Editorial Material
Public, Environmental & Occupational Health
Ghassan B. Hamra
Summary: This article highlights the issue of heuristic nondifferential misclassification biases towards null results and argues against its widespread use. It provides a brief history of this heuristic and discusses its limited applicability.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2022)
Article
Public, Environmental & Occupational Health
Nathaniel Stockham, Peter Washington, Brianna Chrisman, Kelley Paskov, Jae-Yoon Jung, Dennis Paul Wall
Summary: This study demonstrates the practical utility of causal modeling in addressing selection bias and unmeasured confounding in epidemiological research. By constructing various causal models and comparing them with collected data, the study identifies the most compatible causal model and successfully estimates the infection rate.
JMIR PUBLIC HEALTH AND SURVEILLANCE
(2022)
Article
Public, Environmental & Occupational Health
Louisa H. Smith, Maya B. Mathur, Tyler J. VanderWeele
Summary: This study demonstrates a method to bound the total composite bias due to confounding, selection bias, and measurement error, and uses that bound to assess the sensitivity of a risk ratio to any combination of these biases. The approach is conservative and provides a simpler alternative to quantitative bias analysis.
Review
Public, Environmental & Occupational Health
Maya B. Mathur, Tyler J. VanderWeele
Summary: Meta-analyses play a critical role in cumulative science, but can lead to misleading conclusions if the primary studies they include are biased. This article provides practical guidance for addressing biases that affect the internal validity of studies in meta-analyses, focusing on sensitivity analyses to quantify potential biases. Various sensitivity analysis methods are reviewed, with a focus on recent developments that are easy to implement and interpret. The importance of routinely reporting sensitivity analyses in meta-analyses of potentially biased studies is emphasized.
ANNUAL REVIEW OF PUBLIC HEALTH
(2022)
Article
Public, Environmental & Occupational Health
Timothy L. Lash, Thomas P. Ahern, Lindsay J. Collin, Matthew P. Fox, Richard F. MacLehose
Summary: The importance of bias analysis in epidemiologic research is highlighted in the text, with a mention of the lack of quantitative estimates of bias impacts in reports despite the availability of methods and tools. Three suboptimal bias analysis examples are identified, and common shortcomings in these examples are summarized.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2021)
Article
Public, Environmental & Occupational Health
Jennifer J. Yland, Amelia K. Wesselink, Timothy L. Lash, Matthew P. Fox
Summary: Measurement error is common in epidemiologic research, and it is often assumed that nondifferential misclassification biases estimates towards the null. However, there are exceptions to this assumption, and a more critical and nuanced approach is needed to evaluate and discuss bias from nondifferential mismeasurement.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2022)
Article
Public, Environmental & Occupational Health
Jin Liu, Shiyuan Wang, Fang Shao
Summary: Prevalence estimates in epidemiological studies are important but susceptible to misclassification bias. Quantitative bias analysis can effectively estimate this bias, but its usage is limited due to lack of knowledge and tools. This study proposes three indicators based on quantitative bias analysis principles to express the magnitude, direction, and uncertainty of misclassification bias in prevalence estimates, highlighting the importance of using these analysis for epidemiological decision making.
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
(2023)
Review
Public, Environmental & Occupational Health
Julie M. Petersen, Lynsie R. Ranker, Ruby Barnard-Mayers, Richard F. MacLehose, Matthew P. Fox
Summary: QBA applications in epidemiological research were rare but increasing over time. Most studies used QBA as secondary analyses to conventional methods or to assess the extent of bias. Common types of biases included misclassification, uncontrolled confounders, and selection bias. Many studies did not consider multiple biases or correlations between errors.
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
(2021)
Article
Health Care Sciences & Services
Yingrui Yang, Molin Wang
Summary: In epidemiology, incorporating propensity score in the Cox regression model can effectively control for confounding, but determining exposure effect using propensity score remains challenging in situations with moderate to substantial error in exposure measurement. This paper proposes an estimating equation method to correct bias caused by exposure misclassification, providing more accurate estimation of exposure-outcome associations. Simulation studies are conducted to evaluate the performance of the proposed estimators in various settings, with an application to correct bias in estimating the association of PM2.5 levels with lung cancer mortality.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2021)
Article
Public, Environmental & Occupational Health
Paul Brendel, Aracelis Torres, Onyebuchi A. Arah
Summary: Traditional multiple-bias adjustment involves adjusting for biases one at a time, while a novel alternative approach is to simultaneously adjust for all biases using imputation and/or regression weighting. A simulation study showed that using correct bias parameters can yield unbiased effect estimates, and even incorrect specification of bias parameters still resulted in less bias compared to observed biased effects. Simultaneous multi-bias analysis is a useful method to investigate and understand how multiple biases can affect initial effect estimates, enhancing the validity and transparency of real-world evidence obtained from observational, longitudinal studies.
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
(2023)
Article
Public, Environmental & Occupational Health
Erin E. Bennett, Katie M. Lynch, Xiaohui Xu, Eun Sug Park, Qi Ying, Jingkai Wei, Richard L. Smith, James D. Stewart, Eric A. Whitsel, Melinda C. Power
Summary: Current research on the characteristics and predictors of movers is limited. In this study, we analyzed the ARIC cohort to identify important predictors of moving for different types of movers, and found that interaction between characteristics plays a crucial role. This work has important implications for epidemiological research and studies focusing on residential mobility as an exposure.
Article
Public, Environmental & Occupational Health
Maya B. Mathur
Summary: We propose sensitivity analyses for complete-case estimates of treatment effects to address biases caused by non-random missing data. These analyses use simple summary data and avoid distributional assumptions. The proposed methods bound the overall treatment effect by considering unobserved treatment effects among nonretained participants and the strengths of confounding associations among retained participants. The introduction of the M-value as an analog to the E-value provides a measure of the strength of confounding associations required to mitigate the treatment effect. These methods can help evaluate the robustness of complete-case analyses to potential biases due to missing data.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2023)
Article
Dentistry, Oral Surgery & Medicine
Talal S. Alshihayb, Praveen Sharma, Thomas Dietrich, Brenda Heaton
Summary: This study investigated the reasons for misclassification of periodontitis under partial-mouth protocols (PMPs) and proposed a tooth selection method based on population rankings of clinical severity to enhance protocol validity.
JOURNAL OF CLINICAL PERIODONTOLOGY
(2022)
Article
Public, Environmental & Occupational Health
David B. Richardson, Alexander P. Keil, Stephen R. Cole
Summary: This study highlights that the net bias in an estimation of the association of interest may increase when adjusting for confounders in the presence of classical exposure measurement error.
AMERICAN JOURNAL OF EPIDEMIOLOGY
(2022)
Article
Health Care Sciences & Services
Bas B. L. Penning de Vries, Maarten van Smeden, Rolf H. H. Groenwold
Summary: The paper introduces a new estimator for marginal causal effects that takes into account confounding and joint misclassification of exposure and outcome variables, relying on validation data for weight construction. Simulation studies show favorable large sample properties of the new estimator.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2021)