Depression and other forms of psychopathology are highly prevalent in society and very debilitating. The mechanisms underlying psychopathology are very complex and highly person-specific. At the same time, there is an urgent need for improved insight in the prevention of...
Depression and other forms of psychopathology are highly prevalent in society and very debilitating. The mechanisms underlying psychopathology are very complex and highly person-specific. At the same time, there is an urgent need for improved insight in the prevention of psychopathology and for a much more accurate assessment of personalized risk. The TRANS-ID project tries to solve this discrepancy by taking an out-of-the box approach. We hypothesize that depression, and also other forms of psychopathology, behave according to principles of complex systems. In these systems it has been shown earlier that typical early warning signals (EWS) can be detected that anticipate major and sudden shifts in the system. If similar EWS are present before important shifts occur in psychopathology (e.g., depressive relapse or recovery from depression) then this opens a new avenue for finding accurate personalized early warning signals for upcoming symptom shifts. The aim of TRANS-ID, therefore, is to set up entirely novel sorts of data collections, in which individuals are prospectively followed over the course of important symptom transitions, in order to examine whether such EWS can indeed be detected. If we succeed, then these findings are of direct use in clinical practice to monitor an individualâ€™s risk for approaching major symptom shifts.
Important objectives are to examine the presence of EWS before people experience sudden transitions towards higher levels of depressive symptoms (in people who taper their antidepressants), and towards lower levels of depressive symptoms (in people with depression who start therapy). Also, we want to examine whether EWS can inform us on the type of symptoms that will dominate a nearby symptom transition (such as transitions in depressive, manic, anxiety or psychotic symptoms) in young adults at risk.
To examine the above aims, the TRANS-ID project needed to collect time-series datasets on mood and behavior with extremely large numbers of observations per person and with fine-grained measurements and with the inclusion of precisely those periods in which participants experience important symptom transitions. We have successfully managed to carry out these challenging time-series studies in the aimed number of participants . For this, we have developed the necessary experience sampling diaries for each of the three studies. We have developed novel sampling procedures and tested novel instruments to optimally measure heart rate variability over these periods (by carrying out a pilot study in healthy participants and evaluating validity and feasibility of these measurements). We have recruited eligible participants for each of these studies. We have finished inclusion of participants for two of these studies and have almost finished inclusion (end: April 2019 ) for the other study.
Furthermore, we worked on the examination of theories and assumptions related to the aims of the TRANS-ID project. For example, we have shown that the assumption is correct that sudden shifts in depressive symptoms are present during the process of recovery from depression. We have also elaborated, in an invited paper in psychological medicine, on the hypotheses of work package 4 that EWS may inform us on the type of symptoms that will dominate nearby transitions and how we could empirically test that assumption. Furthermore, we carried out a first empirical examination which supported the latter assumption. We found that EWS were more strongly present in mood states that were congruent with the type of symptoms showing the largest shift a year later. Additionally, we have critically reflected on the status of the network theory and best practices to carry out network research in this field. Finally, we have worked on the optimization of statistical techniques that we can use to detect EWS. By applying statistical process control techniques to our data, previously used only for industrial purposes, we have developed a novel way to prospectively detect EWS, based only on data collected up to that point. Such techniques will help us to translate scientific findings into useful clinical applications in which patients can be informed in time that they are likely approaching a transition.
Progress beyond the state of the art:
The TRANS-ID project is the first to test systematically, empirically and in repeated experiments whether psychopathology behaves according to principles of complex systems in terms of the presence of early warning signals. The project thereby makes an important step forward in the application of complex system theory to the field of psychiatry.
Second, the TRANS-ID project has moved considerably beyond the state of the art by delivering datasets with very precise and fine-grained data over extended periods of time that directly precede important symptom transitions. Such datasets now give the field of psychiatry the opportunity to directly examine what happens in individuals before such transitions occur. Although this question is vital to understanding psychopathology, datasets to prospective examine this have never been available before. These datasets will therefore not only provide answers to the main question of the proposal on whether personalized early warning signals can detect nearby symptom transitions, but that these datasets will also be crucial in further examination of vital questions in the field: why do symptoms arise and how will they disappear.
Third, the TRANS-ID project has also contributed strongly to the paradigm shift towards personalized analyses and patient-specific outcomes. This contribution is not only reflected in the delivery of the above mentioned datasets, but also in the testing of novel instruments and statistical tools. We refer, for example, to the implementation of a novel instrument and procedure to repeatedly measure heart rate variability in daily life over extended periods of time that the TRANS-ID project carried out. Second, TRANS-ID has tested a novel application of statistical process control techniques. This technique may aid future translation of time-series data to application in clinical practice.
Expected results until the end of the project:
For WP1, we expect to have performed statistical analyses on the collection of separate time-series studies of these individuals who tapered their antidepressant medication during this period of measurements. From these analyses we expect to be able to conclude to what extent sudden transitions occur and whether EWS, as hypothesized, systematically anticipate these sudden transitions towards higher level of symptoms. We expect to gain further insight into the frequency and timing of the presence hereof and in whether certain variables are more powerful than others in showing these signals. And we expect to have applied the novel statistical control techniques on these data as well, to test not only whether EWS can be detected backwards (using all available research data), but whether they can also be detected prospectively. These results will be written down for publication in peer-reviewed international journals.
For WP2, we expect to finish the data collection, which includes the intensive monitoring period for the last individuals that entered the study and the monthly follow-up measures of depressive symptoms up to a year after that period. Furthermore, we expect to have performed statistical analyses on the collection of time-series studies to answer the main research questions of WP2. We will examine whether sudden transitions occur and whether EWS systematically anticipate transitions towards improvements of symptoms. Also, we expect to get insight into the frequency and timing of these EWS and in the question whether certain variables are more powerful than other variables in showing these signals. These results will be written down for publication in peer-reviewed international journals.
For WP3 we expect to finish the pre-processing of all collected heart rate data of the TRANS-ID data collections, which is currently ongoing. The preprocessing of heart rate is very time-consuming and is performed using specific software for this purpose (CARSPAN). We expect to further analyze and w
More info: https://www.transid.nl/.