Machine learning methods for propensity and disease risk score estimation in high-dimensional data: a plasmode simulation and real-world data cohort analysis
We found that ML methods could be reliable alternatives for PS estimation. ML-based DRS methods performed worse than PS ones, likely given the rarity of outcomes.
Authors
- Yuchen Guo (ORCID: https://orcid.org/0000-0002-0847-4855)
- Martí Català (ORCID: https://orcid.org/0000-0003-3308-9905)
- Victoria Y. Strauss (ORCID: https://orcid.org/0000-0002-5172-512X)
- Daniel Prieto‐Alhambra (ORCID: https://orcid.org/0000-0002-3950-6346)
- Sara Khalid (ORCID: https://orcid.org/0000-0002-2845-5731)
- Annika M. Jödicke (ORCID: https://orcid.org/0000-0002-0000-0110)
Institutions
- Boehringer Ingelheim (Germany) (DE)
- Nuffield Orthopaedic Centre (GB)
- Erasmus MC (NL)
- University of Oxford (GB)
Publication Details
- Journal
- Frontiers in Pharmacology
- Published
- 2024-10-28
- DOI
- https://doi.org/10.3389/fphar.2024.1395707
- Citations
- 3
- Primary Topic
- Advanced Causal Inference Techniques
- Type
- article
- Field-Weighted Citation Impact
- 2.13
Funders
- National Institute for Health and Care Research
- NIHR Oxford Biomedical Research Centre