Multiple imputation for handling missing outcome data in randomized trials involving a mixture of independent and paired data

2021 ◽  
Author(s):  
Thomas R. Sullivan ◽  
Lisa N. Yelland ◽  
Margarita Moreno‐Betancur ◽  
Katherine J. Lee
Keyword(s):  
Multiple Imputation ◽  
Randomized Trials ◽  
Outcome Data ◽  
Paired Data

scholarly journals Should multiple imputation be the method of choice for handling missing data in randomized trials?

2016 ◽  
Vol 27 (9) ◽  
pp. 2610-2626 ◽  
Author(s):  
Thomas R Sullivan ◽  
Ian R White ◽  
Amy B Salter ◽  
Philip Ryan ◽  
Katherine J Lee
Keyword(s):  
Missing Data ◽  
Multiple Imputation ◽  
Treatment Effect ◽  
Randomized Trials ◽  
Missing At Random ◽  
Outcome Data ◽  
Average Treatment Effect ◽  
Analysis Model ◽  
Data Simulation ◽  
Using Data

The use of multiple imputation has increased markedly in recent years, and journal reviewers may expect to see multiple imputation used to handle missing data. However in randomized trials, where treatment group is always observed and independent of baseline covariates, other approaches may be preferable. Using data simulation we evaluated multiple imputation, performed both overall and separately by randomized group, across a range of commonly encountered scenarios. We considered both missing outcome and missing baseline data, with missing outcome data induced under missing at random mechanisms. Provided the analysis model was correctly specified, multiple imputation produced unbiased treatment effect estimates, but alternative unbiased approaches were often more efficient. When the analysis model overlooked an interaction effect involving randomized group, multiple imputation produced biased estimates of the average treatment effect when applied to missing outcome data, unless imputation was performed separately by randomized group. Based on these results, we conclude that multiple imputation should not be seen as the only acceptable way to handle missing data in randomized trials. In settings where multiple imputation is adopted, we recommend that imputation is carried out separately by randomized group.

Properties and pitfalls of weighting as an alternative to multilevel multiple imputation in cluster randomized trials with missing binary outcomes under covariate-dependent missingness

2019 ◽  
Vol 29 (5) ◽  
pp. 1338-1353
Author(s):  
Elizabeth L Turner ◽  
Lanqiu Yao ◽  
Fan Li ◽  
Melanie Prague
Keyword(s):  
Multiple Imputation ◽  
Randomized Trials ◽  
Simulated Data ◽  
Real Data ◽  
Cluster Randomized Trial ◽  
Outcome Data ◽  
Binary Outcomes ◽  
Cluster Randomized Trials ◽  
Cluster Randomized ◽  
Multilevel Multiple Imputation

The generalized estimating equation (GEE) approach can be used to analyze cluster randomized trial data to obtain population-averaged intervention effects. However, most cluster randomized trials have some missing outcome data and a GEE analysis of available data may be biased when outcome data are not missing completely at random. Although multilevel multiple imputation for GEE (MMI-GEE) has been widely used, alternative approaches such as weighted GEE are less common in practice. Using both simulations and a real data example, we evaluate the performance of inverse probability weighted GEE vs. MMI-GEE for binary outcomes. Simulated data are generated assuming a covariate-dependent missing data pattern across a range of missingness clustering (from none to high), where all covariates are measured at baseline and are fully observed (i.e. a type of missing-at-random mechanism). Two types of weights are estimated and used in the weighted GEE: (1) assuming no clustering of missingness (W-GEE) and (2) accounting for such clustering (CW-GEE). Results show that, even in settings with high missingness clustering, CW-GEE can lead to more bias and lower coverage than W-GEE, whereas W-GEE and MMI-GEE provide comparable results. W-GEE should be considered a viable strategy to account for missing outcomes in cluster randomized trials.

Using generalized estimating equations and extensions in randomized trials with missing longitudinal patient reported outcome data

2018 ◽  
Vol 27 (9) ◽  
pp. 2125-2131 ◽  
Author(s):  
Melanie L. Bell ◽  
Nicholas J. Horton ◽  
Haryana M. Dhillon ◽  
Victoria J. Bray ◽  
Janette Vardy
Keyword(s):  
Generalized Estimating Equations ◽  
Randomized Trials ◽  
Estimating Equations ◽  
Patient Reported Outcome ◽  
Outcome Data ◽  
Patient Reported ◽  
Generalized Estimating

Antithrombotic Therapy After Revascularization in Patients With Peripheral Arterial Disease: What Is Here, What Is Next

2019 ◽  
Vol 53 (4) ◽  
pp. 325-336 ◽  
Author(s):  
Mostafa R. Amer ◽  
Surya Teja Chaturvedula ◽  
Saurabh Joshi ◽  
Joseph Ingrassia
Keyword(s):  
Clinical Practice ◽  
Peripheral Arterial Disease ◽  
Antithrombotic Therapy ◽  
Randomized Trials ◽  
Arterial Disease ◽  
Outcome Data ◽  
Significant Heterogeneity ◽  
Treatment Protocols ◽  
Peripheral Arterial

Objective: The optimal antithrombotic regimen in peripheral arterial disease (PAD) is not known, leading to significant variations in antithrombotic treatment protocols in randomized trials and clinical practice. In device trials, antithrombotic regimens in patients receiving peripheral vascular interventions have not been clearly reported on. This review summarizes and discusses the most recent evidence on this topic to provide a potential guide to clinical practice. Methods: A search of the literature was done for publications that reported outcomes of major PAD device trials. Reported outcomes and various antithrombotic regimens were studied. Results: Use of antithrombotic therapy varied significantly between various device trials. Reporting of antithrombotic regimens at the time of follow-up is lacking. Conclusion: Outcome data on optimal antithrombotic regimens are presently lacking largely due to the significant heterogeneity and underreporting of antithrombotic regimens at follow-up among prior clinical trials. Standardization and reporting of precise antithrombotic regimens at various points of follow-up in device trials of patients with PAD should be attempted so as to minimize differences in treatment patterns when evaluating new devices.

scholarly journals Randomized trials with missing outcome data: how to analyze and what to report

2014 ◽  
Vol 186 (15) ◽  
pp. 1153-1157 ◽  
Author(s):  
Rolf H.H. Groenwold ◽  
Karel G.M. Moons ◽  
Jan P. Vandenbroucke
Keyword(s):  
Randomized Trials ◽  
Outcome Data
Sign in / Sign up

Export Citation Format

Share Document