Neural correlates of negative life events and their relationships with alcohol and cannabis use initiation

Figures & data

Table 1. Demographic characteristics of the study sample.

Variables	NU	AUI	CUI
Sample size, n	128	207	168
Age at imaging collection, mean (SD)	16.41 (1.74)	16.36 (1.73)	16.43 (1.49)
Onset age	–	17.14 (1.71)	17.17 (1.50)
Gender (female), No. (%)	68 (53)	101 (49)	86 (51)
Race, No. (%)*
White	81 (63)	169 (82)	129 (77)
Black	23 (18)	15 (7)	17 (10)
Other	24 (19)	23 (11)	22 (13)
Family income, No. (%)*
<$35,000	22 (17)	10 (5)	12 (7)
$35,000–$99,999	41 (32)	63 (31)	44 (27)
≥$100,000	64 (50)	129 (64)	108 (66)
Parental education, No. (%)
≤12 years	13 (10)	7 (3)	9 (5)
13–16 years	49 (38)	85 (41)	72 (43)
≥17 years	66 (52)	115 (56)	87 (52)
Parental marital status (Yes), No. (%)	93 (73)	166 (80)	130 (77)
Number of controllable NLEs, No. (%)*
0	67 (52)	73 (35)	50 (30)
1	38 (30)	63 (31)	49 (29)
≥2	23 (18)	71 (34)	69 (41)
Number of uncontrollable NLEs, No. (%)
0	77 (60)	119 (58)	94 (56)
1	36 (28)	50 (24)	47 (28)
≥2	15 (12)	38 (18)	27 (16)

AUI: alcohol use initiation; CUI: cannabis use initiation; NLEs: negative life events; NU: non-using; SD: standard deviation.

Onset age: self-reported age at substance use initiation; available only for AUI and CUI.

* p < 0.05.

Compared to the NU group, the AUI group had more White participants; both AUI and CUI groups had more participants from high income families (≥$100k) and more controllable NLEs than NUs.

Figure 1. Mediation analysis shows how brain features explain relationships between controllable NLEs and initiation of alcohol and cannabis, respectively. Potential confounding effects due to age, gender, race/ethnicity, parental highest education level, pubertal stage, scanner type, and estimated intracrine volume were controlled in the mediation analyses. Cortical thickness of the left banks of the superior temporal sulcus mediates the effects of controllable NLEs on initiations of alcohol (A) and cannabis (B) use, respectively, (C) left caudate gray-matter volume mediates the effect of controllable NLEs on initiation of cannabis use. The figure includes the following regression coefficients: (1) the effects of controllable NLEs on each brain mediator; (2) the effects of brain mediators on initiation of alcohol and cannabis first use when adjusting for controllable NLEs; (3) the direct and indirect associations along with the 95%CIs between controllable NLEs and the onset of alcohol or cannabis use, respectively. *p < 0.05. **p < 0.001. The right column shows the location of each corresponding brain region. Note, NLEs and substance initiation data were collected at the same visit. While brain feature measures were collected at the visit before substance initiation, therefore, brain immaturity should be regarded as a pre-existing condition in the mediation analyses.

Data availability statement

The data used in this paper was retrieved on 21 October 2021, from Sage Bionetworks Synapse (https://doi.org/10.7303/syn22213272).