Case number calculation in clinical studies

In a case number calculation, the sample size is determined which is necessary to be able to prove a previously determined medically relevant difference at a defined significance level α with a defined test strength. With the help of the sample size calculation, effects that actually exist should be recorded and at the same time there should be a certainty that no effect actually exists if no statistical difference has been calculated.

In order to prove significance, the sample size must therefore not be too small, otherwise no final conclusion can be drawn at the end of the study. On the other hand, too large a sample size is neither sensible nor ethically justifiable, but only ties up unnecessary resources.

According to GCP, case number calculations are required in studies, so that ethics commissions check their implementation in study protocols and in some cases also control them. Recently, more and more often a case number calculation is also required in doctoral theses, which is also controlled by the ethics commissions.

The calculation of case numbers requires both mathematical and medical knowledge. In principle, the more rarely an event occurs or the greater the spread (standard deviation) of samples, the larger the required case number. For a better assessment, these estimates should be based on a previous literature search.

Calculations of case numbers are usually carried out by statisticians using suitable software such as BiAS.