Table 4.

Parameter estimates of direct and indirect effects (N = 130)

Measurement Model	B (SE)	β	95% CI
CU → Callousness	1.00 (.00)	.76	[1.00, 1.00]***
CU → Uncaring	1.11 (.26)	.53	[.61, 1.61]***
CU → Unemotional	1.18 (.28)	.60	[.63, 1.74]***
Mediation Model	B (SE)	β	95% CI
CAN → CU	.59 (.20)	.27	[.20, .98]**
CU → Alcohol	.62 (.30)	.23	[.04, 1.20]*
CU → Drug	.56 (.19)	.35	[.18, .94]**
CAN → Alcohol	−.09 (.51)	−.02	[−1.09, .91]
CAN → Drug	−.21 (.32)	−.06	[−.83, .42]
Covariates
Minority → Alcohol	−2.73 (.86)	−.26	[−4.41, −1.05]***
Gender → Alcohol	−2.23 (1.18)	−.17	[−4.54, .07]
Minority → Drug	−.78 (.54)	−.12	[−1.83, .27]
Gender → Drug	−.13 (.74)	−.02	[−1.58, 1.32]
Gender → CU	−1.79 (.45)	−.37	[−2.67, −.91]***
Moderated Mediation Model	B (SE)	β	95% CI
CAN → CU	.68 (.19)	.34	[.32, 1.05]***
CU → Alcohol	1.14 (.61)	.44	[−.05, 2.34]
ΔHRV → Alcohol	−.50 (.68)	−.14	[−1.82, .82]
CU*ΔHRV → Alcohol	−.33 (.29)	−.17	[−.89, .24]
CAN → Alcohol	−.40 (.56)	−.08	[−1.40, .69]
CU → Drug	1.17 (.46)	.67	[.27, 2.06]*
ΔHRV → Drug	−.99 (.37)	−.42	[−1.73, −.25]**
CU*ΔHRV → Drug	−.39 (.15)	−.29	[−.68, −.09]*
CAN → Drug	−.08 (.36)	−.02	[−.78, .62]
Covariates
Minority → Alcohol	−2.08 (.79)	−.22	[−3.62, −.54]**
Gender → Alcohol	−2.08 (1.45)	−.17	[−4.95, .80]
Minority → Drug	−.65 (.55)	−.10	[−1.73, .42]
Gender → Drug	−.27 (.95)	−.03	[−2.13, 1.60]
Gender → CU	−1.67 (.57)	−.35	[−2.78, −.56]**
Conditional Indirect Effect
CAN → CU*ΔHRV → Drug	−.26 (.13)	−.11	[−.55, −.05]*

ΔHRV Residualized change score of heart rate variability; CAN Child abuse and neglect; CU Callous-unemotional traits; SE Standard Error; CI Confidence Interval for unstandardized coefficients.

* p < .05; ** p < .01; *** p < .001. Gender was coded as 0 = Male, 1 = Female