EUVIMEDAtrasti EUVIMED
Open access

Targeted learning in observational studies with multi‐valued treatments: An evaluation of antipsychotic drug treatment safety

Jason Poulos iD, Marcela Horvitz-Lennon, Katya Zelevinsky, Tudor Cristea-Platon, Thomas Huijskens, Pooja Tyagi, Jiaju Yan, Jordi Diaz, Sharon-Lise Normand iD

Statistics in Medicine · 2024

DOI: 10.1002/sim.10003 ↗

At a glance

Journal article in Statistics in Medicine, 2024. Written by Jason Poulos and 8 co-authors. Institution: Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts, USA. Major MEDLINE topics: Antipsychotic Agents, Cardiovascular Diseases. Cited 3 times so far (Crossref); 78 references. The full text is freely available in PubMed Central. Funded by: NIMH NIH HHS, National Institute of Mental Health.

SantraukaSubjectsBibliografiniai duomenysCiteRelated studies

Abstract (original language)

We investigate estimation of causal effects of multiple competing (multi‐valued) treatments in the absence of randomization. Our work is motivated by an intention‐to‐treat study of the relative cardiometabolic risk of assignment to one of six commonly prescribed antipsychotic drugs in a cohort of nearly 39 000 adults with serious mental illnesses. Doubly‐robust estimators, such as targeted minimum loss‐based estimation (TMLE), require correct specification of either the treatment model or outcome model to ensure consistent estimation; however, common TMLE implementations estimate treatment probabilities using multiple binomial regressions rather than multinomial regression. We implement a TMLE estimator that uses multinomial treatment assignment and ensemble machine learning to estimate average treatment effects. Our multinomial implementation improves coverage, but does not necessarily reduce bias, relative to the binomial implementation in simulation experiments with varying treatment propensity overlap and event rates. Evaluating the causal effects of the antipsychotics on 3‐year diabetes risk or death, we find a safety benefit of moving from a second‐generation drug considered among the safest of the second‐generation drugs to an infrequently prescribed first‐generation drug known for having low cardiometabolic risk.

Subjects

MeSH terms (MEDLINE)

Antipsychotic AgentsCardiovascular DiseasesHumansComputer SimulationLikelihood FunctionsModels, StatisticalAdultObservational Studies as Topic

Author keywords

antipsychotic drugscardiometabolic riskcausal inferencemulti-valued treatmentsserious mental illnesstargeted minimum-loss based estimation

Substances

Antipsychotic Agents

Authors and institutions

  • Department of Health Care Policy, Harvard Medical School, Boston, Massachusetts, USA.
  • RAND Corporation, Boston, Massachusetts, USA.
  • QuantumBlack, London, UK.
  • Department of Biostatistics, Harvard Chan School of Public Health, Boston, Massachusetts, USA.

ORCID profiles: Jason Poulos, Sharon-Lise Normand

Funding

  • NIMH NIH HHS
  • National Institute of Mental Health

Cite

APA

Poulos, J., Horvitz-Lennon, M., Zelevinsky, K., Cristea-Platon, T., Huijskens, T., Tyagi, P., Yan, J., Diaz, J., & Normand, S. L. (2024). Targeted learning in observational studies with multi‐valued treatments: An evaluation of antipsychotic drug treatment safety. Statistics in Medicine. https://doi.org/10.1002/sim.10003

Vancouver

Poulos J, Horvitz-Lennon M, Zelevinsky K, Cristea-Platon T, Huijskens T, Tyagi P, et al. Targeted learning in observational studies with multi‐valued treatments: An evaluation of antipsychotic drug treatment safety. Statistics in Medicine. 2024. doi:10.1002/sim.10003

BibTeX
@article{Poulos2024,
  author = {Poulos, Jason and Horvitz-Lennon, Marcela and Zelevinsky, Katya and Cristea-Platon, Tudor and Huijskens, Thomas and Tyagi, Pooja and Yan, Jiaju and Diaz, Jordi and Normand, Sharon-Lise},
  title = {Targeted learning in observational studies with multi‐valued treatments: An evaluation of antipsychotic drug treatment safety},
  journal = {Statistics in Medicine},
  year = {2024},
  doi = {10.1002/sim.10003}
}
Start for free →