Mostafa Kamal Mallick

Advisors
Mohamad Wehbi (M.Sc.), Imrana Abdullahi Yari (M.Sc.) Dr. Dario Zanca, Prof. Dr. Björn Eskofier

Duration
12 / 2021 – 06 / 2022

Abstract
With advancements in technology and its integration with healthcare, internetbased cognitive behavioral therapy (iCBT) has become a fast-growing intervention channel compared to conventional psychotherapy [1, 2]. From these channels, a large volume of data is generated and the hidden potential of the data is yet to realize in the true sense for improving and personalizing the iCBT [3]. At the same time, the enterprises developing these iCBTs face a serious issue of dropout i.e., not completing all the modules in the prescribed intervention. Therefore, it diminishes the effect of the therapy on the patients and leads to mental health risks [4]. Predicting the dropouts statistically and figuring out the reasons, major factors influencing the dropout would improve decisionmaking and also in developing a hypothesis to offer personalized interventions.

Therefore, there is a need to analyze the behavior and engagement of the individual patients and predict the possibility of the dropout with a data-driven approach. The dataset shows a dropout of approximately 87%, therefore, a science of user attrition and to identify the factors influencing the dropout is the need of the hour for organizations developing iCBTs. Identifying dropouts at an earlier stage can be a basis for the development of micro-interventions, help in personalizing the therapy, and incorporate strategies for the betterment of the patients. The primary objective of the project includes, but is not limited to, data cleaning, data preparation, feature engineering, and using machine learning algorithms to predict dropout at various stages of the patient journey.

References:
[1] Elias Aboujaoude, Lina Gega, Michelle B Parish, and Donald M Hilty. Digital interventions in mental health: current status and future directions Frontiers in psychiatry, 11:111, 2020.
[2] David Daniel Ebert, Mathias Harrer, Jennifer Apolinário-Hagen, and Harald Baumeister. Digital interventions for mental disorders: key features, efficacy, and potential for artificial intelligence applications. In Frontiers in Psychiatry, pages 583–627. Springer, 2019.
[3] Vincent Bremer, Philip I Chow, Burkhardt Funk, Frances P Thorndike, and Lee M Ritterband. Developing a process for the analysis of user journeys and the prediction of dropout in digital health interventions: Machine learning approach. Journal of Medical Internet Research, 22(10):e17738, 2020.
[4] Wing-Fai Yeung, Ka-Fai Chung, Fiona Yan-Yee Ho, and Lai-Ming Ho. Predictors of dropout from internet-based self-help cognitive behavioral therapy for insomnia. Behaviour Research and Therapy, 73:19–24, 2015.