Skip to main content
. 2018 Jan 18;8:753. doi: 10.3389/fneur.2017.00753

Table 5.

Linear regression model estimates of demographic and concussion-related factors.

Estimate (SE)
Heart rate variability variable Sex Age PCSI physical PCSI cognitive PCSI emotional PCSI fatigue Previous concussion
log (SDNN) (N = 294) −0.071 (0.03)* 0.031 (0.01)* −0.013 (0.01) −0.021 (0.009)* 0.015 (0.01) 0.015 (0.01) 0.061 (0.04)
RMSSD (N = 294) −2.746 (3.60) −0.133 (1.46) −0.991 (1.13) −2.195 (1.00)* 1.429 (1.43) 1.097 (1.58) 6.585 (3.90)
pNN50a (N = 294) −2.676 (1.85) 2.997 (1.67)
HF (N = 147) −153.73 (64.71)* −5.98 (20.33) −31.41 (17.38) −11.31 (14.99) −17.77 (23.89) 4.57 (23.74) 43.47 (55.09)
HFnu (N = 147) 0.123 (1.62) −0.731 (0.65) 0.001 (0.56) −0.581 (0.48) −0.217 (0.76) 1.018 (0.56) 2.394 (1.78)
log (LF) (N = 147) −2.172 (1.09)* 0.247 (0.44) −0.621 (0.38) −0.049 (0.33) 0.137 (0.51) 0.325 (0.38) 0.471 (1.2)
LFnu (N = 147) −0.123 (1.62) 0.731 (0.65) −0.001 (0.56) 0.581 (0.48) 0.216 (0.76) −1.018 (0.56) −2.394 (1.78)
log (Total power) (N = 147) −0.203 (0.08)* 0.017 (0.03) −0.035 (0.03) −0.021 (0.02) 0.018 (0.04) 0.037 (0.03) 0.110 (0.09)

*A significant p-value < 0.05.

aThe uniqueness of the pNN50 model. Level of fit indicated a removal of age as a factor and removal of each PCSI domain. Rather, PSCI total score enhanced the fit of the model and is described within the results section.