Skip to main content
. 2021 May 6;35(6):681–690. doi: 10.1007/s40263-021-00817-w

Table 5.

Subgroup interaction analyses for recombinant human erythropoietin on Mental Development Index < 70

Subgroups Placebo rhEPO RR (95% CI) Interaction p value
Sex
 Male 12/75 (16.0) 7/72 (9.7) 0.565 (0.209–1.528)
 Female 7/49 (14.3) 2/53 (3.8) 0.235 (0.046–1.193)
 Total 0.429 (0.186–0.989) 0.361
Gestational age
 < 28 weeks 2/15 (13.3) 3/19 (15.8) 1.219 (0.176–8.423)
 28–296/7 weeks 9/64 (14.1) 2/60 (3.3)* 0.211 (0.044–1.019)
 30–32 weeks 8/45 (17.8) 4/46 (8.7) 0.440 (0.123–1.583)
 Total 0.429 (0.186–0.989) 0.366
Birth weight, g
 < 1000 1/12 (8.3) 2/15 (13.3) 1.692 (0.135–21.270)
 1000–1499 14/93 (15.1) 5/91 (5.5)* 0.328 (0.113–0.952)
 ≥ 1500 4/19 (21.1) 2/19 (10.5) 0.441 (0.070–2.761)
 Total 0.429 (0.186–0.989) 0.470
Degree of IVH
 Grade I–II 14/112 (12.5) 7/120 (5.8) 0.434 (0.168–1.118)
 Grade III–IV 5/12 (41.7) 2/5 (40.0) 0.933 (0.111–7.820)
 Total 0.429 (0.186–0.989) 0.516

Data are presented as n/N (%) unless otherwise indicated. The p value is for the interaction analysis in subgroups using the Mantel–Haenszel test. p < 0.05 was considered statistically significant

CI confidence interval, IVH intraventricular hemorrhage, rhEPO recombinant human erythropoietin, RR relative risk

*p < 0.05 using Fisher’s exact test to compare the placebo and rhEPO groups