High-dimensional propensity score algorithm in comparative effectiveness research with time-varying interventions.

@article{Neugebauer2015HighdimensionalPS,
  title={High-dimensional propensity score algorithm in comparative effectiveness research with time-varying interventions.},
  author={Romain Neugebauer and Julie A. Schmittdiel and Zheng Zhu and Jeremy A. Rassen and John D. Seeger and Sebastian Schneeweiss},
  journal={Statistics in medicine},
  year={2015},
  volume={34 5},
  pages={
          753-81
        }
}
The high-dimensional propensity score (hdPS) algorithm was proposed for automation of confounding adjustment in problems involving large healthcare databases. [] Key Method We describe the application and performance of the hdPS algorithm to improve covariate selection in CER with time-varying interventions based on IPW estimation and explore stabilization of the resulting estimates using Super Learning. The evaluation is based on both the analysis of electronic health records data in a real-world CER study…
View on PubMed
diabetestranslation.org
35 Citations
Background Citations
7
Methods Citations
5
Results Citations
1

35 Citations

Citation Type

Citation Type

Evaluation of adaptive treatment strategies in an observational study where time-varying covariates are not monitored systematically
In studies based on electronic health records (EHR), the frequency of covariate monitoring can vary by covariate type, across patients, and over time. This can lead to major challenges: first, the…
Can We Train Machine Learning Methods to Outperform the High-dimensional Propensity Score Algorithm?
TLDR
This work uses patient’s health status–related information from claims datasets as surrogates or proxies for mismeasured and unobserved confounders, and shows that the high-dimensional propensity score algorithm enables us to reduce bias.
Propensity score prediction for electronic healthcare databases using super learner and high-dimensional propensity score methods
TLDR
Combining the SL with the hdPS was the most consistent prediction method and may be promising for PS estimation and prediction modeling in electronic healthcare databases.
Automated data-adaptive analytics for electronic healthcare data to study causal treatment effects
TLDR
Semiautomated and optimized confounding adjustment in health care database analyses has proven successful across a wide range of settings and machine-learning extensions further automate its use in estimating causal treatment effects across a range of data scenarios.
Exploiting non-systematic covariate monitoring to broaden the scope of evidence about the causal effects of adaptive treatment strategies.
TLDR
How to exploit non-systematic covariate monitoring in EHR-based studies to both improve the generalizability of causal inferences and to evaluate the health impact of monitoring when evaluating adaptive treatment strategies is demonstrated.
Matching with time‐dependent treatments: A review and look forward
TLDR
This tutorial defines a class of longitudinal matching methods that emulate this experiment and provides a review of existing variations, with guidance regarding study design, execution, and analysis.
On the role of marginal confounder prevalence – implications for the high‐dimensional propensity score algorithm
TLDR
The role of the marginal prevalence of confounder variables on potentially caused bias magnitudes when estimating risk ratios in point exposure studies with binary outcomes is investigated.
Comparative Effectiveness of Adaptive Treatment Strategies
In this chapter, we describe secondary analyses of electronic health record (EHR) data from a type 2 diabetes mellitus (T2DM) study of the effect of four adaptive treatment strategies on a…
Collaborative-controlled LASSO for constructing propensity score-based estimators in high-dimensional data
TLDR
A novel approach for collaborative model selection when using the LASSO estimator for propensity score estimation in high-dimensional covariate settings is introduced and the results showed that the collaborative minimum loss-based estimation algorithm outperformed other competing estimators for both point estimation and confidence interval coverage.
Targeted learning with daily EHR data.
TLDR
A novel TMLE implementation is proposed, the "long-format TMLE," and the latest advances in scalable data-adaptive machine-learning software, xgboost and h2o, are relied on, for estimation of the TMLE nuisance parameters.
...
1
2
3
4
...

References

SHOWING 1-10 OF 87 REFERENCES
Targeted learning in real-world comparative effectiveness research with time-varying interventions.
TLDR
This application establishes the feasibility of TMLE in real-world CER based on large healthcare databases and provides evidence of proper confounding and selection bias adjustment with TMLE and SL, and motivates their application for improving estimation efficiency.
High-dimensional Propensity Score Adjustment in Studies of Treatment Effects Using Health Care Claims Data
TLDR
In typical pharmacoepidemiologic studies, the proposed high-dimensional propensity score resulted in improved effect estimates compared with adjustment limited to predefined covariates, when benchmarked against results expected from randomized trials.
Covariate selection in high-dimensional propensity score analyses of treatment effects in small samples.
TLDR
Hd-PS is a flexible analytical tool for nonrandomized research across a range of study sizes and event frequencies and allowed to function robustly with few events in 4 North American pharmacoepidemiologic cohort studies between 1995 and 2005.
Using high‐dimensional propensity scores to automate confounding control in a distributed medical product safety surveillance system
TLDR
The use of hd‐PS for automating confounding control in sequential database cohort studies, as applied to safety monitoring systems is discussed and it is concluded that despite certain limitations, hd-PS offers substantial advantages over non‐automated alternatives in active productSafety monitoring systems.
Marginal structural models for analyzing causal effects of time-dependent treatments: an application in perinatal epidemiology.
TLDR
The objective of this paper is to illustrate the fitting of an MSM for the causal effect of iron supplement use during pregnancy (time-varying treatment) on odds of anemia at delivery in the presence of time-dependent confounding.
Super learning to hedge against incorrect inference from arbitrary parametric assumptions in marginal structural modeling.
A Marginal Structural Modeling Approach with Super Learning for a Study on Oral Bisphosphonate Therapy and Atrial Fibrillation
TLDR
An effort to specifically address challenges with marginal structural modeling based on inverse probability weighting with data reduction and super learning is described.
A Marginal Structural Modeling Approach with Super Learning for a Study on Oral Bisphosphonate Therapy and Atrial Fibrillation
TLDR
An effort to specifically address challenges with marginal structural modeling based on inverse probability weighting with data reduction and super learning to provide a feasible alternative to extended Cox modeling or the point treatment analytic approaches that are often adopted in safety research with large data sets.
Comparison of dynamic treatment regimes via inverse probability weighting.
TLDR
A simple method to compare dynamic treatment regimes by artificially censoring subjects and then using inverse probability weighting (IPW) to adjust for any selection bias introduced by the artificial censoring is described.
Dynamic marginal structural modeling to evaluate the comparative effectiveness of more or less aggressive treatment intensification strategies in adults with type 2 diabetes
TLDR
Under explicit assumptions, the application of inverse probability weighting estimation to fit dynamic marginal structural models (MSMs) in observational studies to address pragmatic CER questions and properly adjust for time‐dependent confounding and informative loss to follow-up is motivated.
...
1
2
3
4
5
...