Racial and ethnic differences in the associations of sensory impairments with cognition in older United States adults

Abstract

Background

Sensory impairments are associated with cognitive decline and Alzheimer's disease risk.

Objective

We studied how sensory function and cognitive change varied by race/ethnicity in Hispanic, non-Hispanic Black (NHB), and non-Hispanic White (NHW) older US adults.

Methods

We studied 2799 individuals, 57–85 years-old, in the National Social Life, Health, and Aging Project with baseline sensory and follow-up Montreal Cognitive Assessment (MoCA) information from 2005–2016. Hearing was interviewer-rated; vision, smell, taste, and touch were objectively measured. Senses were categorized (poor, moderate, normal); number of poor senses (impairment) was summed. Multivariable linear regressions evaluated the association between sensory impairments and cognition, adjusting for demographic and clinical variables. We tested for differences by race/ethnicity using interaction terms.

Results

Hispanic (n = 180) and NHB (n = 509) older adults had higher prevalence of ≥1 sensory impairments and lower cognition scores versus NHW respondents (n = 2110). Having multiple impairments significantly interacted with race/ethnicity such that estimates for the association with cognition was significantly lower for NHB versus NHW respondents (beta = −3.08, 95% CI: −5.14, −1.02; p = 0.004). Interactions were borderline significant for smell, where estimated difference in cognition associated with moderate (β=−0.91, 95% CI: −1.91, 0.10, p = 0.077) and poor (β=−1.33, 95% CI: −2.77, 0.11, p = 0.071) smell was lower for NHB versus NHW participants.

Conclusions

Our findings suggest racial/ethnic differences in how strongly sensory impairments predict cognition, especially for number of impairments and possibly smell. Further insight may reduce disparities in cognitive aging.

Keywords

Alzheimer's disease cognition epidemiology sensory impairment vision

Introduction

Up to 94% of United States (US) adults aged 65 and older have loss of function in at least one sensory domain (e.g., hearing, vision, smell, taste, touch).¹ Sensory impairments affect quality of life and are associated with increased risk for cognitive decline, dementia, and other poor health outcomes.^2–8 Previous studies have identified associations between cognition and individual sensory impairments in hearing,^9,10 vision,^11,12 smell,¹³ and touch,¹⁴ as well as multiple sensory deficits.¹⁵ However, many prior studies have been conducted in White and European samples, or White and Black Americans. Few studies with sensory assessments have included Hispanic/Latino individuals or examined whether associations differ by race/ethnicity.^16,17

Rates of dementia are 1.5 to 2 times higher in non-Hispanic Black Americans and Hispanic Americans compared to White individuals, and the prevalence of sensory impairments may also differ by race/ethnicity.¹⁸ Additional research on racial differences in prevalence of sensory impairments might help clarify which groups are impacted disproportionately. From a public health standpoint, exploring racial and ethnic associations with health outcomes may accelerate efforts to reduce risk for cognitive decline in patients who are known to have sensory impairments. It is also important to understand the extent to which differences by race/ethnicity are driven by social determinants of health and/or higher late-life health burden of cardiovascular or other comorbidities.

The objective of this research was to study how sensory function (in vision, hearing, smell, taste, touch) and cognition varied by race and ethnicity in older, community-dwelling US adults. We studied Hispanic, non-Hispanic Black, and non-Hispanic White adults in the National Social Life, Health, and Aging Project (NSHAP) and examined correlations and interactions between race/ethnicity, sensory impairments, and cognitive function and decline.

Methods

Study sample

NSHAP is a population-based longitudinal study that examines the health and social well-being of community-dwelling older Americans.^19,20 By obtaining a wide range of assessments, including on sensory function, chronic disease, and physical and cognitive health of older adults, NSHAP can provide healthcare professionals with insights into the multilevel factors that affect the overall health of older adults.²⁰ It might also be the first national survey to contain information on all five senses concurrently.²⁰ The National Opinion Research Center and University of Chicago conducted in-home interviews of older adults who were born in the United States between 1920–1947.^1,20 2799 older male and female American adults aged 57 to 85 at time of recruitment comprised our analytic sample. Respondent data was collected in 2005 to 2006 through NSHAP Round 1 interviews.²¹ Subsequent interviews of the participants were taken in Round 2 (conducted in 2010 to 2011)²² and Round 3 (2015 to 2016).²³ Participants were excluded due to reporting Asian, Pacific Islander, American Indian, and/or Alaskan Native race (n = 193) because of the small count or not reporting race (n = 13).

Sensory function

Vision, hearing, smell, taste, and touch were tested in Round 1. We categorized participants into “normal”, “moderate”, and “poor” sensory function based on previously published cut-points.¹

Vision. Visual acuity was tested binocularly based on the Snellen eye chart²⁴; participants were instructed to wear glasses if they normally used them for distance vision. Visual acuity from 20/200 to 20/80 was defined as ‘poor’, 20/63 and 20/50 as ‘moderate’, and 20/40 to 20/10 as ‘normal’.

Hearing. Interviewers were instructed to “Please rate the respondent's functional health and behavior during the interview on the following scales: practically deaf versus normal hearing?” where 1 indicated practically deaf and 5 indicated normal. Cut points were as follows: a score of 1 or 2 was identified as ‘poor’, 3 was ‘moderate’, and 4 or 5 were ‘normal’ functioning.²⁴

Smell. To assess olfactory function, participants smelled scented pens of common odors: rose, leather, orange, fish, and peppermint.^25,26 After smelling the scent, participants were shown four answer choices and asked to identify the presented scent. Odor identification was scored based on the number of errors from 0 to 5. Individuals who made 3, 4, or 5 errors were defined as having ‘poor’ functioning, 2 errors were defined as ‘moderate’, and 0 or 1 errors were defined as having ‘normal’ smell functioning.

Taste. Gustatory function was measured via tasting flavored paper strips that were sour, sweet, bitter, or salty. Respondents selected their response from these four choices. Like olfactory function scoring, cut points for taste were based on the number of errors. ‘Poor’ taste was defined as 3 or 4 errors, ‘moderate’ was 1 or 2 errors, and ‘normal’ functioning was 0 errors.²⁴

Touch. Tactile acuity was evaluated using single-point discrimination and two-point discrimination.²⁴ Participants were asked to indicate the number of points they felt. 2-point tests were administered at distances of 4 mm, 8 mm, and 12 mm on the skin. Individuals who correctly identified the single-point test, and the two-point 12 mm and two-point 8 mm tests, but not the two-point 4 mm test were identified as having ‘moderate’ touch functioning. Individuals who had correct responses on all four tests were categorized into ‘normal’ touch functioning. All other individuals were indicated as having ‘poor’ tactile functioning as long as they had no missing data.

Multiple. Among participants who completed all sensory tests, we categorized participants as having none, one, or multiple (2+) sensory impairments based on the number of senses with “poor” function. Participants also completed self-reported questionnaires for each sense, which we examined in secondary analyses.

Cognition

Cognition was measured using the Montreal Cognitive Assessment (MoCA) in Round 2 (∼5 years after Round 1) and Round 3 (∼5 years after Round 2)²⁷ (it was not assessed in Round 1). The MoCA was the only measurement of cognitive function administered in Round 2 and Round 3. The MoCA is a test of overall cognitive functioning, with components testing several cognitive domains, including memory, visuo-constructional skills, calculations, and others.²⁸ The MoCA has been validated as a sensitive tool for early detection of cognitive decline and mild cognitive impairment.²⁸ We examined cognition using Round 2 MoCA scores and cognitive change between Round 2 and Round 3 as separate continuous outcomes.

Other characteristics

Age at baseline of the respondent was calculated at Round 1 from date of birth. Sex, race, and Hispanic/Latino identity were self-reported. Race categories included ‘White/Caucasian’, ‘Black/African American’, ‘Asian, Pacific Islander, American Indian, or Alaskan Native’, ‘don’t know’, ‘refused’, or ‘not applicable’. Respondents were also asked whether they considered themselves Hispanic or Latino. Based on participant responses to race and Hispanic/Latino identity, ethnicity was designated in NSHAP as ‘White’, ‘Black’, ‘Hispanic, non-Black’, and ‘Other.’ We then categorized race/ethnicity as non-Hispanic White, Hispanic, and non-Hispanic Black. Education was categorized as the highest degree obtained, with options being ‘less than high school’, ‘high school/equivalent’, ‘vocational certificate / some college / association’, and ‘bachelors or more’. One of eight options was reported for household income: ‘0–24,999’, ‘25,000–49,999’, ‘50,000–99,999’, ‘100k or higher’, ‘refused’, ‘don’t know’, ‘not applicable’, ‘and ‘incomplete interview’. We pooled the last four responses (refused, don’t know, not applicable, and incomplete interview) into one category called ‘unknown’. Similarly, participants were asked whether they are covered by private insurance, with options being ‘no’, ‘yes,’ and we categorized all other responses as ‘unknown’ (refused, don’t know, not applicable, no answer, not returned). History of hypertension, diabetes, and stroke were self-reported at Round 1. Participants reported whether or not they ever smoked cigarettes regularly and if they smoke (coded as never, former, current). Similarly, respondents were asked if they ever drank alcohol regularly and if they drink alcohol (coded as never, former, current).

Statistical analyses

We performed descriptive statistics of demographic (age, sex, education, household income, private insurance) and clinical (hypertension, diabetes, stroke, smoking, drinking) characteristics by race/ethnicity. We also examined the prevalence of sensory impairments (vision, hearing, smell, taste, touch) by race/ethnicity. To inform whether any differences in prevalence of sensory impairments were driven by other demographic factors beyond race/ethnicity, we used logistic regressions to evaluate the association between race/ethnicity and each poor sensory impairment (versus normal or moderate) adjusting for age, sex, education, household income, and private insurance. Participants in the analytic sample had complete race, sensory, and cognitive test data.

We used linear regression models to evaluate associations between each sensory function and cognition at Round 2 for participants overall. Then, we further tested interactions between race/ethnicity and each sense. We examined race/ethnicity stratified estimates by each sensory level for any significant or borderline significant interactions, as well as overall interactions using a likelihood ratio test. We repeated analyses for the change in cognition from Round 2–3 (∼5 years). All regression models additionally adjusted for age, sex, education, race/ethnicity, comorbidities (hypertension, diabetes, stroke), smoking and alcohol use, household income, private insurance status, and time between Round 1 and Round 2; models of change in cognition also included adjustment for time between Round 2 and Round 3. The mean time between Round 3 and 2 was 5.05 years, with a standard deviation of 0.40. The mean time between Round 2 and 1 was 5.15 years, with a standard deviation of 0.42. Self-reported sensory function may capture different aspects of sensory loss and have a different relationship with cognition, so we examined self-reported sensory function as a secondary analysis. Analyses were conducted in R (version 4.3.0). All tests were two-sided with α = 0.05; we report 95% confidence intervals.

Institutional Review Boards at National Opinion Research Center at University of Chicago approved this study.

Data availability

All data from NSHAP Rounds 1, 2, and 3 are publicly available through the National Archive of Computerized Data on Aging (NACDA) https://www.icpsr.umich.edu/web/pages/NACDA/colectica-portal.html.

Results

Differences in participant characteristics by race/ethnicity

There were 2799 respondents that comprised our analytic sample, including Hispanic (N = 180), non-Hispanic Black (N = 509), and non-Hispanic White (N = 2110) participants, with mean age of 69. Table 1 shows demographics (age, sex, education, household income, private insurance) and prevalence of clinical variables (hypertension, diabetes, stroke, smoking cigarettes, drinking alcohol) for participants overall and by race/ethnicity. Hispanic and non-Hispanic Black participants were on average younger and less educated than non-Hispanic White participants; non-Hispanic Black participants also had higher prevalence of diabetes, stroke, and hypertension (Table 1). Of the respondents, non-Hispanic White individuals had the greatest prevalence of private insurance and the highest household income brackets. Habits for smoking cigarettes and drinking alcohol also varied by race/ethnicity. Hispanic respondents had the lowest prevalence of current smoking, and non-Hispanic Black respondents had the lowest prevalence of current drinking.

Table 1.

Participant characteristics.

	Race and Ethnicity
Characteristic¹	Overall N = 2799	Hispanic N = 180	Non-Hispanic Black N = 509	Non-Hispanic White N = 2110	p²
Age, Mean (SD)	69 (8)	68 (7)	69 (7)	70 (8)	0.023
Female Sex, n (%)	1444 (52%)	80 (44%)	285 (56%)	1079 (51%)	0.020
Education, n (%)					<0.001
<high school	599 (21%)	95 (53%)	208 (41%)	296 (14%)
high school/equivalent	764 (27%)	26 (14%)	120 (24%)	618 (29%)
Some college/vocational	809 (29%)	35 (19%)	118 (23%)	656 (31%)
Bachelors or more	627 (22%)	24 (13%)	63 (12%)	540 (26%)
Income, n (%)					<0.001
unknown	818 (29%)	52 (29%)	170 (33%)	596 (28%)
0–24,999	681 (24%)	71 (39%)	198 (39%)	412 (20%)
25,000–49,999	561 (20%)	24 (13%)	81 (16%)	456 (22%)
50,000–99,999	506 (18%)	26 (14%)	51 (10%)	429 (20%)
100k or higher	233 (8.3%)	7 (3.9%)	9 (1.8%)	217 (10%)
Private Insurance, n (%)	1361 (49%)	46 (26%)	123 (24%)	1192 (56%)	<0.001
Hypertension, n (%)	1604 (57%)	98 (54%)	371 (73%)	1135 (54%)	<0.001
Diabetes, n (%)	583 (21%)	39 (22%)	167 (33%)	377 (18%)	<0.001
Stroke, n (%)	252 (9.0%)	3 (1.7%)	62 (12%)	187 (8.9%)	<0.001
Smoking Cigarettes, n (%)					<0.001
current smoker	414 (15%)	15 (8.4%)	113 (22%)	286 (14%)
former smoker	1221 (44%)	63 (35%)	192 (38%)	966 (46%)
never smoked	1161 (42%)	101 (56%)	202 (40%)	858 (41%)
Drinking Alcohol, n (%)
current drinker	1556 (56%)	89 (50%)	206 (41%)	1261 (60%)
former drinker	779 (29%)	47 (26%)	189 (37%)	563 (27%)
never drank	439 (16%)	43 (24%)	112 (22%)	284 (13%)

8 total observations of 2799 missing for Smoking Cigarettes and Drinking Alcohol.

Kruskal-Wallis rank sum test; Pearson's Chi-squared test.

Hispanic and non-Hispanic Black participants tended to have worse sensory outcomes compared to non-Hispanic White participants (Table 2). Non-Hispanic Black participants had a higher prevalence of poor vision (10%), lower prevalence of poor hearing (4.9%), and higher prevalence of poor smell (14%) compared to Hispanic (4.9%, 5.0%, and 9.2%, respectively) and non-Hispanic White participants (5.3%, 6.4%, 6.2%). Hispanic and non-Hispanic Black individuals both had higher prevalence of poor touch (54% and 47%, respectively) and one or multiple impairments (61% and 58%) compared to non-Hispanic White individuals (poor touch: 30%, one or multiple impairments: 44%). Overall, moderate or poor sensory function was common, especially for touch and taste tests.

Table 2.

Prevalence of sensory impairments by race.

	Hispanic N = 180	non-Hispanic Black N = 509	non-Hispanic White N = 2110
Characteristic	n (%)	n (%)	n (%)
Vision
poor	4 (4.9%)	23 (10%)	53 (5.3%)
moderate	15 (19%)	45 (20%)	151 (15%)
normal	62 (77%)	156 (70%)	802 (80%)
Hearing
poor	9 (5.0%)	25 (4.9%)	135 (6.4%)
moderate	24 (13%)	68 (13%)	277 (13%)
normal	146 (82%)	415 (82%)	1698 (80%)
Smell
poor	16 (9.2%)	65 (14%)	122 (6.2%)
moderate	23 (13%)	102 (23%)	261 (13%)
normal	134 (77%)	284 (63%)	1581 (80%)
Taste
poor	17 (11%)	57 (14%)	259 (15%)
moderate	83 (55%)	251 (63%)	969 (57%)
normal	51 (34%)	89 (22%)	483 (28%)
Touch
poor	45 (54%)	114 (47%)	308 (30%)
moderate	23 (28%)	66 (27%)	400 (39%)
normal	15 (18%)	61 (25%)	321 (31%)
Number of Impairments
none	25 (39%)	69 (41%)	452 (56%)
one	25 (39%)	68 (40%)	273 (34%)
multiple	14 (22%)	31 (18%)	83 (10%)

In exploratory models adjusting for age, sex, education, household income, and private insurance, there were still several significant disparities in sensory impairment prevalence by race, primarily comparing non-Hispanic Black individuals compared to non-Hispanic White individuals (Table 3). Hispanic individuals had significantly greater odds of touch impairment compared to non-Hispanic White individuals (2.03, 95% CI: 1.14, 3.81, p < 0.05). Non-Hispanic Black individuals had significantly greater odds of impaired vision (1.56, 95% CI: 1.07, 2.26, p < 0.05), smell (2.36, 95% CI: 1.83, 3.04, p < 0.05), and taste (1.37, 95% CI: 1.04, 1.82, p < 0.05), with odds of smell dysfunction being markedly higher compared to non-Hispanic White participants.

Table 3.

Associations of race/ethnicity with poor sensory function (outcomes).

	Poor Vision (N = 1311)	Poor Hearing (N = 2797)	Poor Smell (N = 2588)	Poor Taste (N = 2259)	Poor Touch (N = 1353)
Race/Ethnicity	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
Hispanic	1.18 (0.64, 2.09)	0.78 (0.50, 1.18)	1.10 (0.73, 1.63)	0.71 (0.49, 1.04)	2.03 (1.14, 3.81)*
Non-Hispanic Black	1.56 (1.07, 2.26)*	0.77 (0.58, 1.02)	2.36 (1.83, 3.04)*	1.37 (1.04, 1.82)*	1.22 (0.86, 1.74)
Non-Hispanic White	Ref	Ref	Ref	Ref	Ref

*p < 0.05.

Based on logistic regression models with poor sense as the outcome and adjusted for age, sex, education, income, and health insurance status.

Overall associations between sensory function and cognition

Among participants overall, moderate hearing function (−0.57, 95% CI: −0.99, −0.15, p < 0.01), moderate smell function (−0.77, 95% CI: −1.19, −0.36, p < 0.001), poor smell function (−0.84, 95% CI: −1.46, −0.21, p < 0.01), and poor taste function (−0.57, 95% CI: −1.03, −0.10, p < 0.05) were associated with worse cognition compared to normal function at Round 2 (Table 4). There were no significant associations between touch and vision with cognition at Round 2 (Table 4). Number of impairments was not overall associated with cognitive function, but we observed trends linking increased number of impairments with worse cognition. Trends were mostly negative for sensory impairments and cognitive outcomes (Table 4). The strength of the association was greatest for smell. For cognitive decline from Round 2 to 3, sample sizes were small, and estimates were inconsistent (Table 4). Only poor vision was associated with a significant decline in cognition from Round 2 to 3 (−1.92, 95% CI: −3.33, −0.51, p < 0.01).

Table 4.

Association between sensory impairments and cognitive outcomes at round 2 and 3 analyzed with linear regressions.

Sense	Level	Round 2 Cognition^a	Decline from Round 2 to Round 3^b
Vision	Normal	Ref	Ref
	Moderate	−0.16 (−0.76, 0.44)	0.18 (−0.65, 1.00)
	Poor	−0.23 (−1.17, 0.70)	−1.92 (−3.33, −0.51)**
Hearing	Normal	Ref	Ref
	Moderate	−0.57 (−0.99, −0.15)**	−0.19 (−0.73, 0.36)
	Poor	−0.34 (−1.02, 0.34)	0.46 (−0.56, 1.48)
Smell	Normal	Ref	Ref
	Moderate	−0.77 (−1.19, −0.36)***	−0.23 (−0.77, 0.31)
	Poor	−0.84 (−1.46, −0.21)**	−0.56 (−1.52, 0.41)
Taste	Normal	Ref	Ref
	Moderate	−0.26 (−0.58, 0.06)	−0.25 (−0.64, 0.13)
	Poor	−0.57 (−1.03, −0.10)*	0.25 (−0.33, 0.83)
Touch	Normal	Ref	Ref
	Moderate	−0.31 (−0.75, 0.13)	0.24 (−0.31, 0.79)
	Poor	−0.20 (−0.68, 0.27)	0.26 (−0.35, 0.88)
Number of Impairments	None	Ref	Ref
	One	−0.32 (−0.76, 0.12)	−0.07 (−0.62, 0.49)
	Multiple	−0.70 (−1.47, 0.08)	−0.22 (−1.34, 0.90)

*p < 0.05, **p < 0.01, ***p < 0.001.

^aSample sizes in models: Vision = 831, Hearing = 1699, Smell = 1615, Taste = 1414, Touch = 843, Number of impairments = 696.

^bSample sizes in models: Vision = 499, Hearing = 1047, Smell = 1015, Taste = 897, Touch = 505, Number of impairments = 435.

Models included: Age, Education, Sex, Race/Ethnicity, Hypertension, Diabetes, Stroke, Household Income, Private Insurance, Smoking Cigarettes, Drinking Alcohol, and time between Round 1 and Round 2 for Round 2 cognition estimates or time between Round 2 and Round 3 for cognitive decline estimates.

We examined self-reported sensory function in a secondary analysis; results were generally not significantly associated with cognitive outcomes at Round 2 or cognitive decline from Round 2 to 3 (Supplemental Table 1), with the exception of poor vision for participants overall at Round 2 and moderate vision for participants overall from Round 2 to 3. Very few participants (n = 35) reported Parkinson's disease in Round 2, so we did not examine this as an explanatory variable further.

Interactions by race or ethnicity for sensory function and cognition

In models with Round 2 cognition as an outcome, we tested interactions with race/ethnicity for each of the sensory impairments at each level of functioning (Supplemental Table 2). The only trends for interactions by race/ethnicity were for smell and multiple sensory impairments, so we focused on these below. Moderate (−0.91, 95% CI: −1.91, 0.10, p = 0.077) and poor (−1.33, 95% CI: −2.77, 0.11, p = 0.071) smell functioning were borderline associated with lower cognition for non-Hispanic Black participants, compared to non-Hispanic White participants (Supplemental Table 2). Likewise, estimates for having one sensory impairment compared to none was also borderline associated with lower cognition for non-Hispanic Black compared to non-Hispanic White respondents (−1.19, 95% CI: −2.50, 0.12, p = 0.074) (Supplemental Table 3). Having multiple impairments was associated with significantly lower cognition for non-Hispanic Black participants (−3.08, 95% CI: −5.14, −1.02; p = 0.004), compared to non-Hispanic White participants (Supplemental Table 3).

Figure 1A and 1B illustrate the associations of smell and number of impairments, respectively, with cognition stratified by race/ethnicity. For both smell and multiple impairments, associations with cognition were minimal among non-Hispanic White participants and trends for worse sensory impairment and cognition were most pronounced among non-Hispanic Black participants with Hispanic participants in between. Furthermore, trends were graded such that the poor smell and multiple impairments had the highest estimated magnitude of association. Confidence intervals were wider for the smaller sample of Hispanic participants. We also examined adjustments for additional demographic and clinical variables (e.g., hypertension, diabetes, stroke, private insurance, smoking cigarettes, and drinking alcohol) to determine if interactions were explained by other factors; while estimates after adjustment for clinical variables were less precise, trends were similar.

Discussion

We studied the relationships between sensory impairments, cognition, and race/ethnicity in a well-described population-based study of older US adults with multiple measures of sensory function. Our results demonstrated that traditionally underrepresented populations, including non-Hispanic Black and Hispanic participants, trended towards worse levels of sensory impairments compared to non-Hispanic White individuals; these differences were not fully explained by age, sex, education, household income, private insurance, or clinical comorbidities. We also confirmed associations between vision, smell, and multiple sensory impairments and lower cognition. We further studied interactions between race and sensory impairment and found that smell and number of sensory impairments were more strongly associated with lower cognition among Hispanic and especially non-Hispanic Black participants compared to non-Hispanic White participants. These findings suggest that sensory impairments may exacerbate racial disparities in cognitive outcomes among older adults.

This work builds on additional evidence showing overall associations between sensory impairments and cognitive function. We found associations among participants overall for vision, smell, and multiple sensory impairments, similar to other work.^4,5,29 Sensory impairments may be associated with lower cognitive function through several mechanisms.³⁰ First sensory loss and cognitive decline are strongly age related and may share other common causes including vascular disease, inflammation, and other biologic aging processes.^31,32 Second, sensory processing and closely linked peripheral regions (e.g., olfactory bulb) may be impacted by neurodegeneration from Alzheimer's disease and Parkinson's or Lewy body disease, which also impairs overall cognition.^33–35 Finally, some senses, such as hearing and vision, may hasten cognitive decline either directly (e.g., through changes to brain structure or function)³⁶ or indirectly (e.g., through loneliness, depression, physical mobility)³⁷ contributing to worse cognition and faster cognitive decline.⁷

This study addresses an important gap in prior literature on race/ethnicity differences in the link between sensory function and cognition. There are studies that have examined differences by race or ethnicity in prevalence of sensory function, without measuring associations with cognition.³⁸ We build on these studies, as well as on a number of studies showing associations between individual and multiple sensory impairments and worse cognitive outcomes among older adults and some in different geographic locations.^39–45 Some studies on the association between sensory impairments and cognitive decline do not study differences by race and ethnicity.^3,40,46 One study found associations between multisensory impairment and dementia for both Black and White older adults, but did not have power to test interactions.¹⁵ Multiple studies have found associations between hearing, vision, and dual sensory impairment and cognitive decline in Asian populations.^47–50 Few previous studies have examined taste or touch in association with cognition or evaluated differences by race/ethnicity.^14,51,52 Although some work has found associations with touch overall,¹⁴ we did not find overall associations with touch or interactions by race/ethnicity. However, we did find that poor taste was associated with lower cognition at Round 2, although there was no evidence for interaction by race/ethnicity.

Our study suggests that racial differences may exist in the association between sensory impairment and cognitive decline. In particular, we found that associations between smell and multiple sensory impairments were elevated for non-Hispanic Blacks and borderline elevated for Hispanics, compared to non-Hispanic Whites. Interestingly, these results were not explained by adjustment for clinical variables, income, education or health insurance type. There could be other important aspects of social determinants of health that were not measured in our study⁵³ that may drive our findings. For example, nutritional health,⁵⁴ racism and discrimination,^55,56 physical risk and hazard of respondents’ occupation,^57,58 neighborhood pollution,^59,60 residential condition,^61,62 geographical location as urban or rural,^63,64 quality of education rather than just educational level,^65,66 satisfaction of healthcare,^67,68 and cultural beliefs^69,70 have been associated with sensory impairments and/or worse cognition. These factors were not captured in our study and could account for our findings if any were associated with both sensory impairments and cognition. The exact social determinants are unclear as they have not been comprehensively studied for all senses. In general, though, negative social determinants are associated with worse cognition,⁶⁵ poorer sensory function,⁷¹ and other poor health outcomes.⁷² Imperfect measurement of any of the variables we included (e.g., education) could also lead to residual confounding and explain associations.⁷³ Further, prior research suggests sensory loss may intersect with race/ethnicity to exacerbate health disparities, such as health care access.^53,74 Sensory impairments are associated with decreased self-rated mental health and frequent loneliness, so increased awareness of this is important to provide patient-centered care.^75,76 Additional larger studies with robust social determinants of health measures should be conducted to validate our findings in other populations and to determine drivers of these differences.

Our study has several important limitations. Sensory impairments for all senses were only measured at Round 1 so we could not examine changes in sensory function. For hearing in particular, measurements were interviewer-rated and thus not objective, so ratings could have been influenced by external factors such as how well the interviewer communicated with the respondent. This may affect our results by misestimating the prevalence of hearing impairment in our study population. For participants who wore corrective eyewear, vision results could have varied between study waves based on any improvements in their eyewear prescriptions and differences in income or insurance by race and ethnicity that impact access to corrective wear.⁷⁷ There was more missing data for vision, touch, and number of impairments in models examining Round 2 cognition and decline from Round 2 to 3, which may impact generalizability. Only one measure of cognition (MoCA) was available in NSHAP, but it was not administered in Round 1, limiting options for longitudinal modeling from baseline. A different measure was assessed at Round 1 and has been shown to have limited sensitivity.^78,79 96% scored cognitively normal in Round 1 thus limiting ability to use this in models due to lack of variation.⁸⁰ Although we considered demographics and potential comorbidities, there is still some potential that other life factors or variables that were not considered in our study could have influenced the cognitive and sensory outcomes of participants over time. Though we detected some interactions by race/ethnicity, the sample sizes, especially for Hispanic participants may not have been sufficient to detect all interactions. We were also only able to study Hispanic, non-Hispanic Black, and non-Hispanic White participants, thus findings may not be generalizable to other racial and ethnic backgrounds. Future large-scale studies should collect objective sensory and cognitive data to examine associations among US older adults from other racial and ethnic backgrounds.

Nevertheless, our study is robust due to the longitudinal nature of the data which examined participants over 10 to 11 years. We leveraged majority objective sensory measures and examined understudied sensory measures including smell, touch, taste, and number of sensory impairments, in addition to hearing and vision, which are more commonly studied. Our study also goes beyond current research by further testing race/ethnicity and sensory interactions in the association with cognition and examining the extent to which demographic and clinical comorbidities could explain differences by race/ethnicity.

Our research indicates that there are racial and ethnic differences in the prevalence of sensory impairments and their association with cognition in older US adults. For smell and number of impairments, the association with cognition was stronger among non-Hispanic Black participants compared to non-Hispanic White participants with a similar trend among Hispanic participants. Studying demographic variables that influence cognitive and sensory function is important for identifying disparities in health status, determining a group's potential risk for health decline, and targeting interventions to improve health outcomes. Future work should determine whether diagnosing and treating sensory impairments in early stages may help prevent cognitive decline in individuals from minoritized groups. Further research is also necessary to identify the underlying mechanisms of differences in associations, such as incorporating information on health inequities.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251387115 - Supplemental material for Racial and ethnic differences in the associations of sensory impairments with cognition in older United States adults

Supplemental material, sj-docx-1-alz-10.1177_13872877251387115 for Racial and ethnic differences in the associations of sensory impairments with cognition in older United States adults by Faaizah Arshad, Jayant M Pinto, Kristine Yaffe and Willa D Brenowitz in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

We deeply appreciate NSHAP participants for their generous participation and acknowledge the individuals at the National Opinion Research Center at the University of Chicago for their efforts in collecting the data and making it available for public use.

ORCID iD

Faaizah Arshad

Ethical considerations

An IRB at National Opinion Research Center at University of Chicago approved this study, and participants gave written informed consent.

Consent to participate

Not applicable

Consent for publication

Not applicable

Author contribution(s)

Faaizah Arshad: Conceptualization; Data curation; Formal analysis; Methodology; Software; Visualization; Writing – original draft; Writing – review & editing.

Jayant M Pinto: Conceptualization; Data curation; Methodology; Writing – review & editing.

Kristine Yaffe: Conceptualization; Methodology; Writing – review & editing.

Willa D Brenowitz: Conceptualization; Funding acquisition; Methodology; Supervision; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute on Aging (K01AG062722); and a grant from the Alzheimer's Association (AARF-18-565846). The National Social Life, Health and Aging Project is supported by the National Institutes of Health, National Institute on Aging (R01AG021487, R37AG030481, R01AG033903, R01AG043538, R01AG048511), and National Institute on Drug Abuse (AG043538-08S1). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

All data from NSHAP Rounds 1, 2, and 3 supporting the findings of this study are openly available through the National Archive of Computerized Data on Aging (NACDA) at .

Supplemental material

Supplemental material for this article is available online.

References

Correia

Lopez

Wroblewski

, et al. Global sensory impairment among older adults in the United States. J Am Geriatr Soc 2016; 64: 306.

Dintica

Calderón-Larrañaga

Vetrano

, et al. Association between sensory impairment and dementia: the roles of social network and leisure activity. J Alzheimers Dis 2023; 94: 585.

Schubert

Cruickshanks

Fischer

, et al. Sensory impairments and cognitive function in middle-aged adults. J Gerontol A Biol Sci Med Sci 2017; 72: 1087.

Wang

, et al. Association of visual, hearing, and dual sensory impairment with incident dementia. Front Aging Neurosci 2022; 14: 872967.

Hwang

Longstreth

Jr Thielke

, et al. Longitudinal changes in hearing and visual impairments and risk of dementia in older adults in the United States. JAMA Network Open 2022; 5: e2210734.

Armstrong

Teixeira

CVL

Gendron

, et al. Associations of dual sensory impairment with incident mobility and ADL difficulty. J Am Geriatr Soc 2022; 70: 1997–2007.

Pinto

Wroblewski

Huisingh-Scheetz

, et al. Global sensory impairment predicts morbidity and mortality in older US adults. J Am Geriatr Soc 2017; 65: 2587–2595.

Fischer

Cruickshanks

Klein

, et al. Multiple sensory impairment and quality of life. Ophthalmic Epidemiol 2009; 16: 346–353.

Choi

Adams

Crimmins

, et al. Association between hearing aid use and mortality in adults with hearing loss in the USA: a mortality follow-up study of a cross-sectional cohort. Lancet Healthy Longev 2024; 5: e66–e75.

10.

Loughrey

Kelly

Kelley

, et al. Association of age-related hearing loss with cognitive function, cognitive impairment, and dementia: a systematic review and meta-analysis. JAMA Otolaryngol Head Neck Surg 2018; 144: 115–126.

11.

Varadaraj

Munoz

Deal

, et al. Association of vision impairment with cognitive decline across multiple domains in older adults. JAMA Network Open 2021; 4: e2117416.

12.

Aubin

Phillips

Jaiswal

, et al. Visual and cognitive functioning among older adults with low vision before vision rehabilitation: a pilot study. Front Psychol 2023; 14: 1058951.

13.

Dintica

Marseglia

Rizzuto

, et al. Impaired olfaction is associated with cognitive decline and neurodegeneration in the brain. Neurology 2019; 92: e700–e709.

14.

Löffler

Beier

Bekrater-Bodmann

, et al. Reduced tactile sensitivity is associated with mild cognitive impairment. eBioMedicine 2024; 99: 104896.

15.

Brenowitz

Kaup

Lin

, et al. Multiple sensory impairment is associated with increased risk of dementia among black and white older adults. J Gerontol A Biol Sci Med Sci 2019; 74: 890–896.

16.

Tanner

Yorgasaon

Fankhouser

, et al. Sensory difficulties and social isolation among Mexican, US Hispanic and, US non-Hispanic individuals. Innov Aging 2024; 8(Suppl 1): 720–721.

17.

Cruickshanks

Dhar

Dinces

, et al. Hearing impairment prevalence and associated risk factors in the Hispanic community health study/study of Latinos. JAMA Otolaryngol Head Neck Surg 2015; 141: 641–648.

18.

Lin

Daly

Olchanski

, et al. Dementia diagnosis disparities by race and ethnicity. Med Care 2021; 59: 679–686.

19.

Smith

Jaszczak

Graber

, et al. Instrument development, study design implementation, and survey conduct for the national social life, health, and aging project. J Gerontol B Psychol Sci Soc Sci 2009; 64: i20–i29.

20.

Suzman

. The national social life, health, and aging project: an Introduction. J Gerontol B Psychol Sci Soc Sci 2009; 64: i5–i11.

21.

Waite

Laumann

Levinson

, et al. National Social Life, Health, and Aging Project (NSHAP): Round 1, [United States], 2005-2006. Inter-university Consortium for Political and Social Research [distributor], 2023-01–30. 10.3886/ICPSR20541.v10.

22.

Hayward

Wallace

. Wave 2 of the national social life, health, and aging project: an overview. J Gerontol B Psychol Sci Soc Sci 2014; 69: S1–S3.

23.

Waite

Cagney

Dale

, et al. National Social Life, Health, and Aging Project (NSHAP): Round 3 and COVID-19 Study, [United States], 2015-2016, 2020-2021. Inter-university Consortium for Political and Social Research [distributor], 2024-09-09. 10.3886/ICPSR36873.v9.

24.

Schumm

McClintock

Williams

, et al. Assessment of sensory function in the national social life, health, and aging project. J Gerontol B Psychol Sci Soc Sci 2009; 64B: i76–i85.

25.

Mueller

Renner

. A new procedure for the short screening of olfactory function using five items from the “Sniffin’ Sticks” identification test kit. Am J Rhinol 2006; 20: 113–116.

26.

Pinto

Schumm

Wroblewski

, et al. Racial disparities in olfactory loss among older adults in the United States. J Gerontol A Biol Sci Med Sci 2014; 69: 323–329.

27.

Kotwal

Philip Schumm

Kern

, et al. Evaluation of a brief survey instrument for assessing subtle differences in cognitive function among older adults. Alzheimer Dis Assoc Disord 2015; 29: 317–324.

28.

Nasreddine

Phillips

Bédirian

, et al. The Montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005; 53: 695–699.

29.

Dintica

Marseglia

Rizzuto

, et al. Impaired olfaction is associated with cognitive decline and neurodegeneration in the brain. Neurology 2019; 92: e700–e709.

30.

Lin

Albert

. Hearing loss and dementia - who is listening? Aging Ment Health 2014; 18: 671–673.

31.

Gonzales

Garbarino

Pollet

, et al. Biological aging processes underlying cognitive decline and neurodegenerative disease. J Clin Invest 2022; 132: e158453.

32.

Zimmerman

Rypma

Gratton

, et al. Age-related changes in cerebrovascular health and their effects on neural function and cognition: a comprehensive review. Psychophysiology 2021; 58: e13796.

33.

Sui

Zhou

, et al. Hyposmia as a predictive marker of Parkinson’s disease: a systematic review and meta-analysis. Biomed Res Int 2019; 2019: 3753786.

34.

Dan

Wechter

Gray

, et al. Olfactory dysfunction in aging and neurodegenerative diseases. Ageing Res Rev 2021; 70: 101416.

35.

Zhang

, et al. Sensory processing deficits and related cortical pathological changes in Alzheimer’s disease. Front Aging Neurosci 2023; 15: 1213379.

36.

Lin

Ferrucci

, et al. Association of hearing impairment with brain volume changes in older adults. Neuroimage 2014; 90: 84–92.

37.

Liu

Liang

Hong

, et al. Sensory impairment, loneliness, and the discordance between subjective and objective cognitive function among older adults: a multi-cohort study. Arch Gerontol Geriatr 2024; 124: 105482.

38.

Liu

Zong

Doty

, et al. Prevalence and risk factors of taste and smell impairment in a nationwide representative sample of the US population: a cross-sectional study. BMJ Open 2016; 6: e013246.

39.

Rong

Lai

Jing

, et al. Association of sensory impairments with cognitive decline and depression among older adults in China. JAMA Netw Open 2020; 3: e2014186.

40.

Fischer

Cruickshanks

Schubert

, et al. Age-related sensory impairments and risk of cognitive impairment. J Am Geriatr Soc 2016; 64: 1981–1987.

41.

Yamada

Denkinger

Onder

, et al. Dual sensory impairment and cognitive decline: the results from the shelter study. J Gerontol A Biol Sci Med Sci 2015; 71: 117–123.

42.

Pan

, et al. Sensory impairment and cognitive decline among older adults: an analysis of mediation and moderation effects of loneliness. Front Neurosci 2022; 16: 1092297.

43.

Xiao

Zhao

, et al. Sensory impairments and cognitive decline in older adults: a review from a population-based perspective. Aging Health Res 2021; 1: 100002.

44.

Cai

Schrack

Gross

, et al. Sensory impairment and algorithmic classification of early cognitive impairment. Alzheimers Dement (Amst) 2023; 15: e12400.

45.

Maharani

Dawes

Nazroo

, et al. Associations between self-reported sensory impairment and risk of cognitive decline and impairment in the health and retirement study cohort. J Gerontol B Psychol Sci Soc Sci 2019; 75: 1230–1242.

46.

Alattar

Bergstrom

Laughlin

, et al. Hearing impairment and cognitive decline in older, community-dwelling adults. J Gerontol A Biol Sci Med Sci 2019; 75: 567–573.

47.

Zhao

Zhou

Wei

, et al. Associations of sensory impairment and cognitive function in middle-aged and older Chinese population: the China health and retirement longitudinal study. J Glob Health 2021; 11: 08008.

48.

Zhou

Wei

Sun

, et al. Do sensory impairments portend cognitive decline in older Chinese adults? Longitudinal evidence from a nationally representative survey, 2011–2018. J Clin Med Res 2023; 12: 430.

49.

Lim

Chee

Da Soh

, et al. Association between visual impairment and decline in cognitive function in a multiethnic Asian population. JAMA Netw Open 2020; 3: e203560.

50.

Mitoku

Masaki

Ogata

, et al. Vision and hearing impairments, cognitive impairment and mortality among long-term care recipients: a population-based cohort study. BMC Geriatr 2016; 16: 112.

51.

Mattos

Hasan

Schlosser

, et al. The association of gustatory dysfunction, olfactory dysfunction, and cognition in older adults. Int Forum Allergy Rhinol 2023; 13: 1577–1583.

52.

Steinbach

Hundt

Vaitl

, et al. Taste in mild cognitive impairment and Alzheimer’s disease. J Neurol 2010; 257: 238–246.

53.

Besagar

Yonekawa

Sridhar

, et al. Association of socioeconomic, demographic, and health care access disparities with severe visual impairment in the US. JAMA Ophthalmol 2022; 140: 1219–1226.

54.

Vohra

Assi

Kamath

, et al. Potential role for diet in mediating the association of olfactory dysfunction and cognitive decline: a nationally representative study. Nutrients 2023; 15: 3890.

55.

Shakarchi

Assi

Ehrlich

, et al. Dual sensory impairment and perceived everyday discrimination in the United States. JAMA Ophthalmol 2020; 138: 1227–1233.

56.

Williams

Lawrence

Davis

. Racism and health: evidence and needed research. Annu Rev Public Health 2019; 40: 105–125.

57.

Huang

Wang

, et al. Association of occupational factors and dementia or cognitive impairment: a systematic review and meta-analysis. J Alzheimers Dis 2020; 78: 217–227.

58.

Davila

Caban-Martinez

Muennig

, et al. Sensory impairment among older US workers. Am J Public Health 2009; 99: 1378–1385.

59.

Ajmani

Suh

Pinto

. Effects of ambient air pollution exposure on olfaction: a review. Environ Health Perspect 2016; 124: 1683–1693.

60.

Yuan

Halabicky

Rao

, et al. Lifetime air pollution exposure, cognitive deficits, and brain imaging outcomes: a systematic review. Neurotoxicology 2023; 96: 69–80.

61.

Zhang

Zhou

Liu

, et al. Association between residential environment quality with mild cognitive impairment among middle and elderly adults in China. J Neurol Sci 2024; 467: 123318.

62.

Goldman

Pinto

. Sensory health among older adults in the United States: a neighborhood context approach. J Gerontol B Psychol Sci Soc Sci 2024; 79: gbae003.

63.

Steinberg

Parisi

Feger

, et al. Rural-urban differences in cognition: findings from the advanced cognitive training for independent and vital elderly trial. J Aging Health 2023; 35: 107S–118S.

64.

Hirst

Cassarino

Kenny

, et al. Urban and rural environments differentially shape multisensory perception in ageing. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn 2022; 29: 197–212.

65.

Majoka

Schimming

. Effect of social determinants of health on cognition and risk of Alzheimer disease and related dementias. Clin Ther 2021; 43: 922–929.

66.

Liu

Murchland

VanderWeele

, et al. Eliminating racial disparities in dementia risk by equalizing education quality: a sensitivity analysis. Soc Sci Med 2022; 312: 115347.

67.

Assi

Shakarchi

Sheehan

, et al. Assessment of sensory impairment and health care satisfaction among medicare beneficiaries. JAMA Netw Open 2020; 3: e2025522.

68.

McPherson

Resnick

Boltz

, et al. The association between patient engagement and quality of care interactions among acute care patients with dementia. Geriatr Nurs 2024; 57: 117–122.

69.

Sayegh

Knight

. Cross-cultural differences in dementia: the sociocultural health belief model. Int Psychogeriatr 2013; 25: 517–530.

70.

Barnett

Hixon

Okwiri

, et al. Factors involved in access and utilization of adult hearing healthcare: a systematic review. Laryngoscope 2017; 127: 1187–1194.

71.

Williams

Sahel

. Addressing social determinants of vision health. Ophthalmol Ther 2022; 11: 1371–1382.

72.

World Health Organization . Closing the gap in a generation: health equity through action on the social determinants of health - Final report of the commission on social determinants of health. Geneva, Switzerland: World Health Organization, 2008. https://www.who.int/publications/i/item/WHO-IER-CSDH-08.1 (accessed 17 July 2025).

73.

Sorjonen

Falkstedt

Wallin

, et al. Dangers of residual confounding: a cautionary tale featuring cognitive ability, socioeconomic background, and education. BMC Psychol 2021; 9: 145.

74.

Cheng

Okoro

Mendez

, et al. Health care access and use among adults with and without vision impairment: behavioral risk factor surveillance system, 2018. Prev Chronic Dis 2022; 19: E70.

75.

Wang

Wroblewski

Hawkley

, et al. Types of sensory disability are differentially associated with mental health in older US adults over time. J Am Geriatr Soc 2024; 72: 2770–2781.

76.

Jimenez

Gause

Sims

, et al. Racial and ethnic differences in the association between depressive symptoms and cognitive outcomes in older adults: findings from KHANDLE and STAR. Alzheimers Dement 2024; 20: 3147–3156.

77.

Zhang

Cotch

Ryskulova

, et al. Vision health disparities in the United States by race/ethnicity, education, and economic status: findings from two nationally representative surveys. Am J Ophthalmol 2012; 154: S53–62.e1.

78.

Malhotra

Chan

Matchar

, et al. Diagnostic performance of short portable mental status questionnaire for screening dementia among patients attending cognitive assessment clinics in Singapore. Ann Acad Med Singap 2013; 42: 315–319.

79.

Schönstein

Wahl

Katus

, et al. SPMSQ For risk stratification of older patients in the emergency department: an exploratory prospective cohort study. Z Gerontol Geriatr 2019; 52: 222–228.

80.

Shega

Sunkara

Kotwal

, et al. Measuring cognition: the Chicago cognitive function measure in the national social life, health and aging project, wave 2. J Gerontol B Psychol Sci Soc Sci 2014; 69: S166–S176.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB