Skip to main content

Table 5 Total number of households (children)a required for 80% and 90% power for varying control arm event rates and intra-cluster correlation coefficients

From: Levofloxacin versus placebo for the prevention of tuberculosis disease in child contacts of multidrug-resistant tuberculosis: study protocol for a phase III cluster randomised controlled trial (TB-CHAMP)

Power

Contacts who develop TB disease by 48 weeks (control arm) (%)

Intra-cluster correlation coefficients

0.05

0.1

0.15

80%

5

1056 (2122)

1108 (2216)

1158 (2316)

7

742 (1484)

778 (1556)

814 (1628)

10

508 (1016)

532 (1064)

556 (1112)

90%

5

1414 (2828)

1480 (2960)

1548 (3096)

7

994 (1988)

1040 (2080)

1088 (2174)

10

678 (1356)

710 (1420)

742 (1484)

  1. aAssumes 2 contacts per household, 10% loss to follow-up, two-sided 5% significance level test and a 50% risk reduction in the intervention arm
  2. The number in italics indicates the chosen trial sample size