take into account patient preferences, mean that treatments for depression are currently delivered inefficiently, with frequent non-adherence to medication and non-attendance for psychological treatments. The NICE guidelines recommend a ‘stepped care’ approach the aim of which is to treat patients with the least intensive intervention necessary, and offer more intensive treatments only to non-responders to less intensive treatments. However, it may be more efficient to refer patients who are unlikely to respond to less intensive treatments straight to more intensive treatments, assuming valid predictors of non-response to less intensive treatments can be identified. First we need to establish whether such predictors exist. We propose: (i) to analyse predictors of non-response to low intensity treatments for depression in the Doncaster Improving Access to Psychological Therapies (IAPT) cohort for which we already have the data, (ii) to update a systematic review of predictors of non-response in reported secondary analyses of published trials of low intensity treatments, and (iii) to carry out a meta-analysis of individual patient data gathered from authors of the published trials where possible. Successful achievement of these objectives would lead on to an application for funding for a trial of stepped care versus stratified care, assuming predictors of non-response can be identified.

Dissemination

Published articles

Bower P, Kontopantelis E, Sutton A, Kendrick T, Richards D, Gilbody S et al. Influence of initial severity of depression on effectiveness of low intensity interventions: meta-analysis of individual patient data.

British Medical Journal 2013;346 https://www.bmj.com/content/346/bmj.f540

Kontopantelis E, Reeves D. A short guide and a forest plot command (ipdforest) for one-stage meta-analysis Stata Journal, 2013, 13(3), 574-587.

Public involvement

The original plan was to involve a service user linked to the Southampton group, but this was not viable when the project moved. Given that the project had been funded, involving new PPI members in Manchester was not seen as optimal as they would not have chance for genuine involvement, and given the nature of the project, the lack of PPI involvement was not considered critical.

Impact

We expect that the results will be picked up by the next iteration of the NICE depression guidelines and will inform developments in the stepped care framework

The work has led to additional links with international collaborators in the Netherlands, through Dr Kontopantelis, which has led to another paper using similar methods submitted to PLOS Medicine (see below)

The methods have also been adopted by Dr Coventry in the NIHR School funded project ‘Characteristics of patient populations that determine the effectiveness of collaborative care for depression: meta-analysis of individual participant data in people with long-term conditions’

This project was funded by the National Institute for Health Research School for Primary Care Research (project number 76)

Department of Health Disclaimer

The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR School for Primary Care Research, NIHR, NHS or the Department of Health.