Harvard University Biostatistics Working Paper SeriesCopyright (c) 2014 Harvard University All rights reserved.
http://biostats.bepress.com/harvardbiostat
Recent documents in Harvard University Biostatistics Working Paper Seriesen-usThu, 14 Aug 2014 01:37:10 PDT3600Instrumental Variable Estimation in a Survival Context
http://biostats.bepress.com/harvardbiostat/paper179
http://biostats.bepress.com/harvardbiostat/paper179Tue, 12 Aug 2014 07:33:27 PDTEric J. Tchetgen Tchetgen et al.Likelihood Based Estimation of Logistic Structural Nested Mean Models with an Instrumental Variable
http://biostats.bepress.com/harvardbiostat/paper178
http://biostats.bepress.com/harvardbiostat/paper178Mon, 04 Aug 2014 07:13:37 PDTRoland A. Matsouaka et al.A General Approach to Detect Gene (G)-environment (E) Additive Interaction Leveraging G-E Independence in Case-control Studies
http://biostats.bepress.com/harvardbiostat/paper177
http://biostats.bepress.com/harvardbiostat/paper177Wed, 30 Jul 2014 09:55:20 PDTEric Tchetgen Tchetgen et al.A Simple Regression-based Approach to Account for Survival Bias in Birth Outcomes Research
http://biostats.bepress.com/harvardbiostat/paper176
http://biostats.bepress.com/harvardbiostat/paper176Mon, 21 Jul 2014 06:49:06 PDTEric J. Tchetgen Tchetgen et al.A Note on the Control Function Approach with an Instrumental Variable and a Binary Outcome
http://biostats.bepress.com/harvardbiostat/paper175
http://biostats.bepress.com/harvardbiostat/paper175Mon, 21 Jul 2014 06:49:02 PDTEric Tchetgen TchetgenControl Function Assisted IPW Estimation with a Secondary Outcome in Case-Control Studies
http://biostats.bepress.com/harvardbiostat/paper174
http://biostats.bepress.com/harvardbiostat/paper174Wed, 16 Jul 2014 07:45:51 PDTTamar Sofer et al.Predicting the Future Subject's Outcome via an Optimal Stratification Procedure with Baseline Information
http://biostats.bepress.com/harvardbiostat/paper173
http://biostats.bepress.com/harvardbiostat/paper173Tue, 01 Jul 2014 07:17:06 PDTFlorence H. Yong et al.Adjustment for Mismeasured Exposure using Validation Data and Propensity Scores
http://biostats.bepress.com/harvardbiostat/paper172
http://biostats.bepress.com/harvardbiostat/paper172Tue, 27 May 2014 05:48:01 PDT
Propensity score methods are widely used to analyze observational studies in which patient characteristics might not be balanced by treatment group. These methods assume that exposure, or treatment assignment, is error-free, but in reality these variables can be subject to measurement error. This arises in the context of comparative effectiveness research, using observational administrative claims data in which accurate procedural codes are not always available. When using propensity score based methods, this error affects both the exposure variable directly, as well as the propensity score. We propose a two step maximum likelihood approach using validation data to adjust for the measurement error. First, we use a likelihood approach to estimate an adjusted propensity score. Using the adjusted propensity score, we then use a likelihood approach on the outcome model to adjust for measurement error in the exposure variable directly. In addition, we show the bias introduced when using error-prone treatment in the inverse probability weighting (IPW) estimator and propose an approach to eliminate this bias. Simulations show our proposed approaches reduce the bias and mean squared error (MSE) of the treatment effect estimator compared to using the error-prone treatment assignment.
]]>
Danielle Braun et al.Bounds to Evaluate the Pure/natural Direct Effect without Cross-world Counterfactual Independence
http://biostats.bepress.com/harvardbiostat/paper171
http://biostats.bepress.com/harvardbiostat/paper171Mon, 31 Mar 2014 10:52:13 PDTEric Tchetgen Tchetgen et al.A unification of mediation and interaction: a four-way decomposition
http://biostats.bepress.com/harvardbiostat/paper170
http://biostats.bepress.com/harvardbiostat/paper170Tue, 25 Mar 2014 12:13:39 PDT
It is shown that the overall effect of an exposure on an outcome, in the presence of a mediator with which the exposure may interact, can be decomposed into four components: (i) the effect of the exposure in the absence of the mediator, (ii) the interactive effect when the mediator is left to what it would be in the absence of exposure, (iii) a mediated interaction, and (iv) a pure mediated effect. These four components, respectively, correspond to the portion of the effect that is due to neither mediation nor interaction, to just interaction (but not mediation), to both mediation and interaction, and to just mediation (but not interaction). This four-way decomposition unites methods that attribute effects to interactions and methods that assess mediation. Certain combinations of these four components correspond to measures for mediation, while other combinations correspond to measures of interaction previously proposed in the literature. Prior decompositions in the literature are in essence special cases of this four-way decomposition. The four-way decomposition can be carried out using standard statistical models, and software is provided to estimate each of the four components. The four-way decomposition provides maximum insight into how much of an effect is mediated, how much is due to interaction, how much is due to both mediation and interaction together, and how much is due to neither.
]]>
Tyler J. VanderWeeleA Predictive Enrichment Procedure to Identify Potential Responders to a New Therapy for Randomized, Comparative, Controlled Clinical Studies
http://biostats.bepress.com/harvardbiostat/paper169
http://biostats.bepress.com/harvardbiostat/paper169Tue, 11 Mar 2014 07:48:15 PDTJunlong Li et al.Mediation Analysis with Time-Varying Exposures and Mediators
http://biostats.bepress.com/harvardbiostat/paper168
http://biostats.bepress.com/harvardbiostat/paper168Tue, 04 Mar 2014 10:38:34 PST
In this paper we consider mediation analysis when exposures and mediators vary over time. We give non-parametric identification results, discuss parametric implementation, and also provide a weighting approach to direct and indirect effects based on combining the results of two marginal structural models. We also discuss how our results give rise to a causal interpretation of the effect estimates produced from longitudinal structural equation models. When there are no time-varying confounders affected by prior exposure and mediator values, identification of direct and indirect effects is achieved by a longitudinal version of Pearl's mediation formula. When there are time-varying confounders affected by prior exposure and mediator, natural direct and indirect effects are not identified. We define a randomized interventional analogue of natural direct and indirect effects that are identified in this setting. The formula that identifies these effects we refer to as the "mediational g-formula." When there is no mediation, the mediational g-formula reduces to Robins' regular g-formula for longitudinal data. When there are no time-varying confouders affected by prior exposure and mediator values, then the mediational g-formula reduces to a longitudinal version of Pearl's mediation formula. However, the mediational g-formula itself can accomodate both mediation and time-varying confounders.
]]>
Tyler J. VanderWeele et al.Identification and Estimation of Survivor Average Causal Effects
http://biostats.bepress.com/harvardbiostat/paper167
http://biostats.bepress.com/harvardbiostat/paper167Thu, 14 Nov 2013 06:50:32 PSTEric J. Tchetgen TchetgenAlternative Identification and Inference for the Effect of Treatment on the Treated with an Instrumental Variable
http://biostats.bepress.com/harvardbiostat/paper166
http://biostats.bepress.com/harvardbiostat/paper166Thu, 14 Nov 2013 06:49:14 PSTEric J. Tchetgen Tchetgen et al.A General Instrumental Variable Framework for Regression Analysis with Outcome Missing Not at Random
http://biostats.bepress.com/harvardbiostat/paper165
http://biostats.bepress.com/harvardbiostat/paper165Thu, 14 Nov 2013 06:43:13 PSTEric J. Tchetgen Tchetgen et al.A unification of mediation and interaction
http://biostats.bepress.com/harvardbiostat/paper164
http://biostats.bepress.com/harvardbiostat/paper164Wed, 13 Nov 2013 05:37:33 PST
We show that the overall effect of an exposure on an outcome, in the presence of a mediator with which the exposure may interact, can be decomposed into four components: (i) the effect of the exposure in the absence of the mediator, (ii) the interactive effect when the mediator is left to what is would be in the absence of exposure, (iii) a mediated interaction and (iv) a pure mediated effect. These four components respectively correspond to the portion of the effect that is due to neither mediation nor interaction, to just interaction (but not mediation), to both mediation and interaction, and to just mediation (but not interaction). It is shown that this four-way decomposition unites methods that attribute effects to interactions and methods that assess mediation. Different combinations of these four components correspond to measures for mediation, while other combinations correspond to measures of interaction. The decomposition can be carried out using standard statistical models and software is provided to estimate each of the four components. The four-way decomposition provides the greatest insight into how much of an effect is mediated, how much is due to interaction, how much is due to both mediation and interaction together, and how much is due to neither.
]]>
Tyler J. VanderWeeleOn the causal interpretation of race in regressions adjusting for confounding and mediating variables
http://biostats.bepress.com/harvardbiostat/paper163
http://biostats.bepress.com/harvardbiostat/paper163Wed, 13 Nov 2013 05:37:32 PST
We consider different possible interpretations of the “effect of race” when regressions are run with race as an exposure variable, controlling also for various confounding and mediating variables. When adjustment is made for socioeconomic status early in a person's life, we discuss under what contexts the regression coefficients for race can be interpreted as corresponding to the extent to which a racial disparity would remain if various socioeconomic distributions early in life across racial groups could be equalized. When adjustment is also made for adult socioeconomic status, we note how the overall disparity can be decomposed into the portion that would be eliminated by equalizing adult socioeconomic status across racial groups and the portion of the disparity that would remain even if adult socioeconomic status across racial groups were equalized. We also discuss a stronger interpretation of the “effect of race” involving the joint effects of skin color, parental skin color, genetic background and cultural context when such variables are thought to be hypothetically manipulable and if adequate control for confounding were possible. We discuss some of the challenges with such an interpretation. Further discussion is given as to how the use of selected populations in examining racial disparities can additionally complicate the interpretation of the effects.
]]>
Tyler J. VanderWeele et al.Attributing effects to interactions
http://biostats.bepress.com/harvardbiostat/paper162
http://biostats.bepress.com/harvardbiostat/paper162Fri, 26 Jul 2013 06:41:56 PDT
A framework is presented which allows an investigator to estimate the portion of the effect of one exposure that is attributable to an interaction with a second exposure. We show that when the two exposures are independent, the total effect of one exposure can be decomposed into a conditional effect of that exposure and a component due to interaction. The decomposition applies on difference or ratio scales. We discuss how the components can be estimated using standard regression models, and how these components can be used to evaluate the proportion of the total effect of the primary exposure attributable to the interaction with the second exposure. In the setting in which one of the exposures affects the other, so that the two are no longer independent, alternative decompositions are discussed. The various decompositions are illustrated with an example in genetic epidemiology.
]]>
Tyler J. VanderWeele et al.Sample Size Considerations in the Design of Cluster Randomized Trials of Combination HIV Prevention
http://biostats.bepress.com/harvardbiostat/paper161
http://biostats.bepress.com/harvardbiostat/paper161Wed, 10 Jul 2013 11:03:03 PDTRui Wang et al.Phylogenetic Linkage Among HIV-infected Village Residents in Botswana: Estimation of Clustering Rates in the Presence of Missing Data
http://biostats.bepress.com/harvardbiostat/paper160
http://biostats.bepress.com/harvardbiostat/paper160Thu, 20 Jun 2013 06:06:59 PDTNicole Bohme Carnegie et al.