Demystifying sample-size calculation for clinical trials and comparative effectiveness research: the impact of low-event frequency in surgical clinical research

Type II error, or not meeting sample-size requirement, has been identified as an issue in the surgical literature. The root of this problem is the low frequency of events in the majority of surgical clinical research. This exponentially increases the sample size needed to achieve statistical significance.

The methodology and mechanics of sample-size calculations are presented to demonstrate how sample-size requirements change based on baseline event rate and relative reduction in event rate. These concepts are then illustrated using real-life clinical scenarios.

If a hypothetical baseline event rate is 1 % and the event rate in the experimental group is 0.5 % (therefore representing a 50 % reduction), then the total number of study patients required is 10,130. If the baseline event rate is 1 %, and the event rate in the experimental group is 0.9 % (a 10 % reduction), then the total number of study patients required is 299,410.

Sample-size calculations are affected by the frequency of the event or events of interest. Given advances in clinical medicine, many clinical outcomes of interest occur at very low frequencies. These low frequencies exponentially increase the sample size required to find statistically significant differences, making randomized clinical trials difficult to conduct properly. Surgical clinical researchers should advocate for the establishment of robust, prospective, large, multi-institutional clinical databases along with the establishment of proper outcomes research methodology as a way to augment randomized trials.