Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders
Abstract Background Digital phenotyping promises to unobtrusively obtaining a continuous and objective input of symptomatology from patients’ daily lives. The prime example are bipolar disorders, as smartphone parameters directly reflect bipolar symptomatology. Empirical studies, however, have yield...
Main Authors: | , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
SpringerOpen
2020-11-01
|
Series: | International Journal of Bipolar Disorders |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s40345-020-00210-4 |
_version_ | 1818162926737948672 |
---|---|
author | Ulrich W. Ebner-Priemer Esther Mühlbauer Andreas B. Neubauer Holger Hill Fabrice Beier Philip S. Santangelo Philipp Ritter Nikolaus Kleindienst Michael Bauer Florian Schmiedek Emanuel Severus |
author_facet | Ulrich W. Ebner-Priemer Esther Mühlbauer Andreas B. Neubauer Holger Hill Fabrice Beier Philip S. Santangelo Philipp Ritter Nikolaus Kleindienst Michael Bauer Florian Schmiedek Emanuel Severus |
author_sort | Ulrich W. Ebner-Priemer |
collection | DOAJ |
description | Abstract Background Digital phenotyping promises to unobtrusively obtaining a continuous and objective input of symptomatology from patients’ daily lives. The prime example are bipolar disorders, as smartphone parameters directly reflect bipolar symptomatology. Empirical studies, however, have yielded inconsistent findings. We believe that three main shortcomings have to be addressed to fully leverage the potential of digital phenotyping: short assessment periods, rare outcome assessments, and an extreme fragmentation of parameters without an integrative analytical strategy. Methods To demonstrate how to overcome these shortcomings, we conducted frequent (biweekly) dimensional and categorical expert ratings and daily self-ratings over an extensive assessment period (12 months) in 29 patients with bipolar disorder. Digital phenotypes were monitored continuously. As an integrative analytical strategy, we used structural equation modelling to build latent psychopathological outcomes (mania, depression) and latent digital phenotype predictors (sleep, activity, communicativeness). Outcomes Combining gold-standard categorical expert ratings with dimensional self and expert ratings resulted in two latent outcomes (mania and depression) with statistically meaningful factor loadings that dynamically varied over 299 days. Latent digital phenotypes of sleep and activity were associated with same-day latent manic psychopathology, suggesting that psychopathological alterations in bipolar disorders relate to domains (latent variables of sleep and activity) and not only to specific behaviors (such as the number of declined incoming calls). The identification of latent psychopathological outcomes that dimensionally vary on a daily basis will enable to empirically determine which combination of digital phenotypes at which days prior to an upcoming episode are viable as digital prodromal predictors. |
first_indexed | 2024-12-11T16:41:26Z |
format | Article |
id | doaj.art-73f9cc4ee45a44a6bf81c48f69049355 |
institution | Directory Open Access Journal |
issn | 2194-7511 |
language | English |
last_indexed | 2024-12-11T16:41:26Z |
publishDate | 2020-11-01 |
publisher | SpringerOpen |
record_format | Article |
series | International Journal of Bipolar Disorders |
spelling | doaj.art-73f9cc4ee45a44a6bf81c48f690493552022-12-22T00:58:19ZengSpringerOpenInternational Journal of Bipolar Disorders2194-75112020-11-01811910.1186/s40345-020-00210-4Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disordersUlrich W. Ebner-Priemer0Esther Mühlbauer1Andreas B. Neubauer2Holger Hill3Fabrice Beier4Philip S. Santangelo5Philipp Ritter6Nikolaus Kleindienst7Michael Bauer8Florian Schmiedek9Emanuel Severus10Mental mHealth Lab, Institute of Sport and Sport Sciences, Karlsruhe Institute of TechnologyDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenDIPF - Leibniz Institute for Research and Information in EducationMental mHealth Lab, Institute of Sport and Sport Sciences, Karlsruhe Institute of TechnologyDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenMental mHealth Lab, Institute of Sport and Sport Sciences, Karlsruhe Institute of TechnologyDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenInstitute of Psychiatric and Psychosomatic Psychotherapy, Central Institute of Mental Health, Mannheim / Medical Faculty Mannheim, Heidelberg UniversityDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenDIPF - Leibniz Institute for Research and Information in EducationDepartment of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität DresdenAbstract Background Digital phenotyping promises to unobtrusively obtaining a continuous and objective input of symptomatology from patients’ daily lives. The prime example are bipolar disorders, as smartphone parameters directly reflect bipolar symptomatology. Empirical studies, however, have yielded inconsistent findings. We believe that three main shortcomings have to be addressed to fully leverage the potential of digital phenotyping: short assessment periods, rare outcome assessments, and an extreme fragmentation of parameters without an integrative analytical strategy. Methods To demonstrate how to overcome these shortcomings, we conducted frequent (biweekly) dimensional and categorical expert ratings and daily self-ratings over an extensive assessment period (12 months) in 29 patients with bipolar disorder. Digital phenotypes were monitored continuously. As an integrative analytical strategy, we used structural equation modelling to build latent psychopathological outcomes (mania, depression) and latent digital phenotype predictors (sleep, activity, communicativeness). Outcomes Combining gold-standard categorical expert ratings with dimensional self and expert ratings resulted in two latent outcomes (mania and depression) with statistically meaningful factor loadings that dynamically varied over 299 days. Latent digital phenotypes of sleep and activity were associated with same-day latent manic psychopathology, suggesting that psychopathological alterations in bipolar disorders relate to domains (latent variables of sleep and activity) and not only to specific behaviors (such as the number of declined incoming calls). The identification of latent psychopathological outcomes that dimensionally vary on a daily basis will enable to empirically determine which combination of digital phenotypes at which days prior to an upcoming episode are viable as digital prodromal predictors.http://link.springer.com/article/10.1186/s40345-020-00210-4Bipolar disordersDigital phenotypingMobile sensingAmbulatory assessmentSmartphone sensing |
spellingShingle | Ulrich W. Ebner-Priemer Esther Mühlbauer Andreas B. Neubauer Holger Hill Fabrice Beier Philip S. Santangelo Philipp Ritter Nikolaus Kleindienst Michael Bauer Florian Schmiedek Emanuel Severus Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders International Journal of Bipolar Disorders Bipolar disorders Digital phenotyping Mobile sensing Ambulatory assessment Smartphone sensing |
title | Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
title_full | Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
title_fullStr | Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
title_full_unstemmed | Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
title_short | Digital phenotyping: towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
title_sort | digital phenotyping towards replicable findings with comprehensive assessments and integrative models in bipolar disorders |
topic | Bipolar disorders Digital phenotyping Mobile sensing Ambulatory assessment Smartphone sensing |
url | http://link.springer.com/article/10.1186/s40345-020-00210-4 |
work_keys_str_mv | AT ulrichwebnerpriemer digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT esthermuhlbauer digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT andreasbneubauer digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT holgerhill digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT fabricebeier digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT philipssantangelo digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT philippritter digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT nikolauskleindienst digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT michaelbauer digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT florianschmiedek digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders AT emanuelseverus digitalphenotypingtowardsreplicablefindingswithcomprehensiveassessmentsandintegrativemodelsinbipolardisorders |