Instrumental variable with competing risk model

Paravati

et al. 2018

Statistics in Medicine

Competing risk analysis considers event times due to multiple causes or of more than one event types. Commonly used regression models for such data include (1) cause-specific hazards model, which focuses on modeling one type of event while acknowledging other event types simultaneously, and (2) subdistribution hazards model, which links the covariate effects directly to the cumulative incidence function. Their use in the presence of high-dimensional predictors are largely unexplored. Motivated by an analysis using the linked SEER-Medicare database for the purposes of predicting cancer versus noncancer mortality for patients with prostate cancer, we study the accuracy of prediction and variable selection of existing machine learning methods under both models using extensive simulation experiments, including different approaches to choosing penalty parameters in each method. We then apply the optimal approaches to the analysis of the SEER-Medicare data.

Section: Discussionmentioning

confidence: 99%

High‐dimensional variable selection and prediction under competing risks with application to SEER‐Medicare linked data

Paravati

et al. 2018

Statistics in Medicine

Proceedings of the Annual Hawaii International Conference on System Sciences

“…However, IVs and matching approaches are not feasible for most of traditional survival models for competing events including the models used in this study. A few recent studies discuss about the application of the two approaches [55,56], and we may address the econometric concern following their suggestions.…”

Section: Discussionmentioning

confidence: 99%

Repairing the Digital Divide Can Increase the Service Divide: The Effects of Patient Portals on Kidney Allocation

Kim

Ayvaci

Raghunathan

et al. 2019

The severe shortage of organs combined with increasing demand for them characterize the outcomes for the kidney allocation process. Despite the efforts to improve the allocation of kidneys, notable inefficiencies and unequal access to available organs persist across patient populations. The goal of this study is to examine (i) whether the adoption of a patient-oriented information technology (IT), namely the patient portals, can mitigate inefficiencies in the allocation of these scarce resources (kidneys) in general; (ii) whether the adoption of patient portals magnify or alleviate the disparity issues around access to transplants. Using a rich dataset of all the kidney transplant records in the U.S. from 2011 to 2014, we show that the likelihood that the patient receives deceased donor transplant at a given point in time increases in the presence of patient portals. However, the varying impact of IT across sub-populations may indicate that the efforts to bridge the digital divide may benefit some groups of patients at the expense of other groups, leading to further disparities.

“…Or we can obtain a bound for the causal effect with weaker assumptions [29][30][31]. Another way to handle noncompliance is to consider compliance state as a mediator and use an instrumental variable approach to handle potential un-measured confounders between compliance and outcome [32][33][34][35][36]. Another way is to assume sequential ignorability.…”

Section: Non-compliancementioning

confidence: 99%

Causal inference in randomized clinical trials

Zheng

Dai

Gale

et al. 2019

Bone Marrow Transplant

Self Cite

When we do a clinical trial in which we randomize for one variable, say adding pretransplant anti-thymocyte globulin (ATG), and we see a benefit, say less graft-versus-host disease (GvHD), most people assume receiving ATG caused the benefit. This reasoning, termed causal inference, is common but wrong. Reasons why is described in the accompanying typescript. What we observe is an association or correlation between ATG and less GvHD, not necessarily the cause. This incorrect reasoning is referred to as the association-causation fallacy. A good example is the correlation between US per capita cheese consumption and deaths by strangulation from bedsheets with a Pearson correlation coefficient of 0.95 (see below). This and other problems of human cognition can be found in Thinking, Fast and Slow by Daniel Kahneman. How can we reconcile this discordance between the goal of the clinical trialist who wants to know why GvHD is decreased and the rigor of the statistician? In the following typescript Zheng and colleagues describe the difference between causality and association. They describe statistical methods by which we can plausibly infer causality to results of a randomized clinical trial. We hope this typescript and others will prompt a dialogue between readers and statisticians interested in analyses of data from clinical trials of haematopoietic cell transplants. We welcome comments at #BMTStats.