Importance of depression trial outcomes
Research type
Research Study
Full title
A survey of the importance of outcomes for depression trials
IRAS ID
344064
Contact name
Til Wykes
Contact email
Sponsor organisation
King's College London
Duration of Study in the UK
1 years, 6 months, 9 days
Research summary
Research Summary: Depression is a severe and enduring illness and is expected to be a leading worldwide health problem by 2030. Numerous treatments are currently provided, including antidepressants with a lower uptake for cognitive therapies, such as talking therapies, despite evidence showing they are similar in effectiveness. But, a treatment may not always be effective. Research outcomes are used to evaluate how well any treatment works and are typically decided on by researchers, with little input from those using the services or providing them. This means that outcomes set by the research team may not reflect those outcomes valued by service-users or service providers.
This study aims to bring some balance to this by collecting service-users and service staffs views in which outcomes are important to them. There are various outcomes and can include clinical (e.g., symptoms, severity of illness), cognitive (e.g., memory, attention) or functional (e.g., quality of life). Researchers usually choose one as their primary outcome to determine if a treatment is successful.
This study seeks to present service-users and service staff up to two lists of a total of 13 possible outcomes and will ask them to rank each outcome in order of importance to them. They will be asked to do this by assigning a total of 100 points to each outcome list. This information will then be used to give each of those outcomes a score that can then produce a model that future research can use to aid not only in choosing appropriate outcomes that are valued by service-users and service staff but also to give each of them an appropriate ‘weight’.
We will invite a total of 100 service-users who meet the eligibility criteria and have given their consent and 50 service providers to complete both online questionnaires.
Summary of Results: Depression is a common but serious condition that can affect many areas of a person’s life. Research studies usually judge whether a treatment works by assessing changes in depression symptoms. However, people with lived experience of depression often say that other outcomes such as quality-of-life or social functioning matter just as much, if not more. This means current research may not always reflect what the most important outcome is for the people receiving treatment.
To explore this assumption, we asked three groups of people (people with a few or no current symptoms, those currently experiencing depression, and clinicians who support people with depression such as therapists, psychologists or doctors) to tell us which outcomes they consider most important. These outcomes were taken from two completed UK depression trials evaluating cognitive behavioural therapy (CBT), a recommended first choice treatment for depression. In consultation with people with lived experience, we used a simple point-allocation method, where each person allocated 100 points across two lists of possible outcomes. A total of 229 people took part.
Depression symptoms were important but were given no more points on average than other outcomes including anxiety symptoms, quality-of-life, and impact on daily activities due to mental health issues. This shows that people value a broader range of outcomes equally than those typically used to judge whether a treatment works in depression trials. Quality-of-life was rated as highly important, but it is not measured often in depression trials.
Across the three groups, there were both similarities and differences in what participants prioritised. For example, clinicians placed less importance on anxiety symptoms relative to depression symptoms, service-use and employment outcomes compared to currently depressed participants. Differences linked to personal characteristics only appeared in the full sample and people currently experiencing depression, who are also the group usually recruited into treatment trials. Age could predict outcome importance in the full sample and currently depressed group, with older adults consistently prioritising depression symptoms relative to service-use and employment outcomes than younger adults.
This suggests that the treatment outcomes chosen for research may not fully represent the diversity of what matters to people receiving treatment. This finding challenges the way current trials judge treatment benefit and suggests that research needs to measure a wider range of outcomes to better reflect what matters most to the people who are being treated for depression as well as to those who provide depression treatments. However, asking stakeholders their priorities alone does not influence trial analyses and usual practice remains to decide if a treatment works on a single depression symptom based outcome.
To address this, we also explored whether a method that has not been used much in mental health trials may improve the way we assess trials from this stakeholder perspective. Using the information we collected about the importance of outcomes, we used a method called Multi-Criteria Decision Modelling (MCDM) to combine the importance of the outcomes rated by the three stakeholder groups, with the original trial outcome information. This produced a composite outcome, that combines several outcomes into one single measure, weighted by how important each one is to the stakeholders. We then re-ran the original trial analyses, replacing the single symptom-based primary outcome with our stakeholder-weighted composite, taking into account how important each is to the stakeholders.
For both trials, we did not find that using the composite, changed the overall conclusions of the original trials. Our findings showed that the groups who received CBT performed better than the group who did not at the end of the trial. However, this does not mean that all future trials using this method will not find differences. For one trial, we saw a slightly larger improvement compared to the group who did not receive CBT when using the composite compared to the original primary outcome analyses, whereas the other trial the improvement was smaller than when looking at only the original primary outcome.
Overall, the conclusion that CBT works remained when assessed against a broader set of outcomes, weighted by how much they mattered to stakeholders. This is important as there have been concerns about the quality of the research evidence for CBT.
These studies provide the first evidence in depression of using a point-allocation method to collect information about the importance of outcomes from stakeholders and using a multi-criteria decision model to create stakeholder-weighted composite outcomes to assess how well a treatment works from stakeholder perspectives. This method is intended as complementary alongside traditional ways of looking to trials, and not a replacement for meaningful patient and public involvement across the research lifecycle. This method offers a new perspective on how well a treatment works by reflecting how much multiple outcomes matter to those who receive and provide the treatment.
REC name
West Midlands - South Birmingham Research Ethics Committee
REC reference
25/WM/0067
Date of REC Opinion
17 Apr 2025
REC opinion
Favourable Opinion