The utility of the Cognitive Reserve Index questionnaire in chronic traumatic brain injury

Figures & data

Table 1. Demographic and injury characteristics for participants.

	TBI cohort		Non-TBI cohort		p-value
n	105		91
Gender; %male	76; 72.4%		47; 51.6%		0.003
Age	57.68 (11.89)	40–88	61.54 (12.99)	40–89	0.031
Country of birth; % Australia	83.81%		71.43%
Years of education	12.70 (2.59)	1–18	13.36 (2.59)	7–18	0.074
Premorbid IQ	100.57 (9.09)	74–116	105.06 (8.11)	85–117	<0.001
Age at injury	36.36 (14.71)	18–69
Time since injury (years)	21.18 (6.51)	10–33
Duration of PTA (days)	31.84 (30.79)	1–140 (Median = 23.5)
GOSE overall Injury mechanism	6.42 (1.26)	3–8
Motor vehicle crash	54 (27.6%)
Pedestrian hit by vehicle	18 (9.2%)
Motorcycle crash	11 (5.6%)
Bicycle crash	6 (3.1%)
Fall	6 (3.1%)
Fall from horse	2 (1.0%)
Work-related injury	6 (3.1%)
Assault	2 (1.0%)

Duration of PTA was available for 104/105 individuals with TBI. GOSE was available for 104/105 individuals with TBI. GOSE: Glasgow Outcome Scale Extended; IQ: intelligence quotient; PTA: post-traumatic amnesia; TBI = traumatic brain injury.

Table 2. Cognitive Reserve Index questionnaire total and subscale scores for participants.

	TBI cohort		Non-TBI cohort
	M (SD)	Range	M (SD)	Range	p-value
Total	114.19 (16.43)	83–170	125.44 (17.54)	91–181	<0.001
Education	105.68 (10.18)	77–133	109.64 (11.88)	78–161	0.013
Work	105.25 (16.30)	75–153	109.99 (16.50)	67–147	0.045
Leisure	121.26 (22.45)	75–195	138.44 (23.51)	93–205	<0.001

M: mean; SD: standard deviation; CRIq: Cognitive Reserve Index questionnaire; TBI: traumatic brain injury. TBI cohort n = 105. Non-TBI cohort n = 91.

Table 3. Correlations between Cognitive Reserve Index questionnaire subscale scores.

TBI cohort
	CRIq Ed		CRIq work		CRIq leisure		Premorbid IQ		Years of Ed
	t	p-value	t	p-value	t	p-value	t	p-value	t	p-value
CRIq total	0.57	<0.001	0.78	<0.001	0.84	<0.001	0.47	<0.001	0.45	<0.001
CRIq Ed	–	–	0.37	<0.001	0.22	0.022	0.46	<0.001	0.69	<0.001
CRIq work	–	–	–	–	0.40	<0.001	0.35	<0.001	0.46	<0.001
CRIq leisure	–	–	–	–	–	–	0.31	0.001	0.09	0.348
Premorbid IQ	–	–	–	–	–	–	–	–	0.54	<0.001
Non-TBI cohort
	CRIq Ed		CRIq work		CRIq leisure		IQ		Years of Ed
	t	p-value	t	p-value	t	p-value	t	p-value	t	p-value
CRIq total	0.61	<0.001	0.71	<0.001	0.86	< 0.001	0.54	<0.001	0.44	<0.001
CRIq Ed	–	–	0.37	<0.001	0.41	< 0.001	0.37	<0.001	0.57	<0.001
CRIq work	–	–	–	–	0.30	0.004	0.45	<0.001	0.44	<0.001
CRIq leisure	–	–	–	–	–	–	0.37	<0.001	0.17	0.112
IQ	–	–	–	–	–	–	–	–	0.54	<0.001

CRIq: Cognitive Reserve Index questionnaire; IQ: intelligence quotient; TBI: traumatic brain injury; Ed: Education. TBI cohort n = 105. Non-TBI cohort n = 91.

Figure 1. Traumatic brain injury cohort (n = 105) model-fit statistics comparing models with CRIq subscales scores as independent variables to a traditional proxies model with premorbid IQ and years of education as independent variables, for all outcome variables. AIC, Akaike Information criterion; BIC, Bayesian Information criterion; CRIq, Cognitive Reserve Index questionnaire; TBI, traumatic brain injury. ΔAIC and ΔBIC represent the differences in AIC and BIC between the two models for each outcome, respectively. ΔAIC or ΔBIC > 2 were considered supportive of a better model fit. ΔAIC of 3–7 or ΔBIC of 2–6 indicate small/moderate evidence, and values > 10 indicate strong evidence for a better model fit.

Table 4. Best fit-model statistics for regression models with either Cognitive Reserve Index questionnaire subscales or traditional proxies as independent variables.

	CRIq model		Proxy model		Model comparison
Outcome variable	AIC	BIC	AIC	BIC	ΔAIC	ΔBIC
TBI cohort
Verbal memory	−4.754	10.996	−4.006	9.119	−0.748	1.877
Visual memory	−2.254	13.496	−1.393	11.731	−0.861	1.765
Executive attention	−21.030	−5.280	−26.437	−13.312	5.407	8.032
Episodic memory	8.168	23.917	−5.110	8.015	13.278	15.902
GOSE	53.550	69.416	49.757	62.979	3.793	6.437
Non-TBI cohort
Verbal memory	−17.630	−2.698	−18.235	−5.792	0.605	3.094
Visual memory	−18.080	−3.148	−18.138	−5.694	0.058	2.546
Executive attention	−21.878	−6.946	−30.452	−18.009	8.574	11.063
Episodic memory	−41.980	−27.048	−32.973	−20.530	−9.007	−6.518

CRIq: Cognitive Reserve Index questionnaire; AIC: Akaike Information criterion; BIC: Bayesian Information criterion; GOSE: Glasgow Outcome Scale Extended; TBI: traumatic brain injury.

ΔAIC and ΔBIC represent the difference in AIC or BIC values from the CRIq to proxy models. ΔAIC and ΔBIC values > 2 indicate evidence for a better model, with ΔAIC of 3–7 or ΔBIC of 2–6 indicative of small/moderate evidence, and values > 10 indicative of strong evidence for a better model fit. TBI cohort n = 105. Non-TBI cohort n = 91.

Figure 2. Non-traumatic brain injury cohort (n = 91) model-fit statistics comparing models with CRIq subscales scores as independent variables to a traditional proxies model with IQ and years of education as independent variables, for all outcome variables. AIC, Akaike Information criterion; BIC, Bayesian Information criterion; CRIq, Cognitive Reserve Index questionnaire. ΔAIC and ΔBIC represent the differences in AIC and BIC between the two models for each outcome, respectively. ΔAIC or ΔBIC > 2 were considered supportive of a better model fit. ΔAIC of 3–7 or ΔBIC of 2–6 indicate small/moderate evidence, and values > 10 indicate strong evidence for a better model fit.

Data availability statement

De-identified data from this study are not available in a public archive. De-identified data from this study will be made available (as allowable according to institutional IRB standards) by emailing the corresponding author as of April 2022.