Search

Provider profiling has the goal of identifying healthcare providers with exceptional patient outcomes. When evaluating providers, adjustment is necessary to control for differences in case-mix between different providers. Direct and indirect standardization are two popular risk adjustment methods. In causal terms, direct standardization examines a counterfactual in which the entire target population is treated by one provider. Indirect standardization, commonly expressed as a standardized outcome ratio, examines the counterfactual in which the population treated by a provider had instead been randomly assigned to another provider. Our first contribution is to present nonparametric efficiency bound for direct and indirectly standardized provider metrics by deriving their efficient influence functions. Our second contribution is to propose fully nonparametric estimators based on targeted minimum loss-based estimation that achieve the efficiency bounds. The finite-sample performance of the estimator is investigated through simulation studies. We apply our methods to evaluate dialysis facilities in New York State in terms of unplanned readmission rates using a large Medicare claims dataset. A software implementation of our methods is available in the R package TargetedRisk., Comment: 20 pages, 2 figures

Comparing outcomes across treatments is essential in medicine and public policy. To do so, researchers typically estimate a set of parameters, possibly counterfactual, with each targeting a different treatment. Treatment-specific means (TSMs) are commonly used, but their identification requires a positivity assumption -- that every subject has a non-zero probability of receiving each treatment. This assumption is often implausible, especially when treatment can take many values. Causal parameters based on dynamic stochastic interventions can be robust to positivity violations. However, comparing these parameters may be unfair because they may depend on outcomes under non-target treatments. To address this, and clarify when fair comparisons are possible, we propose a fairness criterion: if the conditional TSM for one treatment is greater than that for another, then the corresponding causal parameter should also be greater. We derive two intuitive properties equivalent to this criterion and show that only a mild positivity assumption is needed to identify fair parameters. We then provide examples that satisfy this criterion and are identifiable under the milder positivity assumption. These parameters are non-smooth, making standard nonparametric efficiency theory inapplicable, so we propose smooth approximations of them. We then develop doubly robust-style estimators that attain parametric convergence rates under nonparametric conditions. We illustrate our methods with an analysis of dialysis providers in New York State.

Marginal structural models are a popular method for estimating causal effects in the presence of time-varying exposures. In spite of their popularity, no scalable non-parametric estimator exist for marginal structural models with multi-valued and time-varying treatments. In this paper, we use machine learning together with recent developments in semiparametric efficiency theory for longitudinal studies to propose such an estimator. The proposed estimator is based on a study of the non-parametric identifying functional, including first order von-Mises expansions as well as the efficient influence function and the efficiency bound. We show conditions under which the proposed estimator is efficient, asymptotically normal, and sequentially doubly robust in the sense that it is consistent if, for each time point, either the outcome or the treatment mechanism is consistently estimated. We perform a simulation study to illustrate the properties of the estimators, and present the results of our motivating study on a COVID-19 dataset studying the impact of mobility on the cumulative number of observed cases., Comment: 15 pages, 1 figure, 3 tables

Causal mediation analyses investigate the mechanisms through which causes exert their effects, and are therefore central to scientific progress. The literature on the non-parametric definition and identification of mediational effects in rigourous causal models has grown significantly in recent years, and there has been important progress to address challenges in the interpretation and identification of such effects. Despite great progress in the causal inference front, statistical methodology for non-parametric estimation has lagged behind, with few or no methods available for tackling non-parametric estimation in the presence of multiple, continuous, or high-dimensional mediators. In this paper we show that the identification formulas for six popular non-parametric approaches to mediation analysis proposed in recent years can be recovered from just two statistical estimands. We leverage this finding to propose an all-purpose one-step estimation algorithm that can be coupled with machine learning in any mediation study that uses any of these six definitions of mediation. The estimators have desirable properties, such as $\sqrt{n}$-convergence and asymptotic normality. Estimating the first-order correction for the one-step estimator requires estimation of complex density ratios on the potentially high-dimensional mediators, a challenge that is solved using recent advancements in so-called Riesz learning. We illustrate the properties of our methods in a simulation study and illustrate its use on real data to estimate the extent to which pain management practices mediate the total effect of having a chronic pain disorder on opioid use disorder.

For many rare diseases with no approved preventive interventions, promising interventions exist, yet it has been difficult to conduct a pivotal phase 3 trial that could provide direct evidence demonstrating a beneficial effect on the target disease outcome. When a promising putative surrogate endpoint(s) for the target outcome is available, surrogate-based provisional approval of an intervention may be pursued. We apply the Causal Roadmap rubric to define a surrogate endpoint based provisional approval causal roadmap, which combines observational study data that estimates the relationship between the putative surrogate and the target outcome, with a phase 3 surrogate endpoint study that collects the same data but is very under-powered to assess the treatment effect (TE) on the target outcome. The objective is conservative estimation/inference for the TE with an estimated lower uncertainty bound that allows (through two bias functions) for an imperfect surrogate and imperfect transport of the conditional target outcome risk in the untreated between the observational and phase 3 studies. Two estimators of TE (plug-in, nonparametric efficient one-step) with corresponding inference procedures are developed. Finite-sample performance of the plug-in estimator is evaluated in two simulation studies, with R code provided. The roadmap is illustrated with contemporary Group B Streptococcus vaccine development., Comment: 33 pages, 4 figures

Difference-in-differences is based on a parallel trends assumption, which states that changes over time in average potential outcomes are independent of treatment assignment, possibly conditional on covariates. With time-varying treatments, parallel trends assumptions can identify many types of parameters, but most work has focused on group-time average treatment effects and similar parameters conditional on the treatment trajectory. This paper focuses instead on identification and estimation of the intervention-specific mean - the mean potential outcome had everyone been exposed to a proposed intervention - which may be directly policy-relevant in some settings. Previous estimators for this parameter under parallel trends have relied on correctly-specified parametric models, which may be difficult to guarantee in applications. We develop multiply-robust and efficient estimators of the intervention-specific mean based on the efficient influence function, and derive conditions under which data-adaptive machine learning methods can be used to relax modeling assumptions. Our approach allows the parallel trends assumption to be conditional on the history of time-varying covariates, thus allowing for adjustment for time-varying covariates possibly impacted by prior treatments. Simulation results support the use of the proposed methods at modest sample sizes. As an example, we estimate the effect of a hypothetical federal minimum wage increase on self-rated health in the US.

The Average Treatment Effect on the Treated (ATT) is a common causal parameter defined as the average effect of a binary treatment among the subset of the population receiving treatment. We propose a novel family of parameters, Generalized ATTs (GATTs), that generalize the concept of the ATT to longitudinal data structures, multi-valued or continuous treatments, and conditioning on arbitrary treatment subsets. We provide a formal causal identification result that expresses the GATT in terms of sequential regressions, and derive the efficient influence function of the parameter, which defines its semi-parametric efficiency bound. Efficient semi-parametric inference of the GATT requires estimating the ratios of functions of conditional probabilities (or densities); we propose directly estimating these ratios via empirical loss minimization, drawing on the theory of Riesz representers. Simulations suggest that estimation of the density ratios using Riesz representation have better stability in finite samples. Lastly, we illustrate the use of our methods to evaluate the effect of chronic pain management strategies on the development of opioid use disorder among Medicare patients with chronic pain., Comment: 35 pages, 1 figure

Platform trials are multi-arm designs that simultaneously evaluate multiple treatments for a single disease within the same overall trial structure. Unlike traditional randomized controlled trials, they allow treatment arms to enter and exit the trial at distinct times while maintaining a control arm throughout. This control arm comprises both concurrent controls, where participants are randomized concurrently to either the treatment or control arm, and non-concurrent controls, who enter the trial when the treatment arm under study is unavailable. While flexible, platform trials introduce the challenge of using non-concurrent controls, raising questions about estimating treatment effects. Specifically, which estimands should be targeted? Under what assumptions can these estimands be identified and estimated? Are there any efficiency gains? In this paper, we discuss issues related to the identification and estimation assumptions of common choices of estimand. We conclude that the most robust strategy to increase efficiency without imposing unwarranted assumptions is to target the concurrent average treatment effect (cATE), the ATE among only concurrent units, using a covariate-adjusted doubly robust estimator. Our studies suggest that, for the purpose of obtaining efficiency gains, collecting important prognostic variables is more important than relying on non-concurrent controls. We also discuss the perils of targeting ATE due to an untestable extrapolation assumption that will often be invalid. We provide simulations illustrating our points and an application to the ACTT platform trial, resulting in a 20% improvement in precision compared to the naive estimator that ignores non-concurrent controls and prognostic variables.

Introduction: Chronic pain patients are at increased risk of opioid-misuse. Less is known about the unique risk conferred by each pain-management treatment, as treatments are typically implemented together, confounding their independent effects. We estimated the extent to which pain-management strategies were associated with risk of incident opioid use disorder (OUD) for those with chronic pain, controlling for baseline demographic and clinical confounding variables and holding other pain-management treatments at their observed levels. Methods: We used data from two chronic pain subgroups within a cohort of non-pregnant Medicaid patients aged 35-64 years, 2016-2019, from 25 states: 1) those with a chronic pain condition co-morbid with physical disability (N=6,133) or 2) those with chronic pain without disability (N=67,438). We considered 9 pain-management treatments: prescription opioid i) dose and ii) duration; iii) number of opioid prescribers; opioid co-prescription with iv) benzodiazepines, v) muscle relaxants, and vi) gabapentinoids; vii) non-opioid pain prescription, viii) physical therapy, and ix) other pain treatment modality. Our outcome was incident OUD. Results: Having an opioid and gabapentin co-prescription or an opioid and benzodiazepine co-prescription was statistically significantly associated with a 16-46% increased risk of OUD. Opioid dose and duration also were significantly associated with increased risk of OUD. Physical therapy was significantly associated with an 11% decreased risk of OUD in the subgroup with chronic pain but no disability. Conclusions: Co-prescription of opioids with either gabapentin or benzodiazepines may substantially increase risk of OUD. More positively, physical therapy may be a relatively accessible and safe pain-management strategy.

In May 2023, the U.S. Food and Drug Administration (FDA) released guidance for industry on "Adjustment for Covariates in Randomized Clinical Trials for Drugs and Biological Products". Covariate adjustment is a statistical analysis method for improving precision and power in clinical trials by adjusting for pre-specified, prognostic baseline variables. Though recommended by the FDA and the European Medicines Agency (EMA), many trials do not exploit the available information in baseline variables or make use only of the baseline measurement of the outcome. This is likely (partly) due to the regulatory mandate to pre-specify baseline covariates for adjustment, leading to challenges in determining appropriate covariates and their functional forms. We will explore the potential of automated data-adaptive methods, such as machine learning algorithms, for covariate adjustment, addressing the challenge of pre-specification. Specifically, our approach allows the use of complex models or machine learning algorithms without compromising the interpretation or validity of the treatment effect estimate and its corresponding standard error, even in the presence of misspecified outcome working models. This contrasts the majority of competing works which assume correct model specification for the validity of standard errors. Our proposed estimators either necessitate ultra-sparsity in the outcome model (which can be relaxed by limiting the number of predictors in the model) or necessitate integration with sample splitting to enhance their performance. As such, we will arrive at simple estimators and standard errors for the marginal treatment effect in randomized clinical trials, which exploit data-adaptive outcome predictions based on prognostic baseline covariates, and have low (or no) bias in finite samples even when those predictions are themselves biased.

We demonstrate a comprehensive semiparametric approach to causal mediation analysis, addressing the complexities inherent in settings with longitudinal and continuous treatments, confounders, and mediators. Our methodology utilizes a nonparametric structural equation model and a cross-fitted sequential regression technique based on doubly robust pseudo-outcomes, yielding an efficient, asymptotically normal estimator without relying on restrictive parametric modeling assumptions. We are motivated by a recent scientific controversy regarding the effects of invasive mechanical ventilation (IMV) on the survival of COVID-19 patients, considering acute kidney injury (AKI) as a mediating factor. We highlight the possibility of "inconsistent mediation," in which the direct and indirect effects of the exposure operate in opposite directions. We discuss the significance of mediation analysis for scientific understanding and its potential utility in treatment decisions., Comment: revision including simulation study, more on interpretation of estimands

The presence of intermediate confounders, also called recanting witnesses, is a fundamental challenge to the investigation of causal mechanisms in mediation analysis, preventing the identification of natural path-specific effects. Proposed alternative parameters (such as randomizational interventional effects) are problematic because they can be non-null even when there is no mediation for any individual in the population; i.e., they are not an average of underlying individual-level mechanisms. In this paper we develop a novel method for mediation analysis in settings with intermediate confounding, with guarantees that the causal parameters are summaries of the individual-level mechanisms of interest. The method is based on recently proposed ideas that view causality as the transfer of information, and thus replace recanting witnesses by draws from their conditional distribution, what we call "recanting twins". We show that, in the absence of intermediate confounding, recanting twin effects recover natural path-specific effects. We present the assumptions required for identification of recanting twins effects under a standard structural causal model, as well as the assumptions under which the recanting twin identification formulas can be interpreted in the context of the recently proposed separable effects models. To estimate recanting-twin effects, we develop efficient semi-parametric estimators that allow the use of data driven methods in the estimation of the nuisance parameters. We present numerical studies of the methods using synthetic data, as well as an application to evaluate the role of new-onset anxiety and depressive disorder in explaining the relationship between gabapentin/pregabalin prescription and incident opioid use disorder among Medicaid beneficiaries with chronic pain.

Non-negative two-part outcomes are defined as outcomes with a density function that have a zero point mass but are otherwise positive. Examples, such as healthcare expenditure and hospital length of stay, are common in healthcare utilization research. Despite the practical relevance of non-negative two-part outcomes, very few methods exist to leverage knowledge of their semicontinuity to achieve improved performance in estimating causal effects. In this paper, we develop a nonparametric two-step targeted minimum-loss based estimator (denoted as hTMLE) for non-negative two-part outcomes. We present methods for a general class of interventions referred to as modified treatment policies, which can accommodate continuous, categorical, and binary exposures. The two-step TMLE uses a targeted estimate of the intensity component of the outcome to produce a targeted estimate of the binary component of the outcome that may improve finite sample efficiency. We demonstrate the efficiency gains achieved by the two-step TMLE with simulated examples and then apply it to a cohort of Medicaid beneficiaries to estimate the effect of chronic pain and physical disability on days' supply of opioids.

Recognizing the importance of real-world data (RWD) for regulatory purposes, the United States (US) Congress passed the 21st Century Cures Act1 mandating the development of Food and Drug Administration (FDA) guidance on regulatory use of real-world evidence. The Forum on the Integration of Observational and Randomized Data (FIORD) conducted a meeting bringing together various stakeholder groups to build consensus around best practices for the use of RWD to support regulatory science. Our companion paper describes in detail the context and discussion carried out in the meeting, which includes a recommendation to use a causal roadmap for complete pre-specification of study designs using RWD. This article discusses one step of the roadmap: the specification of a procedure for sensitivity analysis, defined as a procedure for testing the robustness of substantive conclusions to violations of assumptions made in the causal roadmap. We include a worked-out example of a sensitivity analysis from a RWD study on the effectiveness of Nifurtimox in treating Chagas disease, as well as an overview of various methods available for sensitivity analysis in causal inference, emphasizing practical considerations on their use for regulatory purposes.

Encompassing numerous nationwide, statewide, and institutional initiatives in the United States, provider profiling has evolved into a major health care undertaking with ubiquitous applications, profound implications, and high-stakes consequences. In line with such a significant profile, the literature has accumulated a number of developments dedicated to enhancing the statistical paradigm of provider profiling. Tackling wide-ranging profiling issues, these methods typically adjust for risk factors using linear predictors. While this approach is simple, it can be too restrictive to characterize complex and dynamic factor-outcome associations in certain contexts. One such example arises from evaluating dialysis facilities treating Medicare beneficiaries with end-stage renal disease. It is of primary interest to consider how the coronavirus disease (COVID-19) affected 30-day unplanned readmissions in 2020. The impact of COVID-19 on the risk of readmission varied dramatically across pandemic phases. To efficiently capture the variation while profiling facilities, we develop a generalized partially linear model (GPLM) that incorporates a neural network. Considering provider-level clustering, we implement the GPLM as a stratified sampling-based stochastic optimization algorithm that features accelerated convergence. Furthermore, an exact test is designed to identify under- and over-performing facilities, with an accompanying funnel plot to visualize profiles. The advantages of the proposed methods are demonstrated through simulation experiments and profiling dialysis facilities using 2020 Medicare claims from the United States Renal Data System., Comment: 8 figures, 6 tables

Increasing emphasis on the use of real-world evidence (RWE) to support clinical policy and regulatory decision-making has led to a proliferation of guidance, advice, and frameworks from regulatory agencies, academia, professional societies, and industry. A broad spectrum of studies use real-world data (RWD) to produce RWE, ranging from randomized controlled trials with outcomes assessed using RWD to fully observational studies. Yet many RWE study proposals lack sufficient detail to evaluate adequacy, and many analyses of RWD suffer from implausible assumptions, other methodological flaws, or inappropriate interpretations. The Causal Roadmap is an explicit, itemized, iterative process that guides investigators to pre-specify analytic study designs; it addresses a wide range of guidance within a single framework. By requiring transparent evaluation of causal assumptions and facilitating objective comparisons of design and analysis choices based on pre-specified criteria, the Roadmap can help investigators to evaluate the quality of evidence that a given study is likely to produce, specify a study to generate high-quality RWE, and communicate effectively with regulatory agencies and other stakeholders. This paper aims to disseminate and extend the Causal Roadmap framework for use by clinical and translational researchers, with companion papers demonstrating application of the Causal Roadmap for specific use cases., Comment: 51 pages, 2 figures, 4 tables

There are limited options to estimate the treatment effects of variables which are continuous and measured at multiple time points, particularly if the true dose-response curve should be estimated as closely as possible. However, these situations may be of relevance: in pharmacology, one may be interested in how outcomes of people living with -- and treated for -- HIV, such as viral failure, would vary for time-varying interventions such as different drug concentration trajectories. A challenge for doing causal inference with continuous interventions is that the positivity assumption is typically violated. To address positivity violations, we develop projection functions, which reweigh and redefine the estimand of interest based on functions of the conditional support for the respective interventions. With these functions, we obtain the desired dose-response curve in areas of enough support, and otherwise a meaningful estimand that does not require the positivity assumption. We develop $g$-computation type plug-in estimators for this case. Those are contrasted with g-computation estimators which are applied to continuous interventions without specifically addressing positivity violations, which we propose to be presented with diagnostics. The ideas are illustrated with longitudinal data from HIV positive children treated with an efavirenz-based regimen as part of the CHAPAS-3 trial, which enrolled children $<13$ years in Zambia/Uganda. Simulations show in which situations a standard g-computation approach is appropriate, and in which it leads to bias and how the proposed weighted estimation approach then recovers the alternative estimand of interest.

This tutorial discusses methodology for causal inference using longitudinal modified treatment policies. This method facilitates the mathematical formalization, identification, and estimation of many novel parameters, and mathematically generalizes many commonly used parameters, such as the average treatment effect. Longitudinal modified treatment policies apply to a wide variety of exposures, including binary, multivariate, and continuous, and can accommodate time-varying treatments and confounders, competing risks, loss-to-follow-up, as well as survival, binary, or continuous outcomes. Longitudinal modified treatment policies can be seen as an extension of static and dynamic interventions to involve the natural value of treatment, and, like dynamic interventions, can be used to define alternative estimands with a positivity assumption that is more likely to be satisfied than estimands corresponding to static interventions. This tutorial aims to illustrate several practical uses of the longitudinal modified treatment policy methodology, including describing different estimation strategies and their corresponding advantages and disadvantages. We provide numerous examples of types of research questions which can be answered using longitudinal modified treatment policies. We go into more depth with one of these examples--specifically, estimating the effect of delaying intubation on critically ill COVID-19 patients' mortality. We demonstrate the use of the open-source R package lmtp to estimate the effects, and we provide code on https://github.com/kathoffman/lmtp-tutorial.

We develop flexible, semiparametric estimators of the average treatment effect (ATE) transported to a new population ("target population") that offer potential efficiency gains. Transport may be of value when the ATE may differ across populations. We consider the setting where differences in the ATE are due to differences in the distribution of baseline covariates that modify the treatment effect ("effect modifiers"). First, we propose a collaborative one-step semiparametric estimator that can improve efficiency. This approach does not require researchers to have knowledge about which covariates are effect modifiers and which differ in distribution between the populations, but does require all covariates to be measured in the target population. Second, we propose two one-step semiparametric estimators that assume knowledge of which covariates are effect modifiers and which are both effect modifiers and differentially distributed between the populations. These estimators can be used even when not all covariates are observed in the target population; one requires that only effect modifiers are observed, and the other requires that only those modifiers that are also differentially distributed are observed. We use simulation to compare finite sample performance across our proposed estimators and an existing semiparametric estimator of the transported ATE, including in the presence of practical violations of the positivity assumption. Lastly, we apply our proposed estimators to a large-scale housing trial.

Convolutional neural networks (CNNs) allow for parameter sharing and translational equivariance by using convolutional kernels in their linear layers. By restricting these kernels to be SO(3)-steerable, CNNs can further improve parameter sharing. These rotationally-equivariant convolutional layers have several advantages over standard convolutional layers, including increased robustness to unseen poses, smaller network size, and improved sample efficiency. Despite this, most segmentation networks used in medical image analysis continue to rely on standard convolutional kernels. In this paper, we present a new family of segmentation networks that use equivariant voxel convolutions based on spherical harmonics. These networks are robust to data poses not seen during training, and do not require rotation-based data augmentation during training. In addition, we demonstrate improved segmentation performance in MRI brain tumor and healthy brain structure segmentation tasks, with enhanced robustness to reduced amounts of training data and improved parameter efficiency. Code to reproduce our results, and to implement the equivariant segmentation networks for other tasks is available at http://github.com/SCAN-NRAD/e3nn_Unet, Comment: Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) https://melba-journal.org/2024:010

Mediation analysis is appealing for its ability to improve understanding of the mechanistic drivers of causal effects, but real-world data complexities challenge its successful implementation, including: 1) the existence of post-exposure variables that also affect mediators and outcomes (thus, confounding the mediator-outcome relationship), that may also be 2) multivariate, and 3) the existence of multivariate mediators. Interventional direct and indirect effects (IDE/IIE) accommodate post-exposure variables that confound the mediator-outcome relationship, but currently, no estimator for IDE/IIE exists that allows for both multivariate mediators and multivariate post-exposure intermediate confounders. This, again, represents a significant limitation for real-world analyses. We address this gap by extending two recently developed nonparametric estimators -- one that estimates the IDE/IIE and another that estimates the IDE/IIE transported to a new, target population -- to allow for multivariate mediators and multivariate intermediate confounders simultaneously. We use simulation to examine finite sample performance, and apply these estimators to longitudinal data from the Moving to Opportunity trial. In the application, we walk through a strategy for separating indirect effects into mediator- or mediator-group-specific indirect effects, while appropriately accounting for other, possibly co-occurring intermediate variables.

There is a long-standing debate in the statistical, epidemiological and econometric fields as to whether nonparametric estimation that uses data-adaptive methods, like machine learning algorithms in model fitting, confer any meaningful advantage over simpler, parametric approaches in real-world, finite sample estimation of causal effects. We address the question: when trying to estimate the effect of a treatment on an outcome, across a universe of reasonable data distributions, how much does the choice of nonparametric vs.~parametric estimation matter? Instead of answering this question with simulations that reflect a few chosen data scenarios, we propose a novel approach evaluating performance across thousands of data-generating mechanisms drawn from non-parametric models with semi-informative priors. We call this approach a Universal Monte-Carlo Simulation. We compare performance of estimating the average treatment effect across two parametric estimators (a g-computation estimator that uses a parametric outcome model and an inverse probability of treatment weighted estimator) and two nonparametric estimators (Bayesian additive regression trees and a targeted minimum loss-based estimator that uses an ensemble of machine learning algorithms in model fitting). We summarize estimator performance in terms of bias, confidence interval coverage, and mean squared error. We find that the nonparametric estimators nearly always outperform the parametric estimators with the exception of having similar performance in terms of bias and similar-to-slightly-worse performance in terms of coverage under the smallest sample size of N=100.

This paper introduces a novel decomposition framework to explain heterogeneity in causal effects observed across different studies, considering both observational and randomized settings. We present a formal decomposition of between-study heterogeneity, identifying sources of variability in treatment effects across studies. The proposed methodology allows for robust estimation of causal parameters under various assumptions, addressing differences in pre-treatment covariate distributions, mediating variables, and the outcome mechanism. Our approach is validated through a simulation study and applied to data from the Moving to Opportunity (MTO) study, demonstrating its practical relevance. This work contributes to the broader understanding of causal inference in multi-study environments, with potential applications in evidence synthesis and policy-making.

Recent approaches to causal inference have focused on causal effects defined as contrasts between the distribution of counterfactual outcomes under hypothetical interventions on the nodes of a graphical model. In this article we develop theory for causal effects defined with respect to a different type of intervention, one which alters the information propagated through the edges of the graph. These information transfer interventions may be more useful than node interventions in settings in which causes are non-manipulable, for example when considering race or genetics as a causal agent. Furthermore, information transfer interventions allow us to define path-specific decompositions which are identified in the presence of treatment-induced mediator-outcome confounding, a practical problem whose general solution remains elusive. We prove that the proposed effects provide valid statistical tests of mechanisms, unlike popular methods based on randomized interventions on the mediator. We propose efficient non-parametric estimators for a covariance version of the proposed effects, using data-adaptive regression coupled with semi-parametric efficiency theory to address model misspecification bias while retaining $\sqrt{n}$-consistency and asymptotic normality. We illustrate the use of our methods in two examples using publicly available data.

Continuous treatments have posed a significant challenge for causal inference, both in the formulation and identification of scientifically meaningful effects and in their robust estimation. Traditionally, focus has been placed on techniques applicable to binary or categorical treatments with few levels, allowing for the application of propensity score-based methodology with relative ease. Efforts to accommodate continuous treatments introduced the generalized propensity score, yet estimators of this nuisance parameter commonly utilize parametric regression strategies that sharply limit the robustness and efficiency of inverse probability weighted estimators of causal effect parameters. We formulate and investigate a novel, flexible estimator of the generalized propensity score based on a nonparametric function estimator that provably converges at a suitably fast rate to the target functional so as to facilitate statistical inference. With this estimator, we demonstrate the construction of nonparametric inverse probability weighted estimators of a class of causal effect estimands tailored to continuous treatments. To ensure the asymptotic efficiency of our proposed estimators, we outline several non-restrictive selection procedures for utilizing a sieve estimation framework to undersmooth estimators of the generalized propensity score. We provide the first characterization of such inverse probability weighted estimators achieving the nonparametric efficiency bound in a setting with continuous treatments, demonstrating this in numerical experiments. We further evaluate the higher-order efficiency of our proposed estimators by deriving and numerically examining the second-order remainder of the corresponding efficient influence function in the nonparametric model. Open source software implementing our proposed estimation techniques, the haldensify R package, is briefly discussed.

Natural direct and indirect effects are mediational estimands that decompose the average treatment effect and describe how outcomes would be affected by contrasting levels of a treatment through changes induced in mediator values (in the case of the indirect effect) or not through induced changes in the mediator values (in the case of the direct effect). Natural direct and indirect effects are not generally point-identifiable in the presence of a treatment-induced confounder, however they may still be identified if one is willing to assume monotonicity between a treatment and the treatment-induced confounder. We argue that this assumption may be reasonable in the relatively common encouragement-design trial setting where intervention is randomized treatment assignment and the treatment-induced confounder is whether or not treatment was actually taken/adhered to. We develop efficiency theory for the natural direct and indirect effects under this monotonicity assumption, and use it to propose a nonparametric, multiply robust estimator. We demonstrate the finite sample properties of this estimator using a simulation study, and apply it to data from the Moving to Opportunity Study to estimate the natural direct and indirect effects of being randomly assigned to receive a Section 8 housing voucher -- the most common form of federal housing assistance -- on risk developing any mood or externalizing disorder among adolescent boys, possibly operating through various school and community characteristics.

Interventional effects have been proposed as a solution to the unidentifiability of natural (in)direct effects under mediator-outcome confounders affected by the exposure. Such confounders are an intrinsic characteristic of studies with time-varying exposures and mediators, yet the generalization of the interventional effect framework to the time-varying case has received little attention in the literature. We present an identification result for interventional effects in a general longitudinal data structure that allows flexibility in the specification of treatment-outcome, treatment-mediator, and mediator-outcome relationships. Identification is achieved under the standard no-unmeasured-confounders and positivity assumptions. We also present a theoretical and computational study of the properties of the identifying functional based on the efficient influence function (EIF). We use the EIF to propose a sequential regression estimation algorithm that yields doubly robust, $\sqrt{n}$-consistent, asymptotically Gaussian, and efficient estimators under slow convergence rates for the regression algorithms used. This allows the use of flexible machine learning for regression while permitting uncertainty quantification through confidence intervals and p-values. A free and open source \texttt{R} package implementing our proposed estimators is made available on GitHub. We apply the proposed estimator to an application from a comparative effectiveness trial of two medications for opioid use disorder. In the application, we estimate the extent to which differences between the two treatments' on subsequent risk of opioid use is mediated by craving symptoms.

Longitudinal modified treatment policies (LMTP) have been recently developed as a novel method to define and estimate causal parameters that depend on the natural value of treatment. LMTPs represent an important advancement in causal inference for longitudinal studies as they allow the non-parametric definition and estimation of the joint effect of multiple categorical, numerical, or continuous exposures measured at several time points. We extend the LMTP methodology to problems in which the outcome is a time-to-event variable subject to right-censoring and competing risks. We present identification results and non-parametric locally efficient estimators that use flexible data-adaptive regression techniques to alleviate model misspecification bias, while retaining important asymptotic properties such as $\sqrt{n}$-consistency. We present an application to the estimation of the effect of the time-to-intubation on acute kidney injury amongst COVID-19 hospitalized patients, where death by other causes is taken to be the competing event.

Mediation analysis is a strategy for understanding the mechanisms by which treatments or interventions affect later outcomes. Mediation analysis is frequently applied in randomized trial settings, but typically assumes: a) that randomized assignment is the exposure of interest as opposed to actual take-up of the intervention, and b) no unobserved confounding of the mediator-outcome relationship. In contrast to the rich literature on instrumental variable (IV) methods to estimate a total effect of a non-randomized exposure, there has been almost no research into using IV as an identification strategy in the presence of both exposure-outcome and mediator-outcome unobserved confounding. In response, we define and identify novel estimands -- complier interventional direct and indirect effects (i.e., IV mediational effects) in two scenarios: 1) with a single IV for the exposure, and 2) with two IVs, one for the exposure and another for the mediator, that may be related. We propose nonparametric, robust, efficient estimators, and apply them to a housing voucher experiment.

The rapid finding of effective therapeutics requires the efficient use of available resources in clinical trials. The use of covariate adjustment can yield statistical estimates with improved precision, resulting in a reduction in the number of participants required to draw futility or efficacy conclusions. We focus on time-to-event and ordinal outcomes. A key question for covariate adjustment in randomized studies is how to fit a model relating the outcome and the baseline covariates to maximize precision. We present a novel theoretical result establishing conditions for asymptotic normality of a variety of covariate-adjusted estimators that rely on machine learning (e.g., l1-regularization, Random Forests, XGBoost, and Multivariate Adaptive Regression Splines), under the assumption that outcome data is missing completely at random. We further present a consistent estimator of the asymptotic variance. Importantly, the conditions do not require the machine learning methods to converge to the true outcome distribution conditional on baseline variables, as long as they converge to some (possibly incorrect) limit. We conducted a simulation study to evaluate the performance of the aforementioned prediction methods in COVID-19 trials using longitudinal data from over 1,500 patients hospitalized with COVID-19 at Weill Cornell Medicine New York Presbyterian Hospital. We found that using l1-regularization led to estimators and corresponding hypothesis tests that control type 1 error and are more precise than an unadjusted estimator across all sample sizes tested. We also show that when covariates are not prognostic of the outcome, l1-regularization remains as precise as the unadjusted estimator, even at small sample sizes (n = 100). We give an R package adjrct that performs model-robust covariate adjustment for ordinal and time-to-event outcomes.

Violations of the positivity assumption (also called the common support condition) challenge health policy research, and can result in significant bias, large variance, and invalid inference. We define positivity in the single- and multiple-timepoint (i.e., longitudinal) health policy evaluation setting, and discuss real-world threats to positivity. We show empirical evidence of the practical positivity violations that can result when attempting to estimate effects of health policies (in this case, Naloxone Access Laws). In such scenarios, an alternative is to estimate the effect of a shift in law enactment (e.g., the effect if enactment had been delayed by some number of years). Such an effect corresponds to what is called a modified treatment policy, and dramatically weakens the required positivity assumption, thereby offering a means to estimate policy effects even in scenarios with serious positivity problems. We apply the approach to define and estimate longitudinal effects of Naloxone Access Laws on opioid overdose rates.

Combating the SARS-CoV2 pandemic will require the fast development of effective preventive vaccines. Regulatory agencies may open accelerated approval pathways for vaccines if an immunological marker can be established as a mediator of a vaccine's protection. A rich source of information for identifying such correlates are large-scale efficacy trials of COVID-19 vaccines, where immune responses are measured subject to a case-cohort sampling design. We propose two approaches to estimation of mediation parameters in the context of case-cohort sampling designs. We establish the theoretical large-sample efficiency of our proposed estimators and evaluate them in a realistic simulation to understand whether they can be employed in the analysis of COVID-19 vaccine efficacy trials., Comment: 26 pages, 6 tables, 2 figures

There is a growing literature on finding so-called optimal treatment rules, which are rules by which to assign treatment to individuals based on an individual's characteristics, such that a desired outcome is maximized. A related goal entails identifying individuals who are predicted to have a harmful indirect effect (the effect of treatment on an outcome through mediators) even in the presence of an overall beneficial effect of the treatment on the outcome. In some cases, the likelihood of a harmful indirect effect may outweigh a likely beneficial overall effect, and would be reason to caution against treatment for indicated individuals. We build on both the current mediation and optimal treatment rule literature to propose a method of identifying a subgroup for which the treatment effect through the mediator is harmful. Our approach is nonparametric, incorporates post-treatment variables that may confound the mediator-outcome relationship, and does not make restrictions on the distribution of baseline covariates, mediating variables (considered jointly), or outcomes. We apply the proposed approach to identify a subgroup of boys in the Moving to Opportunity housing voucher experiment who are predicted to have harmful indirect effects, though the average total effect is beneficial.

Causal mediation analysis has historically been limited in two important ways: (i) a focus has traditionally been placed on binary treatments and static interventions, and (ii) direct and indirect effect decompositions have been pursued that are only identifiable in the absence of intermediate confounders affected by treatment. We present a theoretical study of an (in)direct effect decomposition of the population intervention effect, defined by stochastic interventions jointly applied to the treatment and mediators. In contrast to existing proposals, our causal effects can be evaluated regardless of whether a treatment is categorical or continuous and remain well-defined even in the presence of intermediate confounders affected by treatment. Our (in)direct effects are identifiable without a restrictive assumption on cross-world counterfactual independencies, allowing for substantive conclusions drawn from them to be validated in randomized controlled trials. Beyond the novel effects introduced, we provide a careful study of nonparametric efficiency theory relevant for the construction of flexible, multiply robust estimators of our (in)direct effects, while avoiding undue restrictions induced by assuming parametric models of nuisance parameter functionals. To complement our nonparametric estimation strategy, we introduce inferential techniques for constructing confidence intervals and hypothesis tests, and discuss open source software implementing the proposed methodology.

The same intervention can produce different effects in different sites. Transport mediation estimators can estimate the extent to which such differences can be explained by differences in compositional factors and the mechanisms by which mediating or intermediate variables are produced; however, they are limited to consider a single, binary mediator. We propose novel nonparametric estimators of transported stochastic (in)direct effects that consider multiple, high-dimensional mediators and intermediate variables. They are multiply robust, efficient, asymptotically normal, and can incorporate data-adaptive estimation of nuisance parameters. They can be applied to understand differences in treatment effects across sites and/or to predict treatment effects in a target site based on outcome data in source sites.

Most causal inference methods consider counterfactual variables under interventions that set the treatment deterministically. With continuous or multi-valued treatments or exposures, such counterfactuals may be of little practical interest because no feasible intervention can be implemented that would bring them about. Furthermore, violations to the positivity assumption, necessary for identification, are exacerbated with continuous and multi-valued treatments and deterministic interventions. In this paper we propose longitudinal modified treatment policies (LMTPs) as a non-parametric alternative. LMTPs can be designed to guarantee positivity, and yield effects of immediate practical relevance with an interpretation that is familiar to regular users of linear regression adjustment. We study the identification of the LMTP parameter, study properties of the statistical estimand such as the efficient influence function, and propose four different estimators. Two of our estimators are efficient, and one is sequentially doubly robust in the sense that it is consistent if, for each time point, either an outcome regression or a treatment mechanism is consistently estimated. We perform a simulation study to illustrate the properties of the estimators, and present the results of our motivating study on hypoxemia and mortality in Intensive Care Unit (ICU) patients. Software implementing our methods is provided in the form of the open source \texttt{R} package \texttt{lmtp} freely available on GitHub (\url{https://github.com/nt-williams/lmtp}).

Interventional effects for mediation analysis were proposed as a solution to the lack of identifiability of natural (in)direct effects in the presence of a mediator-outcome confounder affected by exposure. We present a theoretical and computational study of the properties of the interventional (in)direct effect estimands based on the efficient influence fucntion (EIF) in the non-parametric statistical model. We use the EIF to develop two asymptotically optimal, non-parametric estimators that leverage data-adaptive regression for estimation of the nuisance parameters: a one-step estimator and a targeted minimum loss estimator. A free and open source \texttt{R} package implementing our proposed estimators is made available on GitHub. We further present results establishing the conditions under which these estimators are consistent, multiply robust, $n^{1/2}$-consistent and efficient. We illustrate the finite-sample performance of the estimators and corroborate our theoretical results in a simulation study. We also demonstrate the use of the estimators in our motivating application to elucidate the mechanisms behind the unintended harmful effects that a housing intervention had on adolescent girls' risk behavior.

We introduce a novel Bayesian estimator for the class proportion in an unlabeled dataset, based on the targeted learning framework. Our procedure requires the specification of a prior (and outputs a posterior) only for the target of inference, instead of the prior (and posterior) on the full-data distribution employed by classical non-parametric Bayesian methods .When the scientific question can be characterized by a low-dimensional parameter functional, focus on such a prior and posterior distributions is more aligned with Bayesian subjectivism, compared to focus on entire data distributions. We prove a Bernstein-von Mises-type result for our proposed Bayesian procedure, which guarantees that the posterior distribution converges to the distribution of an efficient, asymptotically linear estimator. In particular, the posterior is Gaussian, doubly robust, and efficient in the limit, under the only assumption that certain nuisance parameters are estimated at slow rates. We perform numerical studies illustrating the frequentist properties of the method. We also illustrate their use in a motivating application to estimate the proportion of embolic strokes of undetermined source arising from occult cardiac sources or large-artery atherosclerotic lesions. Though we focus on the motivating example of the proportion of cases in an unlabeled dataset, the procedure is general and can be adapted to estimate any pathwise differentiable parameter in a non-parametric model.

Mediation analysis in causal inference has traditionally focused on binary exposures and deterministic interventions, and a decomposition of the average treatment effect in terms of direct and indirect effects. In this paper we present an analogous decomposition of the \textit{population intervention effect}, defined through stochastic interventions on the exposure. Population intervention effects provide a generalized framework in which a variety of interesting causal contrasts can be defined, including effects for continuous and categorical exposures. We show that identification of direct and indirect effects for the population intervention effect requires weaker assumptions than its average treatment effect counterpart, under the assumption of no mediator-outcome confounders affected by exposure. In particular, identification of direct effects is guaranteed in experiments that randomize the exposure and the mediator. We discuss various estimators of the direct and indirect effects, including substitution, re-weighted, and efficient estimators based on flexible regression techniques, allowing for multivariate mediators. Our efficient estimator is asymptotically linear under a condition requiring $n^{1/4}$-consistency of certain regression functions. We perform a simulation study in which we assess the finite-sample properties of our proposed estimators. We present the results of an illustrative study where we assess the effect of participation in a sports team on BMI among children, using mediators such as exercise habits, daily consumption of snacks, and overweight status.

Estimation of causal parameters from observational data requires complete confounder adjustment, as well as positivity of the propensity score for each treatment arm. There is often a trade-off between these two assumptions: confounding bias may be reduced through adjustment for a large number of pre-treatment covariates, but positivity is less likely in analyses with irrelevant predictors of treatment such as instrumental variables. Under empirical positivity violations, propensity score weights are highly variable, and doubly robust estimators suffer from high variance and large finite sample bias. To solve this problem, we introduce the $e$-score, which is defined through a dimension reduction for the propensity score. This dimension reduction is based on a result known as collaborative double robustness, which roughly states that a propensity score conditioning only on the bias of the outcome regression estimator is sufficient to attain double robustness. We propose methods to construct doubly robust estimators based on the $e$-score, and discuss their properties such as consistency, efficiency, and asymptotic distribution. This allows the construction of asymptotically valid Wald-type confidence intervals and hypothesis tests. We present an illustrative application on estimating the effect of smoking on bone mineral content in adolescent girls well as a synthetic data simulation illustrating the bias and variance reduction and asymptotic normality achieved by our proposed estimators.