Correspondence Between Cognitive and Audiological Evaluations Among the Elderly: A Preliminary Report of an Audiological Screening Model of Subjects at Risk of Cognitive Decline With Slight to Moderate Hearing Loss

Abstract

Epidemiological studies show increasing prevalence rates of cognitive decline and hearing loss with age, particularly after the age of 65 years. These conditions are reported to be associated, although conclusive evidence of causality and implications is lacking. Nevertheless, audiological and cognitive assessment among elderly people is a key target for comprehensive and multidisciplinary evaluation of the subject's frailty status. To evaluate the use of tools for identifying older adults at risk of hearing loss and cognitive decline and to compare skills and abilities in terms of hearing and cognitive performances between older adults and young subjects, we performed a prospective cross-sectional study using supraliminal auditory tests. The relationship between cognitive assessment results and audiometric results was investigated, and reference ranges for different ages or stages of disease were determined. Patients older than 65 years with different degrees of hearing function were enrolled. Each subject underwent an extensive audiological assessment, including tonal and speech audiometry, Italian Matrix Sentence Test, and speech audiometry with logatomes in quiet. Cognitive function was screened and then verified by experienced clinicians using the Montreal Cognitive Assessment Score, the Geriatric Depression Scale, and further investigations in some. One hundred twenty-three subjects were finally enrolled during 2016-2019: 103 were >65 years of age and 20 were younger participants (as controls). Cognitive functions showed a correlation with the audiological results in post-lingual hearing-impaired patients, in particular in those affected by slight to moderate hearing loss and aged more than 70 years. Audiological testing can thus be useful in clinical assessment and identification of patients at risk of cognitive impairment. The study was limited by its sample size (CI 95%; CL 10%), strict dependence on language, and hearing threshold. Further investigations should be conducted to confirm the reported results and to verify similar screening models.

Keywords: Italian Matrix Sentence Test; cognitive decline; hearing loss; logatomes; screening; signal-to-noise ratio; slope; speech in noise.

FIGURE 1

Significant differences among the healthy aging group and the at-risk mild cognitive impairment (MCI) group. This graph shows an analysis of variance for parametric and non-parametric variables. All comparisons were significant; thus, the groups identified through the Montreal Cognitive Assessment (MoCA) test (healthy aging and at risk of MCI) were in fact distinct. They also differed significantly in terms of their audiological profile, although they were not differentiated on the basis of their hearing ability. Healthy aging individuals were also statistically distinct from younger subjects. This suggests cognitive and auditory difficulty/fatigability that physiologically accompanies advancing years. These data, already reported in the literature, were not documented by audiometric comparisons that allow determination of whether loss is paraphysiological. In this way, a healthy aging subject is clearly defined with certain audiometric characteristics: (1) the ability to recognize a signal with a signal-to-noise ratio (SNR) < 0; (2) the pure tone audiometry of less than 30 dB HL; (3) an SRT at rest of less than or equal to 30 dB SPL; (4) a difference between max% words in quiet and logatomes of less than 10; and (5) a slope of psychometric function of greater than 10%. S.I., speech intelligibility in noise. See also Tables 1–4.

FIGURE 2

Schematic representation of cognitive decline as defined by different audiological tests. The diagrams summarize the present study. On the y-axis, the results of the Montreal Cognitive Assessment (MoCA) score are shown as the average result of the groups shown in the figure as pentagons (ideally defined by five variables, four audiological parameters plus age). On the x-axis, there are reference levels for the signal-to-noise ratio (SNR) available in the literature, such as a test for matrices with speech in noise sentences. The test was recently chosen for diffusion in clinical practice in Europe for its easy automatic execution and because it yields easily comparable values regardless of age and language. Starting from the available and official reference levels, and moving in steps of 3 standard deviations, so as to include 100% of the study population, it was possible to identify four different populations, which will by definition be statistically significantly different. The groups used in this study fall exactly within those theoretical values: young subjects, elderly but healthy subjects, and elderly subjects suffering from mild cognitive decline either because they were diagnosed or because they are rated using cognitive tests, and finally, at the opposite extreme, patients suffering from dementia who were so severely affected that they were not able to complete the tests.