OBJECTIVES:Older Hispanic adults in the United States and Mexico face a high burden of diabetes and dementia, yet cross-national differences in absolute survival associated with these conditions remain unclear. This study compared 5-year survival among Mexican-origin adults aged 74 and older by condition profile and assessed gender differences. METHODS:Two representative longitudinal surveys were harmonized: the Hispanic Established Populations for Epidemiologic Study of the Elderly-Centers for Medicare & Medicaid Services (H-EPESE-CMS) linked data for the United States and the Mexican Health and Aging Study (MHAS) for Mexico. Flexible parametric survival models and g-computation estimated standardized 5-year survival, risk differences, and risk ratios across four groups: no condition, diabetes only, dementia only, and dual conditions. RESULTS:Survival declined steadily with condition burden in both countries. In the United States, standardized 5-year survival was 73.3% (no condition), 57.9% (diabetes only), 49.2% (dementia only), and 36.6% (dual conditions). In Mexico, absolute survival was higher across all groups: 77.5%, 67.9%, 56.8%, and 40.9%, respectively. Hazard ratios confirmed elevated mortality risks, particularly when both were present. Women had lower mortality than men in the United States, but higher mortality in Mexico. However, gender did not significantly modify the association between condition burden and survival in either country. DISCUSSION:Five-year survival declines with condition burden in both countries, with lower absolute survival and steeper relative penalties in the U.S. Divergent gender patterns across countries reflect overall mortality differences rather than condition-specific modification, highlighting the need for tailored and integrated approaches to chronic disease and dementia care.

Unmet care needs (lack of assistance with daily activities) among older adults are common and linked to adverse health outcomes, greater healthcare utilization, and caregiver burden. While effective patient-provider communication is known to improve care quality, less is understood about the role of caregiver-provider communication, particularly across sociodemographic, health, and caregiving contexts. We conducted a cross-sectional secondary analysis of pooled 2021–2023 National Health and Aging Trends Study (NHATS) and National Study of Caregiving (NSOC) data, restricted to community-dwelling older adults (65 + years) who received assistance from caregivers due to difficulty with daily activities, and had at least 1 caregiver who communicated with medical providers in the past year. The unit of analysis was the caregiver-older adult pair, with clustering at the older adult level. The outcome was older adults’ unmet care needs (0, 1, or 2 + , across 12 daily activities). The primary predictor was caregiver-reported provider communication quality (range: 3–12). Weighted ordered logistic regression estimated associations between communication quality and unmet care needs, adjusting for older adult and caregiver sociodemographic, health, and caregiving characteristics. The analytic sample included 1,414 older adults (weighted N = 9.34 M) and 1,910 caregivers (weighted N = 21.05 M). Half (50.5

Objectives:This study explored the mediation and moderation effects of social engagement on the relationship between hearing status and cognitive function among Mexican adults aged ≥ 50 years across a six-year period (2015-2021). Methods:Data were drawn from the nationally representative Mexican Health and Aging Study (MHAS). Social engagement was calculated by summing the number of social activities and frequencies undertaken in the past year. A global measure of cognitive function was derived by summing z-scores of verbal memory, orientation, verbal fluency, visual scanning, figure copy, and figure recall. Multilevel generalized models with lagged predictors were used to assess the mediation and moderation effects of social engagement on the association between self-reported hearing status and cognitive function over time. Results:There were 10,561 people in the analytic sample. Greater social engagement was independently associated with better cognitive function over time. Better hearing status was associated with higher cognitive functioning in the following survey wave. Fair or better hearing was significantly associated with increased social engagement. The association was attenuated when covariates were adjusted. The association between hearing difficulty and cognitive function varied across levels of social engagement. Discussion:The mediation and moderation effects of social engagement on the association between hearing difficulty and cognitive function were affected by the population's socioeconomic status among middle-aged and older adults in Mexico. Future research should investigate other unknown mediators between hearing difficulty and cognitive health of middle-aged and older adults in low- and middle-income countries.

INTRODUCTION:We examined whether bilingualism was associated with a cognitive advantage among older urban- or rural-dwelling Mexican adults. METHODS:Participants were from the Mexican Health and Aging Study Ancillary Study on Cognitive Aging (Mex-Cog) from urban (N = 1063, 12% Spanish-English bilingual adults) and rural (N = 814, 19% Spanish-Indigenous bilingual adults) areas. Memory, language, and executive functioning were assessed. Weighted linear models stratified by locality evaluated effects of bilingualism on cognitive domains, adjusting for sociodemographic factors. RESULTS:In urban settings, Spanish-English bilingualism was not associated with cognition in any domain (all Ps > 0.05). In rural settings, Spanish-Indigenous bilingual adults had lower scores across all domains (Ps < 0.01). DISCUSSION:There was no evidence of a cognitive advantage among older bilingual adults in Mexico. Indigenous bilingual adults performed worse cognitively compared to Spanish-monolingual peers. Further work is needed to understand the linguistic and sociocultural characterization of older Mexican adults to better evaluate bilingualism and cognitive aging.

This paper examines whether social activities mediate the association between hearing difficulty and cognitive function across contexts. Data come from the Health and Retirement Study (2012-2020), Mexican Health and Aging Study (2015-2021), and China Health and Retirement Longitudinal Study (2011-2018). Weighted multilevel generalized models were used. Hearing difficulty was associated with lower verbal memory in each database while the MHAS showed a comparatively smaller effect (bHRS = -0.094, p < .001; bMHAS = -0.080, p < 0.01, and bCHARLS = -0.093, p < 0.001). These associations remained statistically significant after adjusting for social activities. Hearing difficulty is associated with decreased social activities in HRS and CHARLS, but not in MHAS. Social activities partially mediate the hearing-verbal memory association in HRS and CHARLS, but not in MHAS, which highlights a non-uniform association between hearing difficulty, social activities, and cognition across contexts.

BACKGROUND:Physical activity fragmentation is associated with poor physical function in older adulthood, but the relationship between changes in physical activity fragmentation with concurrent and future changes in physical function is unclear, and this information is important for guiding clinical decision making. We therefore characterized this relationship in a nationally representative sample of older adults. METHODS:Data were drawn from a subset of the National Health and Aging Trends Study. Activity fragmentation was defined as the probability of transitioning from an active to a sedentary bout, captured using wrist-worn actigraphy, while function was assessed using the Short Physical Performance Battery (SPPB). The association between physical activity fragmentation and any instance of longitudinal decline in SPPB was evaluated using multivariable logistic regression. RESULTS:The sample included 531 participants aged ≥70 years (55.4% female). Among those with baseline SPPB ≤ 6, higher baseline activity fragmentation was associated with lower baseline function overall and with a 1-year future decline in function (adjusted OR: 1.32; 95% CI: 1.01-1.72 per 10% higher baseline fragmentation). A 1-year increase in fragmentation was associated with a greater likelihood of experiencing a concurrent decline in SPPB, particularly among those with baseline SPPB = 7-10 (OR: 2.21; 95% CI: 1.17-4.17). The 1-year change in activity fragmentation was not associated with declines in SPPB the following year. CONCLUSIONS:Physical activity fragmentation may represent a digital biomarker, rather than an antecedent, of changes in physical function, highlighting the potential utility of accelerometry for identifying at-risk phenotypes likely to benefit from early interventions.

Drought represents a climate-related exposure impacting communities across the globe. Drought exposure has been linked with adverse human health, including poorer mental health and nutritional outcomes. It is unknown whether drought exposure relates to cognitive function. We evaluated cognitive function in adults aged 50+ by length of exposure to a major drought occurring between 2010 and 2012 in Mexico. We used individual-level data from the Mexican Health and Aging Study (n = 6988), drawing from pre-drought (2003) and post-drought (2012) waves, linked with monthly municipality-level information from the Mexican Drought Monitor on drought exposure produced by the Mexican National Water Commission. We employ multilevel regression models, with inverse probability of attrition weighting, to examine how length of drought exposure is related with post-drought cognition, controlling for pre-drought cognition and nutritional/mental health covariates. Whether quantified as the total number of months of drought exposure or the longest streak of consecutive months of drought exposure, longer drought exposure was negatively related with Verbal Learning and Verbal Recall performance over time but exhibited an inverse U-shaped association with Verbal Fluency. Findings were similar when using various thresholds of drought severity (i.e., severe to exceptional drought). Associations between drought and cognition were not explained by nutrition or mental health covariates. Public health and policy efforts should seek to build community-level resilience and infrastructure to enable effective coping with persistent environmental stressors, especially among older adults, to mitigate effects on health and well-being.

Research Objectives To determine if the associations between self-reported head or neck injury and walking difficulty differs by gender. Design Cross-Sectional analysis of the 2014 interview wave of the U.S. Health and Retirement Study. Setting Community-dwelling middle-aged and older adults in the United States. Participants We identified 1,478 participants who responded to a questionnaire about whether they have experienced a head or neck injury from a vehicle accident, fall, sports, or violent act (i.e. fight, shot, explosion) Half (n=742) of the participants responded "yes" to a questionnaire about whether they have experienced a head or neck injury. Main Outcome Measures Walking Difficulty was the main outcome measure. All participants were asked, “Because of a health problem, do you have difficulty walking several blocks”? Participants could respond with “yes”, “no”, “can’t do”, or “doesn’t do”. We categorized participants who responded yes, can’t do, or don’t do as having difficulty walking. All other participants were categorized as not having difficulty walking several blocks. Results Nearly 60% of participants (59.8%) were female, and the, mean age was of 67.0. Those who reported having experienced a head or neck injury, compared to participants who did not report having experienced a head or neck injury, were approximately three years younger on average (65.8 years vs. 68.4 years, p< 0.01) and were more likely to be male (47.0% vs. 33.4%, p< 0.01). Among the 465 (31.5%) reporting walking difficulty, ages ranged from 32 to 99 years of age. Thirty-two percent of all participants reported they had difficulty walking. In a logistic regression model adjusting for age, sex, and education, having head or neck injury was associated with 1.73 (95% CI=1.36 – 2.20) higher odds of having difficulty walking. This association was similar for men (OR=1.79, 95% CI=1.20-2.69) and women (OR=1.69, 95% CI=1.26-2.28. Conclusions These findings are evidence that having experienced a head or neck injury is associated with difficulty walking several blocks among middle-aged and older adults, but that this association is similar for men and women.

ObjectiveThis study intends to examine the education gradients in functional limitations for older adults in Mexico by comparing effect size between two birth cohorts age 60-76 in 2001 versus 2018. Data were drawn from the Mexican Health and Aging Study (MHAS) 2001 and 2018 waves. Tobit model was used to estimate the education gradient in functional limitations. Living conditions in childhood (pluming in house, parents’ education, and serious health conditions), chronic conditions, and the longest work were adjusted for. The cohort difference was tested by including interaction effects and propensity matching scores. Compared to the 2001 cohort, the 2018 cohort presented lower proportion of no education, and higher proportion in each level of education, especially the secondary and tertiary education. Education gradients were significant in functional limitations for each birth cohort, but the education gradient was attenuated for the 2018 cohort compared to the 2001 cohort. Living conditions in childhood have long-term effects on functional limitations in later life. Agricultural and industrial work contribute to greater functional limitations. No different effects of primary job/work and early life conditions between the cohorts were detected. Disregarding the average higher levels of education within the 2018 wave birth cohort, the education gradients in functional limitations decreased compared to the early birth cohort in 2001 wave.

BackgroundVision impairment is a risk factor for mild cognitive impairment (MCI) among stroke survivors, but it is unclear if this association is driven by vision impairment present before or due to the stroke, and if similar associations exist with dementia.ObjectiveTo (1) characterize the prevalence of pre-stroke and stroke-related vision impairment(s) among stroke survivors, and (2) quantify associations of vision impairment with dementia and cognitive impairment (MCI/dementia).MethodsUsing participants from the Atherosclerosis Risk in Communities (ARIC) dataset with adjudicated incident strokes, we gathered descriptive statistics on the cohort, assessed if vision impairment was present at the time of incident stroke, and classified the impairments as pre-stroke or stroke-related. Multivariable logistic regression was used to estimate the association between these types of vision impairment and post-stroke cognitive impairment.ResultsAmong 233 incident stroke survivors (mean = 69 years old and 50.2% female sex), 23.2% with pre-stroke vision impairment and 18.9% with stroke-related vision impairment, there were 124 (53%) cases of cognitive impairment (n = 76 MCI, n = 48 dementia). Stroke-related vision impairment was significantly associated with higher odds of dementia (ref = normal/MCI) (adjusted odds ratio (aOR) = 2.32, 95% confidence interval (CI) = 1.08-4.92, p = 0.029), but not any cognitive impairment (ref = normal) (aOR = 1.33 95% CI = 0.67-2.70, p = 0.425). Further adjusting for stroke severity score attenuated the association of stroke-related vision impairment with dementia (aOR = 2.0, 95% CI = 0.90, 4.32, p = 0.08).ConclusionsStroke-related vision impairment, but not pre-stroke vision impairment, was associated with higher odds of dementia. There is evidence that stroke severity could, at least partially, explain the observed association.

This study explores the mediating effect of social activities on the association between hearing status and cognitive function across three countries. Data from three waves of the HRS (United States), MHAS (Mexico), and CHARLS (China), collected from 2011 to 2021, were analyzed. Multilevel models were used to examine the association between self-reported hearing status (fair/poor vs. good/excellent) and verbal memory. The mediating effects of social activities were assessed using the Karlson-Holm-Breen (KHB) method. Fair/poor hearing was associated with lower verbal memory in the HRS (b=-0.145, SE = 0.023) and CHARLS (b=-0.056, SE = 0.020), but not in MHAS (b=-0.001, SE = 0.018). Social activities were positively associated with verbal memory in all three datasets. Fair/poor hearing was associated with reduced social activities in the HRS (OR = 0.838, p < 0.001), MHAS (OR = 0.847, p < 0.01), and CHARLS (OR = 0.779, p < 0.001). The association between self-rated hearing and verbal memory remained statistically significant after adjusting for social activities in the HRS (b=-0.144, SE = 0.023) and CHARLS (b=-0.045, SE = 0.021). We did not find universal mediation effects of social activities on the association between hearing status and verbal memory. These findings underscore the need for context-specific approaches to addressing sensory loss and cognitive health.

BACKGROUND:As extreme heat events become increasingly common, it is crucial to understand their health effects on older adults, who are particularly vulnerable to adverse outcomes associated with heat exposure. One area of concern is cognition, which may be affected directly, through heat-related physical health deterioration, and indirectly, via factors such as social isolation and depression. Our objective is to assess the association between extreme heat and cognition among older Mexican American adults METHOD: We used data from the Hispanic Established Population for the Epidemiologic Study of the Elderly (HEPESE; 2004/2005- 2015/2016), a longitudinal study of older Mexican American adults living in the Southwestern US. We limited this analysis to participants residing in Texas or California. Heat data was sourced from the GridMET database, which was linked to the HEPESE through the Contextual Data Resource at the University of Southern California. We used daily maximum heat index values to create a yearly count of days with 1) a maximum heat index of 90°F+ (extreme heat) and 2) a maximum heat index of 104+ (dangerous heat). Cognition was measured at each wave with the Mini-Mental State Examination (MMSE). We used mixed effects models to assess the longitudinal association between yearly count of extreme/dangerous heat days and MMSE score, adjusting for demographic and health covariates. RESULT:Our sample (n = 1,723) had an average age of 81.7 [standard deviation (SD): 5.1] and was 61.5% female. At baseline, the average MMSE score was 20.8 (SD: 7.2). On average, participants were exposed to 96.6 extreme heat days (SD: 60.6) and 18.1 dangerous heat days (SD: 21.6) at baseline. Extreme heat was associated with worse cognition at baseline [β: -0.008; 95% confidence interval (CI): -0.016, -0.001] and with decline over time (β: -0.003; 95% CI: -0.004, -0.002). Similarly, dangerous heat was associated with worse cognition at baseline (β: -0.033; 95% CI: -0.050, -0.017) and decline over time (β: -0.005; 95% CI: -0.007, -0.003). CONCLUSION:We found that daily heat indices greater than 90°F were associated with worse cognition over time among older Mexican Americans. Future work is needed to understand the pathways through which heat influences cognition.

This research investigated the relationship between cognitive performance and an individual's educational attainment as well as occupational mental demands among Mexican adults aged 50 or older. We hypothesized that cognitively demanding work boosts cognitive performance for older adults regardless of their education level. To test our hypothesis, we analyzed data on 12,939 individuals in the 2012 Mexican Health and Aging Study using a Generalized Linear Model with a Gaussian family and identity link function. We assessed cognitive demands of occupations with the National Information Network's descriptors, focusing on worker-oriented and job-oriented mental demands. We found that greater worker-oriented ((3 = 0.5; CI = 0.45, 0.55) and job-oriented ((3 = 0.49; CI = 0.45, 0.53) mental demands predicted better cognitive performance. Educational attainment correlated even more strongly with better cognitive performance ((3 = 0.9; CI = 0.87, 0.92). Both our models showed a statistically significant negative interaction between medium occupational mental demands and medium education level (job-oriented: (3 = -0.09; CI = -0.14, -0.05; worker-oriented: (3 = -0.07; CI = -0.12, -0.02). Other interaction terms were not significant. This study highlighted a significant effect of educational attainment on cognitive function, which is more pronounced than that of occupational mental demands. The association of higher occupational mental demands with higher cognitive function appeared to be largely independent of educational background. The similarity in cognitive scores using worker-oriented or job-oriented metrics suggests that both are useful for assessing occupational mental demands. Education and cognitive engagement at work are crucial for promoting cognitive health in aging populations.

ObjectivesFew studies have investigated health outcomes associated with arthritis in low-middle-income countries. Our objective was to examine the association between arthritis and all-cause hospitalization and mortality among middle-aged and older adults in Mexico.MethodsOur sample included 12,106 participants aged >50 years from the 2012, 2015, and 2018 waves of the Mexican Health and Aging Study. Logistic regression was used for the associations between arthritis and hospitalization. Kaplan-Meier and Cox proportional hazard models were used for the association between arthritis and mortality.ResultsArthritis was associated with higher odds of hospitalization (OR = 1.23; 95% CI = 1.09-1.38), but not mortality. Arthritis with physical limitations had the highest odds of hospitalization (OR = 1.48; 95% CI = 1.27-1.73). Arthritis with joint pain (OR = 1.24; 95% CI = 1.10-1.41) and medication use (OR = 1.28; 95% CI = 1.11-1.48) had higher odds of hospitalization.ConclusionAmong middle-aged and older adults in Mexico, arthritis was associated with a high risk of hospitalization.

Depression, pain, and sleep problems commonly co-occur (overlap) among older adults. Data are lacking on the extent, pattern and sex/gender differences of overlap of these symptoms in adults aged ≥80 and living with dementia. Our objective was to examine patterns and sex/gender differences in overlapping depression-pain-sleep symptoms among older Mexican Americans with and without probable dementia. We used data from Wave 7 (2010/2011) of the Hispanic Established Population for the Epidemiologic Study of the Elderly, a study of Mexican Americans aged ≥75, residing in the Southwestern US. Participants were considered to have probable dementia if they scored <21 on the Mini-Mental State Examination and had ≥1 activity of daily living limitation. Pain was defined as pain on weight-bearing. Depressive symptoms were defined as having scores of 16 or more on the Center for Epidemiologic Studies Depression Scale. Clinically relevant sleep problems were defined as having trouble falling asleep, waking up several times per night, trouble staying asleep, or waking up feeling tired for at least 15 days in the past month. We used descriptive statistics to describe the prevalence and overlap in these three symptoms, by dementia status and sex/gender. Our sample was 85.7 years old on average and 65% were women (n = 974). Almost one-third of participants (32.4%) had dementia. About 16% of participants with probable dementia reported overlapping depression-pain-sleep symptoms, compared to 5.5% of those without dementia (Figure 1). When stratified by cognitive status and sex/gender, 14.3% of men and 17.0% of women with probable dementia reported all three symptoms, compared to 2.0% of men and 7.6% of women without dementia. Participants with probable dementia reported three times the frequency of co-occurrence of clinically relevant depressive symptoms, pain, and sleep problems compared to those without dementia. Women, regardless of cognitive status, more frequently reported depression-pain-sleep symptom overlap, compared to men, but the gender disparity was smaller for those with probable dementia. The presence of one symptom should alert clinicians to screen for, treat, and manage the other two. Evidence supports common underlying mechanisms for these three highly co-occurring conditions, potentially informing therapeutic decision-making in patients with dementia.

Hispanic older adults living near the U.S.-Mexico border encounter considerable healthcare challenges that impact the management of diabetes and Alzheimer’s Disease or Related Dementias (ADRD). Studies have found that Hispanic older adults living on the U.S.-Mexico border are disproportionately affected by financial strain, limited healthcare access, and a higher prevalence of chronic diseases. Individual diagnoses of either diabetes or ADRD already present significant health challenges for Hispanic older adults, and dual diagnoses exacerbates these challenges, increasing the complexity of medical management. Using the Medicare-linked HEPESE survey (n = 718) we conduct statistical analyses to explore the relationship between dual diagnoses of ADRD and diabetes, with self-reported health measures and healthcare utilization. Our key stratifying variable is proximity to the border, given our previous findings that it explains caregiver and community resources that affect outcomes for Hispanic older adults aged 80+. Preliminary results indicate that Hispanic older adults with dual diagnoses near the US-Mexico border exhibit worse measures of healthcare utilization compared to those with a single diagnosis. Similarly, we find that Hispanic older adults with dual diagnoses near the US-Mexico border exhibit worse measures of self-reported health related to their conditions, compared to those with a single diagnosis. Structural and policy-driven factors, particularly healthcare access, influence healthcare utilization and health outcomes, with proximity to the border affecting access to care. This project is significant as it highlights the unique challenges faced by older adults with dual diagnoses in border region, aiming to inform policies that address healthcare issues for vulnerable populations.

BackgroundSocial determinants of health (SDoH), such as financial resources and housing stability, account for between 30% and 55% of people’s health outcomes. While many studies have identified strong associations between specific SDoH and health outcomes, little is known about how SDoH co-occur to form subtypes critical for designing targeted interventions. Such analysis has only now become possible through the All of Us program. ObjectiveThis study aims to analyze the All of Us dataset for addressing two research questions: (1) What are the range of and responses to survey questions related to SDoH? and (2) How do SDoH co-occur to form subtypes, and what are their risks for adverse health outcomes? MethodsFor question 1, an expert panel analyzed the range of and responses to SDoH questions across 6 surveys in the full All of Us dataset (N=372,397; version 6). For question 2, due to systematic missingness and uneven granularity of questions across the surveys, we selected all participants with valid and complete SDoH data and used inverse probability weighting to adjust their imbalance in demographics. Next, an expert panel grouped the SDoH questions into SDoH factors to enable more consistent granularity. To identify the subtypes, we used bipartite modularity maximization for identifying SDoH biclusters and measured their significance and replicability. Next, we measured their association with 3 outcomes (depression, delayed medical care, and emergency room visits in the last year). Finally, the expert panel inferred the subtype labels, potential mechanisms, and targeted interventions. ResultsThe question 1 analysis identified 110 SDoH questions across 4 surveys covering all 5 domains in Healthy People 2030. As the SDoH questions varied in granularity, they were categorized by an expert panel into 18 SDoH factors. The question 2 analysis (n=12,913; d=18) identified 4 biclusters with significant biclusteredness (Q=0.13; random-Q=0.11; z=7.5; P<.001) and significant replication (real Rand index=0.88; random Rand index=0.62; P<.001). Each subtype had significant associations with specific outcomes and had meaningful interpretations and potential targeted interventions. For example, the Socioeconomic barriers subtype included 6 SDoH factors (eg, not employed and food insecurity) and had a significantly higher odds ratio (4.2, 95% CI 3.5-5.1; P<.001) for depression when compared to other subtypes. The expert panel inferred implications of the results for designing interventions and health care policies based on SDoH subtypes. ConclusionsThis study identified SDoH subtypes that had statistically significant biclusteredness and replicability, each of which had significant associations with specific adverse health outcomes and with translational implications for targeted SDoH interventions and health care policies. However, the high degree of systematic missingness requires repeating the analysis as the data become more complete by using our generalizable and scalable machine learning code available on the All of Us workbench.

Vision impairment is a common complication of stroke, which has been linked to cognitive decline in stroke survivors. Little is known about how specific types of vision impairment influence the relationship between stroke and dementia. The aims of this project were: (1) to characterize the prevalence of both pre-stroke and stroke-related vision impairment(s) among stroke survivors with and without mild cognitive impairment (MCI) or dementia, and (2) to determine associations of vision impairment with cognitive impairment, depression, and physical performance. Data from the Atherosclerosis Risk in Communities Study (ARIC) were studied. Our analytic cohort included ARIC participants who had both a stroke hospitalization captured via surveillance prior to dementia and cognitive measures from Visits 5-6. Participants without a retinal exam prior to their stroke were excluded. We determined the prevalence of specific types of vision impairments and ocular conditions as well as patient-level demographics, stratified by cognitive status. Multivariable linear and logistic regressions, controlling for age, sex, and education, modelled the relationship between types of vision impairment and each outcome. Our final cohort contained n = 227 individuals, with mean (standard deviation [SD]) age of 78.5 (5.56) years. Participants were primarily male (54.2%) and white (73.6%), with a mean of 14.5 (SD = 4.24) years of education. 15.9% had a pre-existing vision impairment (e.g., glaucoma), and 18.5% developed a stroke-related vision impairment (e.g., diplopia). Vision impairment (all types) was significantly associated with dementia (adjusted odds ratio (OR) = 2.17, 95% confidence interval (CI) = 1.07-4.41), but not MCI (OR = 1.23, 95% CI = 0.07-2.20). In assessing pre-stroke vision impairment and stroke-related vision impairment separately, there were no statistically significant associations with any outcomes. This study informs the rates of types of well-characterized vision impairments in a group of stroke survivors. There is a 2-fold increase in the likelihood of dementia among stroke survivors with vision impairment as compared to those without vision impairment, defined as either pre-stroke or stroke-related impairment. Future data collection in larger, more diverse populations is needed to determine how specific types of vision impairments in stroke survivors associate with cognitive impairment as well the likelihood of depression or worse functional status.

Pain is associated with cognitive decline, but prior research has not considered differences between high- and low-impact chronic pain or low-and middle-income countries, such as Mexico. Our objective was to evaluate the association between high-impact chronic pain and cognitive decline among older adults in Mexico. We used data from the 2012-2018 Mexican Health and Aging Study (MHAS), a nationally representative study of older Mexican adults (60+). Pain, measured at each wave, was categorized as high-impact chronic pain (pain that limits daily activities), low-impact chronic pain (pain that did not limit daily activities), and no chronic pain. Cognition was measured at each wave using the MHAS cognitive battery, and scores were standardized for global cognitive performance, memory, and non-memory domains. We used linear mixed models with inverse probability weights to evaluate the association between pain and cognition over time. At baseline (n=9056), 19.3% of participants reported high-impact pain and 20.1% reported low-impact chronic pain. Participants with high-impact pain had lower baseline global cognition [β: -0.024, 95% confidence interval (CI): -0.045, -0.004] and memory scores (β: -0.042, 95% CI: -0.074, -0.010) compared to those with no chronic pain. Change in cognition over time did not differ by pain, except for the memory domain, where those with high-impact pain experienced slower decline compared to those with no chronic pain. Low-impact chronic pain was not associated with any cognitive outcome. Better pain treatment may result in benefits in cognitive performance for those with pain. PERSPECTIVE: We found that high-impact chronic pain was associated with worse cognition at baseline, but not with cognitive decline. Memory domains were particularly affected by high-impact chronic pain. Early pain treatment and management may provide cognitive benefits for those living with pain.

OBJECTIVES:The nursing home (NH) population has become increasingly diverse, yet many facilities remain de facto racially segregated. This study examines whether a high proportion of Black, Indigenous, and People of Color (BIPOC) residents is associated with nursing staff levels. DESIGN:We constructed a longitudinal cohort of NHs (2013-2019) by linking Certification and Survey Provider Enhanced Reports, LTCFocUS.org, Medicare Cost Reports, and Payroll-Based Journal data. Separate multivariable random effects linear regressions were conducted. SETTING AND PARTICIPANTS:14,075 Medicare- and Medicaid-certified NHs in the United States. METHODS:The proportion of BIPOC residents was categorized as the 10% of nursing homes serving the highest minority residents in each state each year (High-BIPOC) and the remaining 90% (Low-BIPOC). Total nursing staff levels in hours per resident-day (HPRD) included both hours paid (2013-2019) and hours worked (2017-2019). The total staff included registered nurses, licensed practical nurses, and certified nurse aides. RESULTS:The unadjusted difference in total staff levels between High-BIPOC and Low-BIPOC NHs increased from -0.23 HPRD (4.19 vs 4.42) in 2013 to -0.35 HPRD (3.94 vs 4.29) in 2019 for hours paid. The difference in hours worked increased from -0.19 (3.55 vs 3.74) in 2017 to -0.23 (3.50 vs 3.73) in 2019. The difference became smaller but remained significant after controlling for covariates (-0.037 HPRD for hours paid, and -0.038 for hours worked). Analyses of individual staff types found lower levels of registered nurses and certified nurse aides (but not licensed practical nurses) among High-BIPOC nursing homes. Findings were robust to treating racial and ethnic composition as a continuous variable or excluding payer mix from the models. CONCLUSIONS AND IMPLICATIONS:NHs with high concentrations of minority residents reported lower nursing staff levels. Improving staffing in NHs serving primarily marginalized racial and ethnic groups remains a policy priority.