Dataset Information

Missing data and sensitivity analysis for binary data with implications for sample size and power of randomized clinical trials.

ABSTRACT: Despite our best efforts, missing outcomes are common in randomized controlled clinical trials. The National Research Council's Committee on National Statistics panel report titled The Prevention and Treatment of Missing Data in Clinical Trials noted that further research is required to assess the impact of missing data on the power of clinical trials and how to set useful target rates and acceptable rates of missing data in clinical trials. In this article, using binary responses for illustration, we establish that conclusions based on statistical analyses that include only complete cases can be seriously misleading, and that the adverse impact of missing data grows not only with increasing rates of missingness but also with increasing sample size. We illustrate how principled sensitivity analysis can be used to assess the robustness of the conclusions. Finally, we illustrate how sample sizes can be adjusted to account for expected rates of missingness. We find that when sensitivity analyses are considered as part of the primary analysis, the required adjustments to the sample size are dramatically larger than those that are traditionally used. Furthermore, in some cases, especially in large trials with small target effect sizes, it is impossible to achieve the desired power.

SUBMITTER: Cook T

PROVIDER: S-EPMC6916420 | biostudies-literature | 2020 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Missing data and sensitivity analysis for binary data with implications for sample size and power of randomized clinical trials.

Cook Thomas T Zea Ryan R

Statistics in medicine 20191114 2

Despite our best efforts, missing outcomes are common in randomized controlled clinical trials. The National Research Council's Committee on National Statistics panel report titled The Prevention and Treatment of Missing Data in Clinical Trials noted that further research is required to assess the impact of missing data on the power of clinical trials and how to set useful target rates and acceptable rates of missing data in clinical trials. In this article, using binary responses for illustrati ...[more]

PMID: 31726480

Similar Datasets

Project description:Background/aimsGeneralized estimating equations are commonly used to fit logistic regression models to clustered binary data from cluster randomized trials. A commonly used correlation structure assumes that the intracluster correlation coefficient does not vary by treatment arm or other covariates, but the consequences of this assumption are understudied. We aim to evaluate the effect of allowing variation of the intracluster correlation coefficient by treatment or other covariates on the efficiency of analysis and show how to account for such variation in sample size calculations.MethodsWe develop formulae for the asymptotic variance of the estimated difference in outcome between treatment arms obtained when the true exchangeable correlation structure depends on the treatment arm and the working correlation structure used in the generalized estimating equations analysis is: (i) correctly specified, (ii) independent, or (iii) exchangeable with no dependence on treatment arm. These formulae require a known distribution of cluster sizes; we also develop simplifications for the case when cluster sizes do not vary and approximations that can be used when the first two moments of the cluster size distribution are known. We then extend the results to settings with adjustment for a second binary cluster-level covariate. We provide formulae to calculate the required sample size for cluster randomized trials using these variances.ResultsWe show that the asymptotic variance of the estimated difference in outcome between treatment arms using these three working correlation structures is the same if all clusters have the same size, and this asymptotic variance is approximately the same when intracluster correlation coefficient values are small. We illustrate these results using data from a recent cluster randomized trial for infectious disease prevention in which the clusters are groups of households and modest in size (mean 9.6 individuals), with intracluster correlation coefficient values of 0.078 in the control arm and 0.057 in an intervention arm. In this application, we found a negligible difference between the variances calculated using structures (i) and (iii) and only a small increase (typically <5%) for the independent correlation structure (ii), and hence minimal effect on power or sample size requirements. The impact may be larger in other applications if there is greater variation in the ICC between treatment arms or with an additional covariate.ConclusionThe common approach of fitting generalized estimating equations with an exchangeable working correlation structure with a common intracluster correlation coefficient across arms likely does not substantially reduce the power or efficiency of the analysis in the setting of a large number of small or modest-sized clusters, even if the intracluster correlation coefficient varies by treatment arm. Our formulae, however, allow formal evaluation of this and may identify situations in which variation in intracluster correlation coefficient by treatment arm or another binary covariate may have a more substantial impact on power and hence sample size requirements.

Project description:BackgroundWith the increasing number of expression profiling technologies, researchers today are confronted with choosing the technology that has sufficient power with minimal sample size, in order to reduce cost and time. These depend on data variability, partly determined by sample type, preparation and processing. Objective measures that help experimental design, given own pilot data, are thus fundamental.ResultsRelative power and sample size analysis were performed on two distinct data sets. The first set consisted of Affymetrix array data derived from a nutrigenomics experiment in which weak, intermediate and strong PPARalpha agonists were administered to wild-type and PPARalpha-null mice. Our analysis confirms the hierarchy of PPARalpha-activating compounds previously reported and the general idea that larger effect sizes positively contribute to the average power of the experiment. A simulation experiment was performed that mimicked the effect sizes seen in the first data set. The relative power was predicted but the estimates were slightly conservative. The second, more challenging, data set describes a microarray platform comparison study using hippocampal deltaC-doublecortin-like kinase transgenic mice that were compared to wild-type mice, which was combined with results from Solexa/Illumina deep sequencing runs. As expected, the choice of technology greatly influences the performance of the experiment. Solexa/Illumina deep sequencing has the highest overall power followed by the microarray platforms Agilent and Affymetrix. Interestingly, Solexa/Illumina deep sequencing displays comparable power across all intensity ranges, in contrast with microarray platforms that have decreased power in the low intensity range due to background noise. This means that deep sequencing technology is especially more powerful in detecting differences in the low intensity range, compared to microarray platforms.ConclusionPower and sample size analysis based on pilot data give valuable information on the performance of the experiment and can thereby guide further decisions on experimental design. Solexa/Illumina deep sequencing is the technology of choice if interest lies in genes expressed in the low-intensity range. Researchers can get guidance on experimental design using our approach on their own pilot data implemented as a BioConductor package, SSPA http://bioconductor.org/packages/release/bioc/html/SSPA.html.

Dataset Information

Missing data and sensitivity analysis for binary data with implications for sample size and power of randomized clinical trials.

Publications

Missing data and sensitivity analysis for binary data with implications for sample size and power of randomized clinical trials.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets