Depression Detectives - User-led Citizen Science
Abstract
Depression Detectives was a pilot, online, citizen-science project that brought together people with lived experience (PWLE) of depression, and data-science researchers. The question chosen by the group was: “Do people report all episodes of depression to their GP? And if not, why not?”
Within UK Biobank, 67% of people whose questionnaire answers suggested depression had no mention of it in their GP records. Similarly, 84% of survey respondents saw a doctor for ≤50% of episodes. Focus-group discussions revealed many people only went to the GP for more serious episodes or for practical help (such as to be signed off work). Half regretted not going and 50% of respondents used private talking therapy, exercise, meditation, yoga, social contact and so on, which did not appear in official records. These findings have implications for data scientists, doctors and policymakers.
Introduction
The team that designed, led and facilitated this project comprised: Iona Beange (a knowledge exchange and impact officer working on a Medical Research Council Mental Health Data Pathfinder grant); Andrew McIntosh and his big data and imaging team at the Division of Psychiatry, University of Edinburgh; Sophia Collins (a public engagement consultant); Christine Kupfer, (a researcher in social sciences, counselling and psychology); and Naomi Oppenheim (who has a background in social media moderation).
Edinburgh Neuroscience, an interdisciplinary cross-college community at the University of Edinburgh, helped locate other researchers with an interest in depression, particularly topic experts for Q&A sessions (such as social scientists with a knowledge of trauma). Edinburgh Neuroscience also helped advertise the project and recruit public participants, and allowed us to present our work at their annual conference.
Who should benefit?
Depression is the number one cause of ill-health and disability worldwide, according to the World Health Organization (WHO). Over 5% of adults globally (~300 million people) were affected pre-pandemic and the WHO estimates that coronavirus disease 2019 (COVID-19) has contributed to an additional 53.2 million cases worldwide—an increase of 28%.
Depression can affect people of any nationality or socio-economic status. However, those who have lived through trauma and stressful life events are at increased risk.
Depression affects every aspect of a person’s life. Lost productivity alone is estimated to cost the global economy US$1 trillion per year and is forecast to reach $16 trillion by 2030.
Symptoms include low mood and loss of interest in activities; disrupted concentration, sleep and appetite; and feelings of excessive guilt, low self-worth and hopelessness that can lead to suicide.
Despite substantial advances in research, and demonstrations of the cost-effectiveness of medications and psychosocial interventions, delivery at scale and translation into real-world benefits are slow. The UK Government has invested millions in mental health research over the last few years, but the Royal College of Psychiatrists has identified an urgent need for more. What should this research be looking at? How will it translate into real-world benefits? Have we asked people with depression what they think?
User-led citizen science (CS) can create people-centred research to address the questions that matter to participants. Depression Detectives, which ran between February and September 2021, brought together 70 PWLE of depression and 30 data-science researchers as equal partners. The project took place online, within a Facebook group. Public recruitment was via Facebook adverts and posts in mental health groups. All participants read a patient information sheet/completed a survey before accessing the private group. Participants were mainly from the UK.
Engagement
The project began with a series of live, text-based Q&A sessions. Topics were selected based on suggestions from the community, researcher expertise and the practical needs of the project.
Through these conversations, volunteers found out more about the science of depression, and researchers heard insights from a PWLE perspective. The text-based format allowed people to participate at convenient times. Transcripts of the Q&A sessions were put on a public blog (researchers were identified but public participants were anonymized).
In the second phase, the group discussed research gaps and chose the final question. This was investigated via a data-science project by Melissa Lewins, focus group questions in the private Facebook group and an anonymous survey. For ethical reasons, group members could not assist with the analysis of UK Biobank data. However, they discussed things like the 96 different codes that are available to GPs to record depression; the measures that are used, such as the Patient Health Questionnaire (PHQ-9) screening questionnaire; and the questions that are asked (such as “How many periods did you have in your life lasting two or more weeks where you felt like this?”). Participants expressed how difficult it could be to provide accurate answers, and some felt their condition was chronic rather than episodic.
Of the 1,342 UK Biobank participants with questionnaire results that suggested depression, 67% had no mention of depression in their general practitioner (GP) records.
Of the 26 people who answered our Depression Detectives survey, 84% had gone to the doctor for ≤50% of their episodes, 50% had used private talking therapy, and many had managed episodes through exercise, meditation, yoga or social contact.
The focus group (10 people) gave more expansive answers on topics such as episodic versus chronic depression and help-seeking behaviour.
Participants chose a pseudonym for reporting.
Research
Our key finding is that most episodes of depression are not taken to a doctor. As this has implications for a variety of stakeholders we produced and distributed four targeted infographics. For data scientists, our results indicate that electronic health records (eHRs) may provide an incomplete picture of depression. Many of the ways that people deal with depression will never appear in official records. For researchers using eHRs or prescription records as a way to estimate the number of episodes of depression a person has experienced, our results have highlighted some additional limitations. Within McIntosh’s research group, the results of this CS project have led to in-depth discussions on topics such as the differences between ‘episodic’ and ‘chronic’ depression. They have appeared in Lewins’s PhD thesis and influenced several grant applications. The findings have been disseminated through several presentations both within our department and to organizations such as Mental Health Research UK and the Psychiatric Genomics Consortium. Informal discussions have also been held with research cohorts including Generation Scotland, UK Biobank and Generation Malawi.
Translating into impact
For medical professionals, our results highlighted reasons why people did not go to their GP. These included how ‘bad’ they thought their depression was at the time and their past experiences of receptionists and doctors. Several people mentioned the dominance of drug treatment, and that they would like to be offered counselling. However, it is important to note that half of our respondents regretted not seeking help afterwards. Small actions could make a difference; for example, one participant suggested a ‘mental health’ button within automated telephone services. These results have been distributed via GP networks and social media. For policymakers, our results highlight the need for additional funding and service provision beyond drug treatments. People requested longer appointments for mental health conditions and the ability to consult with the same doctor every time. This has been communicated via direct correspondence with Members of the Scottish Parliament (MSPs) and the Health, Social Care and Sport Committee of the Scottish Parliament.
We created an infographic which communicated the project results to patients. This was disseminated via the Facebook groups used for recruitment and other social media channels, particularly during Mental Health Awareness Week and World Mental Health Day. We have also disseminated the practicalities of our project to patient and public involvement and engagement professionals both within our institution and nationally via the Scottish Public Engagement Network and the National Co-ordinating Centre for Public Engagement.
Creating infographics targeted at specific audiences allowed us to garner attention in a way that may not have been possible with more generalized outputs. The infographics, social media posts and word-of-mouth recommendations led to several invitations to speak and allowed us to disseminate the project and its findings to a range of different audiences.
We have captured some of the impacts the project had on the individuals who took part. Public participants felt more motivated to read research literature after the project, and more confident to critique it and recognize its limitations. Some also mentioned that they would approach their healthcare provider differently. All stated they would take part in similar projects again.
Researchers remarked on the benefits of engaging over a longer period of time versus one-off events, reported an increase in confidence engaging with patients, and suggested that the findings provided insightful directions for future research and to bring more translational aspects into their work.
