Methods to adjust for multiple comparisons in the analysis and sample size calculation of randomised controlled trials with multiple primary outcomes

Table 5 Sample size per group, assuming four outcomes, 90% disjunctive power, after applying a Bonferroni correction

Standardised effect sizes for each of the 4 outcomes				Sample size required to obtain 90% DISJUNCTIVE power				Sample size required to obtain 90% MARGINAL power for each outcome
Standardised effect sizes for each of the 4 outcomes				Correlation between outcomes
Out.^a 1	Out. 2	Out. 3	Out. 4	0.2	0.4	0.6	0.8	Out. 1	Out. 2	Out. 3	Out. 4
0.2	0.2	0.2	0.2	325	382	447	529	716	716	716	716
0.2	0.2	0.3	0.3	189	215	242	270	716	716	319	319
0.2	0.2	0.4	0.4	114	127	129	152	716	716	181	181
0.2	0.2	0.5	0.5	75	82	89	98	716	716	116	116
0.3	0.3	0.3	0.3	145	170	199	235	319	319	319	319
0.3	0.3	0.4	0.4	101	117	133	151	319	319	181	181
0.3	0.3	0.5	0.5	71	80	88	98	319	319	116	116
0.4	0.4	0.4	0.4	82	96	112	133	181	181	181	181
0.4	0.4	0.5	0.5	63	73	84	96	181	181	116	116
0.5	0.5	0.5	0.5	52	61	72	85	116	116	116	116

Sample sizes provided are required per arm. A Bonferroni correction is applied for all calculations to account for the multiple comparisons. Key: ^a‘Out’ Outcome

ISSN: 1471-2288