A sex-specific pathway linking early life maltreatment, vagal activity, and depressive symptoms

Figures & data

Table 1. Summary characteristics of key demographics and clinical data for the adolescent and adult sample, respectively, considering exposure to ELM as grouping variable.

	1: ELM exposure reported						2: No ELM exposure reported						Group comparison (1 vs. 2)
	Adolescent Sample (N = 113)			Adult Sample (N = 39)			Adolescent Sample (N = 100)			Adult Sample (N = 46)			Adolescents		Adults
Demo-graphics	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	W / χ2	p	W / χ2	p
Age (y)	14.97	1.39	12–18	24.76	3.70	19–40	14.75	1.49	12–19	24.22	3.12	19–31	5139.5	.247	836.5	.738
Sex: N (%) females	103 (90.27)	–	–	32 (82.05)	–	–	89 (46.35)	–	–	39 (54.93)	–	–	0.09	.762	0.11	.735
Psycho-pathology	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	W	p	W	p
SCL-90-R score	139.55	64.85	8.00–297.00	55.44	12.42	33.00–75.00	73.39	67.59	0.00–281.00	44.20	9.69	33.00–74.00	1634	<.001	414.5	<.001
Depressive Symptoms
DIKJ	29.51	9.64	4.00–47.00	–	–	–	17.23	12.04	0.00–41.00	–	–	–	2408	<.001	–	–
BDI-II	–	–	–	18.26	10.60	3.00–37.00	–	–	–	10.61	8.00	0.00–31.00	–	–	514	<.001
ELM Subtypes
Physical Neglect	2.28	0.81	1.00–4.00	9.00	3.09	5.00–17.00	1.41	0.55	1.00–3.00	5.52	0.96	5.00–9.00	2060.5	<.001	199.5	<.001
Emotional Neglect	2.29	0.70	1.00–3.83	14.21	5.36	5.00–24.00	1.66	0.58	1.00–3.00	7.52	2.25	5.00–13.00	2440.5	<.001	267.5	<.001
Emotional Abuse	3.53	0.81	1.25–5.00	13.62	4.77	5.00–23.00	1.66	0.62	1.00–3.00	6.59	1.76	5.00–12.00	463	<.001	159.5	<.001
Physical Abuse	0.15	0.28	0.00–1.00	7.67	3.44	5.00–19.00	0	0	–	5.15	0.47	5.00–7.00	4200	<.001	383	<.001
Sexual Abuse	0.15	0.24	0.00–1.00	7.97	4.95	5.00–22.00	0	0	–	5.15	0.47	5.00–7.00	3600	<.001	614	<.001
Cardiac autono-mic data	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	W	p	W	p
HR	75.81	10.21	58.40–107.06	80.66	9.68	50.40–100.31	75.96	11.56	51.08–105.14	84.29	17.51	58.57–126.47	5407	.789	907	.933
rMSSD	55.06	27.17	15.15 –142.04	48.33	41.71	11.82–247.15	57.31	34.38	16.73–239.51	45.92	23.30	9.48–90.02	5269	.958	975	.494

 Note. BDI-II = Beck Depression Inventory-II; DIKJ = Depressionsinventar für Kinder und Jugendliche; ELM = Early Life Maltreatment; HR = Heart Rate; rMSSD = root Mean Square of Standardized Differences; SCL-90-R = Symptom Checklist-90-Revised; SD = Standard Deviation.

Figure 1. Visualization of the network models, including nodes representing ELM exposure, depressive symptoms, and vmHRV. Network layout is adjusted on the Fruchterman-Reingold algorithm (Fruchterman & Reingold, 1991), resulting in edges with stronger connections being grouped together. Red lines indicate negative partial correlations, while black lines indicate positive partial correlations. The more saturated the edge, the stronger the partial correlation.

Figure 2. Scatter plots visualizing the relationships of (ln-transformed) vmHRV values, emotional abuse, and emotional neglect with depressive symptoms in adolescent (upper panes) and adult females (lower panes), respectively. In adolescents, the relationships of vmHRV and emotional neglect with depression appeared likely to be linear, in contrast to emotional abuse, which adapted a rather nonlinear function; in adults, the relationships of both vmHRV and emotional abuse with depression adapted a rather quadratic function, while considering emotional neglect, the respective relationship appeared to be linear.

Figure 3. Visualization of interactions between resting-state vmHRV and emotional maltreatment in the prediction of depressive symptoms. In linear models, a marginally statistically significant interaction was observed between emotional abuse and vmHRV in female adolescents (A), and a significant interaction was found between emotional neglect and vmHRV in female adults (B). In a quadratic model, in female adults, the interaction between emotional abuse and vmHRV was found to be statistically significant (C; D)

Fruchterman, T. M., & Reingold, E. M. (1991). Graph drawing by force-directed placement. Software: Practice and Experience, 21(11), 1129–1164.

 Web of Science ®Google Scholar

Data Availability Statement

The data that support the findings of this study are available on reasonable request from the responsible authors (VC, MK). The data are not publicly available due to ethical restrictions. Code used in the analyses is made publicly available by the corresponding author and can be accessed online (https://osf.io/vd7rh/?view_only = d547aa842a92457b9eaa5381941f91a8).