Table 4.

Multiplication factors for the required sample sizes accounting for clustering at group level

	52 visits	20 visits	12 visits	6 visits	4 visits
Model scenario 1 (LS)
no clustering	1.0	1.4	1.9	3.3	4.8
Cluster size n = 5
ICC = 0.04	1.0	1.4	1.8	3.0	4.3
ICC = 0.1	1.0	1.3	1.7	2.7	3.8
ICC = 0.3	1.0	1.2	1.4	2.1	2.8
Cluster size n = 50
ICC = 0.005	1.0	1.3	1.8	2.9	4.2
ICC = 0.02	1.0	1.2	1.5	2.2	3.0
ICC = 0.1	1.0	1.1	1.2	1.4	1.7
Model scenario 2 (LL)
no clustering	1.0	1.1	1.4	2.2	2.9
Cluster size n = 5
ICC = 0.04	1.0	1.1	1.4	2.0	2.7
ICC = 0.1	1.0	1.1	1.3	1.8	2.4
ICC = 0.3	1.0	1.1	1.2	1.5	1.9
Cluster size n = 50
ICC = 0.005	1.0	1.1	1.3	1.9	2.6
ICC = 0.02	1.0	1.1	1.2	1.6	2.0
ICC = 0.1	1.0	1.0	1.1	1.2	1.4
Model scenario 3 (HS)
no clustering	1.0	1.1	1.3	1.7	2.1
Cluster size n = 5
ICC = 0.04	1.0	1.1	1.2	1.6	2.0
ICC = 0.1	1.0	1.1	1.2	1.5	1.8
ICC = 0.3	1.0	1.1	1.1	1.3	1.5
Cluster size n = 50
ICC = 0.005	1.0	1.1	1.2	1.6	1.9
ICC = 0.02	1.0	1.1	1.2	1.4	1.6
ICC = 0.1	1.0	1.0	1.1	1.2	1.3
Model scenario 4 (HL)
no clustering	1.0	1.1	1.2	1.5	1.8
Cluster size n = 5
ICC = 0.04	1.0	1.1	1.2	1.4	1.7
ICC = 0.1	1.0	1.0	1.1	1.3	1.6
ICC = 0.3	1.0	1.0	1.1	1.2	1.4
Cluster size n = 50
ICC = 0.005	1.0	1.1	1.2	1.4	1.6
ICC = 0.02	1.0	1.0	1.1	1.2	1.4
ICC = 0.1	1.0	1.0	1.0	1.1	1.2

The sample size is calculated for a study comparing the mean LP between two groups with a 20% LP reduction in one arm (80% power and p = 0.05). Sampling was simulated as 7 day point prevalence recall at each visit (see Table 3.).