arXiv subjects

Trang Quynh Nguyen

Publications and source records attributed to Trang Quynh Nguyen.

18 recordsLinked to original sources

arXiv2026

Self-separated and self-connected models for mediator and outcome missingness in mediation analysis

Missing data is a common challenge in studying treatment effects. In the context of mediation analysis, this paper addresses missingness in the mediator and outcome, focusing on identification. We first consider self-separated missingness models where identification is achieved by conditional independence assumptions. This model class is somewhat limited as it is constrained by the need to remove a certain number of connections from the model. We then turn to self-connected missingness models where identification relies on information from shadow variables. This model class turns out to contain substantial variation, allowing models with built-in shadow variables (mediator, outcome or covariates) and models with auxiliary shadow variables at different positions in the causal structure. To improve the practical value of the missingness mechanisms, we allow where possible for dependencies due to unobserved causes of the missingness, a feature often neglected. In this exploration, we review existing models, connect to new models, and develop theory where needed. This results in templates for identification in the mediation setting, generally useful identification techniques, and perhaps most importantly a synthesis and substantial extension of shadow variable theory. Two examples relate the models to practical considerations.

stat.ME↗

arXiv2026

Demystifying Proximal Causal Inference

Proximal causal inference (PCI) has emerged as a promising framework for identifying and estimating causal effects in the presence of unobserved confounders. While many traditional causal inference methods rely on the assumption of no unobserved confounding, this assumption is likely often violated. PCI addresses this challenge by relying on an alternative set of assumptions regarding the relationships between treatment, outcome, and auxiliary variables that serve as proxies for unmeasured confounders. We review existing identification results, discuss the assumptions necessary for valid causal effect estimation via PCI, and compare different PCI estimation methods. We offer practical guidance on operationalizing PCI, with a focus on selecting and evaluating proxy variables using domain knowledge, measurement error perspectives, and negative control analogies. Through conceptual examples, we demonstrate tensions in proxy selection and discuss the importance of clearly defining the unobserved confounding mechanism. By bridging formal results with applied considerations, this work aims to demystify PCI, encourage thoughtful use in practice, and identify open directions for methodological development and empirical research.

stat.ME↗

arXiv2025

Sensitivity Analysis when Generalizing Causal Effects from Multiple Studies to a Target Population: Motivation from the ECHO Program

Unobserved effect modifiers can induce bias when generalizing causal effect estimates to target populations. In this work, we extend a sensitivity analysis framework assessing the robustness of study results to unobserved effect modification that adapts to various generalizability scenarios, including multiple (conditionally) randomized trials, observational studies, or combinations thereof. This framework is interpretable and does not rely on distributional or functional assumptions about unknown parameters. We demonstrate how to leverage the multi-study setting to detect violation of the generalizability assumption through hypothesis testing, showing with simulations that the proposed test achieves high power under real-world sample sizes. Finally, we apply our sensitivity analysis framework to analyze the generalized effect estimate of secondhand smoke exposure on birth weight using cohort sites from the Environmental influences on Child Health Outcomes (ECHO) study.

stat.ME↗

arXiv2025

Precision Mental Health: Predicting Heterogeneous Treatment Effects for Depression through Data Integration

When treating depression, clinicians are interested in determining the optimal treatment for a given patient, which is challenging given the amount of treatments available. To advance individualized treatment allocation, integrating data across multiple randomized controlled trials (RCTs) can enhance our understanding of treatment effect heterogeneity by increasing available information. However, extending these inferences to individuals outside of the original RCTs remains crucial for clinical decision-making. We introduce a two-stage meta-analytic method that predicts conditional average treatment effects (CATEs) in target patient populations by leveraging the distribution of CATEs across RCTs. Our approach generates 95\% prediction intervals for CATEs in target settings using first-stage models that can incorporate parametric regression or non-parametric methods such as causal forests or Bayesian additive regression trees (BART). We validate our method through simulation studies and operationalize it to integrate multiple RCTs comparing depression treatments, duloxetine and vortioxetine, to generate prediction intervals for target patient profiles. Our analysis reveals no strong evidence of effect heterogeneity across trials, with the exception of potential age-related variability. Importantly, we show that CATE prediction intervals capture broader uncertainty than study-specific confidence intervals when warranted, reflecting both within-study and between-study variability.

stat.AP↗

arXiv2025

A Cautionary Tale on Integrating Studies with Disparate Outcome Measures for Causal Inference

Data integration approaches are increasingly used to enhance the efficiency and generalizability of studies. However, a key limitation of these methods is the assumption that outcome measures are identical across datasets -- an assumption that often does not hold in practice. Consider the following opioid use disorder (OUD) studies: the XBOT trial and the POAT study, both evaluating the effect of medications for OUD on withdrawal symptom severity (not the primary outcome of either trial). While XBOT measures withdrawal severity using the subjective opiate withdrawal scale, POAT uses the clinical opiate withdrawal scale. We analyze this realistic yet challenging setting where outcome measures differ across studies and where neither study records both types of outcomes. Our paper studies whether and when integrating studies with disparate outcome measures leads to efficiency gains. We introduce three sets of assumptions -- with varying degrees of strength -- linking both outcome measures. Our theoretical and empirical results highlight a cautionary tale: integration can improve asymptotic efficiency only under the strongest assumption linking the outcomes. However, misspecification of this assumption leads to bias. In contrast, a milder assumption may yield finite-sample efficiency gains, yet these benefits diminish as sample size increases. We illustrate these trade-offs via a case study integrating the XBOT and POAT datasets to estimate the comparative effect of two medications for opioid use disorder on withdrawal symptoms. By systematically varying the assumptions linking the SOW and COW scales, we show potential efficiency gains and the risks of bias. Our findings emphasize the need for careful assumption selection when fusing datasets with differing outcome measures, offering guidance for researchers navigating this common challenge in modern data integration.

stat.ME↗

arXiv2024

In defense of MAR over latent ignorability (or latent MAR) for outcome missingness in studying principal causal effects: a causal graph view

This paper concerns outcome missingness in principal stratification analysis. We revisit a common assumption known as latent ignorability or latent missing-at-random (LMAR), often considered a relaxation of missing-at-random (MAR). LMAR posits that the outcome is independent of its missingness if one conditions on principal stratum (which is partially unobservable) in addition to observed variables. The literature has focused on methods assuming LMAR (usually supplemented with a more specific assumption about the missingness), without considering the theoretical plausibility and necessity of LMAR. In this paper, we devise a way to represent principal stratum in causal graphs, and use causal graphs to examine this assumption. We find that LMAR is harder to satisfy than MAR, and for the purpose of breaking the dependence between the outcome and its missingness, no benefit is gained from conditioning on principal stratum on top of conditioning on observed variables. This finding has an important implication: MAR should be preferred over LMAR. This is convenient because MAR is easier to handle and (unlike LMAR) if MAR is assumed no additional assumption is needed. We thus turn to focus on the plausibility of MAR and its implications, with a view to facilitate appropriate use of this assumption. We clarify conditions on the causal structure and on auxiliary variables (if available) that need to hold for MAR to hold, and we use MAR to recover effect identification under two dominant identification assumptions (exclusion restriction and principal ignorability). We briefly comment on cases where MAR does not hold. In terms of broader connections, most of the MAR findings are also relevant to classic instrumental variable analysis that targets the local average treatment effect; and the LMAR finding suggests general caution with assumptions that condition on principal stratum.

stat.ME↗

arXiv2024

Sensitivity analysis for principal ignorability violation in estimating complier and noncomplier average causal effects

An important strategy for identifying principal causal effects, which are often used in settings with noncompliance, is to invoke the principal ignorability (PI) assumption. As PI is untestable, it is important to gauge how sensitive effect estimates are to its violation. We focus on this task for the common one-sided noncompliance setting where there are two principal strata, compliers and noncompliers. Under PI, compliers and noncompliers share the same outcome-mean-given-covariates function under the control condition. For sensitivity analysis, we allow this function to differ between compliers and noncompliers in several ways, indexed by an odds ratio, a generalized odds ratio, a mean ratio, or a standardized mean difference sensitivity parameter. We tailor sensitivity analysis techniques (with any sensitivity parameter choice) to several types of PI-based main analysis methods, including outcome regression, influence function (IF) based and weighting methods. We illustrate the proposed sensitivity analyses using several outcome types from the JOBS II study. This application estimates nuisance functions parametrically -- for simplicity and accessibility. In addition, we establish rate conditions on nonparametric nuisance estimation for IF-based estimators to be asymptotically normal -- with a view to inform nonparametric inference.

stat.ME↗

arXiv2023

Identification of complier and noncomplier average causal effects in the presence of latent missing-at-random (LMAR) outcomes: a unifying view and choices of assumptions

The study of treatment effects is often complicated by noncompliance and missing data. In the one-sided noncompliance setting where of interest are the complier and noncomplier average causal effects (CACE and NACE), we address outcome missingness of the \textit{latent missing at random} type (LMAR, also known as \textit{latent ignorability}). That is, conditional on covariates and treatment assigned, the missingness may depend on compliance type. Within the instrumental variable (IV) approach to noncompliance, methods have been proposed for handling LMAR outcome that additionally invoke an exclusion restriction type assumption on missingness, but no solution has been proposed for when a non-IV approach is used. This paper focuses on effect identification in the presence of LMAR outcome, with a view to flexibly accommodate different principal identification approaches. We show that under treatment assignment ignorability and LMAR only, effect nonidentifiability boils down to a set of two connected mixture equations involving unidentified stratum-specific response probabilities and outcome means. This clarifies that (except for a special case) effect identification generally requires two additional assumptions: a \textit{specific missingness mechanism} assumption and a \textit{principal identification} assumption. This provides a template for identifying effects based on separate choices of these assumptions. We consider a range of specific missingness assumptions, including those that have appeared in the literature and some new ones. Incidentally, we find an issue in the existing assumptions, and propose a modification of the assumptions to avoid the issue. Results under different assumptions are illustrated using data from the Baltimore Experience Corps Trial.

stat.ME↗

arXiv2023

Comparison of Methods that Combine Multiple Randomized Trials to Estimate Heterogeneous Treatment Effects

Individualized treatment decisions can improve health outcomes, but using data to make these decisions in a reliable, precise, and generalizable way is challenging with a single dataset. Leveraging multiple randomized controlled trials allows for the combination of datasets with unconfounded treatment assignment to better estimate heterogeneous treatment effects. This paper discusses several non-parametric approaches for estimating heterogeneous treatment effects using data from multiple trials. We extend single-study methods to a scenario with multiple trials and explore their performance through a simulation study, with data generation scenarios that have differing levels of cross-trial heterogeneity. The simulations demonstrate that methods that directly allow for heterogeneity of the treatment effect across trials perform better than methods that do not, and that the choice of single-study method matters based on the functional form of the treatment effect. Finally, we discuss which methods perform well in each setting and then apply them to four randomized controlled trials to examine effect heterogeneity of treatments for major depressive disorder.

stat.ME↗

arXiv2023

Multiple imputation for propensity score analysis with covariates missing at random: some clarity on within and across methods

In epidemiology and social sciences, propensity score methods are popular for estimating treatment effects using observational data, and multiple imputation is popular for handling covariate missingness. However, how to appropriately use multiple imputation for propensity score analysis is not completely clear. This paper aims to bring clarity on the consistency (or lack thereof) of methods that have been proposed, focusing on the within approach (where the effect is estimated separately in each imputed dataset and then the multiple estimates are combined) and the across approach (where typically propensity scores are averaged across imputed datasets before being used for effect estimation). We show that the within method is valid and can be used with any causal effect estimator that is consistent in the full-data setting. Existing across methods are inconsistent, but a different across method that averages the inverse probability weights across imputed datasets is consistent for propensity score weighting. We also comment on methods that rely on imputing a function of the missing covariate rather than the covariate itself, including imputation of the propensity score and of the probability weight. Based on consistency results and practical flexibility, we recommend generally using the standard within method. Throughout, we provide intuition to make the results meaningful to the broad audience of applied researchers.

stat.ME↗

arXiv2023

Methods for Integrating Trials and Non-Experimental Data to Examine Treatment Effect Heterogeneity

Estimating treatment effects conditional on observed covariates can improve the ability to tailor treatments to particular individuals. Doing so effectively requires dealing with potential confounding, and also enough data to adequately estimate effect moderation. A recent influx of work has looked into estimating treatment effect heterogeneity using data from multiple randomized controlled trials and/or observational datasets. With many new methods available for assessing treatment effect heterogeneity using multiple studies, it is important to understand which methods are best used in which setting, how the methods compare to one another, and what needs to be done to continue progress in this field. This paper reviews these methods broken down by data setting: aggregate-level data, federated learning, and individual participant-level data. We define the conditional average treatment effect and discuss differences between parametric and nonparametric estimators, and we list key assumptions, both those that are required within a single study and those that are necessary for data combination. After describing existing approaches, we compare and contrast them and reveal open areas for future research. This review demonstrates that there are many possible approaches for estimating treatment effect heterogeneity through the combination of datasets, but that there is substantial work to be done to compare these methods through case studies and simulations, extend them to different settings, and refine them to account for various challenges present in real data.

stat.ME↗

arXiv2023

Causal mediation analysis: From simple to more robust strategies for estimation of marginal natural (in)direct effects

This paper aims to provide practitioners of causal mediation analysis with a better understanding of estimation options. We take as inputs two familiar strategies (weighting and model-based prediction) and a simple way of combining them (weighted models), and show how a range of estimators can be generated, with different modeling requirements and robustness properties. The primary goal is to help build intuitive appreciation for robust estimation that is conducive to sound practice. A second goal is to provide a "menu" of estimators that practitioners can choose from for the estimation of marginal natural (in)direct effects. The estimators generated from this exercise include some that coincide or are similar to existing estimators and others that have not previously appeared in the literature. We note several different ways to estimate the weights for cross-world weighting based on three expressions of the weighting function, including one that is novel; and show how to check the resulting covariate and mediator balance. We use a random continuous weights bootstrap to obtain confidence intervals, and also derive general asymptotic variance formulas for the estimators. The estimators are illustrated using data from an adolescent alcohol use prevention study.

stat.ME↗

arXiv2022

Clarifying causal mediation analysis: Effect identification via three assumptions and five potential outcomes

Causal mediation analysis is complicated with multiple effect definitions that require different sets of assumptions for identification. This paper provides a systematic explanation of such assumptions. We define five potential outcome types whose means are involved in various effect definitions. We tackle their mean/distribution's identification, starting with the one that requires the weakest assumptions and gradually building up to the one that requires the strongest assumptions. This presentation shows clearly why an assumption is required for one estimand and not another, and provides a succinct table from which an applied researcher could pick out the assumptions required for identifying the causal effects they target. Using a running example, the paper illustrates the assembling and consideration of identifying assumptions for a range of causal contrasts. For several that are commonly encountered in the literature, this exercise clarifies that identification requires weaker assumptions than those often stated in the literature. This attention to the details also draws attention to the differences in the positivity assumption for different estimands, with practical implications. Clarity on the identifying assumptions of these various estimands will help researchers conduct appropriate mediation analyses and interpret the results with appropriate caution given the plausibility of the assumptions.

stat.ME↗

arXiv2020

Sensitivity analyses for effect modifiers not observed in the target population when generalizing treatment effects from a randomized controlled trial: Assumptions, models, effect scales, data scenarios, and implementation details

Background: Randomized controlled trials are often used to inform policy and practice for broad populations. The average treatment effect (ATE) for a target population, however, may be different from the ATE observed in a trial if there are effect modifiers whose distribution in the target population is different that from that in the trial. Methods exist to use trial data to estimate the target population ATE, provided the distributions of treatment effect modifiers are observed in both the trial and target population -- an assumption that may not hold in practice. Methods: The proposed sensitivity analyses address the situation where a treatment effect modifier is observed in the trial but not the target population. These methods are based on an outcome model or the combination of such a model and weighting adjustment for observed differences between the trial sample and target population. They accommodate several types of outcome models: linear models (including single time outcome and pre- and post-treatment outcomes) for additive effects, and models with log or logit link for multiplicative effects. We clarify the methods' assumptions and provide detailed implementation instructions. Illustration: We illustrate the methods using an example generalizing the effects of an HIV treatment regimen from a randomized trial to a relevant target population. Conclusion: These methods allow researchers and decision-makers to have more appropriate confidence when drawing conclusions about target population effects.

stat.ME↗

arXiv2020

Clarifying causal mediation analysis for the applied researcher: Defining effects based on what we want to learn

The incorporation of causal inference in mediation analysis has led to theoretical and methodological advancements -- effect definitions with causal interpretation, clarification of assumptions required for effect identification, and an expanding array of options for effect estimation. However, the literature on these results is fast-growing and complex, which may be confusing to researchers unfamiliar with causal inference or unfamiliar with mediation. The goal of this paper is to help ease the understanding and adoption of causal mediation analysis. It starts by highlighting a key difference between the causal inference and traditional approaches to mediation analysis and making a case for the need for explicit causal thinking and the causal inference approach in mediation analysis. It then explains in as-plain-as-possible language existing effect types, paying special attention to motivating these effects with different types of research questions, and using concrete examples for illustration. This presentation differentiates two perspectives (or purposes of analysis): the explanatory perspective (aiming to explain the total effect) and the interventional perspective (asking questions about hypothetical interventions on the exposure and mediator, or hypothetically modified exposures). For the latter perspective, the paper proposes tapping into a general class of interventional effects that contains as special cases most of the usual effect types -- interventional direct and indirect effects, controlled direct effects and also a generalized interventional direct effect type, as well as the total effect and overall effect. This general class allows flexible effect definitions which better match many research questions than the standard interventional direct and indirect effects.

stat.ME↗

arXiv2020

Propensity score analysis with latent covariates: Measurement error bias correction using the covariate's posterior mean, aka the inclusive factor score

We address measurement error bias in propensity score (PS) analysis due to covariates that are latent variables. In the setting where latent covariate $X$ is measured via multiple error-prone items $\mathbf{W}$, PS analysis using several proxies for $X$ -- the $\mathbf{W}$ items themselves, a summary score (mean/sum of the items), or the conventional factor score (cFS , i.e., predicted value of $X$ based on the measurement model) -- often results in biased estimation of the causal effect, because balancing the proxy (between exposure conditions) does not balance $X$. We propose an improved proxy: the conditional mean of $X$ given the combination of $\mathbf{W}$, the observed covariates $Z$, and exposure $A$, denoted $X_{WZA}$. The theoretical support, which applies whether $X$ is latent or not (but is unobserved), is that balancing $X_{WZA}$ (e.g., via weighting or matching) implies balancing the mean of $X$. For a latent $X$, we estimate $X_{WZA}$ by the inclusive factor score (iFS) -- predicted value of $X$ from a structural equation model that captures the joint distribution of $(X,\mathbf{W},A)$ given $Z$. Simulation shows that PS analysis using the iFS substantially improves balance on the first five moments of $X$ and reduces bias in the estimated causal effect. Hence, within the proxy variables approach, we recommend this proxy over existing ones. We connect this proxy method to known results about weighting/matching functions (Lockwood & McCaffrey, 2016; McCaffrey, Lockwood, & Setodji, 2013). We illustrate the method in handling latent covariates when estimating the effect of out-of-school suspension on risk of later police arrests using Add Health data.

stat.ME↗

arXiv2019

The magnitude and direction of collider bias for binary variables

Suppose we are interested in the effect of variable $X$ on variable $Y$. If $X$ and $Y$ both influence, or are associated with variables that influence, a common outcome, called a collider, then conditioning on the collider (or on a variable influenced by the collider -- its "child") induces a spurious association between $X$ and $Y$, which is known as collider bias. Characterizing the magnitude and direction of collider bias is crucial for understanding the implications of selection bias and for adjudicating decisions about whether to control for variables that are known to be associated with both exposure and outcome but could be either confounders or colliders. Considering a class of situations where all variables are binary, and where $X$ and $Y$ either are, or are respectively influenced by, two marginally independent causes of a collider, we derive collider bias that results from (i) conditioning on specific levels of, or (ii) linear regression adjustment for, the collider (or its child). We also derive simple conditions that determine the sign of such bias.

stat.ME↗

arXiv2016

Sensitivity analysis for an unobserved moderator in RCT-to-target-population generalization of treatment effects

In the presence of treatment effect heterogeneity, the average treatment effect (ATE) in a randomized controlled trial (RCT) may differ from the average effect of the same treatment if applied to a target population of interest. If all treatment effect moderators are observed in the RCT and in a dataset representing the target population, we can obtain an estimate for the target population ATE by adjusting for the difference in the distribution of the moderators between the two samples. This paper considers sensitivity analyses for two situations: (1) where we cannot adjust for a specific moderator $V$ observed in the RCT because we do not observe it in the target population; and (2) where we are concerned that the treatment effect may be moderated by factors not observed even in the RCT, which we represent as a composite moderator $U$. In both situations, the outcome is not observed in the target population. For situation (1), we offer three sensitivity analysis methods based on (i) an outcome model, (ii) full weighting adjustment, and (iii) partial weighting combined with an outcome model. For situation (2), we offer two sensitivity analyses based on (iv) a bias formula and (v) partial weighting combined with a bias formula. We apply methods (i) and (iii) to an example where the interest is to generalize from a smoking cessation RCT conducted with participants of alcohol/illicit drug use treatment programs to the target population of people who seek treatment for alcohol/illicit drug use in the US who are also cigarette smokers. In this case a treatment effect moderator is observed in the RCT but not in the target population dataset.

stat.ME↗