Estimated Sample Size Calculation in Flow Mediated Dilatation

Crossover Group Study

This routine is based on the work of Sorensen, et. al. (1995) on non-invasive measurement of human endothelium dependent arterial responses. The routine estimates the sample size needed for a parallel group trial design for 80% power and 95% confidence level. It should be noted that the Sorensen, et. al. (1995) dose not seem to follow the typical crossover design, as shown in the model below.

NOTE: Typical crossover design is defined as:

Y(ijk) = u + Subject(i) + Sequence(j) + Treatment(k) + e(ijk)

Where i = 1, 2, ... n; j = 1,2,...,J; and k = 1,2,...,K.

These Estimates are based on a single sequence of

subjects studied once before and after each treatment

Enter your expected %FMD difference:

Smallest %FMD Difference

Estimated Sample Size for crossover design

Reference

Sorensen, K. E., Celermajer, D. S., Spiegelhalter, D. J., Georgakopoulos, D., Robinson, J., Thomas, O., and Deanfield, J. E. (1995). Non-invasive measurement of human endothelium dependent arterrial responses: Accuracy and reproducibility. British Heart Journal. 74, 247-253.

Programmed by Medical Research Consulting - 1998