Hearing loss and cognition: A protocol for ensuring speech understanding in neuropsychological assessment

Abstract

AbstractBackgroundMany neuropsychological tests involve auditory components including test stimuli and instructions. Hearing loss may adversely affect neuropsychological performance through participants not being able to adequately understand auditory stimuli or verbal instructions, thus hampering identification and accurate classification of cognitive impairment. Formal measures to quickly assess speech understanding are lacking. Herein, we describe a protocol developed to determine adequate speech understanding and report on the relationship between hearing loss and speech understanding performance.MethodParticipants were 2864 older adults (mean age: 80.8 years; SD: 4.6) in the Atherosclerosis Risk in Communities Study visit 7 who completed a test to ensure speech understanding, had hearing data, and had covariate data. To ensure speech understanding, a psychometrist verbally presented 5 short sentences (each containing 3 key words) articulating clearly with normal inflections at a volume and pace commensurate with normal cognitive testing methods. After each sentence, the participant reported back what they heard with performance scored as the number of “key” words accurately repeated. A passing score (i.e., adequate speech understanding) was defined as ≥ 13. The protocol included steps on how to proceed with neurocognitive testing if speech understanding was not established. Better‐ear (0.5,1,2,4kHz pure‐tone‐average) hearing was categorized according to World Health Organization Criteria as normal (<25dBHL), or hearing loss (>25dBHL; mild: ≥25dBHL & <40dBHL, moderate or greater: >40dB HL). Logistic regression analyses adjusting for demographic factors (age, sex, race‐center, and education) assessed the association between hearing loss and performance on the ensuring speech understanding test.ResultTest failure was higher with increasing hearing loss severity (normal: 0.23%; mild: 0.52%, moderate or greater: 6.61%). Adjusting for demographic characteristics, hearing loss was associated with reduced odds of passing the ensuring speech understanding test (OR: 0.14, 95% CI: 0.02, 0.44).ConclusionHearing loss is associated with reduced speech understanding in environments commensurate with normal neurocognitive testing. Given the high prevalence of hearing loss among older adults, it is important to ensure neuropsychological performance is accurately captured and not adversely biased by reduced hearing. The protocol described here provides a brief measure to quickly assess speech understanding in the research setting.