Event centrality in social anxiety disorder and major depressive disorder

Figures & data

Figure 1. Mediation analyses of the effect of event centrality on social anxiety symptoms through the different variables in the SAD sample.

Note: None of the variables significantly mediated the link between event centrality and social anxiety symptoms. Dotted lines represent non-significant betas. Total results of the Bayesian mediation analysis can be found in Table A4.

Figure 2. Mediation analyses of the effect of event centrality on depressive symptoms through the different variables in the MDD sample.

Note: Posttraumatic cognitions and Trait Anxiety mediated the link between event centrality and depressive symptoms. Dotted lines represent non-significant betas. Full results of the Bayesian mediation analysis can be found in Table A5.

Figure 3. Mediation analyses of the effect of event centrality on depressive symptoms through the different variables in the combined sample.

Note: All variables, except worry, mediated the relation between event centrality and depressive symptoms. Dotted lines represent non-significant betas. Full results of the Bayesian mediation analysis can be found in Table A6.

Table A1. Means and standard deviations of all variables in the two patient groups and differences between the two groups.

	SAD M (SD)	MDD M (SD)	Test	Value	Cohen’s d
Number of participants	38	21
Age (years)	30.66 (10.94)	49.48 (14.37)	t	5.65***	0.36
Gender (M/F)	15/23	9/12	χ^2	0.06
Reported onset (yes/no)	26/12	18/3	χ^2	2.13
Disorder duration (months)	141.94 (120.51)	128.20 (154.26)	t	0.37	0.28
Education (primary/secondary/higher)	1/22/16	2/15/4	χ^2	3.41
CES	25.27 (4.93)	28.04 (5.39)	t	1.99^a	0.54
LSAS	72.68 (24.65)	–
BDI-II	25.02 (4.93)	32.50 (8.56)	t	2.71**	0.78
IES-I	14.43 (6.68)	23.58 (6.81)	t	4.82***	1.36
IES-A	16.14 (7.44)	22.00 (5.09)	t	2.86**	0.92
PTCI	127.54 (34.44)	123.75 (28.75)	t	0.38	0.12
RRS	51.71 (10.56)	61.35 (9.76)	t	3.33**	0.95
STAI-T	59.91 (8.18)	63.00 (8.28)	t	1.32	0.37
PSWQ	66.31 (9.50)	61.12 (10.21)	t	1.81^a	0.53

Note: CES = Centrality of Event Scale; LSAS = Liebowitz Social Anxiety Scale; BDI-II-NL = Beck Depression Inventory-II; IES-I = Impact of Event Scale – Intrusions; IES-A = Impact of Event Scale – Avoidance; PTCI = Posttraumatic Cognitions Inventory; RRS = Ruminative Response Scale; STAI-T = State-Trait Anxiety Inventory – Trait; PSWQ = Penn State Worry Questionnaire. Standard deviations are noted between brackets.

^a <.10; *<.05; **<.01; ***<.001.

Table A2. Participant’s descriptives and between-group differences of participants who did or not report an onset event.

	Onset – Yes M (SD)	Onset – No M (SD)	Test	Value
Number of participants	44	15
Group (MDD/SAD)	18/26	3/12	χ^2	2.13
Gender (M/F)	17/27	7/8	χ^2	0.30
Age (years)	35.80 (16.80)	37.89 (14.74).	t	0.46
Disorder duration (months)	131.42 (128.23)	153.86 (147.66)	t	0.55

Note: Standard deviations are noted between brackets. None of the differences were significant.

Table A3. Correlations between all variables within the two patient groups.

Variables	CES	LSAS	BDI-II	IES-I	IES-A	PTCI	RRS	STAI-T	PSWQ
CES	1.00	0.10 [−0.26, 0.44]	0.17 [−0.17, 0.49]	0.53 [0.26, 0.76]	0.54 [0.28, 0.77]	0.36 [0.02, 0.63]	0.28 [−0.05, 0.59]	0.25 [−0.10, 0.56]	0.33 [0.00, 0.62]
LSAS	–	1.00	0.54 [0.25, 0.78]	0.20 [−0.16, 0.54]	0.28 [−0.07, 0.58]	0.41 [0.10, 0.71]	0.50 [0.20, 0.77]	0.46 [0.14, 0.73]	0.42 [0.10, 0.70]
BDI-II	0.51 [0.13, 0.80]	–	1.00	0.46 [0.17, 0.72]	0.59 [0.33, 0.80]	0.57 [0.29, 0.78]	0.59 [0.32, 0.80]	0.68 [0.47, 0.86]	0.38 [0.06, 0.67]
IES-Int	0.65 [0.32, 0.87]	–	0.53 [0.15, 0.82]	1.00	0.79 [0.57, 0.93]	0.55 [0.28, 0.76]	0.51 [0.23, 0.756]	0.48 [0.19, 0.73]	0.31 [−0.03, 0.59]
IES-avoi	0.53 [0.16, 0.85]	–	0.71 [0.38, 0.91]	0.49 [0.03, 0.83]	1.00	0.70 [0.50, 0.85]	0.50 [0.22, 0.74]	0.56 [0.29, 0.77]	0.25 [−0.09, 0.56]
PTCI	0.67 [0.32, 0.91]	–	0.64 [0.28, 0.89]	0.63 [0.22, 0.89]	0.55 [0.13, 0.86]	1.00	0.55 [0.28, 0.77]	0.60 [0.34, 0.80]	0.77 [0.60, 0.90]
RRS	0.60 [0.26, 0.85]	–	0.55 [0.20, 0.83]	0.65 [0.35, 0.88]	0.53 [0.11, 0.85]	0.63 [0.26, 0.90]	1.00	0.36 [0.02, 0.64]	0.57 [0.30, 0.79]
STAI-T	0.54 [0.14, 0.84]	–	0.74 [0.48, 0.91]	0.63 [0.28, 0.88]	0.62 [0.25, 0.89]	0.58 [0.19, 0.87]	0.65 [0.32, 0.88]	1.00	0.61 [0.35, 0.81]
PSWQ	0.35 [−0.14, 0.77]	–	0.57 [0.17, 0.86]	0.34 [−0.16, 0.74]	0.53 [0.06, 0.86]	0.30 [−0.21, 0.75]	0.29 [−0.18, 0.72]	0.59 [0.20, 0.87]	1.00

Note: The top half (above the diagonal) shows the results of the SAD group, the bottom half (under the diagonal) shows the results of the MDD group. Values in bold show significant correlations. CES = Centrality of Event Scale; LSAS = Liebowitz Social Anxiety Scale; BDI-II-NL = Beck Depression Inventory-II; IES = Impact of Event Scale; PTCI = Posttraumatic Cognitions Inventory; RRS = Ruminative Response Scale; STAI-T = State-trait anxiety inventory – trait; PSWQ = Penn State Worry Questionnaire. The table’s top half (above the diagonal) shows the correlations within the SAD group, and the bottom half (under the diagonal) shows the correlations within the MDD group. HDI credibility intervals are shown between square/box brackets.

Table A4. Within-group mediation analysis for the relation between event centrality and SAD.

Mediator	A		B		C		Indirect effect
	β1 (S.E.)	95% CI	β2 (SE)	95% CI	β3 (SE)	95% CI	a*b	Standardised	95% CI
IES-I	0.740 (0.196)	0.347, 1.136	0.869 (0.778)	−0.689, 2.426	−0.005 (1.025)	−2.070, 2.036	0.598	.120	−0.538, 1.974
IES-A	0.911 (0.216)	0.477, 1.347	1.428 (0.682)	0.042, 2.777	−0.665 (1.029)	−2.719, 1.394	1.244	.249	−0.023, 2.799
PTCI	2.618 (1.115)	0.288, 4.873	0.283 (0.124)	0.034, 0.530	−0.082 (0.849)	−1.802, 1.608	0.679	.136	−0.066, 1.735
RRS	0.737 (0.349)	0.022, 1.422	1.239 (0.383)	0.453, 1.999	−0.267 (0.799)	−1.884, 1.328	0.841	.168	−0.050, 2.018
STAI-T	0.415 (0.285)	−0.156, 0.991	1.624 (0.432)	−0.749, 2.491	−0.042 (0.746)	−1.544, 1.453	0.632	.126	−0.302, 1.758
PSWQ	0.671 (0.306)	0.059, 1.289	1.153 (0.439)	0.273, 2.035	−0.134 (0.834)	−1.803,, 1.550	0.712	.142	−0.058, 1.774
BDI-II	0.395 (0.366)	−0.334, 1.128	1.117 (0.386)	0.384, 1.925	−0.184 (0.755)	−1.337, 1.715	0.411	.082	−0.441, 1.469

Note: IES = Impact of Event Scale; PTCI = Posttraumatic Cognitions Inventory; RRS = Ruminative Response Scale; STAI-T = State-trait anxiety inventory – trait; PSWQ = Penn State Worry Questionnaire; BDI-II-NL = Beck Depression Inventory II. Equations: Mediator = ${\beta }_0^m$ * Intercept + β₁ * Centrality (A); SAD = ${\beta }_0^c$ * Intercept + β₂ * Mediator (B) + β₃ * Centrality (C). Sample size = 59; Multiply imputed data (m = 100); Priors = T-distribution for the intercept, the coefficients, and sigma. Significant indirect effects are displayed in bold.

Table A5. Within-group mediation analysis for the relation between event centrality and MDD.

Mediator	A		B		C		Indirect effect
	β1 (S.E.)	95% CI	β2 (S.E.)	95% CI	β3 (S.E.)	95% CI	a*b	Standardised	95% CI
IES-I	0.831 (0.220)	0.375, 1.294	0.330 (0.359)	−0.409, 1.055	0.573 (0.432)	−0.316, 1.458	0.254	.160	−0.362, 0.955
IES-A	0.504 (0.189)	0.110, 0.900	0.864 (0.443)	−0.052, 1.734	0.417 (0.353)	−0.307, 1.154	0.396	.249	−0.075, 1.024
PTCI	3.665 (0.905)	1.793, 5.533	0.179 (0.080)	0.014, 0.342	0.197 (0.405)	−0.662, 1.038	0.630	.397	0.003, 1.388
RRS	1.123 (0.335)	0.428, 1.808	0.338 (0.221)	−0.117, 0.793	0.471 (0.390)	−0.332, 1.127	0.351	.221	−0.159, 1.001
STAI-T	0.885 (0.295)	0.280 1.490	0.691 (0.217)	0.226, 1.140	0.239 (0.316)	−0.421, 0.901	0.579	.365	0.061, 1.209
PSWQ	0.510 (0.428)	−0.367, 1.394	0.334 (0.175)	−0.026, 0.690	0.678 (0.309)	0.044, 1.315	0.141	.088	−0.150, 0.605

Note: IES = Impact of Event Scale; PTCI = Posttraumatic Cognitions Inventory; RRS = ruminative response scale; STAI-T = State-trait anxiety inventory – trait; PSWQ = Penn State Worry Questionnaire. Equations: Mediator = ${\beta }_0^m$ * Intercept + β₁ * Centrality (A); MDD = ${\beta }_0^c$ * Intercept + β₂ * Mediator (B) + β₃ * Centrality (C). Sample size = 59; Multiply imputed data (m = 100); Priors = T-distribution for the intercept, the coefficients, and sigma. Significant indirect effects are displayed in bold.

Table A6. Combined sample mediation analysis for the relation between event centrality (CES) and MDD.

Mediator	A		B		C		Indirect effect
	β1 (S.E.)	95% CI	β2 (S.E.)	95% CI	β3 (S.E.)	95% CI	a*b	Standardised	95% CI
IES-I	0.949 (0.162)	0.623, 1.169	0.702 (0.191)	0.321, 1.080	0.049 (0.287)	−0.522, 0.6118	0.652	.328	0.253, 1.100
IES-A	0.836 (0.158)	0.523, 1.157	1.025 (0.174)	0.684, 1.373	−0.143 (0.244)	−0.629, 0.341	0.844	.425	0.440, 1.295
PTCI	2.903 (0.742)	1.420, 4.377	0.171 (0.040)	0.091, 0.251	0.224 (0.249)	−0.273, 0.718	0.482	.243	0.171, 0.854
RRS	1.042 (0.249)	0.544, 1.534	0.582 (0.116)	0.351, 0.872	0.108 (0.239)	−0.368, 0.584	0.593	.299	0.249, 0.998
STAI-T	0.642 (0.200)	0.244, 1.039	0.855 (0.139)	0.597, 1.110	0.165 (0.204)	−0.242, 0.572	0.539	.271	0.180, 0.937
PSWQ	0.503 (0.252)	0.006, 1.009	0.210 (0.137)	−0.063, 0.481	0.609 (0.257)	0.098, 1.118	0.090	.045	−0.049, 0.310

Note: IES = Impact of Event Scale; PTCI = Posttraumatic Cognitions Inventory; RRS = Ruminative Response Scale; STAI-T = State-trait anxiety inventory – trait; PSWQ = Penn State Worry Questionnaire. Equations: Mediator = ${\beta }_0^m$ * Intercept + β₁ * Centrality (A); MDD = ${\beta }_0^c$ * Intercept + β₂ * Mediator (B) + β₃ * Centrality (C). Sample size = 59; Multiply imputed data (m = 100); Priors = T-distribution for the intercept, the coefficients, and sigma. Significant indirect effects are displayed in bold.

Data availability statement

The data, materials, and analysis code that support the findings of this study are available from the corresponding author ([email protected]) upon reasonable request.