Skip to main content

Table 4 Linear regression with psychiatric data as dependent variable and cerebellar volumes (ml) as independent variable in the VLBW group

From: Mental health and cerebellar volume during adolescence in very-low-birth-weight infants: a longitudinal study

 

15 years

19 years

Coefficient

(95 % CI)

p value

Coefficient

(95 % CI)

p value

CGAS (15 years n = 40, 19 years n = 41)

 Cerebellar WM

1.930

(0.841 to 3.020)

0.001 a

1.450

(0.362 to 2.539)

0.010 a

 Cerebellar GM

0.630

(0.187 to 1.072)

0.007 a

0.764

(0.382 to 1.147)

<0.000 a

ASEBA (15 years n = 38, 19 years n = 40)

 Internalizing

  Cerebellar WM

−0.259

(−0.735 to 0.217)

0.276

−0.454

(−1.317 to 0.410)

0.294

  Cerebellar GM

−0.118

(−0.301 to 0.064)

0.196

−0.287

(−0.611 to 0.038)

0.082

 Externalizing

  Cerebellar WM

−0.098

(−0.516 to 0.320)

0.637

−0.223

(−0.70 to 0.254)

0.349

  Cerebellar GM

−0.130

(−0.285 to 0.026)

0.099

−0.099

(−0.283 to 0.085)

0.282

 Total problems

  Cerebellar WM

−0.818

(−2.142 to 0.506)

0.217

−1.190

(−3.201 to 0.820)

0.238

  Cerebellar GM

−0.457

(−0.955 to 0.040)

0.070

−0.683

(−1.441 to 0.074)

0.075

ADHD-RS-IV (15 years n = 36, 19 years n = 29) 

 Hyperactivity

  Cerebellar WM

−0.286

(−0.597 to 0.026)

0.071

0.158

(−0.348 to 0.663)

0.526

  Cerebellar GM

−0.092

(−0.214 to 0.030)

0.135

0.007

(−0.190 to 0.204)

0.940

 Inattention

  Cerebellar WM

−0.528

(−0.950 to 0.105)

0.016

−0.256

(−0.899 to 0.387)

0.420

  Cerebellar GM

−0.222

(−0.382 to 0.061)

0.008 *

−0.243

(−0.473 to 0.012)

0.040

  1. Adjusted for age, sex and estimated intracranial volume, but not for IQ
  2. ADHD-RS-IV attention-deficit/hyperactivity disorder rating scale, ASEBA Achenbach system of empirically based assessment, YSR (Youth Self Report at 14 years) and ARS (Adult Self Report at 19 years), CGAS children’s global assessment scale, CI confidence interval, GM gray matter, IQ intelligence quotient, VLBW very low birth weight, WM white matter
  3. aSignificant results also when corrected for multiple comparisons using the Benjamini–Hochberg procedure