scholarly journals On weighting approaches for missing data

2011 ◽  
Vol 22 (1) ◽  
pp. 14-30 ◽  
Author(s):  
Lingling Li ◽  
Changyu Shen ◽  
Xiaochun Li ◽  
James M Robins
Keyword(s):  
Missing Data ◽  
Selection Bias ◽  
Inverse Probability Weighting ◽  
Probability Weighting ◽  
Full Data ◽  
Inverse Probability ◽  
Intuitive Idea ◽  
Complex Settings ◽  
Conceptual Overview

We review the class of inverse probability weighting (IPW) approaches for the analysis of missing data under various missing data patterns and mechanisms. The IPW methods rely on the intuitive idea of creating a pseudo-population of weighted copies of the complete cases to remove selection bias introduced by the missing data. However, different weighting approaches are required depending on the missing data pattern and mechanism. We begin with a uniform missing data pattern (i.e. a scalar missing indicator indicating whether or not the full data is observed) to motivate the approach. We then generalise to more complex settings. Our goal is to provide a conceptual overview of existing IPW approaches and illustrate the connections and differences among these approaches.

Adjusting for selection bias due to missing data in electronic health records-based research

2021 ◽  
Vol 30 (10) ◽  
pp. 2221-2238
Author(s):  
Sarah B Peskoe ◽  
David Arterburn ◽  
Karen J Coleman ◽  
Lisa J Herrinton ◽  
Michael J Daniels ◽  
...  
Keyword(s):  
Missing Data ◽  
Electronic Health Records ◽  
Selection Bias ◽  
Inverse Probability Weighting ◽  
Small Sample ◽  
Data Provenance ◽  
Probability Weighting ◽  
Health Records ◽  
Inverse Probability ◽  
Electronic Health

While electronic health records data provide unique opportunities for research, numerous methodological issues must be considered. Among these, selection bias due to incomplete/missing data has received far less attention than other issues. Unfortunately, standard missing data approaches (e.g. inverse-probability weighting and multiple imputation) generally fail to acknowledge the complex interplay of heterogeneous decisions made by patients, providers, and health systems that govern whether specific data elements in the electronic health records are observed. This, in turn, renders the missing-at-random assumption difficult to believe in standard approaches. In the clinical literature, the collection of decisions that gives rise to the observed data is referred to as the data provenance. Building on a recently-proposed framework for modularizing the data provenance, we develop a general and scalable framework for estimation and inference with respect to regression models based on inverse-probability weighting that allows for a hierarchy of missingness mechanisms to better align with the complex nature of electronic health records data. We show that the proposed estimator is consistent and asymptotically Normal, derive the form of the asymptotic variance, and propose two consistent estimators. Simulations show that naïve application of standard methods may yield biased point estimates, that the proposed estimators have good small-sample properties, and that researchers may have to contend with a bias-variance trade-off as they consider how to handle missing data. The proposed methods are motivated by an on-going, electronic health records-based study of bariatric surgery.

scholarly journals Contextualizing selection bias in Mendelian randomization: how bad is it likely to be?

2018 ◽  
Vol 48 (3) ◽  
pp. 691-701 ◽  
Author(s):  
Apostolos Gkatzionis ◽  
Stephen Burgess
Keyword(s):  
Risk Factor ◽  
Selection Bias ◽  
Simulation Study ◽  
Cardiovascular Mortality ◽  
Mendelian Randomization ◽  
Inverse Probability Weighting ◽  
Probability Weighting ◽  
Inverse Probability ◽  
Type 1 Error

Abstract Background Selection bias affects Mendelian randomization investigations when selection into the study sample depends on a collider between the genetic variant and confounders of the risk factor–outcome association. However, the relative importance of selection bias for Mendelian randomization compared with other potential biases is unclear. Methods We performed an extensive simulation study to assess the impact of selection bias on a typical Mendelian randomization investigation. We considered inverse probability weighting as a potential method for reducing selection bias. Finally, we investigated whether selection bias may explain a recently reported finding that lipoprotein(a) is not a causal risk factor for cardiovascular mortality in individuals with previous coronary heart disease. Results Selection bias had a severe impact on bias and Type 1 error rates in our simulation study, but only when selection effects were large. For moderate effects of the risk factor on selection, bias was generally small and Type 1 error rate inflation was not considerable. Inverse probability weighting ameliorated bias when the selection model was correctly specified, but increased bias when selection bias was moderate and the model was misspecified. In the example of lipoprotein(a), strong genetic associations and strong confounder effects on selection mean the reported null effect on cardiovascular mortality could plausibly be explained by selection bias. Conclusions Selection bias can adversely affect Mendelian randomization investigations, but its impact is likely to be less than other biases. Selection bias is substantial when the effects of the risk factor and confounders on selection are particularly large.

scholarly journals Responsiveness-informed multiple imputation and inverse probability-weighting in cohort studies with missing data that are non-monotone or not missing at random

2016 ◽  
Vol 27 (2) ◽  
pp. 352-363 ◽  
Author(s):  
James C Doidge
Keyword(s):  
Missing Data ◽  
Data Collection ◽  
Cohort Studies ◽  
Multiple Imputation ◽  
Missing At Random ◽  
Inverse Probability Weighting ◽  
Probability Weighting ◽  
Inverse Probability ◽  
Not Missing At Random ◽  
Over Time

Population-based cohort studies are invaluable to health research because of the breadth of data collection over time, and the representativeness of their samples. However, they are especially prone to missing data, which can compromise the validity of analyses when data are not missing at random. Having many waves of data collection presents opportunity for participants’ responsiveness to be observed over time, which may be informative about missing data mechanisms and thus useful as an auxiliary variable. Modern approaches to handling missing data such as multiple imputation and maximum likelihood can be difficult to implement with the large numbers of auxiliary variables and large amounts of non-monotone missing data that occur in cohort studies. Inverse probability-weighting can be easier to implement but conventional wisdom has stated that it cannot be applied to non-monotone missing data. This paper describes two methods of applying inverse probability-weighting to non-monotone missing data, and explores the potential value of including measures of responsiveness in either inverse probability-weighting or multiple imputation. Simulation studies are used to compare methods and demonstrate that responsiveness in longitudinal studies can be used to mitigate bias induced by missing data, even when data are not missing at random.

Scaled Inverse Probability Weighting: A Method to Assess Potential Bias Due to Event Nonreporting in Ecological Momentary Assessment Studies

2017 ◽  
Vol 43 (3) ◽  
pp. 354-381 ◽  
Author(s):  
Stephanie A. Kovalchik ◽  
Steven C. Martino ◽  
Rebecca L. Collins ◽  
William G. Shadel ◽  
Elizabeth J. D’Amico ◽  
...  
Keyword(s):  
Missing Data ◽  
Ecological Momentary Assessment ◽  
Inverse Probability Weighting ◽  
Assessment Method ◽  
Probability Weighting ◽  
Inverse Probability ◽  
Traditional Assessment ◽  
Urge To Smoke ◽  
Ecological Momentary ◽  
Momentary Assessment

Ecological momentary assessment (EMA) is a popular assessment method in psychology that aims to capture events, emotions, and cognitions in real time, usually repeatedly throughout the day. Because EMA typically involves more intensive monitoring than traditional assessment methods, missing data are commonly an issue and this missingness may bias results. EMA can involve two types of missing data: known missingness, arising from nonresponse to scheduled prompts, and hidden missingness, arising from nonreporting of focal events (e.g., an urge to smoke or a meal). Prior research on missing data in EMA has focused almost exclusively on nonresponse to scheduled prompts. In this study, we introduce a scaled inverse probability weighting approach to assess the risk of bias due to nonreporting of events due to fatigue on estimates of exposure or correlates of exposure. In our proposed approach, the inverse probability is the estimated probability of compliance with random prompts from a model that uses participant and contextual factors to predict this compliance and a fatigue factor that adjusts for attrition in event reporting over time. We demonstrate the use and utility of our bias assessment method with the Tracking and Recording Alcohol Communications Study, an EMA study of adolescent exposure to alcohol advertising.

Predictors of follow-up and assessment of selection bias from dropouts using inverse probability weighting in a cohort of university graduates

2006 ◽  
Vol 21 (5) ◽  
pp. 351-358 ◽  
Author(s):  
Alvaro Alonso ◽  
María Seguí-Gómez ◽  
Jokin de Irala ◽  
Almudena Sánchez-Villegas ◽  
Juan José Beunza ◽  
...  
Keyword(s):  
Selection Bias ◽  
Inverse Probability Weighting ◽  
University Graduates ◽  
Probability Weighting ◽  
Inverse Probability

Abstract P198: Racial Disparities in Cardiovascular Outcomes in REGARDS Persist After Using Inverse Probability Weighting to Account for Missing Data

Circulation ◽  
2017 ◽  
Vol 135 (suppl_1) ◽  
Author(s):  
D. Leann Long ◽  
George Howard ◽  
Suzanne Judd ◽  
Jennifer Manly ◽  
Leslie McClure ◽  
...  
Keyword(s):  
Left Ventricular Hypertrophy ◽  
Selection Bias ◽  
Ventricular Hypertrophy ◽  
Inverse Probability Weighting ◽  
Left Ventricular ◽  
Probability Weighting ◽  
Complete Case ◽  
Inverse Probability ◽  
Weighted Analysis ◽  
Inverse Probability Weighted

Introduction: When data are missing, analysts often choose to perform ‘complete case’ analysis, restricting statistical analysis to those participants with all necessary fields completed. However, this analysis is subject to selection bias if those participants excluded are somehow inherently different from those included. One approach to address the potential selection bias is inverse probability weighting, where participants with complete data are weighted to reflect the original population. We compare estimated racial disparities in hypertension and left ventricular hypertrophy using complete cases to those estimates from inverse probability weighted analysis. Methods: The REasons for Geographic and Racial Differences in Stroke (REGARDS) study enrolled 30,183 participants aged 45+ between 2003 and 2007 to study racial and geographic differences in stroke and cardiovascular health. When the second in-home visits were conducted, 18% (5542) of participants had died and 23% (7071) of participants had withdrawn. For the baseline population, the probabilities of being a complete case are estimated through logistic regression models using baseline characteristics. These predicted probabilities are inverted to create the weights used in statistical analysis, such that the complete participants are weighted to represent the enrolled population. Through logistic regression, we estimate the association between race and hypertension and left ventricular hypertrophy at the second in-home visit for complete data and compare these results to the results from inverse probability weighted analysis. All models are adjusted for sex, age and region. Results: The logistic models for dropout and death have low and moderate predictive abilities (c-statistics 0.602 and 0.811, respectively). For incident hypertension, the estimated odds ratio comparing blacks to whites differs little between the complete case (1.87 (1.66, 2.10)) and the weighted (1.83 (1.61, 2.09)) analysis. For left ventricular hypertrophy, the estimated odds ratio comparing blacks to whites changes little from the complete case analysis (1.54 (1.32, 1.79)) to the weighted analysis (1.45 (1.21, 1.74)). Discussion: Estimated racial inequalities in the odds of incident hypertension and left ventricular hypertrophy were similar in the complete case and inverse probability weighting analyses, indicating little evidence of selection bias in the estimation of racial inequalities for these outcomes.

Sign in / Sign up

Export Citation Format

Share Document