A quick aphasia battery for efficient, reliable, and multidimensional assessment of language function

Abstract

This paper describes a quick aphasia battery (QAB) that aims to provide a reliable and multidimensional assessment of language function in about a quarter of an hour, bridging the gap between comprehensive batteries that are time-consuming to administer, and rapid screening instruments that provide limited detail regarding individual profiles of deficits. The QAB is made up of eight subtests, each comprising sets of items that probe different language domains, vary in difficulty, and are scored with a graded system to maximize the informativeness of each item. From the eight subtests, eight summary measures are derived, which constitute a multidimensional profile of language function, quantifying strengths and weaknesses across core language domains. The QAB was administered to 28 individuals with acute stroke and aphasia, 25 individuals with acute stroke but no aphasia, 16 individuals with chronic post-stroke aphasia, and 14 healthy controls. The patients with chronic post-stroke aphasia were tested 3 times each and scored independently by 2 raters to establish test-retest and inter-rater reliability. The Western Aphasia Battery (WAB) was also administered to these patients to assess concurrent validity. We found that all QAB summary measures were sensitive to aphasic deficits in the two groups with aphasia. All measures showed good or excellent test-retest reliability (overall summary measure: intraclass correlation coefficient (ICC) = 0.98), and excellent inter-rater reliability (overall summary measure: ICC = 0.99). Sensitivity and specificity for diagnosis of aphasia (relative to clinical impression) were 0.91 and 0.95 respectively. All QAB measures were highly correlated with corresponding WAB measures where available. Individual patients showed distinct profiles of spared and impaired function across different language domains. In sum, the QAB efficiently and reliably characterized individual profiles of language deficits.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1. Sample stimulus cards from the QAB.

(A) Action pictures for elicitation of connected speech. (B) Single word comprehension with semantic distractors. (C) Single word comprehension with phonemic distractors. (D) ‘Yes’ and ‘no’. (E) Picture naming. (F) Reading aloud.

Fig 2. Subtest 1: Level of consciousness.

Fig 3. Subtest 2: Connected speech.

Fig 4. Subtest 3: Word comprehension.

Fig 5. Subtest 4: Sentence comprehension.

Fig 6. Subtest 5: Picture naming.

Fig 7. Subtest 6: Repetition.

Fig 8. Subtest 7: Reading aloud.

Fig 9. Subtest 8: Motor speech.

Fig 10. Distributions of the eight QAB summary measures.

Boxes = interquartile range; whiskers = range not including outliers; plusses = outliers; thick horizontal lines = medians; Ac+A = acute stroke patients with aphasia (n = 28); Ac = acute stroke patients without aphasia (n = 25); Ch+A = chronic stroke patients with aphasia (n = 16); HC = healthy control participants (n = 14). Each measure is color-coded to match subsequent figures.

Fig 11. Inter-rater and test-retest reliability.

(A) Inter-rater reliability of each summary measure, across two trained speech-language pathologists who each rated the same 48 evaluations from 16 individuals with chronic post-stroke aphasia. ICC type A-1 = intraclass correlation coefficient, absolute agreement. (B) Test-retest reliability of each measure, based on 16 individuals with chronic post-stroke aphasia who were each evaluated three times. ICC type A-1 = intraclass correlation coefficient, absolute agreement, correlation between any random pair of evaluations. (C) Test-retest reliability of each measure, based on 16 individuals with chronic aphasia who were each evaluated three times. ICC type A-k = intraclass correlation coefficient, absolute agreement, correlation between the means of sets of three random evaluations. Error bars denote 95% confidence intervals. (D) Alternate forms reliability (ICC type A-1).

Fig 12. Determining a cutoff for diagnosis of aphasia.

(A) Receiver operating characteristic (ROC) plot of sensitivity versus specificity. The sensitivity and specificity at the chosen cutoff of 8.9 is indicated with a red circle. (B) Diagnosis based on this cutoff, relative to clinical impression. Green circles = correctly diagnosed patients; Red crosses = incorrectly diagnosed patients.

Fig 13. Concurrent validity with respect to the Western Aphasia Battery (WAB).

Each QAB summary measure was correlated with most similar WAB measure, except for Reading, which was omitted because the written language section of the WAB was not administered. All correlations were significant (p

Fig 14. Profiles of QAB summary measures for each of the 16 individuals with chronic aphasia.

The patients are arranged in groups according to clinical impression, then in ascending order of overall QAB score within each group. Patient scores were averaged across three sessions and two raters, but the scores for each of the three sessions are indicated with black circles to give a sense of test-retest reliability.

Fig 15. Correlation matrix between QAB measures.

Pearson r values for significant correlations are shown (p < 0.05, uncorrected). Included in this analysis were the 28 patients with acute post-stroke aphasia and the 16 patients with chronic post-stroke aphasia.