Article
Economics
Michela Bia, Alessandra Mattei, Andrea Mercatanti
Summary: This study addresses important statistical issues in evaluating the effects of training programs for unemployed individuals by proposing an extended framework to simultaneously tackle truncation and attrition issues. The framework utilizes principal stratification to define causal effects and handle missing data, in addition to applying a Bayesian approach for inference. The findings from evaluating the causal effects of foreign language training programs in Luxembourg may provide insights for improving future employment programs.
JOURNAL OF BUSINESS & ECONOMIC STATISTICS
(2022)
Article
Statistics & Probability
Ying Huang, Yingying Zhuang, Peter Gilbert
Summary: This article addresses the evaluation of postrandomization immune response biomarkers as principal surrogate endpoints of a vaccine's protective effect, based on data from randomized vaccine trials. The study proposes a new sensitivity analysis framework for principal surrogate evaluation allowing for early vaccine efficacy, and uses this framework to assess the surrogacy of postrandomization neutralization titer in a dengue application.
ANNALS OF APPLIED STATISTICS
(2022)
Article
Mathematics, Interdisciplinary Applications
Michael R. Elliott
Summary: Surrogate markers are used in clinical trials for evaluating treatment effectiveness when obtaining final outcomes is time-consuming or costly. This review focuses on approaches using causal inference paradigm to define surrogate marker quality and efforts to evaluate the risk of surrogate paradox. Recent work in robust surrogate marker estimation is also discussed, with suggestions for future research.
ANNUAL REVIEW OF STATISTICS AND ITS APPLICATION
(2023)
Article
Mathematical & Computational Biology
Booil Jo
Summary: This study explores the use of parametric mixture modeling in principal stratification modeling and proposes a method to assess the quality of estimation results. By employing the flexible moving exclusion restriction assumption, the study demonstrates the potential of using parametric mixture modeling as a tool for causal inference.
STATISTICS IN MEDICINE
(2022)
Article
Statistics & Probability
Zhichao Jiang, Peng Ding
Summary: Principal stratification is a framework to deal with posttreatment intermediate variables between treatment and outcomes, but the principal causal effects are not identifiable without additional assumptions due to unobservability. Previous studies have shown that using auxiliary variables can improve the inference of principal causal effects.
STATISTICAL SCIENCE
(2021)
Article
Mathematical & Computational Biology
Yanxun Xu, Daniel Scharfstein, Peter Muller, Michael Daniels
Summary: We propose a Bayesian nonparametric approach to evaluate the causal effect of treatment in a randomized trial with semi-competing risks. Our method introduces a novel estimand based on principal stratification and utilizes identification assumptions indexed by a sensitivity parameter.
Article
Mathematical & Computational Biology
Wei Wang, Guangyu Tong, Shashivadan P. Hirani, Stanton P. Newman, Scott D. Halpern, Dylan S. Small, Fan Li, Michael O. Harhay
Summary: In many medical studies, missing data due to death or dropout pose challenges in estimating treatment effects. This study proposes a mixed model approach and algorithm to estimate the survivor average causal effect (SACE) in cluster-randomized trials, filling the gap in existing methods.
STATISTICS IN MEDICINE
(2023)
Article
Cardiac & Cardiovascular Systems
An Wang, Mengqi Chen, Qi Zhuang, Lihua Guan, Weiping Xie, Lan Wang, Wei Huang, Zhaozhong Cheng, Shiyong Yu, Hongmei Zhou, Jieyan Shen
Summary: This prospective multicenter study evaluated the efficacy of domestic ambrisentan in Chinese PAH patients and observed risk improvement and time to clinical improvement. The results showed that ambrisentan significantly improved exercise capacity and risk status in the patients. Time to clinical improvement can be considered as an appropriate composite surrogate endpoint for PAH medication trials.
FRONTIERS IN CARDIOVASCULAR MEDICINE
(2023)
Article
Statistics & Probability
Zhichao Jiang, Shu Yang, Peng Ding
Summary: Causal inference not only focuses on the average effect of treatment on outcomes, but also on the mechanism through an intermediate variable. Principal stratification aims to identify subgroup causal effects within principal strata, but the latent nature of principal strata makes it challenging. Leveraging the principal ignorability assumption, various nonparametric identification formulas for causal effects within principal strata are derived in observational studies, leading to the development of triply robust estimators.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
(2022)
Article
Statistics & Probability
Mats J. Stensrud, James M. Robins, Aaron Sarveta, Eric J. Tchetgen Tchetgen, Jessica G. Young
Summary: Researchers are interested in treatment effects on outcomes that are only defined based on posttreatment events. However, naive contrasts of outcomes conditional on posttreatment events are not average causal effects. To address this issue, we propose the conditional separable effects to quantify the causal effects of modified treatments. We also provide an applied example.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Mathematical & Computational Biology
Daniel Nevo, Malka Gorfine
Summary: This article investigates the causal effects of the Apolipoprotein E epsilon 4 allele (APOE) on late-onset Alzheimer's disease (AD) and death. A new estimation method is proposed for the scenario in which both AD and death are considered as outcomes, based on a stratification of the population and a novel assumption utilizing the time-to-event nature of the data. Nonparametric and semiparametric estimation methods are implemented to study the complex effect of APOE on AD and death.
Article
Mathematical & Computational Biology
Emily K. Roberts, Michael R. Elliott, Jeremy M. G. Taylor
Summary: This study extends causal inference approaches to validate surrogate endpoints using potential outcomes. The relationship of treatment effects on the surrogate and the true endpoint is assessed using principal surrogacy criteria. Bayesian methods are developed to incorporate modeling assumptions and explore their impact on assessing surrogacy in a muscular dystrophy gene therapy study.
STATISTICS IN MEDICINE
(2021)
Article
Education & Educational Research
Adam C. Sales, John F. Pane
Summary: Randomized evaluations of educational technology produce log data as a by-product, which can shed light on causal mechanisms, effect heterogeneity, or optimal use. However, methodological challenges exist, and this article discusses three approaches to overcome these issues. Analysis of hint data from an evaluation of the Cognitive Tutor Algebra I curriculum using the three approaches yielded conflicting results, indicating a need for further research and discussion.
JOURNAL OF RESEARCH ON EDUCATIONAL EFFECTIVENESS
(2021)
Article
Psychiatry
Chengdong Wang, Dongdong Zhu, Dongjun Zhang, Xiaowei Zuo, Lei Yao, Teng Liu, Xiaodan Ge, Chenlu He, Yuan Zhou, Ziyuan Shen
Summary: This study demonstrated the close connection between immune cells and schizophrenia through genetic means, providing guidance for future clinical research.
Article
Mathematical & Computational Biology
Mats J. Stensrud, Oliver Dukes
Summary: This article discusses how to account for intercurrent events in randomized trials and proposes methods for choosing estimands. The authors argue that the formulation of a research question should reflect current or future decision making and advocate for a more thorough consideration of treatment effects.
STATISTICS IN MEDICINE
(2022)
Article
Mathematical & Computational Biology
Erin E. Gabriel, Martha Nason, Michael P. Fay, Dean A. Follmann
STATISTICS IN MEDICINE
(2018)
Letter
Mathematical & Computational Biology
Erin E. Gabriel, Martha Nason, Michael P. Fay, Dean Follmann
STATISTICS IN MEDICINE
(2018)
Article
Mathematical & Computational Biology
Erin E. Gabriel, Michael C. Sachs, Michael J. Daniels, M. Elizabeth Halloran
STATISTICS IN MEDICINE
(2019)
Article
Immunology
Afroditi Boulougoura, Erin Gabriel, Elizabeth Laidlaw, Vikram Khetani, Ken Arakawa, Jeanette Higgins, Adam Rupert, Robert J. Gorelick, Keith Lumbard, Alice Pau, April Poole, Angela Kibiy, Princy Kumar, Irini Sereti
OPEN FORUM INFECTIOUS DISEASES
(2019)
Article
Immunology
Irini Sereti, Virginia Sheikh, Douglas Shaffer, Nittaya Phanuphak, Erin Gabriel, Jing Wang, Martha C. Nason, Gregg Roby, Hellen Ngeno, Fredrick Kirui, Alice Pau, Joann M. Mican, Adam Rupert, Rachel Bishop, Brian Agan, Nitiya Chomchey, Nipat Teeratakulpisarn, Somsit Tansuphaswadikul, Deborah Langat, Josphat Kosgei, Martyn French, Jintanat Ananworanich, Fredrick Sawe
CLINICAL INFECTIOUS DISEASES
(2020)
Article
Biology
Erin E. Gabriel, Michael C. Sachs, Dean A. Follmann, Therese M-L. Andersson
Article
Mathematical & Computational Biology
Adam Brand, Susanne May, James P. Hughes, Gertrude Nakigozi, Steven J. Reynolds, Erin E. Gabriel
Summary: Regular testing for chronic medical conditions is necessary, but may not be feasible in resource-limited settings. Prediction-driven pooled testing methods incorporating covariate information can improve testing efficiency and sensitivity, reducing the number of testing rounds.
STATISTICS IN MEDICINE
(2021)
Article
Public, Environmental & Occupational Health
Anna Persson, Sofia Lindmark, Kerstin Petersson, Erin Gabriel, Malin Thorsell, Karolina Lindstrom, Mona Goransson, Gunilla Cardell, Asa Magnusson
Summary: The study aimed to estimate the current alcohol and drug use among pregnant women attending antenatal care lectures in Stockholm, Sweden. Approximately 1 in 25 women reported alcohol use during pregnancy, and about 1 in 200 reported using illicit or non-medical prescription drugs while pregnant. Alcohol use during pregnancy may have decreased in Stockholm.
SEXUAL & REPRODUCTIVE HEALTHCARE
(2021)
Article
Pharmacology & Pharmacy
Pablo Gonzalez Ginestet, Erin E. Gabriel, Michael C. Sachs
Summary: This paper proposes a new method for applying machine learning algorithms to right censored survival data, improving risk prediction accuracy through pre-processing steps and ensemble methods.
JOURNAL OF BIOPHARMACEUTICAL STATISTICS
(2022)
Article
Statistics & Probability
Michael C. Sachs, Gustav Jonzon, Arvid Sjolander, Erin E. Gabriel
Summary: A causal query is often unidentifiable from observed data, but symbolic bounds can still be derived based on the distribution of observed variables, which can be more valuable than statistical estimators derived from implausible assumptions. We develop a general approach for calculating symbolic bounds and prove its effectiveness in certain settings. Additionally, we demonstrate that our method can provide valid and potentially informative symbolic bounds in a larger range of problems.
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
(2023)
Article
Statistics & Probability
Erin E. Gabriel, Michael C. Sachs, Arvid Sjolander
Summary: This study focuses on exploring mechanistic effects of intervention through mediators in the setting of two sequential mediators, with several decompositions of the total risk difference obtained. The results provide sharp and valid bounds for these mediation effects. It is also found that simply adding or subtracting the limits of the bounds may not produce sharp and informative results.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Statistics & Probability
Arvid Sjoelander, Jose M. M. Pena, Erin E. E. Gabriel
Summary: This study focuses on estimating the causal effect of an exposure on an outcome while adjusting for a binary confounder with measurement error. The study shows that adjusting for the mismeasured confounder results in a biased parameter between the true and crude parameters under certain assumptions. The study further presents empirical tests for these assumptions and demonstrates that the bias decreases with the sensitivity and specificity of the mismeasured confounder.
STATISTICS & PROBABILITY LETTERS
(2022)
Article
Mathematical & Computational Biology
Erin E. Gabriel, Arvid Sjolander, Dean Follmann, Michael C. Sachs
Summary: When multiple mediators are present, additional effects beyond natural and controlled direct effects are of interest. This study introduces five estimands for cross-CDE and -NDE when measuring two mediators, considering both sequential mediators and non-influential mediators. These estimands may have implications in immunology, specifically in relation to immunological responses to SARS-CoV-2 vaccination.
Article
Mathematical & Computational Biology
Iuliana Ciocanea-Teodorescu, Els Goetghebeur, Ingeborg Waernbaum, Staffan Schon, Erin E. Gabriel
Summary: Long-term register data provide unique opportunities to explore causal effects of treatments on time-to-event outcomes. However, the data structure may pose methodological challenges. In this study, focusing on a specific case in the Swedish Renal Registry, we investigate the consequences of missing covariate data and informative censoring on causal effect estimation. We find that using an imputation model including the entry date as a covariate and applying regression standardization leads to the best estimation results overall.
STATISTICS IN MEDICINE
(2023)
Article
Mathematical & Computational Biology
Erin E. Gabriel, Michael C. Sachs, Torben Martinussen, Ingeborg Waernbaum, Els Goetghebeur, Stijn Vansteelandt, Arvid Sjolander
Summary: There is a concerning trend in the applied literature to believe that the combination of a propensity score and an adjusted outcome model automatically results in a doubly robust estimator and to misuse more complex established doubly robust estimators. This article introduces a simple alternative method and aims to help readers understand why it is doubly robust. It also provides examples and code for implementation.
STATISTICS IN MEDICINE
(2023)