Table 1.

Notation and simple algebraic relationships

	Parallel CRT	CRT-BA	SW-CRT (with t steps)
Number of measurement times	1	2	t + 1
Sample size per cluster per measurement time	M	M/2	m = M/(t + 1)
Total sample size per clustera	M	M	M = m × (t + 1)
Total study sample size	N = M × k	N = M × k	N = m × (t + 1) × k
Total number of clustersb	k = N/M	k = N/M	k = N/[m × (t + 1)]
Number of clusters per step	Not applicable	Not applicable	g = k/t

Abbreviations: CRT, cluster randomized trial; CRT-BA, cluster randomized trial with before and after observations; SW-CRT, stepped wedge cluster randomized trial.

^aFixed under design scenario A.

^bFixed under design scenario B.