Sample Size for Regression Models
Calculations for Common Regression, Mixed-Effects and Prediction Models
In this tutorial, we have reviewed conventional regression models such as linear regression, logistic regression and Cox regression and examined the use of mixed-effects models in cluster randomized trials.
You will learn about:
-
Regression Models and Sample Size Calculation for Common Regression Models
-
Sample Size Determination for Cluster Randomized Designs using Mixed-effects Models
-
How to Compute the Minimum Sample Size for Clinical Prediction Models
We investigate sample size determination for these models and contrast their benefits and limitations in different scenarios. We also explore methods for calculating the minimum sample size for clinical prediction models.
Sample Size for Regression Models
Regression models are types of statistical models often used in clinical trials to describe the relationship between an outcome variable and one or more predictor variable(s).
Simple linear regression, multiple linear regression, logistic regression and Cox regression are all examples of regression models that can be used in various scenarios. We investigate practical examples showing how to calculate the sample size when these models are used.
Mixed-effects or hierarchical models are another example of regression models typically used in clinical trials, particularly in the analysis of Cluster randomized trials (CRTs). CRTs are a type of experimental design which can be more efficient and cost-effective than individually randomized trials, leading to smaller sample sizes and reduced variance in outcome, however they come with their own challenges and limitations.
We have explored the use of Mixed-effects models in these scenarios and provided practical examples of how to determine the sample size for these models.
Minimum sample size requirements for clinical prediction models in the past have often been set using guidelines such as the 10 subjects per predictor variable rule-of-thumb. In recent years many methods have been developed in order to provide a more rigorous approach to determining the minimum sample size for clinical prediction models.
We have provided an analysis of the various regression models that can be used for clinical trials and discussed how the sample size can be computed for each. We have also discussed the benefits and limitations of each model and investigated methods for computing the minimum sample size for clinical prediction models.
nQuery helps make your clinical trials faster, less costly and more successful with tools for sample size calculation - refining Frequentist, Bayesian & Adaptive designs.
Start your unlimited 14-day trial to help design efficient, informative and ethical adaptive trials.
