Multiple robust estimation of marginal structural mean models for unconstrained outcomes.

We consider estimation, from longitudinal observational data, of the parameters of marginal structural mean models for unconstrained outcomes. Current proposals include inverse probability of treatment weighted and double robust (DR) estimators. A difficulty with DR estimation is that it requires postulating a sequence of models, one for the each mean of the counterfactual outcome given covariate and treatment history up to each exposure time point. Most natural models for such means are often incompatible. Robins et al., (2000b) proposed a parameterization of the likelihood which implies compatible parametric models for such means. Their parameterization has not been exploited to construct DR estimators and one goal of this article is to fill this gap. More importantly, exploiting this parameterization we propose a multiple robust (MR) estimator that confers even more protection against model misspecification than DR estimators. Our methods are easy to implement as they are based on the iterative fit of a sequence of weighted regressions.

Initiation of erythropoiesis-stimulating agents and outcomes: a nationwide observational cohort study in anaemic chronic kidney disease patients.

BACKGROUND: In 2012, new clinical guidelines were introduced for use of erythropoiesis-stimulating agents (ESA) in chronic kidney disease (CKD) patients, recommending lower haemoglobin (Hb) target levels and thresholds for ESA initiation. These changes resulted in lower blood levels in these patients. However, there is limited evidence on just when ESA should be initiated and the safety of a low Hb initiation policy. METHODS: In this observational inception cohort study, Swedish, nephology-referred, ESA-naïve CKD patients (n = 6348) were enrolled when their Hb dropped below 12.0 g/L, and they were followed for mortality and cardiovascular events. Four different ESA treatments were evaluated applying dynamic marginal structural models: (i) begin ESA immediately, (ii) begin ESA when Hb <11.0 g/dL, (iii) begin ESA when Hb <10.0 g/dL and (iv) never begin ESA in comparison with 'current practice' [the observed (factual) survival of the entire study cohort]. The adjusted 3-year survival following ESA begun over a range of Hb (from <9.0 to 12.0 g/dL) was evaluated, after adjustment for covariates at baseline and during follow-up. RESULTS: Overall, 36% were treated with ESA. Mortality during follow-up was 33.4% of the ESA-treated and 27.9% of the non-treated subjects. The adjusted 3-year survival associated with ESA initiation improved for subjects with initial Hb <9.0 to 11 g/dL and then decreased again for those with Hb above 11.5 g/dL. Initiating ESA at Hb <11.0 g/dL and <10.0 g/dL was associated with improved survival compared with 'current practice' [hazard ratio (HR) 0.83; 95% confidence interval (CI) 0.79-0.89 and 0.90; 95% CI 0.86-0.94, respectively] and did not increase the risk of a cardiovascular event (HR 0.93; 95% CI 0.87-1.00). CONCLUSION: In non-dialysis patients with CKD, ESA initiation at Hb < 10.0-11.0 g/dL is associated with improved survival in patients otherwise treated according to guidelines.

Doubly robust estimation of the local average treatment effect curve.

We consider estimation of the causal effect of a binary treatment on an outcome, conditionally on covariates, from observational studies or natural experiments in which there is a binary instrument for treatment. We describe a doubly robust, locally efficient estimator of the parameters indexing a model for the local average treatment effect conditionally on covariates V when randomization of the instrument is only true conditionally on a high dimensional vector of covariates X, possibly bigger than V. We discuss the surprising result that inference is identical to inference for the parameters of a model for an additive treatment effect on the treated conditionally on V that assumes no treatment-instrument interaction. We illustrate our methods with the estimation of the local average effect of participating in 401(k) retirement programs on savings by using data from the US Census Bureau's 1991 Survey of Income and Program Participation.

Optimal auxiliary-covariate-based two-phase sampling design for semiparametric efficient estimation of a mean or mean difference, with application to clinical trials.

To address the objective in a clinical trial to estimate the mean or mean difference of an expensive endpoint Y, one approach employs a two-phase sampling design, wherein inexpensive auxiliary variables W predictive of Y are measured in everyone, Y is measured in a random sample, and the semiparametric efficient estimator is applied. This approach is made efficient by specifying the phase two selection probabilities as optimal functions of the auxiliary variables and measurement costs. While this approach is familiar to survey samplers, it apparently has seldom been used in clinical trials, and several novel results practicable for clinical trials are developed. We perform simulations to identify settings where the optimal approach significantly improves efficiency compared to approaches in current practice. We provide proofs and R code. The optimality results are developed to design an HIV vaccine trial, with objective to compare the mean 'importance-weighted' breadth (Y) of the T-cell response between randomized vaccine groups. The trial collects an auxiliary response (W) highly predictive of Y and measures Y in the optimal subset. We show that the optimal design-estimation approach can confer anywhere between absent and large efficiency gain (up to 24 % in the examples) compared to the approach with the same efficient estimator but simple random sampling, where greater variability in the cost-standardized conditional variance of Y given W yields greater efficiency gains. Accurate estimation of E[Y | W] is important for realizing the efficiency gain, which is aided by an ample phase two sample and by using a robust fitting method.

Evaluation of Viable Dynamic Treatment Regimes in a Sequentially Randomized Trial of Advanced Prostate Cancer.

We present new statistical analyses of data arising from a clinical trial designed to compare two-stage dynamic treatment regimes (DTRs) for advanced prostate cancer. The trial protocol mandated that patients were to be initially randomized among four chemotherapies, and that those who responded poorly were to be rerandomized to one of the remaining candidate therapies. The primary aim was to compare the DTRs' overall success rates, with success defined by the occurrence of successful responses in each of two consecutive courses of the patient's therapy. Of the one hundred and fifty study participants, forty seven did not complete their therapy per the algorithm. However, thirty five of them did so for reasons that precluded further chemotherapy; i.e. toxicity and/or progressive disease. Consequently, rather than comparing the overall success rates of the DTRs in the unrealistic event that these patients had remained on their assigned chemotherapies, we conducted an analysis that compared viable switch rules defined by the per-protocol rules but with the additional provision that patients who developed toxicity or progressive disease switch to a non-prespecified therapeutic or palliative strategy. This modification involved consideration of bivariate per-course outcomes encoding both efficacy and toxicity. We used numerical scores elicited from the trial's Principal Investigator to quantify the clinical desirability of each bivariate per-course outcome, and defined one endpoint as their average over all courses of treatment. Two other simpler sets of scores as well as log survival time also were used as endpoints. Estimation of each DTR-specific mean score was conducted using inverse probability weighted methods that assumed that missingness in the twelve remaining drop-outs was informative but explainable in that it only depended on past recorded data. We conducted additional worst-best case analyses to evaluate sensitivity of our findings to extreme departures from the explainable drop-out assumption.

Improved double-robust estimation in missing data and causal inference models.

Recently proposed double-robust estimators for a population mean from incomplete data and for a finite number of counterfactual means can have much higher efficiency than the usual double-robust estimators under misspecification of the outcome model. In this paper, we derive a new class of double-robust estimators for the parameters of regression models with incomplete cross-sectional or longitudinal data, and of marginal structural mean models for cross-sectional data with similar efficiency properties. Unlike the recent proposals, our estimators solve outcome regression estimating equations. In a simulation study, the new estimator shows improvements in variance relative to the standard double-robust estimator that are in agreement with those suggested by asymptotic theory.

Double-robust estimation of an exposure-outcome odds ratio adjusting for confounding in cohort and case-control studies.

Modern epidemiologic studies often aim to evaluate the causal effect of a point exposure on the risk of a disease from cohort or case-control observational data. Because confounding bias is of serious concern in such non-experimental studies, investigators routinely adjust for a large number of potential confounders in a logistic regression analysis of the effect of exposure on disease outcome. Unfortunately, when confounders are not correctly modeled, standard logistic regression is likely biased in its estimate of the effect of exposure, potentially leading to erroneous conclusions. We partially resolve this serious limitation of standard logistic regression analysis with a new iterative approach that we call ProRetroSpective estimation, which carefully combines standard logistic regression with a logistic regression analysis in which exposure is the dependent variable and the outcome and confounders are the independent variables. As a result, we obtain a correct estimate of the exposure-outcome odds ratio, if either thestandard logistic regression of the outcome given exposure and confounding factors is correct, or the regression model of exposure given the outcome and confounding factors is correct but not necessarily both, that is, it is double-robust. In fact, it also has certain advantadgeous efficiency properties. The approach is general in that it applies to both cohort and case-control studies whether the design of the study is matched or unmatched on a subset of covariates. Finally, an application illustrates the methods using data from the National Cancer Institute's Black/White Cancer Survival Study.

Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, Part II: proofs of results.

In this companion article to "Dynamic Regime Marginal Structural Mean Models for Estimation of Optimal Dynamic Treatment Regimes, Part I: Main Content" [Orellana, Rotnitzky and Robins (2010), IJB, Vol. 6, Iss. 2, Art. 7] we present (i) proofs of the claims in that paper, (ii) a proposal for the computation of a confidence set for the optimal index when this lies in a finite set, and (iii) an example to aid the interpretation of the positivity assumption.

Dynamic regime marginal structural mean models for estimation of optimal dynamic treatment regimes, Part I: main content.

Dynamic treatment regimes are set rules for sequential decision making based on patient covariate history. Observational studies are well suited for the investigation of the effects of dynamic treatment regimes because of the variability in treatment decisions found in them. This variability exists because different physicians make different decisions in the face of similar patient histories. In this article we describe an approach to estimate the optimal dynamic treatment regime among a set of enforceable regimes. This set is comprised by regimes defined by simple rules based on a subset of past information. The regimes in the set are indexed by a Euclidean vector. The optimal regime is the one that maximizes the expected counterfactual utility over all regimes in the set. We discuss assumptions under which it is possible to identify the optimal regime from observational longitudinal data. Murphy et al. (2001) developed efficient augmented inverse probability weighted estimators of the expected utility of one fixed regime. Our methods are based on an extension of the marginal structural mean model of Robins (1998, 1999) which incorporate the estimation ideas of Murphy et al. (2001). Our models, which we call dynamic regime marginal structural mean models, are specially suitable for estimating the optimal treatment regime in a moderately small class of enforceable regimes of interest. We consider both parametric and semiparametric dynamic regime marginal structural models. We discuss locally efficient, double-robust estimation of the model parameters and of the index of the optimal treatment regime in the set. In a companion paper in this issue of the journal we provide proofs of the main results.

Nonparametric Regression With Missing Outcomes Using Weighted Kernel Estimating Equations.

We consider nonparametric regression of a scalar outcome on a covariate when the outcome is missing at random (MAR) given the covariate and other observed auxiliary variables. We propose a class of augmented inverse probability weighted (AIPW) kernel estimating equations for nonparametric regression under MAR. We show that AIPW kernel estimators are consistent when the probability that the outcome is observed, that is, the selection probability, is either known by design or estimated under a correctly specified model. In addition, we show that a specific AIPW kernel estimator in our class that employs the fitted values from a model for the conditional mean of the outcome given covariates and auxiliaries is double-robust, that is, it remains consistent if this model is correctly specified even if the selection probabilities are modeled or specified incorrectly. Furthermore, when both models happen to be right, this double-robust estimator attains the smallest possible asymptotic variance of all AIPW kernel estimators and maximally extracts the information in the auxiliary variables. We also describe a simple correction to the AIPW kernel estimating equations that while preserving double-robustness it ensures efficiency improvement over nonaugmented IPW estimation when the selection model is correctly specified regardless of the validity of the second model used in the augmentation term. We perform simulations to evaluate the finite sample performance of the proposed estimators, and apply the methods to the analysis of the AIDS Costs and Services Utilization Survey data. Technical proofs are available online.

A note on overadjustment in inverse probability weighted estimation.

Standardized means, commonly used in observational studies in epidemiology to adjust for potential confounders, are equal to inverse probability weighted means with inverse weights equal to the empirical propensity scores. More refined standardization corresponds with empirical propensity scores computed under more flexible models. Unnecessary standardization induces efficiency loss. However, according to the theory of inverse probability weighted estimation, propensity scores estimated under more flexible models induce improvement in the precision of inverse probability weighted means. This apparent contradiction is clarified by explicitly stating the assumptions under which the improvement in precision is attained.

On doubly robust estimation in a semiparametric odds ratio model.

We consider the doubly robust estimation of the parameters in a semiparametric conditional odds ratio model. Our estimators are consistent and asymptotically normal in a union model that assumes either of two variation independent baseline functions is correctly modelled but not necessarily both. Furthermore, when either outcome has finite support, our estimators are semiparametric efficient in the union model at the intersection submodel where both nuisance functions models are correct. For general outcomes, we obtain doubly robust estimators that are nearly efficient at the intersection submodel. Our methods are easy to implement as they do not require the use of the alternating conditional expectations algorithm of Chen (2007).

Estimation of the ROC curve under verification bias.

The ROC (receiver operating characteristic) curve is the most commonly used statistical tool for describing the discriminatory accuracy of a diagnostic test. Classical estimation of the ROC curve relies on data from a simple random sample from the target population. In practice, estimation is often complicated due to not all subjects undergoing a definitive assessment of disease status (verification). Estimation of the ROC curve based on data only from subjects with verified disease status may be badly biased. In this work we investigate the properties of the doubly robust (DR) method for estimating the ROC curve under verification bias originally developed by Rotnitzky, Faraggi and Schisterman (2006) for estimating the area under the ROC curve. The DR method can be applied for continuous scaled tests and allows for a non-ignorable process of selection to verification. We develop the estimator's asymptotic distribution and examine its finite sample properties via a simulation study. We exemplify the DR procedure for estimation of ROC curves with data collected on patients undergoing electron beam computer tomography, a diagnostic test for calcification of the arteries.

Analysis of quality-of-life adjusted failure time data in the presence of competing, possibly informative, censoring mechanisms.

We derive estimators of the mean of a function of a quality-of-life adjusted failure time, in the presence of competing right censoring mechanisms. Our approach allows for the possibility that some or all of the competing censoring mechanisms are associated with the endpoint, even after adjustment for recorded prognostic factors, with the degree of residual association possibly different for distinct censoring processes. Our methods generalize from a single to many censoring processes and from ignorable to non-ignorable censoring processes.

Estimation of the disease-specific diagnostic marker distribution under verification bias.

We consider the estimation of the parameters indexing a parametric model for the conditional distribution of a diagnostic marker given covariates and disease status. Such models are useful for the evaluation of whether and to what extent a marker's ability to accurately detect or discard disease depends on patient characteristics. A frequent problem that complicates the estimation of the model parameters is that estimation must be conducted from observational studies. Often, in such studies not all patients undergo the gold standard assessment of disease. Furthermore, the decision as to whether a patient undergoes verification is not controlled by study design. In such scenarios, maximum likelihood estimators based on subjects with observed disease status are generally biased. In this paper, we propose estimators for the model parameters that adjust for selection to verification that may depend on measured patient characteristics and additonally adjust for an assumed degree of residual association. Such estimators may be used as part of a sensitivity analysis for plausible degrees of residual association. We describe a doubly robust estimator that has the attractive feature of being consistent if either a model for the probability of selection to verification or a model for the probability of disease among the verified subjects (but not necessarily both) is correct.

Estimation and extrapolation of optimal treatment and testing strategies.

We review recent developments in the estimation of an optimal treatment strategy or regime from longitudinal data collected in an observational study. We also propose novel methods for using the data obtained from an observational database in one health-care system to determine the optimal treatment regime for biologically similar subjects in a second health-care system when, for cultural, logistical, or financial reasons, the two health-care systems differ (and will continue to differ) in the frequency of, and reasons for, both laboratory tests and physician visits. Finally, we propose a novel method for estimating the optimal timing of expensive and/or painful diagnostic or prognostic tests. Diagnostic or prognostic tests are only useful in so far as they help a physician to determine the optimal dosing strategy, by providing information on both the current health state and the prognosis of a patient because, in contrast to drug therapies, these tests have no direct causal effect on disease progression. Our new method explicitly incorporates this no direct effect restriction.

Effect modification by time-varying covariates.

Marginal structural models (MSMs) allow estimation of effect modification by baseline covariates, but they are less useful for estimating effect modification by evolving time-varying covariates. Rather, structural nested models (SNMs) were specifically designed to estimate effect modification by time-varying covariates. In their paper, Petersen et al. (Am J Epidemiol 2007;166:985-993) describe history-adjusted MSMs as a generalized form of MSM and argue that history-adjusted MSMs allow a researcher to easily estimate effect modification by time-varying covariates. However, history-adjusted MSMs can result in logically incompatible parameter estimates and hence in contradictory substantive conclusions. Here the authors propose a more restrictive definition of history-adjusted MSMs than the one provided by Petersen et al. and compare the advantages and disadvantages of using history-adjusted MSMs, as opposed to SNMs, to examine effect modification by time-dependent covariates.