Cardiorespiratory fitness as a mediator of the relationship between birth weight and cognition in school children

Figures & data

Table 1 Characteristics of the study sample

	Total (n=664)	Boys (n=321)	Girls (n=343)	P-value
Age (months)	75.19 (3.45)	75.12 (3.57)	75.22 (3.36)	0.129
Height (cm)	118.64 (5.20)	119.25 (5.19)	118.09 (5.10)	0.046
Weight (Kg)	22.95 (5.00)	23.33 (5.22)	22.61 (4.76)	0.161
BMI (Kg/m^2)	16.19 (2.68)	16.28 (2.77)	16.10 (2.61)	0.492
Birth weight (g)	3279.69 (508.88)	3319.87 (544.74)	3243.72 (472.43)	0.042
CRF	49.85 (2.87)	50.44 (3.30)	49.30 (2.29)	<0.000
Logic reasoning	27.26 (8.75)	27.15 (8.89)	27.22 (8.56)	0.370
Verbal factor	21.70 (5.78)	21.42 (5.89)	21.95 (5.66)	0.191
Numerical factor	17.24 (8.16)	17.81 (8.69)	16.64 (7.68)	0.012
Spatial factor	15.78 (6.56)	15.69 (6.63)	15.76 (6.45)	0.627
General intelligence	54.92 (16.82)	55.08 (17.62)	54.60 (16.07)	0.063

Notes: Data presented as mean (SD). CRF indicates VO₂ max values obtained with Leger’s formula. Boldface type indicates statistical significance p<0.05.

Abbreviations: BMI, Body mass index; CRF, Cardiorespiratory fitness.

Table 2 Pearson correlation coefficients (r) of cognition with birth weight and CRF parameters, controlling for age (in months) and sex

		Logic reasoning	Verbal factor	Numerical factor	Spatial factor	General intelligence
Birth weight	Total	0.12*	0.08	0.09	0.12*	0.11*
	Boys	0.09	0.04	0.06	0.13	0.10
	Girls	0.14*	0.11	0.06	0.09	0.11
CRF	Total	0.27**	0.21**	0.23**	0.18**	0.26**
	Boys	0.26**	0.23**	0.23**	0.14*	0.25**
	Girls	0.24**	0.19**	0.13*	0.21**	0.21**

Notes: *p<0.05; **p<0.001.

Abbreviation: CRF, cardiorespiratory fitness.

Table 3 Mean differences (SE) in cognition variables by cardiorespiratory fitness and birth weight categories

		Cardiorespiratory fitness				Birth weight
		Poor	Medium	Good	P-value	Low birth weight	Normal birth weight	P-value
Logic reasoning	Model 1	26.2 (0.7)^M-G	27.0 (0.7)^G	30.5 (0.8)	<0.001	24.5 (8.6)	27.7 (8.8)	0.036
	Model 2	26.4 (0.7)^M-G	27.0 (0.6)^G	30.3 (0.8)	<0.001	25.5 (9.2)	27.8 (8.8)	0.135
Verbal factor	Model 1	21.1 (0.4)	21.9 (0.4)	23.6 (0.5)	0.001	20.4 (5.5)	22.0 (6.6)	0.084
	Model 2	21.2 (0.4)^M	21.9 (0.4)	23.5 (0.5)	0.002	21.4 (5.5)	22.1 (4.4)	0.434
Numerical factor	Model 1	16.4 (0.6)^M	16.9 (0.6)	19.8 (0.7)	0.001	15.3 (8.0)	17.5 (8.8)	0.119
	Model 2	16.5 (0.6)^M	16.9 (0.6)	19.7 (0.7)	0.002	16.7 (8.6)	17.5 (8.8)	0.519
Spatial factor	Model 1	15.7 (0.5)^M-G	15.4 (0.5)^G	17.1 (0.6)	0.059	13.3 (6.8)	16.1 (6.6)	0.011
	Model 2	15.9 (0.5)^M-G	13.4 (0.4)^G	17.0 (0.6)	0.095	13.7 (6.8)	16.1 (6.6)	0.038
General intelligence	Model 1	53.4 (1.3)^M	54.3 (1.1)	60.8 (1.5)	<0.001	49.7 (17.3)	55.7 (15.4)	0.038
	Model 2	53.7 (1.3)^M	54.3 (1.1)	60.4 (1.5)	0.001	52.6 (17.3)	55.9 (15.4)	0.261

Notes: Model 1: Age (months), sex, socioeconomic status and body mass index. Model 2: Model 1+ Birth weight or CRF depending on fixed factor. Boldface type indicates statistical significance P<0.05. Pairwise mean comparisons using the Bonferroni post hoc test were statistically significant except for superscript letters (M-G, Medium-Good; M, Medium; P<0.05).

Abbreviations: M-G, medium-good; G, good; M, medium.

Figure 1 Mediation models testing whether the relationship between birth weight and cognition variables was mediated by cardiorespiratory fitness (CRF), controlling for age (months), sex, socioeconomic status and body mass index.

Notes: *p<0.05; **p<0.001.

Abbreviations: IE (95%CI), Indirect Effect (95% Confidence Interval); %Med, percentage mediated by proposed mediator. Symbols.