Skip to main content

Advertisement

Table 2 Any hip treatment in the UK Hip trial

From: Assessing subgroup effects with binary data: can the use of different effect measures lead to different conclusions?

Level of clinical suspicion Ultrasound No Ultrasound Risk ratio for harm (95% CI) Risk ratio for benefit (95% CI) Odds ratio (95% CI) Risk difference (95% CI)
All 126/314 (40%) 159/315 (50%) 0.79 (0.67 to 0.95) 1.21 (1.05 to 1.40) 0.66 (0.48 to 0.90) -0.10 (-0.18 to -0.03)
Strong 65/95 (68%) 86/89 (97%) 0.71 (0.61 to 0.82) 9.37 (2.96 to 29.62) 0.08 (0.02 to 0.24) -0.28 (-0.38 to -0.18)
Moderate 61/219 (28%) 73/226 (32%) 0.86 (0.65 to 1.15) 1.07 (0.94 to 1.20) 0.81 (0.54 to 1.21) -0.04 (-0.13 to 0.04)
Test of
interaction1
   P = 0.14 P < 0.001 P < 0.001 P < 0.001
  1. 1 Mantel-Haenszel test.