Table 4.

Example: required total sample sizes per cluster (M) and total study size (N) under each design given fixed number of clusters k and ICC (ρ)

Design constraints			CRT		CRT-BA		SW-CRT
k	ρ	t	M	N	M	N	M	N
30	0.01	2	36	1,080	66	1,980	96	2,880
60	0.01	5	15	900	30	1,800	30	1,800
30	0.25	2	Infeasible (min number clusters 197)		76	2,280	90	2,700
60	0.25	5	38	2,280	30	1,800

Abbreviations: ICC, intracluster correlation coefficient; t, number of steps in the stepped wedge design; CRT, cluster randomized trial; CRT-BA, cluster randomized trial with before and after observations; SW-CRT, stepped wedge cluster randomized trial.

Example relates to a trial requiring 788 observations under individual randomization. Note that both M and N are rounded up to the nearest integer.