Skip to main content
. 2023 Apr 14;33(3):833–845. doi: 10.1007/s00787-023-02197-6

Table 4.

Associations of combined physically active early morning patterns with tract-specific white matter microstructure indicators

FA RD
N β (95% CI) P β (95% CI) P
Cingulate gyrus part of cingulum
 0 pattern 28 Ref Ref
 1 pattern 47 − 0.007 (− 0.249, 0.235) 0.956 − 0.003 (− 0.246, 0.246) 0.982
 2 patterns 14 0.031 (− 0.210, 0.271) 0.801 − 0.095 (− 0.355, 0.150) 0.433
Corticospinal tract
 0 pattern 35 Ref Ref
 1 pattern 53 0.072 (− 0.144, 0.288) 0.512 − 0.076 (− 0.266, 0.133) 0.485
 2 patterns 15 0.157 (− 0.059, 0.373) 0.153 − 0.186 (− 0.386, 0.029) 0.090
Inferior longitudinal fasciculus
 0 pattern 35 Ref Ref
 1 pattern 53 0.034 (− 0.187, 0.255) 0.760 − 0.029 (− 0.205, 0.161) 0.786
 2 patterns 15 0.150 (− 0.071, 0.372) 0.181 − 0.193 (− 0.410, 0,024) 0.077
Superior longitudinal fasciculus
 0 pattern 35 Ref Ref
 1 pattern 53 − 0.001 (− 0.209, 0.206) 0.989 − 0.010 (− 0.216, 0.197) 0.923
 2 patterns 15 0.314 (0.104, 0.523) 0.004* − 0.234 (− 0.441, − 0.028) 0.032
Uncinate fasciculus
 0 pattern 35 Ref Ref
 1 pattern 53 0.002 (− 0.223, 0.227) 0.985 0.020 (− 0.243, 0.284) 0.857
 2 patterns 15 0.039 (− 0.186, 0.264) 0.730 − 0.083 (− 0.309, 0.142) 0.461
Forceps major
 0 pattern 35 Ref Ref
 1  pattern 53 0.047 (− 0.176, 0.269) 0.679 − 0.167 (− 0.392, 0.050) 0.130
 2 patterns 15 0.163 (− 0.060, 0.386) 0.151 − 0.117 (− 0.339, 0.099) 0.288
Forceps minor
 0 pattern 35 Ref Ref
 1 pattern 53 − 0.009 (− 0.234, 0.215) 0.934 0.014 − 0.313, 0.342) 0.901
 2 patterns 15 0.007 (− 0.218, 0.232) 0.951 − 0.020 (− 0.231, 0.198) 0.862

β values are standardized regression coefficients. Analyses were adjusted for sex, peak height velocity (year) and parent education university level (neither/one/both). Statistically significant values are shown in bold (P < 0.05). Combined physically active patterns were calculated as the sum of performing active commuting to school and physical activity before school

FA Fractional anisotropy, RD Radial diffusivity, Ref. reference

*Missing values by preprocessing

*Statistically significant values surpassed multiple comparisons using the Benjamini and Hochberg method