Utility of a Short-Form Phonemic Fluency Task

Abstract

Abstract Objective to establish a proof-of-concept and ascertain the reliability of an abbreviated 30-second (30s) phonemic fluency measure as a cognitive screening tool in older adults. Methods in all, 201 English-speaking individuals with normal cognition (NC; n = 119) or cognitive impairment (CI; mild CI or dementia; n = 82) were administered a standard 60s phonemic fluency task (FAS/CFL) with discrete 30s intervals denoted. Results for all letters, 30s trial scores significantly predicted 60s scores for the same letter, R2 = .7–.9, F(1, 200) = 850–915, p < .001. As with 60s total scores, 30s cumulative scores (for all three trials) were significantly different between NC and CI groups (p < .001). Receiver operating characteristic analyses showed that 30s total scores distinguished NC and CI groups as effectively (AUC = .675) as 60s total scores (AUC = .658). Conclusions these findings support the utility and reliability of a short-form phonemic fluency paradigm, as 30s performance reliably predicted 60s/trial totals and was equally accurate in distinguishing impaired/non-impaired groups.