Using Longitudinal Targeted Maximum Likelihood Estimation in Complex Settings with Dynamic Interventions
Preprint
Abstract
Longitudinal targeted maximum likelihood estimation (LTMLE) has hardly ever been used to estimate dynamic treatment effects in the context of time-dependent confounding affected by prior treatment when faced with long follow-up times, multiple time-varying confounders, and complex associational relationships simultaneously. Reasons for this include the potential computational burden, technical challenges, restricted modeling options for long follow-up times, and limited practical guidance in the literature. However, LTMLE has desirable asymptotic properties, i.e. it is doubly robust, and can yield valid inference when used in conjunction with machine learning. We use a topical and sophisticated question from HIV treatment research to show that LTMLE can be used successfully in complex realistic settings and compare results to competing estimators. Our example illustrates the following practical challenges common to many epidemiological studies 1) long follow-up time (30 months), 2) gradually declining sample size 3) limited support for some intervention rules of interest 4) a high-dimensional set of potential adjustment variables, increasing both the need and the challenge of integrating appropriate machine learning methods 5) consideration of collider bias. Our analyses, as well as simulations, shed new light on the application of LTMLE in complex and realistic settings: we show that (i) LTMLE can yield stable and good estimates, even when confronted with small samples and limited modeling options; (ii) machine learning utilized with a small set of simple learners (if more complex ones can't be fitted) can outperform a single, complex model, which is tailored to incorporate prior clinical knowledge; (iii) performance can vary considerably depending on interventions and their support in the data, and therefore critical quality checks should accompany every LTMLE analysis.Keywords
All Related Versions
This publication has 40 references indexed in Scilit:
- Targeted maximum likelihood estimation for a binary treatment: A tutorialStatistics in Medicine, 2018
- Collaborative targeted learning using regression shrinkageStatistics in Medicine, 2017
- Model selection and model averaging after multiple imputationComputational Statistics & Data Analysis, 2014
- Targeted learning in real‐world comparative effectiveness research with time‐varying interventionsStatistics in Medicine, 2014
- Methods for dealing with time‐dependent confoundingStatistics in Medicine, 2012
- Early versus deferred antiretroviral therapy for children older than 1 year infected with HIV (PREDICT): a multicentre, randomised, open-label trialThe Lancet Infectious Diseases, 2012
- The parametric g‐formula to estimate the effect of highly active antiretroviral therapy on incident AIDS or deathStatistics in Medicine, 2012
- Shrinkage averaging estimationStatistical Papers, 2011
- Comparative Effectiveness of Dynamic Treatment Regimes: An Application of the Parametric G-FormulaStatistics in Biosciences, 2011
- A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effectMathematical Modelling, 1986