Our team led by the brilliant Astrid Chevance just published a new paper, entitled “Identifying outcomes for depression that matter to patients, informal caregivers and healthcare professionals: qualitative content analysis of a large international online survey”. The paper is online in Lancet Psychiatry.
Rationale
When it comes to measuring outcomes in clinical studies of depression, we usually focus on assessing symptoms of depression (similar to assessing symptoms of e.g. COVID19 or the flu), and when there are no symptoms anymore, we conclude depression is gone. There are two core issues with this.
First, a recent review of about 200 psychotherapy studies showed that over 30 different symptom measures were used for depression. That is, it is unclear what symptoms to measure, because there are so many. I actually wrote a paper [blog post] on this a while ago, and visualized the differences of 7 common rating scales for depression that dramatically differ in symptom content. It becomes obvious that this is a problem when you go back to COVID19 or the flu: imagine there are over 30 different checklists available for doctors to see whether you have the flu, and all differ from each other. That would be a problem.
Second, symptoms are not all that matters. For example, we know from studies that even after depressed patients recover in their symptoms, they often still have considerable functional impairment (e.g. struggle in their relationships, cannot go to work, etc). So we cannot just declare patients “fully recovered” just because they show no symptoms anymore.
Our study
For these two reasons, Astrid Chevance started a project to establish a Core Outcome Set of depression measures, i.e. things that we should all measure and that matter. How to best do this?
Historically, this has been done by experts (often older white men) sitting around a table and talking about their patients. And I don’t want to unfair, there are fantastic phenomenological studies and case reports, going back over 150 years. But it’s 2020, and we are not aware of any single study that actually queried a broad range of people with lived experiences of depression, healthcare professionals, and informal caregivers to tell us what about their experiences, and outcomes that matter to them. So we did that, and received over 8183 open-ended answers from 1912 patients, 464 informal caregivers, & 627 health-care professionals from 52 countries.
This led to qualitative content analyses, i.e. trying to group these 8183 answers into categories of answers that overlap. Eventually, we identified 80 outcome domains that relate to treatment benefits, and 53 other outcome domains.
Results
The 80 outcome domains that relate to treatment benefits can be further grouped into symptom-related (red) and functioning-related domains (blue), and are summarized below (the width of lines is proportional to how often this domain was brought up by participants).
Of the 53 additional outcome domains (unrelated to treatment benefits, not included in the figure above), the most common ones were raising awareness of society about depression, reducing the stigma of depression (i.e. depression is not a weakness, a sign of laziness, or a lack of courage), and improving access to mental health services.
The paper also contains a lot of quotes in which people describe their first or second hand experiences with depression. These are not new to clinicians, of course, but these harrowing testimonies of what mental ill health can look like give a good impression of how it is to live with depression, or to live with someone afflicted by depression. I hope it will raise some further awareness for the topic of mental health.
There are many more testimonies in the paper. Thanks to Astrid, the entire team, and–most importantly–all participants of our survey. For sharing their stories and experiences with us, trying to start an initiative to establish a Core Outcome Set of depression measures that actually matter for people afflicted by depression, their caregivers, and their healthcare professionals. Current measures do not meet these criteria.
Pingback: Which depression measure is best? » Eiko Fried