Skip to main content
. 2018 Feb 19;172(4):361–367. doi: 10.1001/jamapediatrics.2017.5323

Table. Results of Univariate Meta-Regression Analyses of the Association Between Perinatal and Demographic Covariates and Intelligence Scores.

Covariate κ Value Change in IQ Points per Unit Change in Covariate (95% CI)a R2 Value
GA, wk 70 1.26 (0.52 to 2.00)b 23.94
BW, g 68 0.02 (0.003 to 0.02)c 19.84
SGA, % 50 −0.04 (−0.15 to 0.07) 0.79
IVH grade I/II, % 50 −0.17 (−0.31 to −0.03)d 13.07
IVH grade III/IV, % 54 −0.26 (−0.54 to 0.03) 9.15
PVL, % 46 −0.26 (−0.48 to −0.04)d 13.08
BPD, % 49 −0.15 (−0.19 to −0.09)b 64.99
NEC, % 41 0.06 (−0.13 to 0.24) 0.26
Sepsis, % 38 −0.05 (−0.12 to 0.03) 1.36
Postnatal corticosteroid use, % 31 −0.09 (−0.17 to −0.01)d 32.68
Sex, % male 66 0.10 (−0.05 to 0.26) 1.62
Race/ethnicity, % nonwhite 33 −0.001 (−0.11 to 0.09) 0.17
Maternal education <12 y, % 15 −0.03 (−0.20 to 0.07) 0

Abbreviations: BPD, bronchopulmonary dysplasia; BW, birth weight; GA, gestational age; IVH, intraventricular hemorrhage; NEC, necrotizing enterocolitis; PVL, periventricular leukomalacia; SGA, small for gestational age.

a

These values indicate the change in IQ points for each unit change in the covariate. For example, a 1-week change in GA is associated with a 1.26-point change in IQ. A negative value indicates a negative association (ie, a decrease in the covariate is associated with an increase in IQ and vice versa). For example, a decrease of 1% in the rate of BPD is associated with an increase in IQ of 0.15 points.

b

P < .001.

c

P < .01.

d

P < .05.