EUVIMEDObjevte EUVIMED
Open access

Causal AI Recommendation System for Digital Mental Health: Bayesian Decision-Theoretic Analysis

Mathew Varidel, Victor An, Ian B Hickie, Sally Cripps, Roman Marchant, Jan Scott, Jacob J Crouse, Adam Poulsen, Bridianne O'Dea, Sarah McKenna, Frank Iorfino

Journal of Medical Internet Research · 2025-06-19 · 27 · e71305-e71305

DOI: 10.2196/71305 ↗

At a glance

Journal article in Journal of Medical Internet Research, published online on 2025-06-19. Written by Mathew Varidel and 10 co-authors. Institution: Brain and Mind Centre, The University of Sydney, 94 Mallett Street, Sydney, 2050, Australia, 61 0293510774. Major MEDLINE topics: Artificial Intelligence, Mental Health. Cited 6 times so far (Crossref); 65 references. The full text is freely available in PubMed Central.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Background

Digital mental health tools promise to enhance the reach and quality of care. Current tools often recommend content to individuals, typically using generic knowledge-based systems or predictive artificial intelligence (AI). However, predictive AI is problematic for interventional recommendations as cause-effect relationships can be confounded in observed data. Therefore, causal AI is required to compare future outcomes under different interventions.

Objective

We aimed to develop a causal AI recommendation system that uses an individual’s current presentation, their preferences, and the learned dynamics between domains to rank interventions.

Methods

We frame the recommendation problem within a Bayesian decision-theoretic framework, whereby a preference ordering of decisions is estimated using the expected utility of outcomes under interventions. The causal processes are assumed to follow a structural causal model, where the posterior distribution of structural causal models is estimated using a Markov chain Monte Carlo method. Expected utilities under interventions are estimated using a do-operation, which estimates the effects of changing a variable on outcomes, while accounting for confounders. We apply our approach to rank domains relating to mental health and well-being as intervention targets for adults (n=619) who used the Innowell Fitness app between September 2021 to September 2023 and completed a questionnaire at 2 time points (1 wk-6 mo from baseline).

Results

The causal AI recommendation system recommends intervention targets as a function of a user’s baseline presentation, the causal effects of the intervention on itself and other domains, and the utility function. In our example, psychological distress was typically the optimal intervention target in complex cases where multiple domains were unhealthy at baseline, due to it affecting multiple domains with paths to personal functioning (probability [p] of path; p path =86%), social support (p path =92%), sleep (p path =88%), and physical activity (p path =86%). The probability of being the optimal intervention target was personal functioning (p opt =30%), psychological distress (p opt =29%), social support (p opt =18%), nutrition (p opt =9.6%), substance use (p opt =6.7%), sleep (p opt =4.5%), and physical activity (p opt =2.2%).

Conclusions

This work illustrates the use of causality and decision-theoretic principles to personalize interventions in digital mental health tools.

Subjects

MeSH terms (MEDLINE)

Artificial IntelligenceMental HealthBayes TheoremHumansAdultFemaleMaleMiddle AgedMobile ApplicationsSurveys and QuestionnairesMarkov Chains

Author keywords

causal inferencecausal structure learningdecision analysisdecision theorydigital healthfunctioningpsychological distresssleepsocial supportwell-being

Authors and institutions

  • Brain and Mind Centre, The University of Sydney, 94 Mallett Street, Sydney, 2050, Australia, 61 0293510774.
  • Human Technology Institute, University of Technology Sydney, Sydney, Australia.
  • Institute of Neuroscience, Academic Psychiatry, Newcastle University, Newcastle, United Kingdom.
  • Institute for Mental Health and Wellbeing, Flinders University, Adelaide, Australia.

Cite

APA

Varidel, M., An, V., Hickie, I. B., Cripps, S., Marchant, R., Scott, J., Crouse, J. J., Poulsen, A., O'Dea, B., McKenna, S., & Iorfino, F. (2025). Causal AI Recommendation System for Digital Mental Health: Bayesian Decision-Theoretic Analysis. Journal of Medical Internet Research, 27, e71305-e71305. https://doi.org/10.2196/71305

Vancouver

Varidel M, An V, Hickie IB, Cripps S, Marchant R, Scott J, et al. Causal AI Recommendation System for Digital Mental Health: Bayesian Decision-Theoretic Analysis. Journal of Medical Internet Research. 2025;27:e71305-e71305. doi:10.2196/71305

BibTeX
@article{Varidel2025,
  author = {Varidel, Mathew and An, Victor and Hickie, Ian B and Cripps, Sally and Marchant, Roman and Scott, Jan and Crouse, Jacob J and Poulsen, Adam and O'Dea, Bridianne and McKenna, Sarah and Iorfino, Frank},
  title = {Causal AI Recommendation System for Digital Mental Health: Bayesian Decision-Theoretic Analysis},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  volume = {27},
  pages = {e71305--e71305},
  doi = {10.2196/71305}
}
Start for free →