Page 23 - IJPS-8-1

International Journal of
Population Studies Hearing loss, hearing aids, and cognitive function

risk factor for cognitive decline (Livingston, Sommerlad, associated with trajectories of cognitive function in U.S.
Orgeta, et al., 2017). In the United States, hearing loss is one older adults. Using nationally-representative longitudinal
of the most common health problems in later life (Whitson, data of adults aged 65 and older from 1998 to 2018, we
Cronin-Golomb, Cruickshanks, et al., 2018), impacting characterize how levels of self-reported hearing and
more than 25% of individuals aged 65–74 and upward of hearing aid use are related to changes in cognitive function
50% of individuals aged 75 and older (CDC, 2017). for 20 years in non-Hispanic White, non-Hispanic Black,
and Hispanic older adults. We also account for a wide
Although there is accumulating evidence that hearing
loss is linked to cognitive decline (Whitson, Cronin- array of sociodemographic, behavioral, and health-related
factors that may contribute to the associations.
Golomb, Cruickshanks, et al., 2018), the findings have
been disparate, and the explanations for the associations 2. Data and methods
have varied in the literature (see Wayne and Johnsrude
2015 for a comprehensive review). In particular, most of 2.1. Data
the research on hearing and cognition has examined cross- We analyzed 11 waves of publicly-available RAND
sectional measures of hearing loss – either at baseline Health and Retirement Study (HRS) data from 1998
(Alattar, Bergstrom, Laughlin, et al., 2019; Deal, Betz, to 2018 (RAND Center for the Study of Aging, 2021).
Yaffe, et al., 2017; Ge, McConnell, Wu, et al., 2021; Golub, Sponsored by the National Institute on Aging (grant
Brickman, Ciarleglio, et al., 2020; Golub, Luchsinger, number U01AG009740), the HRS is a nationally-
Manly, et al., 2017; Lin, Yaffe, Xia, et al., 2013) or at the representative prospective study of U.S. adults over the
end of the follow-up period (Deal, Sharrett, Albert, et al., age of 50 that has collected biennial data over the past
2015). Consequently, these studies do not account for the 30 years. Specific details of the multistage sampling
co-occurring changes in hearing status and cognition that design, data collection techniques, and response rates
can occur over time. Indeed, research using longitudinal have been documented extensively elsewhere (Sonnega,
measures of hearing and cognitive function has shown that Faul, Ofstedal, et al., 2014). The current analysis is limited
hearing loss is associated with lower baseline performance to 38,231 participants who were eligible to participate in
on cognitive tests, as well as accelerated declines in 1998-2018. We limited our analysis to adults aged 65 and
cognition compared to those with no hearing loss older who were administered the measures for cognitive
(Maharani, Dawes, Nazroo, et al., 2018b; 2019). However, functioning in the HRS (described below) and aged 85
more research is needed from a life course perspective and younger to minimize the potential influence of
to better understand the short- and long-term impact of selective survival at advanced ages. We further limited
hearing loss to potentially identify individuals who may be our analysis to participants who identified themselves
at greater risk of cognitive decline over time. as Hispanic, non-Hispanic Black or African American
(hereafter referred to as Black), or non-Hispanic White (n
Studies have shown that the prevalence of hearing loss
in the United States is highest in White adults, followed = 21,076). Approximately 3% of the sample had missing
data on at least one measure of hearing or cognitive
by Hispanic and Black adults, respectively (Agrawal, Platz, function and were omitted. Full details of the inclusion/
Niparko, et al., 2008). Studies have also shown that Black exclusion criteria for the analytic sample are provided in
and Hispanic older adults have higher risks for ADRD a flow chart (Figure 1). The final sample included 20,545
compared with White older adults (Babulal, Quiroz, Albensi, individuals who provided a total of 90,990 observations
et al., 2019). However, most studies that have examined for analysis.
racial/ethnic differences in the impact of hearing loss on
cognition have simply controlled for race and/or ethnicity 2.2. Measures
and have not considered possible differences among these
groups (Brenowitz, Kaup, Lin, et al., 2019; Curhan, Willett, Our primary dependent variable was cognitive function.
Cognitive function was ascertained in HRS participants
Grodstein, et al., 2019; Lin, 2011; Lin, Metter, O’Brien, et al., using an adapted version of the Telephone Interview
2011; Wallhagen, Strawbridge, and Shema, 2008). It is also for Cognitive Status (TICS) which was modeled after
unclear whether and to what extent the associations change the Mini-Mental State Examination (Ofstedal, Fisher,
with increasing age among these population groups. Thus, and Herzog, 2005). Beginning in 1998, all age-eligible
there remains limited evidence of racial/ethnic differences participants received the full set of cognitive performance
in the longitudinal association between hearing loss and tests – which included six tasks that measured (i) speed of
cognitive decline at the national level. mental processing, (ii) memory, (iii) working memory, (iv)
The purpose of this report is to examine how age- orientation, (v) knowledge, and (vi) language (Ofstedal,
related changes in hearing loss and hearing-aid use are Fisher, and Herzog, 2005). Correct responses from each

Volume 8 Issue 1 (2022) 17 https://doi.org/10.36922/ijps.v8i1.1308