Validity and Reliability of the Auditory Consonant Trigrams Test in Multiple Sclerosis

Background:   It is well established that cognitive impairment is present in over half of people with MS (PWMS).  While evaluation through neuropsychological assessment is optimal, access to such services is limited.  Thus, there is an increased need for assessment tools that can be administered by non-neuropsychologists in the clinic setting.  The Brief International Cognitive Assessment for MS (BICAMS) was established for just such a purpose.  Nonetheless, the BICAMS fails to evaluate working memory.  The Auditory Consonant Trigrams test (ACT) is a measure of working memory under conditions of interference that takes approximately 5 minutes to administer.  The utility of ACT in MS has yet to be fully established.

Objectives:   The goal of the present study was to determine if ACT can discriminate between PWMS and healthy controls and to establish its psychometric properties.

Methods: Fifty-seven PWMS recruited from the Ottawa Hospital MS Clinic were matched to 51 healthy controls on age, sex and education.  Participants completed the ACT (9sec and 18sec delay intervals) at baseline and two-week follow-up as part of a larger battery of tests which also included the BICAMS.

Results:   PWMS performed significantly worse than HC on the ACT (9 & 18 sec) at both baseline and follow-up.  On the 9sec ACT 35.1% were classified as impaired (1.5 SD below HC mean) at both baseline and follow-up.  On the 18sec ACT 17.5% and 29.8% were classified as impaired at baseline and follow-up, respectively.  The sensitivity of the ACT is comparable to that of the individual BICAMS measures.  Test-retest reliability of both the 9sec (r = 0.52, p <.001) and 18 sec (r = 0.56, p < .001) ACT intervals was statistically significant, though the effect size was only moderate.

Conclusions:   Findings indicate that ACT discriminated between PWMS and HC and was able to identify impairment in a large proportion of the sample to a similar degree as a previously validated measure (BICAMS).  While test-retest reliability was moderate, situational variables may also contribute to the variance in test scores.  Given that ACT taps into an area of cognitive functioning that the BICAMS does not, working memory under conditions of interference (an area of difficulty often reported by PWMS), clinicians may wish to consider supplementing evaluations in the clinic with this measure.