Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1001/jama.2015.18026 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Exposure range matters: considering non-linear associations in the meta-analysis of environmental pollutant exposure using examples of per- and polyfluoroalkyl substances and birth outcomes.
Am J Epidemiol
September 2024
Department of Environmental Health Sciences, Yale School of Public Health, New Haven, USA.
Meta-analysis is a powerful analytic method for summarizing effect estimates across studies. However, conventional meta-analysis often assumes a linear exposure-outcome relationship and does not account for variability over the exposure ranges. In this work, we first used simulation techniques to illustrate that the linear-based meta-analytical approach may result in oversimplistic effect estimation based on three plausible non-linear exposure-outcome curves (S-shape, inverted U-shape, and M-shape).
View Article and Find Full Text PDFTargeted maximum likelihood estimation (TMLE) is increasingly used for doubly robust causal inference, but how missing data should be handled when using TMLE with data-adaptive approaches is unclear. Based on data (1992-1998) from the Victorian Adolescent Health Cohort Study, we conducted a simulation study to evaluate 8 missing-data methods in this context: complete-case analysis, extended TMLE incorporating an outcome-missingness model, the missing covariate missing indicator method, and 5 multiple imputation (MI) approaches using parametric or machine-learning models. We considered 6 scenarios that varied in terms of exposure/outcome generation models (presence of confounder-confounder interactions) and missingness mechanisms (whether outcome influenced missingness in other variables and presence of interaction/nonlinear terms in missingness models).
View Article and Find Full Text PDFLeft ventricular hypertrophy as a risk factor for accelerated brain aging: Results from the Study of Health in Pomerania.
Hum Brain Mapp
February 2024
Department of Internal Medicine B, University Medicine Greifswald, Greifswald, Mecklenburg-Western Pomerania, Germany.
Article Synopsis
- * In a study of 926 individuals without serious heart conditions, it was found that a higher baseline LVMI was directly linked to decreased cortical thickness and an increased brain age after 7 years, even when factoring out hypertension and blood pressure.
- * The findings indicate that LVMI could serve as a simple marker for identifying individuals at risk of cognitive decline and dementia, with stronger effects observed in those with significantly higher LVMI levels.
Mediation Analysis using Semi-parametric Shape-Restricted Regression with Applications.
ArXiv
October 2023
Department of Epidemiology, University of Pittsburgh.
Often linear regression is used to perform mediation analysis. However, in many instances, the underlying relationships may not be linear, as in the case of placentalfetal hormones and fetal development. Although, the exact functional form of the relationship may be unknown, one may hypothesize the general shape of the relationship.
View Article and Find Full Text PDFUniversal Difference-in-Differences for Causal Inference in Epidemiology.
Epidemiology
January 2024
Department of Environmental & Occupational Health, University of California, Irvine, CA.
Difference-in-differences is undoubtedly one of the most widely used methods for evaluating the causal effect of an intervention in observational (i.e., nonrandomized) settings.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!