Psychometric properties and diagnostic utility of a Brazilian version of the PCL-5 (complete and abbreviated versions)

Figures & data

Table 1. Descriptive statistics and Pearson correlation coefficients for the Brazilian PCL-5 items.

Item	Mean	SD	Range	Item 1	Item 2	Item 3	Item 4	Item 5	Item 6	Item 7	Item 8	Item 9	Item 10	Item 11	Item 12	Item 13	Item 14	Item 15	Item 16	Item 17	Item 18	Item 19	Item 20
1	1.86	1.47	0–4	1
2	1.09	1.32	0–4	.49**	1
3	1.62	1.47	0–4	.78**	.51**	1
4	2.25	1.46	0–4	.75**	.45**	.70**	1
5	1.98	1.59	0–4	.67**	.43**	.84**	.64**	1
6	2.06	1.56	0–4	.67**	.41**	.70**	.67**	.71**	1
7	2.00	1.63	0–4	.65**	.50**	.66**	.64**	.68**	.81**	1
8	1.06	1.48	0–4	.26*	.11	.28*	.36**	.32**	.47**	.32**	1
9	2.11	1.60	0–4	.54**	.55**	.59**	.68**	.56**	.51**	.56**	.27*	1
10	1.75	1.61	0–4	.51**	.54**	.59**	.64**	.57**	.53**	.51**	.21	.70**	1
11	2.18	1.60	0–4	.69**	.58**	.71**	.78**	.71**	.64**	.67**	.26*	.75**	.758**	1
12	1.95	1.65	0–4	.62**	.52**	.60**	.61**	.56**	.57**	.56**	.18	.58**	.568**	.70**	1
13	1.52	1.53	0–4	.64**	.54**	.60**	.64**	.65**	.58**	.53**	.33**	.63**	.641**	.73**	.67**	1
14	1.39	1.47	0–4	.59**	.41**	.49**	.62**	.50**	.53**	.50**	.34**	.58**	.596**	.58**	.55**	.75**	1
15	1.51	1.56	0–4	.52**	.54**	.63**	.53**	.68**	.50**	.60**	.18	.62**	.559**	.69**	.60**	.61**	.50**	1
16	0.85	1.19	0–4	.34**	.35**	.44**	.49**	.43**	.37**	.35**	.22*	.47**	.502**	.52**	.36**	.51**	.49**	.52**	1
17	2.15	1.62	0–4	.56**	.54**	.58**	.65**	.56**	.51**	.57**	.16	.68**	.639**	.68**	.65**	.54**	.49**	.54**	.30**	1
18	1.98	1.59	0–4	.60**	.58**	.65**	.70**	.67**	.50**	.52**	.18	.67**	.598**	.74**	.68**	.61**	.45**	.56**	.31**	.84**
19	1.92	1.58	0–4	.65**	.50**	.75**	.67**	.68**	.65**	.64**	.40**	.64**	.607**	.77**	.68**	.66**	.59**	.52**	.54**	.56**	.61**	1
20	1.64	1.63	0–4	.51**	.48**	.52**	.47**	.55**	.36**	.44**	.46**	.52**	.400**	.56**	.56**	.55**	.39**	.65**	.49**	.35**	.47**	.62**	1

* Correlation is significant at the .05 level (2-tailed)

** Correlation is significant at the .01 level (2-tailed)

Table 2. Internal consistency analyses for the complete and abbreviated versions of the Brazilian PCL-5.

Subscale	Complete PCL-5	Abbreviated 8-item PCL-5	Abbreviated 4-item PCL-5
Total PCL-5	α = .96 (20 items)	α = .93 (8 items)	α = .85 (4 items)
Cluster B	α = .90 (5 items)	r = .75 (p < .001)	Not applicable
		α = .86 (2 items)
Cluster C	α = .90 (2 items)	r = .81 (p < .001)	Not applicable
		α = .90 (2 items)
Cluster D	α = .89 (7 items)	r = .58 (p < .001)	Not applicable
		α = .73 (2 items)
Cluster E	α = .87 (6 items)	r = .61 (p < .001)	Not applicable
		α = .76 (2 items)

α = Cronbach’s alpha

r = Pearson correlation coefficient

Table 3. Pearson correlation coefficients between the different clusters of the complete and abbreviated versions of the Brazilian PCL-5.

Correlation between clusters	Complete PCL-5	Abbreviated 8-item PCL-5	Abbreviated 4-item PCL-5
B and C	r = .79 (p < .001)	r = .74 (p < .001)	r = .65 (p < .001)
B and D	r = .84 (p < .001)	r = .73 (p < .001)	r = .57 (p < .001)
B and E	r = .86 (p < .001)	r = .78 (p < .001)	r = .60 (p < .001)
C and D	r = .72 (p < .001)	r = .65 (p < .001)	r = .56 (p < .001)
C and E	r = .68 (p < .001)	r = .68 (p < .001)	r = .52 (p < .001)
D and E	r = .88 (p < .001)	r = .84 (p < .001)	r = .67 (p < .001)

Figure 1. ROC curves of the completed and abbreviated versions of the Brazilian PCL-5.

Table 4. Sensitivity, specificity, likelihood, predictive, and overall efficiency values of the complete and abbreviated versions of the Brazilian PCL-5.

PCL-5 Version/CP	Sn	Sp	+LR	-LR	+PV	-PV	OE
Complete PCL-5
≥ 0	1	0	1	-	.40	-	.40
21	1	.53	2.12	0	.59	1	.72
28	.97	.55	2.15	.05	.59	.97	.72
34	.97	.65	2.75	.05	.65	.97	.78
35	.94	.69	3.00	.09	.67	.95	.79
36^a b	.94	.71	3.20	.08	.68	.95	.80
37	.91	.73	3.32	.12	.69	.93	.80
41	.79	.73	2.89	.28	.66	.84	.75
42	.79	.75	3.12	.28	.68	.84	.77
43	.76	.78	3.55	.30	.70	.83	.78
45	.68	.82	3.83	.39	.72	.79	.77
49	.53	.82	3.00	.57	.67	.72	.71
50	.47	.84	3.00	.63	.67	.71	.69
51	.47	.86	3.43	.61	.70	.71	.71
52	.44	.86	3.21	.65	.68	.70	.69
54	.41	.90	4.20	.65	.74	.70	.71
55	.35	.90	3.60	.72	.71	.68	.68
57	.32	.92	4.12	.73	.73	.67	.68
59	.32	.94	5.50	.72	.79	.68	.69
66	.12	.94	2.00	.94	.57	.62	.61
72	.12	1	-	.88	1	.63	.65
80	0	1	-	1	-	.60	.60
Abbreviated 8-item PCL-5
≥ 0	1	0	1	-	.40	-	.40
13 ^a	1	.61	2.55	0	.63	1	.77
14	.97	.61	2.47	.05	.62	.97	.75
15	.94	.63	2.53	.09	.63	.94	.75
16	.91	.63	2.45	.14	.62	.91	.74
18	.85	.67	2.56	.22	.63	.87	.74
19	.82	.69	2.62	.26	.64	.85	.74
20	.79	.75	3.12	.28	.68	.84	.77
21 ^b	.79	.76	3.37	.27	.69	.85	.78
22	.71	.78	3.27	.38	.69	.80	.75
23	.62	.80	3.15	.48	.68	.76	.73
24	.53	.82	3.00	.57	.67	.72	.71
25	.47	.90	4.80	.59	.76	.72	.73
26	.35	.92	4.50	.70	.75	.68	.69
27	.24	.94	4.00	.81	.73	.65	.66
28	.21	.96	5.25	.83	.78	.65	.66
29	.21	.98	1.50	.81	.88	.65	.67
31	.06	.98	3.00	.96	.67	.61	.61
32	.06	1	-	.94	1	.61	.62
33	0	1	-	1	-	.60	.60
Abbreviated 4-item PCL-5
≥ 0	1	0	1	-	.40	-	.40
4	1	.45	1.82	0	.55	1	.67
5	.97	.47	1.83	.06	.55	.96	.67
6	.97	.51	1.98	.06	.57	.96	.69
7	.94	.55	2.09	.11	.58	.93	.71
8^a b	.94	.65	2.67	.09	.64	.94	.77
9	.82	.65	2.33	.27	.61	.85	.72
10	.74	.75	2.88	.36	.66	.81	.74
11	.59	.78	2.73	.53	.65	.74	.71
12	.53	.82	3.00	.57	.67	.72	.71
13	.41	.86	3.00	.68	.67	.69	.68
14	.32	.96	8.25	.70	.85	.68	.71
15	.18	.98	9.00	.84	.86	.64	.66
16	.12	1	-	.88	1	.63	.65
17	0	1	-	1	-	.60	.60

CP = cutoff point; Sn = sensitivity; Sp = specificity; +LR = positive likelihood ratio; -LR = negative likelihood ratio; +PV = positive predictive value; -PV = negative predictive value; OE = overall efficiency; ^a cutoff point correspondent to the Youden index J, ^b cutoff point with the highest overall efficiency