"Therapie trifft KI"-Studie (TIKI-Studie) im Rahmen der DFG-FOR 5187 "Towards precision psychotherapy for non-respondent patients: from signatures to predictions to clinical utility" (PREACT)

ID

3

DZPG Site

Berlin/Potsdam
General Information
Status

Ongoing (II): Recruitment and data collection ongoing

Description

Although cognitive-behavioral therapy (CBT) is a first-line treatment for internalizing disorders, a substantial proportion of patients fails to benefit - with severe consequences for patients and increasing costs for societies. The paucity of standard clinical features that allow for single-case predictions, limited methodological approaches, and fragmented data levels serve as an impetus to implement state-of-the-art predictive analytics to search for the best predictors of (non-) response. The present project (SP1, TIKI study) aims to set up a prospective-longitudinal observational cohort of n = 500 patients with mental disorders from the internalizing spectrum (specific phobia, social anxiety disorder, panic disorder, agoraphobia, generalized anxiety disorder, obsessive-compulsive disorder, post-traumatic stress disorder, unipolar depressive disorders) who are treated with CBT. Embedded within the Research Unit, patients will be deeply phenotyped for several layers of bio-behavioral markers (SP3-SP6). It will be supported by our clinical (SP1), methods (SP2), and neuroimaging platform (SP3). As such, SP1 contains both a research and a service component. Research-wise, we will test the hypothesis if treatment (non-)response can be predicted with sufficient accuracy based on clinical routine data when state-of-the-art machine learning methods are applied. Service-wise, SP1 will coordinate the clinical trial by serving as a recruitment hub for the other Subprojects, implementing diagnostic assessments, documenting and quality-controlling treatment contents, and calculating primary and secondary outcomes. As an observational trial in the academic outpatient sector, the study design exerts a high degree of external validity as a prerequisite for translating predictive analytics into practice. No randomization applies - all patients receive active treatment. In line with the idea of precision medicine, our overall aim is to enable the prediction of treatment outcomes prior to therapy onset for the early identification and optimized treatment of patients at risk for non-response towards standard treatment.

Trial Number

DRKS00030915

Contact
PI

Prof. Dr. Ulrike Lueken

Further PIs

Norbert Kathmann, Babette Renneberg, Frank Jacobi, Lydia Fehm

Survey Results
Is it an interventional or non-interventional study?
  • Interventional
Specification of study type
  • Single group
Is it a mono- or multicentric study?

Multicentric

Number of sites

4

Is it a single or multi country study?

Single

Which country?

Germany

Who are participants?
  • Patients with specific diseases
Which diseases? (ICD-11 classification)
  • Mental, behavioural or neurodevelopmental disorders (06)
Mental, behavioural or neurodevelopmental disorders (ICD-11 classification)
  • Mood disorders
  • Anxiety or fear-related disorders
  • Obsessive-compulsive or related disorders
  • Disorders specifically associated with stress
Recruitment setting
  • Outpatient
Target sample size

585

Obtained sample size

27

Start of recruitment

2023

Minimum age of participants

18

Data sources for the study
  • Administrative databases
  • Imaging data
  • Medical records
  • Interview
  • Questionnaire
  • Physiological/biochemical measurements
  • Smartphone or wearable based data
Is it planned to share the data?

Yes, there is a plan to make data available

Additional information about data sharing

Pseudonymized data will be shared only within the FOR5187 consortium. We will however anonymize data 1 year after the last patient out, making data sharing possible. It is not planed to upload full first patient datasets into a public repository (sensitive patient data)

Are further follow-ups planned?

No

Is there consent to re-contact participants?

No

Is the implementation of further study components possible?

No

View raw data