The current study included N = 90 participants at baseline and n = 82 participants posttreatment. Of the eight participants that did not complete their assessment post-treatment, two participants (n = 2) withdrew from the study and six (n = 6) participants did not complete their outcome assessment following treatment. The majority of included participants were women (63.3%) and White (75.6%). The median age of the sample was 40.5 years (range = 20-65). See Table 1 for full demographic and injury characteristics.
Descriptive statistics for the full sample are shown in Table 2. In terms of single-item avoidance scale scores at baseline/pre-treatment, the median score was 6 (Q1- Q3: 3-8). Posttreatment, the median score was 3 (Q1- Q3: 1.25-6). The total score of the single-item measure of fear avoidance showed minimal floor (7.8%, n = 7) and ceiling effects (14.4%, n = 13) at baseline. At follow up, a greater percentage demonstrated floor levels (23.3%, n = 21), however the ceiling effects were diminished (3.7%, n = 3). Frequency distributions for single-item avoidance scale (Appendix 1a, 1b) and the FAB-TBI (Appendix 2a, 2b) are available in the supplementary materials.
FAB-TBI scores demonstrate a normal, unimodal distribution both at baseline (Appendix 2a) and following treatment (Appendix 2b). Scores for both FAB-TBI and the single-item measure of fear avoidance decreased on average following treatment, as expected, as treatments aimed to reduce fear avoidance behavior (Appendix 1, Appendix 2).
Table 1 Participant demographic and clinical characteristics (N=90)
Variable
|
Value
|
Age, Mean (SD),
|
41.59 (11.1)
|
Sex, n
(% female)
|
57 (63.3%)
|
Ethnicity, n (%)
White
Asian
Middle Eastern
African
Indigenous
Hispanic
|
67 (75.6%)
15 (16.5%)
5 (5.5%)
1 (1.1%)
1 (1.1%)
1 (1.1%)
|
Years Education, M (SD)
|
15.46 (2.31)
|
Mechanism of Injury, n (%)
Moving head
Whiplash no contact
Moving object
|
63 (70.0%)
18 (20.0%)
9 (10.0%)
|
LOC, n (%)
Yes
Suspected
No
Unknown
|
11 (12.2%)
10 (11.1%)
62 (68.9%)
5 (5.6%)
|
PTA, n (%)
Yes
No
Unknown
|
43 (47.8%)
46 (51.1%)
1 (1.1%)
|
Confusion
Yes
No
Unknown
|
84 (93.3%)
4 (4.4%)
1 (1.1%)
|
Note. M = Mean, SD = Standard Deviation, LOC = Loss of consciousness; PTA = post-traumatic amnesia.
Table 2 Descriptive statistics of psychological self-report measures
Scale
|
Baseline
|
Posttreatment
|
Mean (SD)
|
Min - Max
|
Mean (SD)
|
Min - Max
|
Single-item fear avoidance
|
5.8 (2.87)
|
1 - 10
|
3.9 (2.72)
|
1 - 10
|
FAB-TBI Rasch
|
23.7 (7.23)
|
6.05 - 48.0
|
18.2 (6.89)
|
0 - 34.9
|
GAD-7
|
9.2 (5.15)
|
0 - 21
|
5.8 (5.88)
|
0 - 21
|
PHQ-9
|
12.4 (5.70)
|
1 - 27
|
7.7 (5.79)
|
0 - 26
|
WHODAS Rasch
|
19.8 (3.90)
|
10.09 - 27.90
|
14.3 (5.74)
|
0 - 26.26
|
RPQ baseline
|
33.9 (14.41)
|
0 - 61
|
22.9 (14.82)
|
0 - 54
|
Note.
FAB-TBI = Fear Avoidance Behavior after Traumatic Brain Injury Questionnaire
GAD-7 = General Anxiety Questionnaire 7
PHQ-9 = Patient Health Questionnaire
WHODAS = World Health Organization Disability Assessment
RPQ = Rivermead Postconcussion Symptoms Questionnaire
SD = Standard Deviation
At baseline, a moderately strong positive correlation was observed between the single-item fear avoidance measure and FAB-TBI (ρ = .67, p < .001) The single-item fear avoidance measure correlated moderately with anxiety (GAD-7; ρ = .34, p < .001), depression (PHQ-9; ρ = .43, p < .001), post-concussion symptoms (RPQ; ρ = .50, p < .001), and disability (WHODAS; ρ = .43, p < .001). The FAB-TBI was more strongly correlated with these measures: anxiety (ρ = .53; p < .001), depression (ρ = .71; p < .001), post-concussion symptoms (ρ = .73; p < .001) and disability (ρ = .66; p < .001).
Posttreatment, there was a moderate correlation between the single-item fear avoidance measure and FAB-TBI (ρ = .63, p < .001). Following treatment, the single-item fear avoidance measure correlated moderately with anxiety (GAD-7; ρ = .38, p < .001), depression (PHQ-9; ρ = .47, p < .001), post-concussion symptoms (RPQ; ρ = .53, p < .001), and disability (WHODAS; ρ = .53, p < .001). Again, the FAB-TBI was more strongly correlated with these measures: anxiety (ρ = .59; p < .001), depression (ρ = .64; p < .001), post-concussion symptoms (ρ = .72; p < .001) and disability (ρ = .71; p < .001). The correlation between the change scores from baseline to posttreatment for the single-item fear avoidance measure and FAB-TBI was moderate (ρ = .45, p < .001).
Inspection of the Bland-Altman plots suggested moderate agreement between the single-item fear avoidance measure and FAB-TBI both before (Figure 1) and after (Figure 2) treatment. At both baseline and posttreatment, most values were inside the limits of agreement. A mean standardized difference of -0.02 (-1.66, 1.62) was observed between the single-item fear avoidance measure and FAB-TBI at baseline, indicating that the single-item measure did not systematically under- or overestimate FAB-TBI scores. The mean standardized difference following treatment was .78 (-1.04, 2.62), indicating that participants tended to score slightly lower on the single-item fear avoidance measure as compared to FAB-TBI, however this difference is negligible. Posttreatment, participants scored higher on the single-item measure as compared to FAB-TBI. There was a negligible difference between the standardized change scores [.01 (-2.15, 2.16)], indicating that there is high agreement between in the change in standardized scores from baseline to posttreatment. In addition, the relatively symmetrical distribution of difference scores further supports the agreement of the change scores.
Figure 1a
Bland-Altman Plot Exploring Agreement Between Single-Item and FAB-TBI at Baseline
Figure 1b
Bland-Altman Plot Exploring Agreement Between Single-Item and FAB-TBI Posttreatment
Note. Figures 1a and 1b represent the Bland-Altman plots of the total standardized scores of the single-item measure of fear avoidance and FAB-TBI at baseline and posttreatment, respectively.
Standardization involved subtracting the mean from the total score, and dividing by the standard deviation (sd), or [(value – mean)/sd]. The X axis represents the average of the single-item measure of fear avoidance and FAB-TBI, or [(single-item + FAB-TBI)/2]. The Y axis represents the difference between the single-item measure of fear avoidance and FAB-TBI, or single-item minus FAB-TBI. Positive Y axis values indicate that standardized scores in the single-item measure of fear avoidance are greater than standardized FAB-TBI scores. Negative values on the Y axis indicate that standardized FAB-TBI scores were greater than the standardized single-item scores. The dotted lines display the 95% limits of agreement, and the dashed line represents the mean difference.
Figure 1c
Bland-Altman Plot Exploring Agreement Between Change in Scores between Baseline and Posttreatment in the Single-Item and FAB-TBI
Note. Figure 1c represents the change in total standardized scores of the single-item measure of fear avoidance and FAB-TBI between baseline and posttreatment. The change in scores was calculated by subtracting scores at baseline from the scores posttreatment [(score posttreatment) – (score at baseline)], for both the single-item measure and FAB-TBI. The X axis represents the change in the average score of both the single-item measure of fear avoidance and FAB-TBI, or [(change score of single-item + change score of FAB-TBI)/2]. The Y axis represents the difference between the change in scores between baseline and posttreatment in the single-item measure of fear avoidance and FAB-TBI, or [(change score of single-item - change score of FAB-TBI). Positive Y axis values indicate that standardized scores in the single-item measure of fear avoidance are greater than standardized FAB-TBI scores. Negative values on the Y axis indicate that standardized FAB-TBI scores were greater than the standardized single-item scores. The dotted lines display the 95% limits of agreement, and the dashed line represents the mean difference.