Skip to main content
. 2016 Mar 21;11(3):e0151819. doi: 10.1371/journal.pone.0151819

Table 3. Proportion of children in each group scoring < -2 SD on cognitive parameters of EF.

Total population NSI-free and FSIQ > 70
EPT Control Adjusted ORa (95% CI) P value EPT Control Adjusted ORa (95% CI) P value
N = 128 N = 103 N = 98 N = 99
Verbal conceptual reasoning (Sim) 17 (13) 0 NA < 0.001b 2 (2) 0 NA 0.24b
Verbal working memory (Ds) 26 (20) 2 (2) 12.8 (3–56) < 0.001 7 (7) 2 (2) 3.6 (.7–19) 0.1
Non-verbal memory, attention (Co) 26 (20) 3 (3) 10.0 (2.9–35.0) < 0.001 8 (8) 2 (2) 5.5 (1.1–27) 0.039
Memory, attention, distractibility (Ar) 34 (27) 4 (4) 9.1 (3.1–6.2) < 0.001 13 (13) 2 (2) 7.9 (1.7–37) 0.008
Spatial conceptualization (Bd) 34 (27) 2 (2) 18.0 (4–77) < 0.001 14 (14) 0 NA < 0.001b
Visual reasoning (Pa) 25 (20) 5 (5) 4.7 (1.8–12.7) 0.003 8 (8) 4 (4) 2.1 (.6–7.3) 0.24
Planning ability (Tower test) 26 (20) 1 (1) 26.0 (3.4–192) < 0.002 7 (7) 0 NA 0.007b

EPT, extremely preterm (23–25 weeks); NSI, neurosensory impairment; FSIQ, full-scale intelligence quotient; Sim, similarities; Ds, digit span; Bd, block design; Pa, picture arrangement; Co, coding; Ar, arithmetic; OR, odds ratio; CI, confidence interval; NA, not appropriate for calculation due to wide confidence intervals for exponents.

The values are n (%) unless otherwise stated. The data were adapted from core subtests of the WISC-III-R (Swedish edition, Wechsler 1999) [31] and the Tower test of the Delis-Kaplan Executive Function Scale [30].

a The ORs were derived from logistic regression analyses adjusted for the children’s gender and social risk (any vs none, see ‘Methods‘) and mother’s country of birth (Nordic vs non-Nordic).

b P values were determined by Fisher’s exact test for the EPT children vs control subjects.