scholarly journals Considerations for using multiple imputation in propensity score-weighted analysis

2021 ◽  
Author(s):  
Andreas Halgreen Eiset ◽  
Morten Frydenberg
Keyword(s):  
Missing Data ◽  
Propensity Score ◽  
Confidence Interval ◽  
Multiple Imputation ◽  
Computational Time ◽  
Standardized Mortality Ratio ◽  
Data Set ◽  
Propensity Score Model ◽  
Weighted Analysis ◽  
Percentile Confidence Interval

We present our considerations for using multiple imputation to account for missing data in propensity score-weighted analysis with bootstrap percentile confidence interval. We outline the assumptions underlying each of the methods and discuss the methodological and practical implications of our choices and briefly point to alternatives. We made a number of choices a priori for example to use logistic regression-based propensity scores to produce standardized mortality ratio-weights and Substantive Model Compatible-Full Conditional Specification to multiply impute missing data (given no violation of underlying assumptions). We present a methodology to combine these methods by choosing the propensity score model based on covariate balance, using this model as the substantive model in the multiple imputation, producing and averaging the point estimates from each multiple imputed data set to give the estimate of association and computing the percentile confidence interval by bootstrapping. The described methodology is demanding in both work-load and in computational time, however, we do not consider the prior a draw-back: it makes some of the underlying assumptions explicit and the latter may be a nuisance that will diminish with faster computers and better implementations.

scholarly journals 1216Combining propensity score-weighting and multiple imputation is not a trivial task

2021 ◽  
Vol 50 (Supplement_1) ◽  
Author(s):  
Andreas Halgreen Eiset ◽  
Morten Frydenberg
Keyword(s):  
Propensity Score ◽  
Confidence Interval ◽  
Multiple Imputation ◽  
Computing Time ◽  
Theory And Practice ◽  
Computational Time ◽  
Post Traumatic Stress ◽  
Long Distance ◽  
Data Set ◽  
Propensity Score Weighting

Abstract Background Propensity score (PS)-weighting and multiple imputation are two widely used statistical methods. Combining the two is not trivial and has received little attention in theory and practice. We present our considerations for their combination with application to a study of long-distance migration and post-traumatic stress disorder. We elaborate on the assumptions underlying the methods and discuss the methodological and practical implications of our choices and alternatives. Methods We made a number of choices a priori: to use logistic regression-based PS to produce “standardised mortality ratio”-weights and SMC-FCS to multiply impute missing data. We present a methodology to combine the methods by choosing the PS model based on covariate balance, using this model as the substantive model in the multiple imputation, producing and averaging the point estimates from each multiply imputed data set to give the estimate of association and computing the percentile confidence interval by bootstrapping. Results In our application, a simple PS model was chosen as the substantive model for imputing 10 data sets with 40 iterations and repeating the entirety 999 times to obtain a bootstrap confidence interval. Computing time was approximately 36 hours. Conclusions Our structured approach is demanding in both work-load and computational time. We do not consider the prior a draw-back: it makes some of the underlying assumptions explicit and the latter may be a nuisance that diminishes with time. Key messages Combining propensity score-weighting and multiple imputation is not a trivial task.

scholarly journals How oil income and missing hydrocarbon rents data influence autocratic survival: A response to Lucas and Richter (2016)

2017 ◽  
Vol 4 (3) ◽  
pp. 205316801771979 ◽  
Author(s):  
Joseph Wright ◽  
Erica Frantz
Keyword(s):  
Missing Data ◽  
Multiple Imputation ◽  
Data Set ◽  
Oil Rents

This paper re-examines the findings from a recently published study on hydrocarbon rents and autocratic survival by Lucas and Richter (LR hereafter). LR introduce a new data set on hydrocarbon rents and use it to examine the link between oil income and autocratic survival. Employing a placebo test, we show that the authors’ strategy for dealing with missingness in the new hydrocarbon rents data set – filling in missing data with zeros – creates bias in the reported estimates of interest. Addressing missingness with multiple imputation shows that the LR findings linking oil rents to democratization do not hold. Instead, we find that hydrocarbon rents reduce the chances of transition to a new dictatorship, consistent with the conclusions of Wright et al.

scholarly journals How to Apply Multiple Imputation in Propensity Score Matching with Partially Observed Confounders: A Simulation Study and Practical Recommendations

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Albee Ling ◽  
Maria Montez-Rath ◽  
Maya Mathur ◽  
Kris Kapphahn ◽  
Manisha Desai
Keyword(s):  
Missing Data ◽  
Propensity Score ◽  
Multiple Imputation ◽  
Propensity Score Matching ◽  
Simulation Study ◽  
Observational Studies ◽  
Auxiliary Variables ◽  
Imputation Model ◽  
Partially Observed ◽  
Practical Recommendations

Propensity score matching (PSM) has been widely used to mitigate confounding in observational studies, although complications arise when the covariates used to estimate the PS are only partially observed. Multiple imputation (MI) is a potential solution for handling missing covariates in the estimation of the PS. However, it is not clear how to best apply MI strategies in the context of PSM. We conducted a simulation study to compare the performances of popular non-MI missing data methods and various MI-based strategies under different missing data mechanisms. We found that commonly applied missing data methods resulted in biased and inefficient estimates, and we observed large variation in performance across MI-based strategies. Based on our findings, we recommend 1) estimating the PS after applying MI to impute missing confounders; 2) conducting PSM within each imputed dataset followed by averaging the treatment effects to arrive at one summarized finding; 3) a bootstrapped-based variance to account for uncertainty of PS estimation, matching, and imputation; and 4) inclusion of key auxiliary variables in the imputation model.

scholarly journals Propensity Score Estimation Using Classification and Regression Trees in the Presence of Missing Covariate Data

2018 ◽  
Vol 7 (1) ◽  
Author(s):  
Bas B.L. Penning de Vries ◽  
Maarten van Smeden ◽  
Rolf H.H. Groenwold
Keyword(s):  
Logistic Regression ◽  
Missing Data ◽  
Propensity Score ◽  
Multiple Imputation ◽  
Incomplete Data ◽  
Regression Trees ◽  
Covariate Data ◽  
Missing Covariate Data ◽  
Classification And Regression ◽  
Cart Algorithm

AbstractData mining and machine learning techniques such as classification and regression trees (CART) represent a promising alternative to conventional logistic regression for propensity score estimation. Whereas incomplete data preclude the fitting of a logistic regression on all subjects, CART is appealing in part because some implementations allow for incomplete records to be incorporated in the tree fitting and provide propensity score estimates for all subjects. Based on theoretical considerations, we argue that the automatic handling of missing data by CART may however not be appropriate. Using a series of simulation experiments, we examined the performance of different approaches to handling missing covariate data; (i) applying the CART algorithm directly to the (partially) incomplete data, (ii) complete case analysis, and (iii) multiple imputation. Performance was assessed in terms of bias in estimating exposure-outcome effects among the exposed, standard error, mean squared error and coverage. Applying the CART algorithm directly to incomplete data resulted in bias, even in scenarios where data were missing completely at random. Overall, multiple imputation followed by CART resulted in the best performance. Our study showed that automatic handling of missing data in CART can cause serious bias and does not outperform multiple imputation as a means to account for missing data.

scholarly journals Accounting for missing data caused by drug cessation in observational comparative effectiveness research: a simulation study

2022 ◽  
pp. annrheumdis-2021-221477
Author(s):  
Denis Mongin ◽  
Kim Lauper ◽  
Axel Finckh ◽  
Thomas Frisell ◽  
Delphine Sophie Courvoisier
Keyword(s):  
Missing Data ◽  
Disease Activity ◽  
Multiple Imputation ◽  
Simulation Study ◽  
Comparative Effectiveness ◽  
Real World Data ◽  
Data Set ◽  
Multiple Imputations ◽  
True Value ◽  
The Absolute

ObjectivesTo assess the performance of statistical methods used to compare the effectiveness between drugs in an observational setting in the presence of attrition.MethodsIn this simulation study, we compared the estimations of low disease activity (LDA) at 1 year produced by complete case analysis (CC), last observation carried forward (LOCF), LUNDEX, non-responder imputation (NRI), inverse probability weighting (IPW) and multiple imputations of the outcome. All methods were adjusted for confounders. The reasons to stop the treatments were included in the multiple imputation method (confounder-adjusted response rate with attrition correction, CARRAC) and were either included (IPW2) or not (IPW1) in the IPW method. A realistic simulation data set was generated from a real-world data collection. The amount of missing data caused by attrition and its dependence on the ‘true’ value of the data missing were varied to assess the robustness of each method to these changes.ResultsLUNDEX and NRI strongly underestimated the absolute LDA difference between two treatments, and their estimates were highly sensitive to the amount of attrition. IPW1 and CC overestimated the absolute LDA difference between the two treatments and the overestimation increased with increasing attrition or when missingness depended on disease activity at 1 year. IPW2 and CARRAC produced unbiased estimations, but IPW2 had a greater sensitivity to the missing pattern of data and the amount of attrition than CARRAC.ConclusionsOnly multiple imputation and IPW2, which considered both confounding and treatment cessation reasons, produced accurate comparative effectiveness estimates.

Identification of a Suitable Missing Data Handling Method for Lifelogs-based Wellness Index Estimation with Panel Data (Preprint)

2020 ◽  
Author(s):  
KI-Hun Kim ◽  
Kwang-Jae Kim
Keyword(s):  
Missing Data ◽  
Panel Data ◽  
Health Behavior ◽  
Multiple Imputation ◽  
Reference Data ◽  
Low Rank ◽  
Data Sets ◽  
Data Handling ◽  
Data Set ◽  
Handling Method

BACKGROUND A lifelogs-based wellness index (LWI) is a function to calculate wellness scores from health behavior lifelogs such as daily walking steps and sleep time collected through smartphones. A wellness score intuitively shows a user of a smart wellness service the overall condition of health behaviors. LWI development includes LWI estimation (i.e., estimating coefficients in LWI with data). A panel data set of health behavior lifelogs allows LWI estimation to control for variables unobserved in LWI and hence to be less biased. Such panel data sets are likely to have missing data due to various random events of daily life (e.g., smart devices stop collecting data when they are out of batteries). Missing data can introduce the biases to LWI coefficients. Thus, the choice of appropriate missing data handling method is important to reduce the biases in LWI estimation with a panel data set of health behavior lifelogs. However, relevant studies are scarce in the literature. OBJECTIVE This research aims to identify a suitable missing data handling method for LWI estimation with panel data. Six representative missing data handling methods (i.e., listwise deletion (LD), mean imputation, Expectation-Maximization (EM) based multiple imputation, Predictive-Mean Matching (PMM) based multiple imputation, k-Nearest Neighbors (k-NN) based imputation, and Low-rank Approximation (LA) based imputation) are comparatively evaluated through the simulation of an existing LWI development case. METHODS A panel data set of health behavior lifelogs collected in the existing LWI development case was transformed into a reference data set. 200 simulated data sets were generated by randomly introducing missing data to the reference data set at each of missingness proportions from 1% to 80%. The six methods were applied to transform the simulated data sets into complete data sets by handling missing data. Coefficients in a linear LWI, a linear function, were estimated with each of all the complete data sets by following the case. Coefficient biases of the six methods were calculated by comparing the estimated coefficient values with reference values estimated with the reference data set. RESULTS Based on the coefficient biases, the superior methods changed according to the missingness proportion: LA based imputation, PMM based multiple imputation, and EM based multiple imputation for 1% to 30% missingness proportions; LA based imputation and PMM based multiple imputation for 31% to 60%; and only LA based imputation for over 60%. CONCLUSIONS LA based imputation was superior among the six methods regardless of the missingness proportion. This superiority is generalizable for other panel data sets of health behavior lifelogs because existing works have verified their low-rank nature where LA based imputation works well. This result will guide the missing data handling to reduce the coefficient biases in new development cases of linear LWIs with panel data.

scholarly journals Collaborative-controlled LASSO for constructing propensity score-based estimators in high-dimensional data

2017 ◽  
Vol 28 (4) ◽  
pp. 1044-1063 ◽  
Author(s):  
Cheng Ju ◽  
Richard Wyss ◽  
Jessica M Franklin ◽  
Sebastian Schneeweiss ◽  
Jenny Häggström ◽  
...  
Keyword(s):  
Model Selection ◽  
Propensity Score ◽  
Confidence Interval ◽  
Point Estimation ◽  
High Dimensional ◽  
Minimum Loss ◽  
Propensity Score Model ◽  
Confidence Interval Coverage ◽  
Score Model ◽  
Interval Coverage

Propensity score-based estimators are increasingly used for causal inference in observational studies. However, model selection for propensity score estimation in high-dimensional data has received little attention. In these settings, propensity score models have traditionally been selected based on the goodness-of-fit for the treatment mechanism itself, without consideration of the causal parameter of interest. Collaborative minimum loss-based estimation is a novel methodology for causal inference that takes into account information on the causal parameter of interest when selecting a propensity score model. This “collaborative learning” considers variable associations with both treatment and outcome when selecting a propensity score model in order to minimize a bias-variance tradeoff in the estimated treatment effect. In this study, we introduce a novel approach for collaborative model selection when using the LASSO estimator for propensity score estimation in high-dimensional covariate settings. To demonstrate the importance of selecting the propensity score model collaboratively, we designed quasi-experiments based on a real electronic healthcare database, where only the potential outcomes were manually generated, and the treatment and baseline covariates remained unchanged. Results showed that the collaborative minimum loss-based estimation algorithm outperformed other competing estimators for both point estimation and confidence interval coverage. In addition, the propensity score model selected by collaborative minimum loss-based estimation could be applied to other propensity score-based estimators, which also resulted in substantive improvement for both point estimation and confidence interval coverage. We illustrate the discussed concepts through an empirical example comparing the effects of non-selective nonsteroidal anti-inflammatory drugs with selective COX-2 inhibitors on gastrointestinal complications in a population of Medicare beneficiaries.

scholarly journals Multiple Imputation with Survey Weights: A Multilevel Approach

2019 ◽  
Vol 8 (5) ◽  
pp. 965-989
Author(s):  
M Quartagno ◽  
J R Carpenter ◽  
H Goldstein
Keyword(s):  
Multiple Imputation ◽  
Missing At Random ◽  
Model Parameters ◽  
Imputation Model ◽  
Data Set ◽  
Weighted Analysis ◽  
Empirical Performance ◽  
Uk Millennium Cohort Study ◽  
The Uk ◽  
Variance Estimates

Abstract Multiple imputation is now well established as a practical and flexible method for analyzing partially observed data, particularly under the missing at random assumption. However, when the substantive model is a weighted analysis, there is concern about the empirical performance of Rubin’s rules and also about how to appropriately incorporate possible interaction between the weights and the distribution of the study variables. One approach that has been suggested is to include the weights in the imputation model, potentially also allowing for interactions with the other variables. We show that the theoretical criterion justifying this approach can be approximately satisfied if we stratify the weights to define level-two units in our data set and include random intercepts in the imputation model. Further, if we let the covariance matrix of the variables have a random distribution across the level-two units, we also allow imputation to reflect any interaction between weight strata and the distribution of the variables. We evaluate our proposal in a number of simulation scenarios, showing it has promising performance both in terms of coverage levels of the model parameters and bias of the associated Rubin’s variance estimates. We illustrate its application to a weighted analysis of factors predicting reception-year readiness in children in the UK Millennium Cohort Study.

scholarly journals Assessing the external validity of the VALIDATE-SWEDEHEART trial

2021 ◽  
pp. 174077452110124
Author(s):  
Rebecca T Rylance ◽  
Philippe Wagner ◽  
Elmir Omerovic ◽  
Claes Held ◽  
Stefan James ◽  
...  
Keyword(s):  
Propensity Score ◽  
Confidence Interval ◽  
Cox Regression ◽  
Hazard Ratio ◽  
Inclusion Criteria ◽  
Cox Regression Analysis ◽  
Inverse Probability ◽  
Weighted Analysis ◽  
Score Difference ◽  
Inverse Probability Weighted

Aims: The VALIDATE-SWEDEHEART trial was a registry-based randomized trial comparing bivalirudin and heparin in patients with acute myocardial infarction undergoing percutaneous coronary intervention. It showed no differences in mortality at 30 or 180 days. This study examines how well the trial population results may generalize to the population of all screened patients with fulfilled inclusion criteria in regard to mortality at 30 and 180 days. Methods: The standardized difference in the mean propensity score for trial inclusion between trial population and the screened not-enrolled with fulfilled inclusion criteria was calculated as a metric of similarity. Propensity scores were then used in an inverse-probability weighted Cox regression analysis using the trial population only to estimate the difference in mortality as it would have been had the trial included all screened patients with fulfilled inclusion criteria. Patients who were very likely to be included were weighted down and those who had a very low probability of being in the trial were weighted up. Results: The propensity score difference was 0.61. There were no significant differences in mortality between bivalirudin and heparin in the inverse-probability weighted analysis (hazard ratio 1.11, 95% confidence interval (0.73, 1.68)) at 30 days or 180 days (hazard ratio 0.98, 95% confidence interval (0.70, 1.36)). Conclusion: The propensity score difference demonstrated that the screened not-enrolled with fulfilled inclusion criteria and trial population were not similar. The inverse-probability weighted analysis showed no significant differences in mortality. From this, we conclude that the VALIDATE results may be generalized to the screened not-enrolled with fulfilled inclusion criteria.

Sign in / Sign up

Export Citation Format

Share Document