Journal of Rehabilitation Medicine 51-5 | Page 62

378 T. Ownsworth et al. Table I. Receiver operating characteristic (ROC) curve data for SADI Total Scores (State Variable) and Awareness Questionnaire Discrepancy scores for Individuals with TBI (bold font represents the optimal balance between sensitivity and specificity) SADI ≥ 2 (60%) AUC: 0.82 (CI: 0.72–0.91) SADI ≥ 3 (45%) AUC: 0.88 (CI: 0.80–0.96) SADI ≥ 4 (26%) AUC: 0.87 (CI: 0.79–0.96) SADI ≥ 5 (13%) AUC: 0.88 (CI: 0.79–0.96) SADI ≥ 6 (10%) AUC: 0.88 (CI: 0.78–0.98) AQ Discrepancy (% with ISA ) Sensitivity Specificity Sensitivity Specificity Sensitivity Specificity Sensitivity Specificity Sensitivity Specificity ≥ 2 (59) ≥  3 (56) ≥  4 (48) ≥  5 (46) ≥  6 (43) ≥  7 (41) ≥  8 (39) ≥  9 (34) ≥  10 (31) ≥  11 (29) ≥  12 (26) ≥  13 (21) 0.72 0.78 0.84 0.84 0.91 0.91 0.91 0.97 0.97 0.97 0.97 0.97 0.89 0.89 0.81 0.78 0.75 0.75 0.72 0.67 0.61 0.58 0.53 0.42 0.66 0.70 0.80 0.80 0.84 0.86 0.89 0.93 0.93 0.95 0.95 0.95 0.95 0.95 0.95 0.95 0.91 0.91 0.86 0.81 0.76 0.71 0.67 0.52 0.54 0.58 0.70 0.71 0.75 0.76 0.78 0.83 0.85 0.86 0.82 0.89 1.0 1.0 1.0 1.0 0.90 0.90 0.90 0.90 0.90 0.90 0.90 0.60 0.47 0.50 0.60 0.62 0.64 0.66 0.69 0.74 0.77 0.80 0.83 0.84 1.0 1.0 1.0 1.0 0.88 0.88 0.88 0.88 0.88 0.88 0.88 0.63 0.46 0.49 0.58 0.60 0.62 0.64 0.67 0.72 0.75 0.78 0.81 0.89 AQ: Awareness Questionnaire; AUC: area under the curve; CI: 95 th lower and upper confidence intervals; ISA: impaired self-awareness; Self-Awareness of Deficits Interview; TBI: traumatic brain injury. Table II. Consistency of Classification of Impaired Self-Awareness between the Self-Awareness of Deficits Interview and Awareness Questionnaire at Optimal Levels of Sensitivity and Specificity (bold font indicates consistent classification) SADI score (% with ISA) AQ discrepancy score AUC (95 th CI) Sensitivity Specificity 0.79 0.79 0.69 0.67 0.65 0.63 0.58 0.54 0.46 0.42 0.33 0.25 Classification, n (%) ≥ 2 (60%) AQ ≥ 2 (59%) ≥ 3 (45%) AQ ≥ 4 (48%) ≥ 4 (26%) AQ ≥ 9 (34%) ≥ 5 (13%) AQ ≥ 12 (26%) 0.82 (0.72–0.91) 0.88 (0.80–0.96) 0.87 (0.79–0.96) 0.88 (0.79–0.96) 0.79 0.78 0.81 0.80 0.81 0.83 0.90 0.83 ISA AQ ISA 38 (79.2) AQ Good SA 10 (20.8) Overall classification consistency, n (%) 76.3 Good SA ISA 9 (28.1) 29 (80.6) 23 (71.9) 7 (19.4) 80 Good SA ISA 9 (20.5) 17 (81.0) 35 (79.5) 4 (19.0) 83 Good SA ISA Good SA 10 (16.9) 49 (83.1) 9 (90) 1 (10) 84 12 (17.1) 58 (82.9) AQ: Awareness Questionnaire; AUC: area under the curve; ISA: impaired self-awareness; SA: self-awareness. DISCUSSION Fig. 1. Receiver operating characteristic (ROC) curve indicating optimal sensitivity and specificity of the Awareness Questionnaire discrepancy score ≥ 4 relative to the Self-Awareness of Deficits Interview score ≥ 3. www.medicaljournals.se/jrm Good concordance between the AQ and SADI indica- ted that the tools yield consistent information regarding the presence of ISA in community-based individuals. A 4-point AQ discrepancy corresponded to the SADI score (≥ 3) that previously distinguished between good and poor self-awareness groups and community outco- mes (6). Classification consistency was also high when adopting more conservative scores for ISA; namely, SADI ≥ 4~AQ ≥ 9 or SADI ≥ 5~AQ ≥ 12. These AQ discrepancies are smaller than ISA cut-offs (> 20) in an inpatient sample (7). Given that self-awa- reness typically improves following hospital discharge (1, 8), smaller AQ discrepancies may be more sensi- tive to detect persisting ISA in a community sample. Accurate identification of ISA is imperative to guide appropriate management and use of clinical resources. Due to its brevity, the AQ is more feasible for routine administration to detect ISA and determine the need for metacognitive interventions. The SADI could be used as a more comprehensive assessment to determine the nature of awareness deficits (i.e. specific to a particular functional domain or generalized across multiple do- mains) and the extent to which individuals understand