Association of sensory and motor impairments with dementia: a cross-sectional analysis in US older adults

Abstract

Background

Impairments in sensory and motor function are common in older adults and have been linked with higher prevalence of dementia. However, their independent associations with dementia have not been examined in a nationally representative population.

Methods

Cross-sectional analysis of the National Health and Aging Trends Study (NHATS; Round 11/2021) of U.S. Medicare beneficiaries. Dementia was classified per NHATS protocol as none, possible, or probable dementia. Vision measures included presenting binocular distance visual acuity, near visual acuity, and contrast sensitivity. Hearing measures included the pure tone average of the better ear. Motor function measures included gait speed, chair stands time, balance score, and grip strength. Dichotomized or categorized presence of impairment in each sensory and motor function was defined and used as the main exposures. Multivariable survey-weighted logistic regression was used to assess the independent associations of sensory and motor function impairments with dementia status.

Results

Our final analytic sample included 2,379 participants with 36.6% aged $\ge$ 80 years, 56.2% female, 82.9% non-Hispanic White, and 9.8% having possible or probable dementia. In adjusted models including visual, hearing, and motor function variables, participants with impairment in gait (odds ratio (OR), 2.18; 95% CI, 1.32–3.6; p = 0.004), chair stand performance (OR, 1.69; 95% CI, 1.17–2.44; p < 0.01), or visual contrast sensitivity (OR, 1.52; 95% CI, 1.04–2.22; p = 0.03) had higher odds of dementia. Impaired standing balance, grip strength, hearing, and near visual acuity were not associated with higher odds of dementia (p > 0.05).

Conclusions

Findings suggest current impairment in vision or motor function may be strong indicators of prevalent dementia in older adults.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06082-6.

Introduction

Sensory and motor impairments have been extensively linked with cognitive decline and dementia in older adults. To date, there is a growing body of compelling evidence that age-related impairments in sensory function, such as hearing and vision, are associated with higher prevalence of mild cognitive impairment (MCI) and dementia [1–8]. Likewise, motor deficits, such as slow or slowing gait speed, impaired standing balance, and low grip strength, have also been linked with higher prevalence of cognitive impairment and dementia [9–15]. These sensory and motor impairments are highly prevalent in older adults [1, 16–18], and findings from prospective studies suggest declines in sensory and motor function may occur several years before the onset of a clinical diagnosis of MCI or dementia [19–21] making them noteworthy indicators of future dementia risk. Although several studies have examined the associations of motor function with prevalent dementia [9–11, 14], hearing with prevalent dementia [7, 8], vision with prevalent dementia [1, 4], or dual-sensory impairment with prevalent dementia [1, 2], the strength of the independent associations between each individual sensory or motor impairment and prevalent dementia is not well characterized.

Associations of changes in sensory and motor function with changes in cognitive health may result from neurodegeneration and/or accumulation of pathology in shared regions of the brain, or may represent the progression of independent risk factors, such as increased cognitive load or reduced social engagement, that contribute to adverse cognitive outcomes [19]. Regardless of the cause, this collective body of evidence suggests that the presence of sensory and/or motor deficits may act as important indicators of dementia that warrants a deeper understanding for the development of potential therapeutic or intervention efforts designed to attenuate cognitive decline and prevent dementia onset earlier in the pathological course when interventions may be more effective. Although previous research has investigated the associations of individual or dual sensory or motor impairments with cognitive impairment and/or dementia in separate analyses, the independent association between each sensory or motor impairment and dementia has not been assessed in a single study. Given the high prevalence of sensory and motor impairments with aging, a better understanding of the strength of the independent associations between each type of impairment and dementia status is warranted.

To address this gap, we used data from the National Health and Aging Trends Study (NHATS) [22], a nationally representative study of U.S. Medicare beneficiaries, to evaluate the cross-sectional independent associations of sensory and motor impairments with prevalent dementia. We hypothesized that impairments in hearing, vision, and motor function would be independently associated with dementia prevalence, even after adjusting for important risk factors such as age and education [23–25].

Methods

Study Participants

Data for this analysis are from NHATS, an ongoing national panel study of Medicare beneficiaries aged 65 years and older living in the United States. Participants were recruited into the NHATS sample in either 2011 (study initiation) or 2015 (replenishment) and annual in-home interviews are conducted. For participants unable to answer for themselves (typically 7%−8%), a proxy interview is conducted. We used survey data from Round 11 (2021) of NHATS (N = 3,817) when objective hearing and vision testing were initiated. We included community-dwelling participants and those residing in assisted living (non-nursing home) facilities who had available data on dementia status, sensory function, and motor function, and relevant covariates [26]. The study was approved by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board and all participants provided written informed consent.

Measures

Main outcome: Dementia

Individuals were classified as having (possible/probable) dementia if they met any of the following criteria: (1) a doctor had ever reportedly told the participant he/she had dementia or Alzheimer’s disease (via participant or proxy report); (2) a proxy rating of ≥ 2 on the eight-item AD8 screener [27] or (3) had a score ≤ 1.5 SDs below the mean in at least two cognitive domains based on test items that evaluate memory, orientation, and executive function. Dementia outcome determination was performed using the SAS program provided by NHATS [28] (SAS Version 9.4). In the current study, dementia was defined as either possible or probable dementia, which has an 71.8% sensitivity and 83.7.% specificity relative to the research diagnosis criteria for dementia and cognitive impairment, no dementia in the Aging, Demographics, and Memory Study [29].

Sensory performance

In round 11 of NHATS (2021) interviewers collected objective measures of visual and auditory function using iPad-based software developed by Ridgevue Vision and ShoeBOX, [30] respectively.

Vision

Vision tests were administered with the participants using routinely worn corrective lenses so best corrected vision could be assessed.

Distance acuity vision test score: Participants were asked to read five letters per screen, with the size becoming smaller on each successive screen. Distance acuity score was calculated as 0.02*(55-S_D) where S_D is the sum of correct letters for distance acuity test (0 corresponds to 20/20 “normal” vision and higher values indicate worse visual functioning). Mild distance vision impairment was considered a value ≥ 0.30 logarithm of the minimum angle of resolution (logMAR) (20/60), and moderate/severe distance vision impairment was considered a value ≥ 0.48 logMAR [31].

Near acuity vision test score: Participants were asked to hold the tablet at their usual reading distance and to read five lowercase letters per screen on the tablet, with the size becoming smaller on each successive screen. Near acuity vision test scores were calculated as 0.02*(55-S_N) + log₁₀(40/x), where S_N is the sum of correct letters for near acuity test, and X is the reading distance in centimeters (higher values indicate worse visual functioning). Similar to distance vision impairment, near visual acuity impairment was considered a value ≥ 0.30 logMAR [31].

Contrast sensitivity vision test score: Participants were asked to read two letters per screen on the tablet, with the tone becoming lighter on each successive screen. Contrast sensitivity was computed as 0.40 + (0.05*S_C) where S_C is the sum of correct letters for contrast sensitivity test (higher values indicate better visual functioning). In the absence of standardized definition of contrast sensitivity impairment, a cutoff of < 1.55 logarithm of contrast sensitivity (logCS) was used [31–34].

Hearing

Pure-tone audiometry was performed by technicians in participants’ homes using an iPad (Apple)–based portable audiometer (SHOEBOX Ltd) with sound-attenuating Radio Ear DD450 headphones [17, 35], ambient noise monitoring, and automated software consistent with best practice hearing assessment methodologies. Consistent with previous epidemiologic research and World Health Organization hearing classification, a pure tone average of the frequencies most important for speech understanding (500 Hz, 1000 Hz, 2000 Hz, and 4000 Hz) was calculated for each ear. We used the better ear pure tone average (BPTA) (e.g., lower pure-tone average indicating better peripheral hearing acuity [31]) in the current study. Mild hearing loss was categorized as hearing score ≥ 26 dB BPTA, and the moderate or greater hearing loss was defined as hearing score ≥ 41 dB BPTA [36].

Motor performance

NHATS measures four motor function activities, including gait speed, standing balance, chair stands, and grip strength.

Standing balance

Participants were asked to attempt 5 standing balance activities —side-by-side, semi-tandem, full tandem, one leg eyes open, and one leg eyes closed. Participants were asked to hold each of the two-leg stands for 10 s and the one-leg stand for 30 s. The stands progressed from easy (side-by-side) to hard (one leg, eyes closed). Participants progressed to the next balance activity if they completed the prior (easier) activity. The standing balance components were coded according to the Short Physical Performance Battery (SPPB) ranging from 0 to 4, with higher scores indicating better standing balance. The NHATS balance test differs from the balance test in the original SPPB version as a more difficult balance test was added in NHATS–standing on one leg with eyes open, followed by eyes closed [37]. We used the composite NHATS SPPB score for standing balance independently in our analyses, and standing balance impairment was considered a score less than 3 based on the data distribution. Participants who did not complete the standing balance tests because they were not eligible or did not attempt for safety reasons, or attempted but could not complete any of the tests were also classified as impaired.

Grip strength

Participants were given a handheld dynamometer in a seated position with the dynamometer placed on a table horizontal surface and asked to squeeze as hard as possible. The test was performed twice using the dominant hand unless an exclusion criterion applied (e.g., injury, arthritis, etc.) We used the average of the two grip strength tests in our analyses, with higher values indicating better grip strength performance. Impairment in grip strength was defined as less than 32 kg for men and less than 20 kg for women [18]. In addition, participants who were not eligible for the grip strength test or did not attempt for safety reasons, or attempted but could not complete the test were also classified as impaired.

Gait speed

Gait speed was assessed over a 3-m walking course. Timing was initiated as participants walked from the start to finish and then timed again on the return. Both times were recorded, along with whether a mobility device was used [38]. Our analyses used the average speed for the two tests, which were standardized per meter walked to derive gait speed in meters per second, with higher values indicating faster gait speed. Impairment in gait speed was defined as gait speed slower than 0.8 m/second [39]. In addition, participants who were not eligible or did not attempt because of safety reasons, or attempted but did not complete the walking test were also classified as impaired.

Chair stands

Participants were asked to complete 5 repeated chair stands as quickly as possible. Suitable chairs with standard height were used for the test and excluded furniture such as bar stools, hassocks, ottomans, or footrests [38]. Time to complete the activity was recorded. In the analyses, we used the speed to complete the chair stand test for five repeated chair stands (chair stands per second), with higher values indicating longer time to complete, or worse performance. Impairment in chair stand speed test was defined as chair stand pace < 0.42 stands/s [18]. In addition, participants who were not eligible or did not attempt because of safety reasons, or attempted but did not complete the chair stand test were also classified as impaired.

Table 1 summarizes the sensory and motor definitions, directionality, and impairment thresholds used for analysis.

Table 1.

Sensory and motor definitions, directionality, and impairment thresholds used for analysis

Variable		Description	Direction	Impairment cutoff
Sensory variables
Vision contrast		Contrast sensitivity vision test score	Higher values are better functioning	Vision contrast < 1.55 logCS: impaired
Vision distance		Distance acuity vision test score	Higher values are worse functioning	Vision distance score ≥ 0.48 logMAR: Mild loss Vision distance score ≥ 0.48 logMAR: Moderate/Severe impairment
Vision near		Near acuity vision test score	Higher values are worse functioning	Vision near score ≥ 0.30 logMAR: impaired
Hearing		BPTA	Higher values are worse functioning	Hearing score ≥ 26 dB BPTA: Mild hearing loss Hearing score ≥ 41 dB BPTA: Moderate/Severe hearing loss
Motor variables
Chair stand speed		5chairstand/Chair stand test time	Higher values are better functioning	Chair stand NHATS SPPB = 0: impaired Chair stand speed < 0.42 chairstand/sec: impaired
Gait speed		3-m/(average time of two walking tests)	Higher values are better functioning	Walking test NHATS SPPB = 0: impaired Gait speed < 0.8 m/sec: impaired
Balance score		0–4	Higher values are better functioning	Balance NHATS SPPB < 3: impaired
Grip strength		Average of two grip strength test scores	Higher values are better functioning	Grip strength NHATS SPPB = 0: impaired Grip strength < 32 kg for men & < 20 kg for women: impaired

BPTA Best pure tone average, or the pure tone average calculated for the better hearing ear

dB decibels

Log MAR Logarithm of the minimum angle of resolution

Log CS Logarithm of contrast sensitivity

SPPB Short Physical Performance Battery

kg kilogram

Covariates

Sociodemographic characteristics included age (categorized in 5-year intervals [70–74, 75–79, 80–84, 85–89, and 90 years and older]), sex, race (non-Hispanic White, non-Hispanic Black, Hispanic, and other), and education level (up to high school, some college, bachelor’s degree, and master’s degree and above). Height and weight were self-reported and body mass index (BMI) was calculated as kilograms per meter squared. Health conditions included self-report history of diagnosis of heart attack, heart disease, lung disease, stroke, cancer, diabetes, depression, hypertension, and hospitalization during the previous year. The sum of the number of health conditions was calculated (0, 1, 2, 3, 4, ≥ 5). History of smoking was also included (ever, never).

Statistical analysis

To compare the participant characteristics between those with dementia and those without dementia, the weighted distribution of covariates (age, sex, race, education, BMI, number of health conditions and history of smoking) was calculated by dementia status and their statistical differences examined using t-tests (continuous variables) and chi-squared tests (categorical variables). Survey-weighted cross-sectional logistic regression models were used to assess the odds of dementia by sensory and motor-function impairment status. We fit five models (or sets of models) with sensory impairment or motor impairment as the independent variable(s) and dementia status as the dependent variable. All models were adjusted for demographics and clinical history. Model set 1 included a separate model for each sensory and motor independent variable. Model set 2 included each vision variable with the hearing variable. A separate model was used for each vision variable to avoid potential collinearity. Model 3 included all motor variables together in one model. Model set 4 included each a separate model for each motor function variable and each vision variable, as well as hearing. Model set 5 included all motor variables together, plus each vision variable (separately) and hearing. In separate analyses, we also explored potential interactions between the significant sensory variables and the motor variables. Additionally, a sensitivity analysis was performed to repeat the analyses after removing participants living in residential care. Odds ratios (OR) with 95% confidence intervals (95% CI) are reported. In NHATS participants are chosen based on a stratified three-stage sample-design. The oldest age groups and racial minorities are intentionally oversampled to ensure sufficient representation. Survey weights were used to account for the complex sampling survey design and enhance generalizability. The significance level (α) was set as 0.05. All analyses were conducted using R version 4.2 (R Project for Statistical Computing).

Results

Cohort description

Of the N = 3466 participants alive and eligible to participate in NHATS round 11, N = 2379 had data on dementia, sensory function, motor function, and covariates, of whom 317 had possible or probable dementia (13.3%), and 2062 did not have dementia (86.7%), with a weighted prevalence of 9.8% (Fig. 1). Demographic characteristics are summarized by dementia status in Table 2. Compared to participants without dementia, those living with possible or probable dementia were more likely to be aged 80 years or older (61.7% vs 33.9%, p < 0.0001), non-White (30.6% vs 15.7%, p < 0.0001), less educated (54.6% vs 32.9%, p < 0.0001), and have a lower BMI and a higher number of comorbidities (with 26.4% vs 13.7% having at least four comorbidities, p < 0.0001).

Fig. 1.

Analytic cohort selection diagram

Table 2.

Weighted characteristics of participants by dementia status, National Health and Aging Trends Study, Round 11, 2021

Variable	All (N = 2379)	Dementia (N = 317)	No Dementia (N = 2062)	Pl
Age range, % (95% CI)				< 0.0001
70–74	29.9 (27.4, 32.4)	14.3 (7.9, 20.6)	31.6 (28.7, 34.4)
75–79	33.5 (31, 36)	24 (17.3, 30.8)	34.6 (31.9, 37.2)
80–84	20.3 (18.5, 22.1)	25.1 (19.8, 30.4)	19.8 (17.9, 21.7)
85–89	11.4 (10.4, 12.4)	24.6 (19.8, 29.5)	9.9 (8.8, 11)
90 and above	5 (4.3, 5.6)	12 (8.2, 15.8)	4.2 (3.6, 4.8)
Sex, % (95% CI)				0.10
Male	43.8 (41.2, 46.3)	37.7 (30.6, 44.7)	44.4 (41.7, 47.2)
Female	56.2 (53.7, 58.8)	62.3 (55.3, 69.4)	55.6 (52.8, 58.3)
Race/Ethnicity, % (95% CI)				< 0.0001
Non-Hispanic White	82.8 (80, 85.7)	69.4 (62.2, 76.6)	84.3 (81.6, 87)
Non-Hispanic Black	7.6 (6.4, 8.8)	13.3 (9.9, 16.6)	7 (5.8, 8.1)
Hispanic	6.1 (4, 8.3)	14.9 (8.5, 21.4)	5.2 (3.2, 7.2)
Other	3.4 (2.2, 4.6)	2.4 (0.5, 4.3)	3.5 (2.2, 4.8)
Education, % (95% CI)				< 0.0001
Up to high school	35 (32, 38)	54.6 (48.2, 61)	32.9 (29.8, 35.9)
Some college	29 (26.4, 31.6)	21.8 (15.7, 27.8)	29.8 (26.9, 32.6)
Bachelor’s and above	36 (32.2, 39.8)	23.6 (17.7, 29.5)	37.4 (33.3, 41.4)
BMI categorya, % (95% CI)				< 0.001
Underweight	1.4 (0.8, 2)	4 (1.1, 7)	1.1 (0.5, 1.7)
Healthy weight	32.2 (29.8, 34.6)	41.9 (34.2, 49.7)	31.1 (28.6, 33.6)
Overweight	37.9 (35.8, 40)	30.2 (24.8, 35.6)	38.7 (36.5, 41)
Obese	28.5 (26.3, 30.7)	23.8 (16.9, 30.8)	29 (26.7, 31.4)
Ever smokedb, % (95% CI)				0.34
No	50 (47.6, 52.5)	53.3 (46.1, 60.5)	49.7 (47.2, 52.2)
Yes	50 (47.5, 52.4)	46.7 (39.5, 53.9)	50.3 (47.8, 52.8)
Number of comorbidities, % (95% CI)				< 0.0001
0	11.4 (9.6, 13.1)	6.9 (3.5, 10.3)	11.8 (10, 13.7)
1	26 (23.3, 28.6)	17.8 (12.6, 23)	26.8 (23.9, 29.8)
2	28.3 (25.8, 30.7)	22 (16.6, 27.4)	28.9 (26.2, 31.6)
3	19.5 (17.9, 21.1)	26.8 (20.4, 33.3)	18.7 (17, 20.4)
≥ 4	14.9 (13.1, 16.7)	26.4 (20.7, 32.2)	13.7 (11.9, 15.5)
Distance vision, mean (95% CI)	0.1 (0.1, 0.1)	0.2 (0.2, 0.2)	0.1 (0.1, 0.1)	< 0.0001
Contrast sensitivity, mean (95% CI)	1.7 (1.7, 1.7)	1.6 (1.5, 1.6)	1.7 (1.7, 1.7)	< 0.0001
Near vision, mean (95% CI)	0.2 (0.2, 0.2)	0.3 (0.2, 0.3)	0.2 (0.2, 0.2)	< 0.001
Hearing (dB), mean (95% CI)	33.2 (32.5, 33.9)	38.4 (36.4, 40.4)	32.6 (31.8, 33.4)	< 0.0001
Hearing (10 dB steps), mean (95% CI)	3.8 (3.7, 3.8)	4.3 (4.1, 4.5)	3.7 (3.6, 3.8)	< 0.0001
Distance visionh, % (95% CI)				< 0.0001
Not impaired	91 (89.6, 92.5)	81.9 (76.4, 87.5)	92 (90.7, 93.4)
Mild impairment	4.8 (3.8, 5.8)	7.6 (4.1, 11.1)	4.5 (3.5, 5.4)
Moderate/Severe impairment	4.2 (3.2, 5.2)	10.5 (5.8, 15.2)	3.5 (2.6, 4.4)
Contrast sensitivityi, % (95% CI)				< 0.0001
Not impaired	79.8 (77.6, 82)	62.3 (55.5, 69.1)	81.7 (79.4, 84)
Impaired	20.2 (18, 22.4)	37.7 (30.9, 44.5)	18.3 (16, 20.6)
Near visionj, % (95% CI)				0.0001
Not impaired	79.1 (77.1, 81.1)	67.4 (60.8, 74)	80.4 (78.3, 82.5)
Impaired	20.9 (18.9, 22.9)	32.6 (26, 39.2)	19.6 (17.5, 21.7)
Hearing lossk, % (95% CI)				< 0.0001
No loss	33.2 (30.4, 36)	20.2 (13.7, 26.6)	34.6 (31.6, 37.7)
Mild loss	38 (35.9, 40.2)	36.9 (30.1, 43.6)	38.1 (36, 40.3)
Moderate/Severe loss	28.8 (26.6, 30.9)	43 (36.6, 49.4)	27.2 (24.8, 29.6)
Grip Strength (kg), mean (95% CI)				< 0.0001
All	26.9 (26.3, 27.5)	22.5 (21.3, 23.8)	27.3 (26.7, 28)
Male	34.96 (33.94, 35.98)	29.15 (27.07, 31.22)	35.48 (34.40, 36.56)
Female	20.43 (19.95, 20.91)	18.37 (16.95, 19.79)	20.67 (20.18, 21.16)
NHATS SPPB score, mean (95% CI)c	7.2 (7, 7.4)	4.8 (4.3, 5.3)	7.4 (7.3, 7.6)	< 0.0001
Balance score, mean (95% CI)c	2.5 (2.4, 2.5)	1.8 (1.6, 2)	2.5 (2.5, 2.6)	< 0.0001
Gait speed (m/s), mean (95% CI)c	0.8 (0.8, 0.8)	0.6 (0.5, 0.6)	0.8 (0.8, 0.8)	< 0.0001
Chair stand speed (stands/s), mean (95% CI)c	0.5 (0.5, 0.5)	0.4 (0.4, 0.4)	0.5 (0.5, 0.5)	< 0.0001
Grip Strength impairmentd, % (95% CI)				< 0.0001
Not impaired	47.9 (45.4, 50.3)	32.6 (26.4, 38.8)	49.5 (46.9, 52.1)
Impaired	52.1 (49.7, 54.6)	67.4 (61.2, 73.6)	50.5 (47.9, 53.1)
Balance impairmente, % (95% CI)				< 0.0001
Not impaired	49 (46.1, 52)	25.3 (17.2, 33.4)	51.6 (48.8, 54.4)
Impaired	51 (48, 53.9)	74.7 (66.6, 82.8)	48.4 (45.6, 51.2)
Chair stand speed impairmentf, % (95% CI)				< 0.0001
Not impaired	53.1 (50.6, 55.7)	25.5 (19.6, 31.4)	56.1 (53.6, 58.6)
Impaired	46.9 (44.3, 49.4)	74.5 (68.6, 80.4)	43.9 (41.4, 46.4)
Gait speed impairmentg, % (95% CI)				< 0.0001
Not impaired	47.4 (44.4, 50.4)	16.6 (10.5, 22.7)	50.7 (47.8, 53.6)
Impaired	52.6 (49.6, 55.6)	83.4 (77.3, 89.5)	49.3 (46.4, 52.2)

^aBMI (body mass index) categories are defined as underweight: BMI < 18.5, healthy weight: 18.5 ≤ BMI < 25, overweight: 25 ≤ BMI < 30, obese: BMI ≥ 30

^bEver smoked is defined as self-reported history of smoking

^cThe continuous motor function variables are reported for those participants that had completed the test

^dGrip strength impairment is defined as grip strength < 32 for men and grip strength < 20 for women, or NHATS SPPB grip component = 0

^eStanding balance impairment is defined as NHATS SPPB balance component < 3

^fChair speed impairment is defined as chair stand speed < 0.42 chair stand/sec, or NHATS SPPB chair stand component = 0

^gGait speed impairment is defined as gait speed < 0.8 m/sec, or NHATS SPPB walking component = 0

^hDistance vision impairment is defined as mild impairment with distance acuity vision test score > 0.3, and moderate/severe impairment with distance acuity vision test score ≥ 0.48

ⁱContrast sensitivity impairment is defined as contrast sensitivity test score < 1.55

^jNear vision impairment is defined as near acuity vision test score ≥ 0.30

^kHearing loss or hearing impairment is defined as mild loss: hearing score ≥ 26 dB; moderate/severe loss: hearing score ≥ 41 dB [BPTA (Best pure tone average: the pure tone average calculated for the better hearing ear).]

^l Bold values denote statistical significance at the p < 0.05 level

Those living with dementia also had a higher prevalence of impaired distance vision (10.5% vs 3.5% moderate or severe distance vision impairment, p < 0.0001), contrast sensitivity (37.7% vs 18.3%, p < 0.0001), and near vision (32.6% vs 19.6%, p < 0.001) as well as hearing loss (43.0% vs. 27.2% moderate or severe hearing loss, p < 0.0001) compared to those without dementia. Those living with dementia were also more likely to have impaired grip strength (67.4% vs 50.5%, p < 0.0001), standing balance (74.7% vs 48.4%, p < 0.0001), chair stand time (74.5% vs 43.9%, p < 0.0001), and gait speed (83.4% vs 49.3%, p < 0.0001) than those without dementia (Table 2).

Participants missing sensory function tests had a higher prevalence of dementia and were more likely to have impaired grip strength, standing balance, slow chair stand speed, or slow gait speed. They also tended to be older, non-White, and less educated (Supplement Table 1). Participants missing motor function tests had a higher prevalence of dementia and were more likely to have impaired hearing (Supplement Table 2).

For sensory impairment, in models considering each sensory measure separately (Table 3, Model set 1), moderate or severe distance vision impairment and contrast sensitivity impairment were associated with higher odds of dementia (OR = 2.25 (95% CI: 1.09–4.64), OR = 1.87 (95% CI: 1.32–2.66), respectively). These results remained consistent when independent variables for hearing or vision impairment were placed in the same model (Table 3, Model set 2).

Table 3.

Cross-sectional multivariable-adjusted association between sensory and motor impairment and dementia, National Health and Aging Trends Study, Round 11, 2021

Model set	Model	Variable	Categorical Analyses
			OR (CI)	Pa
Model set 1 (each impairment individually)	Distance vision Mild imp Moderate/Severe imp		1.1 (0.64–1.88) 2.25 (1.09–4.64)	0.73 0.03
	Contrast sensitivity		1.87 (1.32–2.66)	< 0.001
	Near vision		1.25 (0.9–1.73)	0.18
	Hearing Mild loss Moderate/Severe loss		1.22 (0.76–1.96) 1.61 (0.92–2.8)	0.40 0.09
	Grip strength		1.21 (0.87–1.69)	0.26
	Standing balance		2.1 (1.38–3.18)	< 0.01
	Gait speed		3.11 (1.92–5.06)	< 0.0001
	Chair stand speed		2.46 (1.69–3.57)	< 0.0001
Model set 2	Distance vision + hearing	Distance vision Mild imp Moderate/Severe imp	1.06 (0.63–1.79) 2.17 (1.06–4.45)	0.81 0.03
		Hearing Mild loss Moderate/Severe loss	1.19 (0.74–1.9) 1.55 (0.9–2.69)	0.46 0.11
	Contrast sensitivity + hearing	Contrast sensitivity	1.83 (1.28–2.63)	< 0.01
		Hearing Mild loss Moderate/Severe loss	1.18 (0.74–1.89) 1.53 (0.87–2.69)	0.48 0.13
	Near vision + hearing	Near vision	1.23 (0.88–1.72)	0.21
		Hearing Mild loss Moderate/Severe loss	1.22 (0.76–1.96) 1.6 (0.92–2.78)	0.41 0.10
Model 3	Motor	Grip Strength	0.94 (0.67–1.32)	0.71
		Balance	1.49 (0.97–2.3)	0.07
		Gait speed	2.29 (1.4–3.74)	< 0.01
		Chair stand speed	1.77 (1.22–2.56)	< 0.01
Model set 4	Grip + contrast sensitivity + hearing	Grip Strength	1.15 (0.83–1.6)	0.38
		Contrast sensitivity	1.81 (1.26–2.59)	< 0.01
		Hearing Mild loss Moderate/Severe loss	1.18 (0.73–1.89) 1.54 (0.87–2.72)	0.48 0.13
	Grip + near vision + hearing	Grip Strength	1.2 (0.86–1.68)	0.26
		Near vision	1.21 (0.87–1.69)	0.24
		Hearing Mild loss Moderate/Severe loss	1.22 (0.75–1.96) 1.61 (0.92–2.82)	0.41 0.09
	Grip + distance vision + hearing	Grip Strength	1.19 (0.85–1.66)	0.31
		Distance vision Mild imp Moderate/Severe imp	1.06 (0.63–1.77) 2.12 (1.03–4.37)	0.82 0.04
		Hearing Mild loss Moderate/Severe loss	1.19 (0.74–1.91) 1.56 (0.9–2.72)	0.46 0.11
	Balance + contrast sensitivity + hearing	Balance	2 (1.3–3.06)	< 0.01
		Contrast sensitivity	1.77 (1.22–2.55)	< 0.01
		Hearing Mild loss Moderate/Severe loss	1.16 (0.73–1.85) 1.48 (0.86–2.56)	0.52 0.15
	Balance + near vision + hearing	Balance	2.03 (1.35–3.06)	< 0.01
		Near vision	1.17 (0.85–1.61)	0.33
		Hearing Mild loss Moderate/Severe loss	1.2 (0.75–1.92) 1.54 (0.9–2.63)	0.44 0.11
	Balance + distance vision + hearing	Balance	2.04 (1.34–3.12)	< 0.01
		Distance vision Mild imp Moderate/Severe imp	1.03 (0.62–1.72) 2.12 (1.03–4.36)	0.91 0.04
		Hearing Mild loss Moderate/Severe loss	1.17 (0.74–1.87) 1.5 (0.88–2.54)	0.49 0.13
	Gait speed + contrast sensitivity + hearing	Gait speed	2.86 (1.75–4.67)	< 0.001
		Contrast sensitivity	1.6 (1.1–2.34)	0.02
		Hearing Mild loss Moderate/Severe loss	1.11 (0.69–1.79) 1.44 (0.83–2.51)	0.66 0.19
	Gait speed + near vision + hearing	Gait speed	3.04 (1.89–4.9)	< 0.0001
		Near vision	0.402	0.40
		Hearing Mild loss Moderate/Severe loss	1.14 (0.71–1.84) 1.5 (0.87–2.57)	0.59 0.14
	Gait speed + distance vision + hearing	Gait speed	2.98 (1.86–4.79)	< 0.001
		Distance vision Mild imp Moderate/Severe imp	0.97 (0.58–1.62) 1.87 (0.93–3.75)	0.90 0.08
		Hearing Mild loss Moderate/Severe loss	1.11 (0.69–1.79) 1.46 (0.86–2.5)	0.64 0.16
	Chair stand speed + contrast sensitivity + hearing	Chair stand speed	2.27 (1.56–3.31)	< 0.001
		Contrast sensitivity	1.62 (1.12–2.36)	0.01
		Hearing Mild loss Moderate/Severe loss	1.16 (0.71–1.87) 1.5 (0.84–2.68)	0.55 0.17
	Chair stand speed + near vision + hearing	Chair stand speed	2.41 (1.67–3.47)	< 0.0001
		Near vision	1.14 (0.82–1.58)	0.42
		Hearing Mild loss Moderate/Severe loss	1.18 (0.73–1.91) 1.54 (0.87–2.74)	0.49 0.13
	Chair stand speed + distance vision + hearing	Chair stand speed	2.39 (1.65–3.45)	< 0.0001
		Distance vision Mild imp Moderate/Severe imp	1 (0.61–1.64) 1.98 (0.99–3.96)	0.98 0.05
		Hearing Mild loss Moderate/Severe loss	1.15 (0.71–1.86) 1.51 (0.86–2.66)	0.55 0.15
Model set 5	Motor + contrast sensitivity + hearing	Grip Strength	0.92 (0.66–1.28)	0.61
		Balance	1.48 (0.96–2.28)	0.08
		Gait speed	2.18 (1.32–3.6)	< 0.01
		Chair stand speed	1.69 (1.17–2.44)	< 0.01
		Contrast sensitivity	1.52 (1.04–2.22)	0.03
		Hearing Mild loss Moderate/Severe loss	1.09 (0.68–1.75) 1.41 (0.8–2.47)	0.71 0.22
	Motor + distance vision + hearing	Grip Strength	0.93 (0.66–1.3)	0.66
		Balance	1.47 (0.95–2.28)	0.08
		Gait speed	2.22 (1.36–3.63)	< 0.01
		Chair stand speed	1.75 (1.21–2.52)	< 0.01
		Distance vision Mild imp Moderate/Severe imp	0.96 (0.58–1.58) 1.83 (0.91–3.67)	0.85 0.09
		Hearing Mild loss Moderate/Severe loss	1.09 (0.68–1.75) 1.42 (0.82–2.45)	0.71 0.20
	Motor + near vision + hearing	Grip Strength	0.94 (0.68–1.32)	0.73
		Balance	1.46 (0.95–2.24)	0.08
		Gait speed	2.27 (1.38–3.71)	< 0.01
		Chair stand speed	1.76 (1.22–2.53)	< 0.01
		Near vision	1.08 (0.79–1.49)	0.62
		Hearing Mild loss Moderate/Severe loss	1.11 (0.69–1.79) 1.45 (0.84–2.52)	0.66 0.18
Interaction models	Gait speed * contrast	Gait speed	2.54 (1.55–4.14)	< 0.001
		Contrast	0.82 (0.19–3.63)	0.79
		Gait speed * Contrast	2.16 (0.48–9.69)	0.30
	Gait speed * distance	Gait speed	2.86 (1.76–4.65)	< 0.001
		Distance Mild imp Moderate/Severe imp	0.48 (0.10–2.38) 0.62 (0.07–5.60)	0.36 0.66
		Gait speed * Mild distance imp Gait speed * Moderate/severe distance imp	2.19 (0.39–12.34) 3.36 (0.24–46.56	0.36 0.35
	Chair stand speed * contrast	Chair stand speed	2.27 (1.47–3.52)	< 0.001
		Contrast	1.62 (0.73–3.61)	0.23
		Chair stand speed * Contrast	1.02 (0.42–2.48)	0.96
	Chair stand speed * distance	Chair stand speed	2.33 (1.56–3.48)	< 0.001
		Distance Mild imp Moderate/Severe imp	0.75 (0.19–3.01) 1.74 (0.55–5.57)	0.67 0.34
		Chair stand speed * Mild distance imp Chair stand speed * Moderate/severe distance imp	1.47 (0.29–7.57) 1.21 (0.25–5.93)	0.63 0.81
	Standing balance * contrast	Standing balance	2.33 (1.49–3.63)	< 0.001
		Contrast	2.54 (1.22–5.28)	0.01
		Standing balance * Contrast	0.62 (0.27–1.44)	0.26
	Standing balance * distance	Standing balance	2.04 (1.33–3.13)	< 0.01
		Distance Mild imp Moderate/Severe imp	1.2 (0.40–3.59) 1.66 (0.34–8.07)	0.73 0.52
		Standing balance * Mild distance imp Standing balance * Moderate/severe distance imp	0.85 (0.25–2.95) 1.41 (0.22–9.08)	0.79 0.71

^aBold values denote statistical significance at the p < 0.05 level

For motor impairment, in models considering each motor impairment separately (Table 3, Model set 1), impairment in standing balance, gait speed, or chair stand speed was associated with higher odds of dementia (OR = 2.10 (95% CI: 1.38–3.18), OR = 3.11 (95% CI: 1.92–5.06), and OR = 2.46 (95% CI: 1.69–3.57), respectively). Impaired grip strength was not associated with dementia (OR = 1.21 (95% CI: 0.87–1.69)). These results were consistent when independent variables for each motor function were placed in the same model, with the exception of standing balance impairment, which was attenuated (OR = 1.49 (95% CI: 0.97–2.30)) (Table 3, Model 3)).

In models where independent sensory or motor variables were placed in the same model (Table 3, Model set 4), impairments in contrast sensitivity and moderate or severe impairment in distance as well as impairments in balance, gait speed, and chair stand speed were consistently associated with higher odds of dementia (p < 0.05 for all), with the exception of when distance vision impairment was modeled in conjunction with gait speed or chair stand speed impairments. In the models where vision impairment, hearing impairment, and all four motor variables modeled together (Table 3, Model set 5), impairments in gait speed, chair stand speed or contrast sensitivity were associated with higher odds of dementia (p < 0.05 for all). Interactions of either distance vision impairment or contrast vision impairment with gait speed, chair stand speed, or standing balance were not statistically significant (Table 3).

Sensitivity analyses

When the sample was limited to community-dwelling participants (N = 2247, dementia: 283, non-dementia: 1964), results remained consistent for contrast sensitivity, gait speed, standing balance, and chair stand speed, but distance vision was attenuated (Supplemental Table 3).

Discussion

This study examined the independent associations of sensory and motor function with dementia in a cross-sectional study of a nationally representative population of U.S. Medicare beneficiaries. Our results show that impairments in gait speed, chair stand speed, contrast sensitivity, and distance vision are consistently associated with higher odds of concurrent dementia, even after adjusting for other impairments and demographics, which were considered as potential confounders; with the exception of when distance vision impairment was modeled in conjunction with gait speed or chair stand speed impairments. Hearing impairment and near visual acuity impairment were not associated with dementia in any models. Collectively, these cross-sectional findings suggest that lower extremity motor impairments and certain vision impairments are strong indicators of odds of dementia at older ages.

Motor function is a well-established predictor of poor health outcomes, including, disability and death [40, 41]. More recently, interest has grown in the association between motor function and cognitive health [42–44], and previous literature suggests that, in general, better motor function is associated with better cognitive performance [45]. For example, stronger grip strength, better standing balance, faster chair stand performance, and faster gait speed have been associated with better cognitive performance [46–48]. Much of this evidence has been attributed to neurodegeneration and/or accumulation of adverse pathology in shared pathways of the peripheral and central nervous systems [44, 49–53]. In the current study, grip strength was not associated with dementia prevalence. Differences between the findings of our study and the existing literature could be attributed to different population characteristics (age, gender, socioeconomic status, comorbidity burden, etc.), methodological differences (definition of cognitive impairment/dementia, inclusion of both sensory and motor variables, and varying thresholds used for definition of impairments), and inclusion of specific covariates (smoking, clinical history). NHATS has the advantage of being a nationally representative sample of U.S. older adults that provides an objective assessment of motor and sensory function, unlike other clinical cohorts which may be focused by geography or clinical conditions. Additionally, in most cross-sectional studies of dementia, only one motor variable has been considered at a time (e.g., only gait speed or only grip strength) rather than a wider range of motor variables that includes both upper and lower extremity function, and few (if any) have included sensory performance as well [54–57].

A growing body of evidence also suggests that poor hearing and vision function are associated with worse cognitive performance and higher odds of dementia [4, 58–62]. The current results did not find associations of near vision impairment or hearing impairment with dementia, either independently or when modeled with other sensory or motor impairments. Recent evidence from other NHATS work suggests the prevalence of moderate or greater (≥ 41 dB BPTA) hearing loss ranges from 18.2% at ages 71–74 years to more than 75% in those over 85 years [17], and that moderate or greater hearing loss is associated with a higher prevalence of concurrent dementia compared to normal hearing [63]. Similarly, the number of older adults with vision impairment in NHATS is higher with advancing age, with moderate or greater distance impairment (≥ 0.48 logMAR) ranging from 1.6% at ages 71–74 years to over 10% in those over 85 years, and contrast sensitivity impairment (< 1.55 logCS) ranging from 14.1% at ages 71–74 to over 40.2% in those over age 85 [16]. As with hearing, dementia prevalence was higher among those with worse vision impairment, including near vision impairment [4]. There are potential explanations for the difference between our findings and previous NHATS work: [1] the previous studies only examined hearing or vision and did not consider motor impairments or other sensory impairments in their analyses, and [2] varying inclusion criteria contributed to some discrepancies in the participants included in the analytic samples. Longitudinal analyses examining the onset and progression of sensory and motor impairments are needed to better disentangle the temporality of these pathways, their potential for shared and distinct neurodegeneration, and resulting compensatory behaviors. Specifically, how, and when in the pathological process motor and sensory impairments affect neural pathways, as well as potential resulting modifications to social and behavioral factors (e.g., decreased social engagement or physical activity), should be considered when assessing dementia risk.

A few limitations should also be noted. First, a diagnosis of dementia in most clinical and epidemiological research requires a documented history of decline in cognitive functioning and clinical adjudication. Such information is generally unavailable in large-scale population surveys. Although the dementia outcome in NHATS is not based on clinical diagnosis, the algorithm has been validated [29]. Second, the cross-sectional nature of this work prevents temporal assessments of the sensory function, motor function, and dementia association. In addition, our study was limited to those who participated in sensory and motor function testing. This design may have excluded some with worse sensory or motor function who declined the tests and may be at higher risk of dementia. This limitation is not unique to NHATS, and future research is needed to discern how to best assess sensory and motor function information among those with limited capacity for testing. It is possible that those with sensory impairments had difficulty hearing or seeing the material for cognitive testing. This is ubiquitous in scientific research and not unique to NHATS. In NHATS, participants complete the cognitive performance battery using their vision/hearing aids if they have them, and interviewers may repeat the instructions as needed [26]. Whether they had difficulty hearing words was also recorded by the interviewer. Finaly, because we are unable to address questions relating to how long participants have used hearing aids, type of hearing aid used, or consistency of use, we did not include hearing aid use in our analyses, which has been associated with lower dementia prevalence among those with moderate to severe hearing loss [63] Future longitudinal research should explore whether -and how- hearing aid use may mediate the relationship between hearing impairment and dementia.

Conclusions

In this nationally representative sample of U.S. Medicare beneficiaries with motor and sensory impairments, those with impaired gait speed, chair stand time, and contrast sensitivity appeared to have higher odds of concurrent dementia even after controlling for hearing impairment, grip strength impairment, standing balance impairment, and potential confounders such as age and number of comorbidities. With the aging of the population and the high prevalence of motor function and sensory impairments at older ages, further research is needed to understand the temporal association among vision and motor impairments, and the associated risk of incident dementia to identify prevention strategies to reduce dementia risk among older adults.

Authors’ contributions

Study concept and design: JAS, YA. Analysis: AD. Interpretation of data: AD, HW, RJD, AAW, ALG, JAS. Drafting: AD, JAS. Critical revision: All authors.

Funding

NHATS is supported by U01 AG032947 from the National Institute on Aging. RJD is supported by grant K01 AG080122, JRE is supported by grant R01EY034479 and U01 AG032947. NSR is supported by K23 AG065443 and U01 AG032947. ALG is supported by R01 AG070953, R01 AG030153, R01 AG077001, 1R01 AG074355, P30 AG066507, RF1 AG059869, and R01 AG079391. JAS is supported by U01 AG032947, and R01 AG061786. YA is supported by R01 AG061786. AD is supported by R01 AG061786. AAW is supported by K01 AG076967, and R01 AG061786.

Data availability

NHATS data is publicly available at www.nhats.org.

Declarations

Ethics approval and consent to participate

Existing data were collected using an informed consent document. IRB approval for NHATS individual data is held by the Johns Hopkins Bloomberg School of Public Health Institutional Review Board (IRB#2083). This study was conducted in accordance with the guidelines of Declaration of Helsinki.

Clinical Trial Number

Not applicable.

Consent for publication

Not applicable.

Competing interests

NSR reports serving the scientific advisory board of Neosensory (Summer 2022-Fall 2023). JAS is a consultant for Edwards Lifesciences. JAS and AAW are on the editorial board of JGMS. No financial disclosures have been reported by other authors of this study.