Universal gestational age effects on cognitive and basic mathematic processing: 2 cohorts in 2 countries

Abstract

Objective: To determine whether general cognitive ability, basic mathematic processing, and mathematic attainment are universally affected by gestation at birth, as well as whether mathematic attainment is more strongly associated with cohort-specific factors such as schooling than basic cognitive and mathematical abilities.

Study design: The Bavarian Longitudinal Study (BLS, 1289 children, 27-41 weeks gestational age [GA]) was used to estimate effects of GA on IQ, basic mathematic processing, and mathematic attainment. These estimations were used to predict IQ, mathematic processing, and mathematic attainment in the EPICure Study (171 children <26 weeks GA).

Results: For children born <34 weeks GA, each lower week decreased IQ and mathematic attainment scores by 2.34 (95% CI: -2.99, -1.70) and 2.76 (95% CI: -3.40, -2.11) points, respectively. There were no differences among children born 34-41 weeks GA. Similarly, for children born <36 weeks GA, mathematic processing scores decreased by 1.77 (95% CI: -2.20, -1.34) points with each lower GA week. The prediction function generated using BLS data accurately predicted the effect of GA on IQ and mathematic processing among EPICure children. However, these children had better attainment than predicted by BLS.

Conclusions: Prematurity has adverse effects on basic mathematic processing following birth at all gestations <36 weeks and on IQ and mathematic attainment <34 weeks GA. The ability to predict IQ and mathematic processing scores from one cohort to another among children cared for in different eras and countries suggests that universal neurodevelopmental factors may explain the effects of gestation at birth. In contrast, mathematic attainment may be improved by schooling.

Figure 1

Mathematics Estimation Test example items.

Figure 4

Observed and predicted mean change of outcomes according to GA at birth in the EPICure Study (23-25 weeks GA) and BLS (27-41 weeks GA). Blue vertical lines: 95% CIs of observed means (circles: BLS and squares: EPICure); X: GA change points; black solid horizontal lines: predicted means below the GA change point; dashed horizontal lines: predicted means above the GA change point.

Figure 5

Observed and predicted mean change of mathematic attainment scores only for children in mainstream schools according to GA at birth in the BLS (N = 1094; 27-41 weeks GA) and in the EPICure Study (N = 165; 23-25 weeks GA). Blue vertical lines: 95% CIs of observed means (circles: BLS and squares: EPICure); X: GA change points; black solid horizontal lines: predicted means below the GA change point; dashed horizontal lines: predicted means above the GA change point.

Figure 2

Observed and predicted mean change in outcomes according to GA at birth in the BLS (Germany; 27-41 weeks GA). Grey vertical lines: 95% CIs of observed means (circles); X: GA change points; black solid horizontal lines: predicted means below the GA change point; dashed horizontal lines: predicted means above the GA change point.

Figure 3

EPICure Study observed score distributions (box plots^∗) with predicted mean scores (solid lines) and 50% (dashed lines) and 75% (dotted lines) prediction intervals based on the BLS cohort. Precision of prediction was examined by calculating the percentiles of box plots within prediction intervals and show that observed IQ and basic mathematic processing scores were mostly covered within the 50% prediction interval. ^∗The bottom and top of each box are the 25th and 75th percentiles of observed scores, respectively. The line in the middle is the 50th percentile and hollow circles are observed mean scores.