Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT

Objective: Predicting who will not benefit enough from Internet-Based Cognitive Behavioral (ICBT) Therapy early on can assist in better allocation of limited mental health care resources. Repeated measures of symptoms during treatment is the strongest predictor of outcome, and we want to investigate...

Full description

Saved in:
Bibliographic Details
Main Authors: Nils Hentati Isacsson, Kirsten Zantvoort, Erik Forsell, Magnus Boman, Viktor Kaldo
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Internet Interventions
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2214782924000666
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846127207277133824
author Nils Hentati Isacsson
Kirsten Zantvoort
Erik Forsell
Magnus Boman
Viktor Kaldo
author_facet Nils Hentati Isacsson
Kirsten Zantvoort
Erik Forsell
Magnus Boman
Viktor Kaldo
author_sort Nils Hentati Isacsson
collection DOAJ
description Objective: Predicting who will not benefit enough from Internet-Based Cognitive Behavioral (ICBT) Therapy early on can assist in better allocation of limited mental health care resources. Repeated measures of symptoms during treatment is the strongest predictor of outcome, and we want to investigate if methods that explicitly account for time-dependency are superior to methods that do not, with data from (a) only two pre-treatment timepoints and (b) the pre-treatment timepoints and three timepoints during initial treatment. Methods: We use 1) commonly used time-independent methods (i.e., Linear Regression and Random Forest models) and 2) time-dependent methods (i.e., multilevel model regression, mixed-effects random forest, and a Long Short-Term Memory model) to predict symptoms during treatment, including the final outcome. This is done with symptom scores from 6436 ICBT patients from regular care, using robust multiple imputation and nested cross-validation methods. Results: The models had a 14 %–12 % root mean squared error (RMSE) in predicting the post-treatment outcome, corresponding to a balanced accuracy of 67–74 %. Time-dependent models did not have higher accuracies. Using data for the initial treatment period (b) instead of only from before treatment (a) increased prediction results by 1.3 % percentage points (12 % to 10.7 %) RMSE and 6 % percentage points BACC (69 % to 75 %). Conclusion: Training prediction models on only symptom scores of the first few weeks is a promising avenue for symptom predictions in treatment, regardless of which model is used. Further research is necessary to better understand the interaction between model complexity, dataset length and width, and the prediction tasks at hand.
format Article
id doaj-art-05c52c4188774955afb6f3c0ec40962e
institution Kabale University
issn 2214-7829
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Internet Interventions
spelling doaj-art-05c52c4188774955afb6f3c0ec40962e2024-12-12T05:21:53ZengElsevierInternet Interventions2214-78292024-12-0138100773Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBTNils Hentati Isacsson0Kirsten Zantvoort1Erik Forsell2Magnus Boman3Viktor Kaldo4Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, & Stockholm Health Care Services, Region Stockholm, Sweden; Corresponding author.Institute of Information Systems, Leuphana University, Lueneburg, GermanyCentre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, & Stockholm Health Care Services, Region Stockholm, SwedenDivision of Psychiatry, University College London, UK; Department of Medicine Solna, Clinical Epidemiology Division, Karolinska Institutet, Stockholm, SwedenCentre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, & Stockholm Health Care Services, Region Stockholm, Sweden; Department of Psychology, Faculty of Health and Life Sciences, Linnaeus University, Växjö, SwedenObjective: Predicting who will not benefit enough from Internet-Based Cognitive Behavioral (ICBT) Therapy early on can assist in better allocation of limited mental health care resources. Repeated measures of symptoms during treatment is the strongest predictor of outcome, and we want to investigate if methods that explicitly account for time-dependency are superior to methods that do not, with data from (a) only two pre-treatment timepoints and (b) the pre-treatment timepoints and three timepoints during initial treatment. Methods: We use 1) commonly used time-independent methods (i.e., Linear Regression and Random Forest models) and 2) time-dependent methods (i.e., multilevel model regression, mixed-effects random forest, and a Long Short-Term Memory model) to predict symptoms during treatment, including the final outcome. This is done with symptom scores from 6436 ICBT patients from regular care, using robust multiple imputation and nested cross-validation methods. Results: The models had a 14 %–12 % root mean squared error (RMSE) in predicting the post-treatment outcome, corresponding to a balanced accuracy of 67–74 %. Time-dependent models did not have higher accuracies. Using data for the initial treatment period (b) instead of only from before treatment (a) increased prediction results by 1.3 % percentage points (12 % to 10.7 %) RMSE and 6 % percentage points BACC (69 % to 75 %). Conclusion: Training prediction models on only symptom scores of the first few weeks is a promising avenue for symptom predictions in treatment, regardless of which model is used. Further research is necessary to better understand the interaction between model complexity, dataset length and width, and the prediction tasks at hand.http://www.sciencedirect.com/science/article/pii/S2214782924000666Machine learningPredictionTreatment outcomeAdaptive treatment strategyPrecision psychiatryTimeseries symptom
spellingShingle Nils Hentati Isacsson
Kirsten Zantvoort
Erik Forsell
Magnus Boman
Viktor Kaldo
Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
Internet Interventions
Machine learning
Prediction
Treatment outcome
Adaptive treatment strategy
Precision psychiatry
Timeseries symptom
title Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
title_full Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
title_fullStr Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
title_full_unstemmed Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
title_short Making the most out of timeseries symptom data: A machine learning study on symptom predictions of internet-based CBT
title_sort making the most out of timeseries symptom data a machine learning study on symptom predictions of internet based cbt
topic Machine learning
Prediction
Treatment outcome
Adaptive treatment strategy
Precision psychiatry
Timeseries symptom
url http://www.sciencedirect.com/science/article/pii/S2214782924000666
work_keys_str_mv AT nilshentatiisacsson makingthemostoutoftimeseriessymptomdataamachinelearningstudyonsymptompredictionsofinternetbasedcbt
AT kirstenzantvoort makingthemostoutoftimeseriessymptomdataamachinelearningstudyonsymptompredictionsofinternetbasedcbt
AT erikforsell makingthemostoutoftimeseriessymptomdataamachinelearningstudyonsymptompredictionsofinternetbasedcbt
AT magnusboman makingthemostoutoftimeseriessymptomdataamachinelearningstudyonsymptompredictionsofinternetbasedcbt
AT viktorkaldo makingthemostoutoftimeseriessymptomdataamachinelearningstudyonsymptompredictionsofinternetbasedcbt