Sample size calculations in randomised clinical trials: beyond the basics

Potential Attendees

Those working on trials and study designs, which includes trialists, trial statisticians, clinicians and other scientists. Familiarity with the hypothesis testing framework (i.e., type I error rate, power, and various effect sizes) are essential. Familiarity with sample size calculation for simple/basic designs is desirable but not essential.

Learning objectives

This course aims to help participants:

• Understand the underlying statistical principles of sample-size calculations from the hypothesis testing framework.
• Understand the practical issues in calculating sample sizes.
• Know and use different sample-size formulae for different outcome types: continuous, binary, time-to-event, and ordinal outcomes.
• Learn about the projection of power and events in trials with time-to-event outcomes based on realistic trial design scenarios.
• Know about the nuisance parameters and how to specify them in different designs.
• Know about sample size calculations for complex designs:
  • Factorial designs.
  • Non-inferiority designs, including DURATIONS designs.
  • Cluster randomised trial designs.
  • Group sequential designs.
• Know about blinded and unblinded sample size re-estimation.
• Know how to use the programs, written in R and Stata, to calculate the sample size for the above design types.
• Understand how to use simulation to calculate power / sample size.

Sample size calculation for trial design

Sample size calculations arise in planning a study. It is one of the most important aspects of any design since it drives study timelines and the cost of undertaking the protocol. It determines the degree of required information to robustly answer the primary research questions in the study. If the sample size is too small, the study is likely to be inconclusive. Furthermore, increasing sample size during the study can be expensive and time-consuming. Therefore, it is imperative to accurately estimate the required sample size and trial timelines from the outset, and particularly for any funding application.

This course builds on a basic course in sample-size calculations such as the one provided by the ICH, and provides an overview of the underlying statistical theory for sample size calculations within the hypothesis testing framework. It addresses practical issues and statistical considerations when calculating sample sizes for a wide range of advanced trial designs and outcome distributions, including factorial, non-inferiority, DURATIONS, group sequential, and clustered randomised clinical trials. It uses real advanced trial examples to calculate the sample size using the available user-written software in both R and Stata.  

Structure of the course

This one-day course consists of five lectures. The first lecture provides the statistical theory for sample size calculations and covers the methods for continuous and binary outcomes. The second session extends the theory to trials with time-to-event outcomes and presents methods for the prediction of power and trial timelines in complex and realistic scenarios with non-uniform recruitment and follow-up patterns. The third and fourth lectures present methods for sample size calculations with more complex designs, including factorial and cluster, and group sequential trial designs. The final lecture covers sample size calculation methods for non-inferiority designs and presents methods on how to deal with uncertainties in design parameters when calculating the sample size as well as the application of simulations in calculating sample size for more complex designs.

Software and examples

In all lectures, the methods are explored using real trial examples, and using programs R and Stata, including the power and Stata art suites (artbin, artsurv, artcat, and artpep).
Each lecture concludes with a computer practical when participants have the opportunity to implement the methods. For this reason, we encourage those planning to attend the course to bring their own laptop.

Fees

This course is free for staff from Units within ICTM (CRUK-CTC, CCTU, MRC CTU at UCL and PRIMENT) although places are limited. Places on this course are limited, If demand exceeds the number of places, places will be awarded to those who the course presenters believe will benefit most from the course. Fees are based on the attendee's organisation as follows:

• £200 UCL staff, students, alumni; places are limited
• £380 External delegates (non UCL) 

The ICTM CPD Administrator will advise successful applicants of the location and fee payment process.

If you have any questions about the course, please email ictm.cpd@ucl.ac.uk