Periodic Reporting for period 2 - WARN-D (Using Network Theory to Predict Depression Onset and Build a Personalized Early Warning System)
Reporting period: 2022-10-01 to 2024-03-31
Depression, a prevalent and debilitating mental health condition, is a pressing global issue. Despite significant efforts and investments, progress in treating depression has lagged behind other diseases. Effective prevention is considered the most promising approach to reduce the immense burden of depression, especially among young adults. However, a significant challenge in implementing prevention programs is the reliable detection of individuals at risk for depression before they transition into it.
The main objectives of the WARN-D project are to understand and predict depression onset, and to build the personalized early warning system WARN-D. WARN-D aims to identify individuals at risk of depression before its onset, along with potential reasons for why they are at risk. This approach promises to transform the detection of depression onset, facilitating the development and implementation of more effective personalized prevention programs. We are currently in the process of collecting extensive data from almost 2,000 participants, gathered at various time points: a comprehensive baseline survey; a 3-month collection phase consisting of daily surveys and smartwatch data; and regular follow-up surveys conducted over several years. These data will be used to construct WARN-D.
In conclusion, the WARN-D project is a crucial step toward addressing the global burden of depression. By focusing on early detection and prevention, it offers hope for reducing the suffering caused by this debilitating condition. The impact of this project extends beyond depression, as its success could inspire similar initiatives for other mental disorders.
The main objectives of the WARN-D project are to understand and predict depression onset, and to build the personalized early warning system WARN-D. WARN-D aims to identify individuals at risk of depression before its onset, along with potential reasons for why they are at risk. This approach promises to transform the detection of depression onset, facilitating the development and implementation of more effective personalized prevention programs. We are currently in the process of collecting extensive data from almost 2,000 participants, gathered at various time points: a comprehensive baseline survey; a 3-month collection phase consisting of daily surveys and smartwatch data; and regular follow-up surveys conducted over several years. These data will be used to construct WARN-D.
In conclusion, the WARN-D project is a crucial step toward addressing the global burden of depression. By focusing on early detection and prevention, it offers hope for reducing the suffering caused by this debilitating condition. The impact of this project extends beyond depression, as its success could inspire similar initiatives for other mental disorders.
Our project started with the establishment of a data collection infrastructure, including the development and implementation of surveys for the various stages of the study. To ensure that we measure all relevant risk and protective factors, we conducted a detailed literature search, organized a Delphi study, and consulted with various additional experts. Our focus on best measurement practices and open scholarship principles means that we created our measurement batteries with attention to modern, valid and reliable available in English and Dutch, the two languages used in our study.
Simultaneously, we worked with a group of legal, privacy, ethical, and IT experts and set up a data collection infrastructure—and obtain ethical approval for it—that deals efficiently and responsibly with the personal and research data we collect from data numerous sources such as smartphones and smartwatches.
To improve communication with participants and facilitate recruitment, we set up and maintained a website (www.warn-d.com) and a social media presence (Facebook and Instagram). This, combined with a 4-minute animated video explaining details of the WARN-D study to participants in two languages, helped us recruit and retain nearly 2,000 participants over the course of two years.
Our most significant achievement thus far is the ongoing collection of extensive data from our participants—a monumental undertaking that has demanded careful planning and execution and led to a dataset that is unique in the world. After administrating an extensive 90-minute baseline survey collecting detailed data on risk and protective factors for depression, each participant received a smartwatch via mail, which they wore for three months. During this time, participants also responded to multiple daily questionnaires via a smartphone app. Furthermore, we administered and continue to administer follow-up surveys every three months over the course of two years, resulting in a wealth of data. These datasets, which we are still cleaning and organizing, serve as the bedrock of our research efforts, paving the way for in-depth analyses in the future.
We strive to adhere to best practices in terms of open scholarship principles, with a focus on (1) project infrastructure, (2) measurement, (3) code, and (4) data.
(1) All information regarding project infrastructure has been made available as part of our protocol paper.
(2) To ensure the accessibility and transparency of our work, we are nearly finished creating detailed codebooks for all the data collected. The codebooks not only detail all data sources, measures and items, but also document if and (if so) how we adapted them; references to the original scale; and details on translation and validation processes. They are available as part of the protocol paper.
(3) We have built a pipeline to make analysis of the data reproducible both within the team and for potential external users. This pipeline is currently on GitHub, with code set to be released when data becomes publicly available.
(4) Our datasets are one-of-a-kind and promise to be a valuable resource for researchers across a broad range of academic disciplines. We are currently working with experts at Leiden University to make optimal decisions on what data can be shared without compromising anonymity, and how the data can be shared most efficiently with the research community.
Simultaneously, we worked with a group of legal, privacy, ethical, and IT experts and set up a data collection infrastructure—and obtain ethical approval for it—that deals efficiently and responsibly with the personal and research data we collect from data numerous sources such as smartphones and smartwatches.
To improve communication with participants and facilitate recruitment, we set up and maintained a website (www.warn-d.com) and a social media presence (Facebook and Instagram). This, combined with a 4-minute animated video explaining details of the WARN-D study to participants in two languages, helped us recruit and retain nearly 2,000 participants over the course of two years.
Our most significant achievement thus far is the ongoing collection of extensive data from our participants—a monumental undertaking that has demanded careful planning and execution and led to a dataset that is unique in the world. After administrating an extensive 90-minute baseline survey collecting detailed data on risk and protective factors for depression, each participant received a smartwatch via mail, which they wore for three months. During this time, participants also responded to multiple daily questionnaires via a smartphone app. Furthermore, we administered and continue to administer follow-up surveys every three months over the course of two years, resulting in a wealth of data. These datasets, which we are still cleaning and organizing, serve as the bedrock of our research efforts, paving the way for in-depth analyses in the future.
We strive to adhere to best practices in terms of open scholarship principles, with a focus on (1) project infrastructure, (2) measurement, (3) code, and (4) data.
(1) All information regarding project infrastructure has been made available as part of our protocol paper.
(2) To ensure the accessibility and transparency of our work, we are nearly finished creating detailed codebooks for all the data collected. The codebooks not only detail all data sources, measures and items, but also document if and (if so) how we adapted them; references to the original scale; and details on translation and validation processes. They are available as part of the protocol paper.
(3) We have built a pipeline to make analysis of the data reproducible both within the team and for potential external users. This pipeline is currently on GitHub, with code set to be released when data becomes publicly available.
(4) Our datasets are one-of-a-kind and promise to be a valuable resource for researchers across a broad range of academic disciplines. We are currently working with experts at Leiden University to make optimal decisions on what data can be shared without compromising anonymity, and how the data can be shared most efficiently with the research community.
The project goes beyond the state of the art in several ways, with expected results tied to each.
First, regarding establishing whether someone is depressed, the current state of the art in the field is to either conduct one clinical interview per person, or one questionnaire that queries depression symptoms. Regarding interviews (that are not feasible for WARN-D in the first place), inter-rater reliability for depression has consistently been shown to be among the lowest of all mental disorders, meaning that clinicians more often disagree than agree with each other whether a person should be assigned a diagnosis of depression. Regarding the questionnaires that are being used, they are short and only contain a very limited amount of information about how a person is doing, and are usually given only once. In WARN-D, we have therefore collected multidimensional data from various sources, all of which have shown to play a role in deciding whether someone should be considered depressed. Our battery not only integrates wellbeing, impairment, depression and anxiety symptoms, as well as questions on recently obtained diagnoses, therapy status or taking medication, but also information on whether increases in problems could be expected based on e.g. questionnaires on severe or adverse life events. Further, we have continuous data (compared to the one-time assessments usually done), meaning we can utilize data collected every single day to make a more informed decision whether we should consider someone to be depressed. We are currently working on a pipeline to make the optimal decision based on all data sources, which goes considerably beyond the state of the art and will lead to important output regarding the conceptualization, operationalization and measurement of depression.
Second, there is growing agreement that mental health problems like depression are biopsychosocial systems out of which mental illness emerges. However, what this system is and how it operates is largely unknown. One of the PhD projects will map out this mood system in detail, utilizing the enormous amount of relevant data with high temporal resolution from different data sources we have collected in the project. The statistical focus of this project is on utilizing statistical network models that can integrate various data sources into one system, considerably extending the literature on the human mood system.
A third milestone, based on the perspective that mental health problems operate as systems, is to see whether depression onset can be anticipated using early warning signals (EWS). Other fields such as ecology have shown that when e.g. lakes tip from a healthy to a disease state, this transition can be anticipated using statistical features of the system that can be measured and modeled. EWS thus can serve as signals acquired from data that may indicate a potential transition to an alternative system state, such as developing depression, and may serve as a point for preemptive interventions. One PhD project is currently concerned with the question of what particular EWS can be identified in our data, and how well they do in prediction depression onset. As part of this effort, we are testing existing EWS, but also testing novel EWS we developed specifically for psychological questions in the team.
Finally, these findings will be integrated together into a prospective prediction model, which will use state of the art machine learning to create models that can forecast depression onset. These models will be both data-driven, utilizing all information we collect on participants in the project, but also theory-driven, leveraging the impressive EWS literature in other areas where many EWS have been established successfully. We hope to utilize these models to develop a real-time app that can integrate wearables and daily diaries to deliver smartphone-based warning signals.
First, regarding establishing whether someone is depressed, the current state of the art in the field is to either conduct one clinical interview per person, or one questionnaire that queries depression symptoms. Regarding interviews (that are not feasible for WARN-D in the first place), inter-rater reliability for depression has consistently been shown to be among the lowest of all mental disorders, meaning that clinicians more often disagree than agree with each other whether a person should be assigned a diagnosis of depression. Regarding the questionnaires that are being used, they are short and only contain a very limited amount of information about how a person is doing, and are usually given only once. In WARN-D, we have therefore collected multidimensional data from various sources, all of which have shown to play a role in deciding whether someone should be considered depressed. Our battery not only integrates wellbeing, impairment, depression and anxiety symptoms, as well as questions on recently obtained diagnoses, therapy status or taking medication, but also information on whether increases in problems could be expected based on e.g. questionnaires on severe or adverse life events. Further, we have continuous data (compared to the one-time assessments usually done), meaning we can utilize data collected every single day to make a more informed decision whether we should consider someone to be depressed. We are currently working on a pipeline to make the optimal decision based on all data sources, which goes considerably beyond the state of the art and will lead to important output regarding the conceptualization, operationalization and measurement of depression.
Second, there is growing agreement that mental health problems like depression are biopsychosocial systems out of which mental illness emerges. However, what this system is and how it operates is largely unknown. One of the PhD projects will map out this mood system in detail, utilizing the enormous amount of relevant data with high temporal resolution from different data sources we have collected in the project. The statistical focus of this project is on utilizing statistical network models that can integrate various data sources into one system, considerably extending the literature on the human mood system.
A third milestone, based on the perspective that mental health problems operate as systems, is to see whether depression onset can be anticipated using early warning signals (EWS). Other fields such as ecology have shown that when e.g. lakes tip from a healthy to a disease state, this transition can be anticipated using statistical features of the system that can be measured and modeled. EWS thus can serve as signals acquired from data that may indicate a potential transition to an alternative system state, such as developing depression, and may serve as a point for preemptive interventions. One PhD project is currently concerned with the question of what particular EWS can be identified in our data, and how well they do in prediction depression onset. As part of this effort, we are testing existing EWS, but also testing novel EWS we developed specifically for psychological questions in the team.
Finally, these findings will be integrated together into a prospective prediction model, which will use state of the art machine learning to create models that can forecast depression onset. These models will be both data-driven, utilizing all information we collect on participants in the project, but also theory-driven, leveraging the impressive EWS literature in other areas where many EWS have been established successfully. We hope to utilize these models to develop a real-time app that can integrate wearables and daily diaries to deliver smartphone-based warning signals.