Skip to main content

Table 3 Approximate maximum intra-class correlation (ICC) for various cluster sizes and number of intervention hospitals to answer primary hypothesis

From: What is the effect of a formalised trauma tertiary survey procedure on missed injury rates in multi-trauma patients? Study protocol for a randomised controlled trial

Number of intervention (and control) hospitals

Average size of cluster

Maximum ICC to have 80% power to detect 5% difference

8

400

0.01

15

400

0.02

21

400

0.03

9

300

0.01

15

300

0.02

22

300

0.03

10

200

0.01

16

200

0.02

23

200

0.03