gms | German Medical Science

23. Deutscher Kongress für Versorgungsforschung

Deutsches Netzwerk Versorgungsforschung e. V.

24.09. - 27.09.2024, Potsdam

How to digitalize clinical data in routine mental health care – the use case psychiatry in the medical informatics initiatives DECIDE Digital Hub

Meeting Abstract

  • Hauke Felix Wiegand - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Francesca Uhl - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Sophia Hütter - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Nicoletta Momtahen - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Ronja Husemann - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Dirk Riedinger - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Alexander Scherre - Frauenhofer ITWM, Kaiserslautern, Deutschland
  • Torsten Panholzer - IMBEI, Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Oliver Tüscher - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland
  • Klaus Lieb - Klinik für Psychiatrie und Psychotherapie der Universitätsmedizin der Johannes Gutenberg Universität Mainz, Deutschland

23. Deutscher Kongress für Versorgungsforschung (DKVF). Potsdam, 25.-27.09.2024. Düsseldorf: German Medical Science GMS Publishing House; 2024. Doc24dkvf286

doi: 10.3205/24dkvf286, urn:nbn:de:0183-24dkvf2869

Published: September 10, 2024

© 2024 Wiegand et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Background: Digitalized clinical data of routine mental health care is a prerequisite for clinical meaningful longterm observations of large populations, a data-driven personalized medicine, and the use of digital decision support and artificial intelligence based tools. These tools might help to ameliorate both the lack of evidence-based decisions in routine treatment reality and the growing lack of qualified mental health specialists. Furthermore, they might improve the quality of treatment through better allocation of effective therapy interventions. However, it is unclear, which variables are eligible for digitalization in routine mental health care. They should be chosen based on a scientifically informed strategy and not only by expert consensus.

Objective: Within the BMBF supported Digital Hub DECIDE of the Medical Informatics Initiative (MII) we aim to develop a science based strategy to identify candidate variables for digitalization of mental health routine data. Routine treatment of Major Depressive Disorder (MDD) is used as a practical example.

Methods: We first discuss principles that can guide a choice of variables. We then present a systematic review and meta-analysis strategy and its initial results for the identification of candidate variables for MDD treatment.

Results: Principles that can guide the development of a variable set are: First, a variable set should characterize the clinical phenomenon MDD comprehensively; second, variables should have reliably shown to have a predictive value in MDD treatment; third, variables should be easy to gather by routine examinations, and fourth, the set should be compatible with existing dataset projects like the MII core dataset or the core dataset of the DZPG (Deutsches Zentrum für Psychische Gesundheit). Comprehensive literature on the phenomenology of MDD exists. Literature on the reliable predictive value of variables in MDD treatment however is vast and difficult to interpret. We therefore conducted an umbrella review for individual patient data meta-analyses. The review strategy and first results will be presented.

Implication for research and/or (healthcare) practice: The four basic principles and the results of the umbrella review can inform the choice of variables for digitalizing routine MDD care. A scientifically informed countrywide or better Europe-wide dataset for mental health routine care could help to unleash the potentials of AI based personalized medicine in mental health care.

Funding: Other funding; Project name: DECIDE; Grant number: 01ZZ2106A