The child and Adolescent Trauma Screen 2 (CATS-2) – validation of an instrument to measure DSM-5 and ICD-11 PTSD and complex PTSD in children and adolescents

Figures & data

Table 1. Descriptive characteristics of the study samples.

Characteristic	Total sample (N = 283)	U.S. sample (n = 95)	Norwegian sample (n = 100)	German sample (n = 88)
Age (M, SD)	12.21 (3.16)	11.14 (3.17)	11.53 (2.81)	14.15 (2.66)
Gender (% female)	59.0	51.6	58.0	68.2
Living situation (%)
Biological parents	74.6	46.7	91.0	85.2
Foster Parents/Adoptive Parents	20.7	51.1	8.0	3.4
Child Welfare	3.6	0.0	1.0	10.2
Somewhere else	1.1	2.2	0	1.1
Race*
American Indian	–	10.6	–	–
Black or African American	–	13.8	–	–
White	–	69.1	–	–
Other	–	6.4	–	–
Ethnicity*
Hispanic or Latino	–	12.5	–	–
Country of origin*
Not Norwegian/German	–	–	9.3	12.5
Self-report
Number of traumatic events	4.53 (2.33)	5.26 (2.57)	3.81 (2.13)	4.57 (2.04)
CATS-2 DSM-5 PTSD Total Score	27.63 (12.17)	25.66 (11.91)	26.75 (11.30)	30.75 (12.91)
CATS-2 ICD-11 Total Score	8.11 (4.13)	7.92 (4.34)	7.72 (3.84)	8.76 (4.18)
CATS-2 ICD-11 DSO Total Score	7.47 (4.53)	6.66 (4.03)	6.98 (4.25)	8.90 (5.03)
CATS-2 ICD-11 CPTSD Total Score	15.58 (7.82)	14.58 (7.65)	14.70 (7.10)	17.66 (8.43)
MFQ Total Score	11.28 (7.14)	8.72 (6.78)	10.87 (6.37)	14.01 (7.34)
MFQ Depression (%)	46.7%	34.2%	42.4%	62.5%
Caregiver-report
Number of traumatic events	3.84 (2.20)	4.68 (2.14)	3.04 (1.96)	3.78 (2.19)
CATS-2 DSM-5 PTSD Total Score	24.16 (12.26)	22.70 (12.57)	22.36 (11.66)	27.80 (11.91)
CATS-2 ICD-11 Total Score	6.20 (4.14)	6.04 (4.13)	5.84 (3.90)	6.76 (4.39)
CATS-2 ICD-11 DSO Total Score	7.20 (4.50)	6.89 (4.71)	6.18(4.33)	8.67 (4.12)
CATS-2 ICD-11 CPTSD Total Score	13.42 (7.65)	12.93 (7.95)	12.09 (7.22)	15.43 (7.44)
PSC-17 Internalizing	6.02 (2.56)	5.18 (2.53)	6.18 (2.62)	6.74 (2.28)
PSC-17 Externalizing	4.35 (3.24)	5.48 (3.37)	3.67 (2.70)	3.92 (3.40)
PSC-17 Attention	4.91 (2.68)	4.88 (2.66)	4.97 (2.55)	4.88 (2.90)
Interview CAPS-CA-5	n = 121	n = 30	n = 28	n = 63
DSM-5 Total Score	27.79 (15.08)	27.03 (15.64)	24.68 (14.16)	29.54 (15.19)
ICD-11 Total Score	8.45 (5.11)	9.10 (5.72)	7.25 (4.60)	8.68 (5.02)
DSM5 PTSD diagnosis (%)	53.7	50.0	39.3	61.9
ICD-11 PTSD diagnosis (%)	29.8	26.7	14.3	38.1

Note: N = 283 self-reports; N = 255 caregiver reports; values are means (standard deviations) or % as appropriate; *Race and ethnicity were only assessed in the U.S. sample; Country of origin was only assessed in the Norwegian and German sample.

Figure 1. Alternative models of ICD-11 CPTSD using the CATS-2: Loading patterns of confirmatory factor analyses.

Note. PTSD: Posttraumatic Stress Disorder; CPTSD: complex PTSD; RE: Re-experiencing; Av: Avoidance; Th: Sense of Threat; AD: Affect Dysregulation; NSC: Negative Self-concept; DR:  Disturbances in Relationships.

Table 2. Convergent-divergent validity pattern.

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.
1. CATS-2 DSM-5 PTSD Total Score (Self)	(.89)
2. CATS-2 ICD-11 PTSD Total Score (Self)	.88**	(.67)
3. CATS-2 ICD-11 DSO Total Score (Self)	.84**	.63**	(.78)
4. CATS-2 ICD-11 CPTSD Total Score (Self)	.95*	.89**	.91**	(.83)
5. CATS-2 DSM-5 PTSD Total Score (Care)	.34**	.31**	.27**	.33**	(.91)
6. CATS-2 ICD-11 PTSD Total Score (Care)	.25**	.30**	.13*	.23**	.86**	(.79)
7. CATS-2 ICD-11 DSO Total Score (Care)	.30**	.21**	.36**	.32**	.85**	.57**	(.82)
8. CATS-2 ICD-11 CPTSD Total Score (Care)	.31**	.29**	.28**	.31**	.96**	.87**	.90**	(.87)
9. MFQ Total Score	.68**	.51**	.72**	.69**	.25**	.10	.31**	.24**	(.91)
10. PSC-17 Internalizing	.30**	.22**	.31**	.30**	.62**	.48**	.64**	.64**	.32**	(.81)
11. PSC-17 Attention	.03	-.02	.02	.00	.24**	.12	.24**	.21**	.12	.21**	(.79)
12. PSC-17 Externalizing	-.11	-.11	-.06	-.09	.09	-.10	.23**	.08	-.04	.05	.35**	(.81)
13. Traumaload (self)	.42**	.39**	.34**	.40**	.22**	.16**	.19**	.20**	.29**	.04	.10	.15*	-

Note. Bivariate correlations; * = p < .01; ** = p < .001; values on the diagonal = Cronbach α.

Table 3. Goodness of fit indices for the ICD-11 PTSD / CPTSD models.

Model	χ^2	df	P	CFI	TLI	SRMR	RMSEA (90% CI)	BIC
Self-report (n = 283)
Model 1: Unidimensional CPTSD	191.06	54	0.00	0.817	0.776	0.065	0.095 (0.080-0.109)	9731.56
Model 2: Six Factor First-Order Model	54.35	39	0.05	0.979	0.965	0.035	0.037 (0.000-0.059)	9660.04
Model 3: Single-Factor Second-Order Model	66.69	49	0.05	0.976	0.968	0.038	0.036 (0.004-0.056)	9618.09
Model 4: Two-Factor Second-Order	66.71	48	0.04	0.975	0.966	0.038	0.037 (0.009-0.057)	9623.52
Caregiver-report (n  = 255)
Model 1: Unidimensional CPTSD	180.21	54	0.00	0.843	0.808	0.068	0.096 (0.081-0.111)	7878.80
Model 2: Six Factor First-Order Model	58.10	39	0.03	0.976	0.960	0.037	0.044 (0.016-0.066)	7814.27
Model 3: Single-Factor Second-Order Model	85.06	48	0.01	0.954	0.937	0.051	0.055 (0.035-0.074)	7797.25
Model 4: Two-Factor Second-Order	59.95	47	0.10	0.984	0.977	0.039	0.033 (0.000-0.056)	7773.40

Note. χ²: chi-square goodness of fit statistic; df: degrees of freedom; CFI: comparative fit index; TLI: Tucker-Lewis index; SRMR: standardized root mean square residual; RMSEA: root mean square error of approximation; AIC: Akaike information criterion; BIC: Bayesian information criterion.

Table 4. Latent profile analyses and multivariate normal mixture models with fit indices for the ICD-11 CPTSD items.

Model	Log-likelihood	Number of parameters	AIC	BIC	Entropy	LMRA-LRT	BLRT	Class proportions
Latent profile analyses
1 class	−5118.66	24	10285.33	10372.82				100
2 classes	−4803.92	37	9681.84	9816.72	0.862	0.0002	0.0000	60/40
3 classes	−4737.22	50	9574.45	9756.72	0.810	0.0687	0.0000	23/46/31
4 classes	−4679.73	63	9485.47	9715.13	0.875	0.2720	0.0000	21/19/40/18
5 classes	−4634.27	76	9420.53	9697.59	0.874	0.1164	0.0000	23/34/6/18/19
Multivariate normal mixture models
1 class	−4656.11	90	9492.22	9820.31				100
2 classes	−4571.96	103	9349.93	9725.41	0.979	0.0000	0.0000	63/37
3 classes	−4547.24	116	9326.49	9749.36	0.932	0.4442	0.0128	62/24/13
4 classes	−4504.98	129	9267.96	9738.23	0.968	0.0007	0.0000	54/8/14/23
5 classes	−4491.61	142	9267.22	9784.88	0.950	0.9163	1.0000	5/21/44/14/16

Note. n = 283; AIC: Akaike information criterion; BIC: Bayesian information criterion; LMRA-LRT: Lo-Mendell-Rubin-adjusted likelihood ratio test; BLRT: bootstrap likelihood ratio test. The best fitting and interpretable model is indicated by bold values.

Figure 2. Plot of the 2-class multivariate normal mixture model.

Note. 2-class multivariate normal mixture model based on the CATS-2 self-report (n = 283); PTSD: Posttraumatic Stress Disorder; CPTSD: complex PTSD; RE: Re-experiencing; Av: Avoidance; Th: Sense of Threat; AD: Affect Dysregulation; NSC: Negative Self-concept; DR: Disturbances in Relationships.

Table 5. Sensitivity, specificity, positive predictive value and negative predictive value for different CATS-2 cut-off values (n = 121).

Cut-off	Sensitivity	Specificity	PPV	NPV	Youden's J
DSM-5 PTSD Score
19	0.97	0.54	71	94	0.51
20	0.95	0.57	72	90	0.53
21	0.94	0.57	72	89	0.51
22	0.94	0.64	75	90	0.58
23	0.91	0.68	77	87	0.59
24	0.91	0.73	79	87	0.64
25	0.89	0.75	81	85	0.64
26	0.86	0.79	83	83	0.65
27	0.77	0.84	85	76	0.61
28	0.75	0.84	84	74	0.59
29	0.71	0.84	83	71	0.55
30	0.65	0.84	82	67	0.49
ICD-11 PTSD Score
5	0.97	0.35	39	96	0.33
6	0.94	0.48	43	95	0.43
7	0.92	0.57	48	94	0.48
8	0.83	0.67	52	90	0.50
9	0.72	0.75	55	86	0.48
10	0.64	0.85	64	85	0.49

Note. Interview = Clinician-Administered PTSD Scale for DSM-5 (CAPS-CA-5); PPV = Positive Predictive Value; NPV = Negative Predictive Value.

Data availability statement

Data assessment was conducted in the course of a research project under different legal backgrounds. For legal reasons, primary data cannot be published online, as we had not included this in our informed consent and assent for participants. The dataset generated during and analyzed for the current study is available from the corresponding author after complying with data protection agreements.