Abstract
Background:
Latino individuals represent one of the fastest-growing demographic groups in the United States, and the impact of dementia is rising within this population. Despite this growth, most research on healthcare utilization has predominantly focused on non-Hispanic White populations. The limited body of literature that does include Latino populations often treats them as a monolithic racial/ethnic category, which overlooks intra-group heterogeneity. As a result, little is known about how healthcare utilization patterns relate to cognitive impairment status across specific Latino ethnic groups.
Methods:
Using data collected from the Study of Latinos-Investigation of Neurocognitive Aging (SOL-INCA), we examined the odds of overall healthcare utilization and avoidable healthcare utilization according to cognitive impairment and whether these associations vary by Latino ethnicity. Weighted logistic regressions and two-way interactions were conducted.
Results:
This study included 4,908 (unweighted) Latino participants (1,665 Mexican, 931 Cuban, 877 Puerto Rican, 453 Dominican, 485 Central American, 352 South American, and 145 More than One/Other). Overall, participants reported high rates of overall healthcare utilization (88%) and avoidable healthcare utilization (42%). Individuals with cognitive impairment had higher odds of both overall (OR=1.94, p=0.013) and avoidable healthcare utilization (OR=1.51, p=0.004) compared to those without cognitive impairment. Puerto Rican participants were the only ethnic group found to have significant differences in avoidable healthcare utilization and cognitive impairment (OR=2.61, p =0.014).
Discussion:
Disaggregation of Latino data revealed significant healthcare utilization across Latino ethnic groups. Targeted interventions and resources are needed to promote preventive care that may decrease avoidable healthcare utilization and associated expenditures.
Keywords: health disparity, cognitive impairment, data disaggregation, healthcare utilization, Latino/Hispanic, interaction term, preventive care
Introduction
Latino communities are at high risk for Alzheimer’s Disease and Alzheimer’s Disease Related Dementias (AD/ADRD). There is a projected 7-fold increase in the number of Latinos living with AD/ADRD in the United States (US) by 2060, which is the highest among any other racial/ethnic group [1]. This increase poses a significant problem as research has shown that an individual with a diagnosis of AD/ADRD may have significantly greater inpatient healthcare utilization and costs compared to their age- and sex-matched peers without AD/ADRD [2]. Moreover, a nationally representative study found that among 1,853,632 hospital discharges of older adults with an AD/ADRD diagnosis, hospitalizations for potentially preventable conditions amounted to $9.3 billion [3].
There is growing recognition of the need to address social determinants of health (SDOH) (e.g., race, socioeconomic status), as they are key drivers that underscore avoidable healthcare utilization. Avoidable healthcare utilization refers to the preventable use of costly health services, such as hospitalizations or emergency room (ER) visits, through timely and quality primary and preventive care [4]. Reducing avoidable healthcare utilization is critical, as such encounters are generally more costly than preventive care and are associated with adverse outcomes for persons living with dementia [5]. Racial and ethnic groups, particularly Latino populations with mild cognitive impairment (MCI) or AD/ADRD, have been observed to engage in different types and frequencies (e.g., fewer doctor visits) of healthcare services than non-Hispanic Whites, potentially leading to increased avoidable healthcare utilization [6]. SDOHs are often differentially distributed across racial and ethnic populations and specific ethnic groups. However, there remains a paucity of research examining race and ethnicity as determinants of healthcare utilization, primarily due to the lack of representation and research on diverse Latino populations.
Existing studies often aggregate Latinos into one category, further limiting the scope and depth of our understanding of healthcare utilization disparities. Data disaggregation is one strategy that may aid in identifying AD/ADRD-related disparities across specific Latino ethnic groups [7]. Yet, data disaggregation is limited by poor data infrastructures, collection, and organization of data that is largely aggregated into broad racial and ethnic categories (e.g., Latino/Hispanic) [8]. Evaluating healthcare utilization among Latino ethnic groups across different types of cognitive impairment is needed, particularly before the identification of AD/ADRD (e.g., MCI) [9].
There is limited research on differences in healthcare utilization by cognitive impairment in diverse Latino ethnic groups. Older adults may experience cognitive decline several years before a formal AD/ADRD diagnosis that coincides with increased avoidable healthcare utilization, indicating a need to examine the nuances between healthcare utilization and cognitive impairment more closely [10]. Prior research has established that individuals with AD/ADRD or MCI have higher avoidable healthcare utilization, but these studies have largely used data that has consisted of predominantly non-Hispanic White populations [11,12]. One study sought to address this limitation by examining a cohort of Mexican American Medicare beneficiaries with AD/ADRD and found that they had higher odds of one or more hospitalizations and ER admissions than beneficiaries without AD/ADRD [13]. Another study attempted to examine varying levels of cognitive decline (including MCI) as a predictor for avoidable healthcare utilization, but found no significant associations, and most participants were non-Hispanic White (96%) [14].
The purpose of this study was to examine the odds of overall healthcare utilization and avoidable healthcare utilization according to cognitive impairment and whether these associations vary by Latino ethnicity. Two hypotheses guide this study: 1) we hypothesize that participants with cognitive impairment will have higher odds of overall healthcare utilization and avoidable healthcare utilization, and 2) the association between avoidable healthcare utilization and cognitive impairment will be moderated by Latino ethnicity. In turn, these analyses may point to Latino ethnic groups who would benefit from targeted interventions regarding healthcare utilization.
Methods
Study Design
The Study of Latinos-Investigation of Neurocognitive Aging (SOL-INCA) is a Hispanic Community Health Study/Study of Latinos (HCHS-SOL) ancillary study. HCHS-SOL is a multisite, prospective cohort study of N=16,415 self-identified Latino participants at Visit 1 (2008–2011) that collected data from four field centers in California, Florida, Illinois, and New York [15]. SOL-INCA (n=6,377) leveraged the HCHS/SOL cohort for a follow-up (Visit 2; 2014–2017) examination to obtain a second time-point of cognitive testing [15]. Further detailed descriptions of HCHS/SOL and SOL-INCA study designs have been previously published [15,16]. This study did not meet the definition of Human Subjects Research (IRB-FY2024–8548) as determined by the New York University Institutional Review Board.
Analytic Sample
We included SOL-INCA participants aged 50 years and older who completed a baseline cognitive assessment at Visit 1 and follow-up at Visit 2, reported their ethnicity as Latino or Hispanic, and further self-identified as Mexican, Cuban, Puerto Rican, Dominican, Central American, South American, More than One Heritage, or Other, and consented to future use of their data. Of the total SOL-INCA participants (n=6,377), participants were excluded by the Biologic Specimen and Data Repository (BioLINCC), where dataset access was obtained, due to lack of informed consent and/or additional adjudications/data corrections (n=1,460). We excluded participants missing a reported specific Latino ethnicity (n=9) or cognitive impairment variables (n=83). The final analytic (unweighted) sample comprised 4,825 participants.
For our subgroup analysis, we applied the original inclusion criteria and further restricted the sample to participants who reported both a healthcare need and instances of avoidable healthcare utilization. Participants with missing data on the overall healthcare utilization (n=40) and those with missing responses regarding avoidable healthcare utilization (ER or hospitalizations) (n=629) were excluded. The final analytic (unweighted) sample for this subgroup analysis was 4,196 participants.
Primary Exposure: Cognitive Impairment vs No Cognitive Impairment
MCI diagnostic criteria were operationalized into three categories by SOL-INCA: non-MCI, MCI, and severe cognitive impairment aligned with the National Institute on Aging-Alzheimer’s Association criteria: 1) self-reported cognitive decline, 2) any cognitive score in the mildly impaired range, that is, from −1 to −2 standard deviations (SD) below the mean compared to the SOL-INCA internal robust norms (age, education, and sex-adjusted scores), 3) significant cognitive decline (≤–0.055 SD/year) from Visit 1, and 4) no or minimal instrumental activities of daily living impairment [15].
In this manuscript, we refer to participants with cognitive impairment as individuals with MCI and severe cognitive impairment (-2 SD relative to SOL-INCA robust norms and with significant functional impairment) and collapse them into this one category due to small cell sizes [15].
Primary Outcome: Overall and Avoidable Healthcare Utilization
We used healthcare utilization data obtained from self-reported responses to the Health Care Use Questionnaire at Visit 2 [16]. Overall healthcare utilization was defined as any healthcare received in the past 12 months and coded as a binary variable (yes, no). Examples of types of healthcare utilization respondents could select include preventive care and/or chronic care of a disease. Avoidable healthcare utilization was defined as participants who responded yes to the overall healthcare utilization question and further responded that the reason for seeking healthcare was ER or hospitalizations; this was coded as a binary variable (yes, no).
Interaction Term: Latino Ethnicity
We used self-reported Latino ethnicity at Visit 1. Participants were asked to describe their Latino heritage as either Mexican, Cuban, Puerto Rican, Dominican, Central American, South American, More than One Heritage, or Other. More than One Heritage and Other responses were collapsed into one category due to small cell sizes.
Covariates
We included covariates identified in the literature as potential confounders of the associations between cognitive decline, healthcare utilization, and Latino ethnicity. Acculturation factors included nativity, years of residence in the US, and language preference (English, Spanish). Chronic comorbidities that are known risk factors for cognitive decline and increased healthcare utilization, such as diabetes measured using self-report of diabetes, the use of antihyperglycemic medication, or using the American Diabetes Association laboratory criteria [17] and hypertension (yes, no) measured using the American College of Cardiology and American Heart Association definition, individuals met criteria for hypertension if systolic or diastolic blood pressure was greater or equal to 130/80 or if they self-reported currently taking antihypertensive medications [18,19]. Individual health behaviors, including current smoking status [20]. Additionally, we included sociodemographic factors, such as age (continuous), gender (male, female), education (< high school, high school, > high school), health insurance status, and marital status based on prior SOL-INCA studies [21,22].
Data Analysis
All analyses were conducted using R version 4.2.2. Sampling weights were applied in analyses to account for the complex sampling design of HCHS/SOL and SOL-INCA [15]. SOL-INCA calibrated age, gender, and Latino background to the 2010 US Census in each field center’s population. Therefore, all reported values, except for the sample size (which is reported as an unweighted value), were weighted to account for the sampling design, adjust for non-response bias, and allow generalizability [15]. A p-value <0.05 was considered statistically significant in all analyses.
First, we generated weighted descriptive statistics to characterize our study sample across seven disaggregated Latino ethnic groups (Cuban, Dominican, Mexican, Puerto Rican, Central American, South American, and More than One/Other) and cognitive impairment status. Second, we fit a series of weighted logistic regression models to examine the association between cognitive impairment status and both overall and avoidable healthcare utilization. For each healthcare utilization outcome, we tested 1) unadjusted and 2) fully adjusted (including all covariates as described above) models. Lastly, we added an interaction between cognitive impairment and Latino ethnicity for avoidable healthcare utilization to test whether Latino ethnicity moderates the association between cognitive impairment and avoidable healthcare utilization.
Results
Characteristics of Study Participants by Latino Ethnicity
Table 1 presents the sample characteristics and covariates of interest stratified by Latino ethnicity groups (Mexican, Dominican, Cuban, Puerto Rican, Central American, South American, and More Than One/Other). By overall Latino ethnicity, the majority were Female (52%), possessed education beyond the high school level (43%), and had a mean age of 63.46 years ± 8.16 standard deviations (SD). Most participants reported high rates of overall healthcare utilization (88%), as well as avoidable healthcare utilization (42%). By specific Latino ethnicity, Dominican participants (95%) reported higher overall healthcare utilization compared to the overall Latino sample (88%). Relative to other Latino ethnic groups, More than One/Other groups were more likely to report higher ER visits (38%), while individuals identifying as Puerto Ricans were more likely to report higher rates of hospitalizations (22%).
Table 1.
Weighted Descriptive Statistics by Latino Ethnicity for the Overall Sample
| Characteristic | Overall N = 4,825 |
Mexican N = 1,664 |
Dominican N = 437 |
Central American N = 481 |
Cuban N = 917 |
Puerto Rican N = 859 |
South American N = 345 |
More than One/Other N =142 |
|---|---|---|---|---|---|---|---|---|
| Cognitive Impairment | ||||||||
| No Cognitive Impairment | 4,503 (89%) | 1,328 (90%) | 399 (88%) | 319 (88%) | 1,302 (90%) | 712 (86%) | 239 (91%) | 205 (83%) |
| Cognitive Impairment | 570 (11%) | 145 (9.8%) | 52 (12%) | 45 (12%) | 150 (10%) | 111 (14%) | 24 (9.1%) | 43 (17%) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Overall Healthcare Utilization | 4,419 (88%) | 1,240 (85%) | 424 (95%) | 304 (85%) | 1,246 (87%) | 761 (93%) | 220 (84%) | 223 (91%) |
| Missing | 52 | 7 | 4 | 4 | 26 | 6 | 0 | 4 |
| Emergency Care | 1,192 (27%) | 269 (21%) | 138 (32%) | 74 (24%) | 320 (25%) | 274 (35%) | 56 (25%) | 86 (38%) |
| Missing | 661 | 232 | 27 | 61 | 208 | 62 | 45 | 26 |
| Hospitalizations | 670 (15%) | 187 (15%) | 66 (16%) | 36 (12%) | 167 (13%) | 168 (22%) | 20 (9.1%) | 25 (11%) |
| Missing | 667 | 233 | 27 | 60 | 210 | 67 | 42 | 26 |
| Education | ||||||||
| Less Than High School | 1,832 (36%) | 673 (46%) | 207 (47%) | 148 (41%) | 321 (22%) | 345 (42%) | 50 (19%) | 89 (36%) |
| High School or Equivalent | 1,084 (21%) | 305 (21%) | 84 (19%) | 72 (20%) | 364 (25%) | 178 (22%) | 52 (20%) | 29 (12%) |
| Greater Than High School | 2,157 (43%) | 494 (34%) | 161 (36%) | 144 (40%) | 767 (53%) | 299 (36%) | 160 (61%) | 131 (53%) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Age (years) | 63.46 (8.16) | 62.07 (7.63) | 62.38 (8.31) | 62.51 (7.46) | 65.00 (8.58) | 64.13 (8.00) | 62.97 (8.04) | 64.39 (8.32) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Income | ||||||||
| More Than $30,000 | 1,756 (36%) | 650 (45%) | 138 (32%) | 124 (35%) | 413 (29%) | 239 (30%) | 101 (40%) | 90 (38%) |
| Less Than $30,000 | 3,111 (64%) | 776 (54%) | 290 (68%) | 226 (65%) | 979 (70%) | 546 (70%) | 148 (60%) | 146 (62%) |
| Missing | 207 | 47 | 23 | 14 | 59 | 38 | 13 | 12 |
| Blood Pressure Status | ||||||||
| Normotensive | 1,515 (30%) | 599 (41%) | 89 (20%) | 127 (35%) | 332 (23%) | 209 (25%) | 101 (39%) | 57 (23%) |
| Hypertensive | 3,556 (70%) | 873 (59%) | 362 (80%) | 233 (65%) | 1,120 (77%) | 614 (75%) | 161 (61%) | 192 (77%) |
| Missing | 3 | 0 | 0 | 3 | 0 | 0 | 0 | 0 |
| Smoker Status | ||||||||
| Never/Former Smoker | 4,292 (85%) | 1,313 (89%) | 400 (89%) | 334 (92%) | 1,150 (79%) | 645 (78%) | 241 (92%) | 209 (84%) |
| Current Smoker | 779 (15%) | 159 (11%) | 51 (11%) | 30 (8.3%) | 301 (21%) | 177 (22%) | 22 (8.3%) | 40 (16%) |
| Missing | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 |
| Nativity | ||||||||
| Not Born in US or Territory | 4,024 (79%) | 1,324 (90%) | 447 (99%) | 360 (99%) | 1,431 (99%) | 9 (1.1%) | 258 (98%) | 195 (79%) |
| Born in US or Territory | 1,049 (21%) | 149 (10%) | 5 (1.1%) | 3 (0.9%) | 21 (1.5%) | 813 (99%) | 4 (1.6%) | 53 (21%) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Primary Language | ||||||||
| Spanish | 4,398 (87%) | 1,279 (87%) | 441 (98%) | 353 (97%) | 1,414 (97%) | 461 (56%) | 257 (98%) | 194 (78%) |
| English | 675 (13%) | 194 (13%) | 10 (2.2%) | 11 (3.0%) | 38 (2.6%) | 362 (44%) | 6 (2.1%) | 55 (22%) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Gender | ||||||||
| Female | 2,663 (52%) | 786 (53%) | 275 (61%) | 210 (58%) | 699 (49%) | 429 (52%) | 142 (54%) | 122 (49%) |
| Male | 2,441 (48%) | 686 (47%) | 177 (39%) | 153 (42%) | 753 (51%) | 394 (48%) | 120 (46%) | 127 (51%) |
| Missing | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Diabetes Status | ||||||||
| Does Not Have Diabetes | 3,769 (74%) | 1,120 (76%) | 323 (72%) | 271 (74%) | 1,108 (76%) | 543 (66%) | 223 (85%) | 181 (73%) |
| Has Diabetes | 1,303 (26%) | 352 (24%) | 128 (28%) | 93 (26%) | 343 (24%) | 280 (34%) | 39 (15%) | 68 (27%) |
| Missing | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 |
| Marital Status | ||||||||
| Single | 879 (17%) | 147 (10.0%) | 82 (18%) | 93 (26%) | 219 (15%) | 2650 (32%) | 31 (12%) | 47 (19%) |
| Married | 2,437 (48%) | 892 (61%) | 205 (45%) | 160 (44%) | 728 (50%) | 233 (28%) | 144 (55%) | 76 (31%) |
| Separated/Divorced | 1,123 (22%) | 258 (18%) | 111 (25%) | 75 (20%) | 337 (23%) | 194 (24%) | 70 (27%) | 78 (31%) |
| Widowed | 350 (6.9%) | 95 (6.4%) | 30 (6.6%) | 18 (5.0%) | 101 (7.0%) | 67 (8.1%) | 14 (5.3%) | 25 (10.0%) |
| Living with a Partner | 279 (5.5%) | 80 (5.4%) | 24 (5.2%) | 18 (4.8%) | 64 (4.4%) | 69 (8.4%) | 3 (1.3%) | 22 (8.7%) |
| Missing | 5 | 1 | 0 | 0 | 2 | 1 | 0 | 1 |
| Years in the US | ||||||||
| <10 Years in US | 549 (11%) | 87 (5.9%) | 42 (9.4%) | 23 (6.3%) | 346 (24%) | 4 (0.5%) | 21 (8.0%) | 25 (10%) |
| >10 Years in US | 4,514 (89%) | 1,384 (94%) | 405 (91%) | 338 (94%) | 1,103 (76%) | 819 (99.5%) | 241 (92%) | 223 (90%) |
| Missing | 11 | 1 | 4 | 3 | 3 | 0 | 0 | 0 |
Note. Please note that the sample size is unweighted, all other reported values are weighted to represent the targeted population; n (%); Mean (SD)
Characteristics of Study Participants by Cognitive Impairment Status
We observed significant differences in sociodemographic, chronic conditions, and behavioral characteristics between participants who were identified as having cognitive impairment and those who did not. As presented in Table 2, bivariate analyses revealed that, compared to participants in the no cognitive impairment group, those identified as having cognitive impairment were significantly more likely to exhibit higher rates of avoidable healthcare utilization (41% vs 30%). Additionally, they were more likely to be older, female, single, insured, Spanish-speaking, have less than a high school education, report an annual income below $30,000, and have a history of hypertension or diabetes.
Table 2.
Weighted Descriptive Statistics by Cognitive Impairment for the Subgroup Analysis of Participants with Avoidable Healthcare Utilization
| Variables | Overall N = 4,196 |
No Cognitive Impairment N = 3,724 |
Cognitive Impairment N = 472 |
P a |
|---|---|---|---|---|
| Latino Ethnicity | 0.32 | |||
| Dominican | 424 (9.6%) | 375 (9.7%) | 49 (9.3%) | |
| Central American | 303 (6.9%) | 262 (6.8%) | 40 (7.7%) | |
| Cuban | 1,242 (28%) | 1,112 (29%) | 130 (25%) | |
| Mexican | 1,239 (28%) | 1,101 (28%) | 138 (26%) | |
| Puerto Rican | 755 (17%) | 649 (17%) | 106 (20%) | |
| South American | 217 (4.9%) | 196 (5.0%) | 22 (4.1%) | |
| More than One/Other | 223 (5.1%) | 183 (4.7%) | 39 (7.5%) | |
| Missing | 0 | 0 | 0 | |
| Avoidable Healthcare Utilization | <0.001 | |||
| No Avoidable Healthcare Utilization | 3,026 (69%) | 2,717 (70%) | 309 (59%) | |
| Avoidable Healthcare Utilization | 1,377 (31%) | 1,162 (30%) | 215 (41%) | |
| Missing | 0 | 0 | 0 | |
| Health Insurance Status | <0.001 | |||
| No Health Insurance | 456 (10%) | 426 (11%) | 30 (5.7%) | |
| Currently Have Health Insurance | 3,937 (90%) | 3,444 (89%) | 493 (94%) | |
| Missing | 10 | 9 | 1 | |
| Education | <0.001 | |||
| Less Than High School | 1,594 (36%) | 1,349 (35%) | 245 (47%) | |
| High School or Equivalent | 944 (21%) | 849 (22%) | 94 (18%) | |
| Greater Than High School | 1,865 (42%) | 1,681 (43%) | 184 (35%) | |
| Missing | 0 | 0 | 0 | |
| Age (years) | 63.96 (8.18) | 63.56 (8.03) | 66.98 (8.67) | <0.001 |
| Missing | 0 | 0 | 0 | |
| Income | <0.001 | |||
| Less than $30,000 | 2,723 (64%) | 2,332 (62%) | 391 (79%) | |
| More Than $30,000 | 1,518 (36%) | 1,413 (38%) | 105 (21%) | |
| Missing | 162 | 134 | 28 | |
| Blood Pressure Status | <0.001 | |||
| Normotensive | 1,208 (27%) | 1,120 (29%) | 88 (17%) | |
| Hypertensive | 3,191 (73%) | 2,756 (71%) | 436 (83%) | |
| Missing | 3 | 3 | 0 | |
| Smoker Status | 0.83 | |||
| Never/Former Smoker | 3,774 (86%) | 3,327 (86%) | 447 (85%) | |
| Current Smoker | 628 (14%) | 551 (14%) | 77 (15%) | |
| Missing | 1 | 1 | 0 | |
| Nativity | 0.20 | |||
| Not Born in US or Territory | 3,452 (78%) | 3,058 (79%) | 394 (75%) | |
| Born in US or Territory | 951 (22%) | 821 (21%) | 129 (25%) | |
| Missing | 0 | 0 | 0 | |
| Primary Language | 0.04 | |||
| Spanish | 3,801 (86%) | 3,325 (86%) | 476 (91%) | |
| English | 601 (14%) | 554 (14%) | 48 (9.1%) | |
| Missing | 0 | 0 | 0 | |
| Gender | 0.22 | |||
| Female | 2,401 (55%) | 2,096 (54%) | 305 (58%) | |
| Male | 2,001 (45%) | 1,783 (46%) | 219 (42%) | |
| Missing | 0 | 0 | 0 | |
| Diabetes Status | <0.001 | |||
| Does Not Have Diabetes | 3,189 (72%) | 2,891 (75%) | 297 (57%) | |
| Has Diabetes | 1,214 (28%) | 987 (25%) | 226 (43%) | |
| Missing | 0 | 0 | 0 | |
| Marital Status | 0.94 | |||
| Single | 744 (17%) | 655 (17%) | 89 (17%) | |
| Married | 2,117 (48%) | 1,874 (48%) | 243 (46%) | |
| Separated/Divorced | 982 (22%) | 865 (22%) | 118 (23%) | |
| Widowed | 326 (7.4%) | 281 (7.2%) | 45 (8.6%) | |
| Living with a Partner | 228 (5.2%) | 199 (5.1%) | 29 (5.5%) | |
| Missing | 5 | 5 | 0 | |
| Years in the US | 0.69 | |||
| <10 Years in US | 475 (11%) | 423 (11%) | 53 (10%) | |
| >10 Years in US | 3,919 (89%) | 3,449 (89%) | 470 (90%) | |
| Missing | 8 | 7 | 1 |
Note. Please note that the sample size is unweighted, all other reported values are weighted to represent the targeted population; See inclusion criteria for further details. Pa = Comparisons and p-values generated using survey-weighted chi-squared tests for categorical and t-tests for continuous variables; US = United States; n / N (%); Mean (SD)
Overall Healthcare Utilization and Cognitive Impairment
In our unadjusted Model 1, we observed a statistically significant association between overall healthcare utilization and cognitive impairment, as participants with cognitive impairment were three times more likely to report healthcare utilization (OR=3.10, p= <0.001, 95% CI [1.89, 5.08]) (see Table 3). After fully adjusting in Model 2, this association remained significant as participants with cognitive impairment were twice as likely to report healthcare utilization (OR=1.94, p=0.013, 95% CI [1.15, 3.27]).
Table 3.
Crude and Adjusted Models for Overall Healthcare Utilization and Cognitive Impairment
| Model 1 | Model 2 | |||||
|---|---|---|---|---|---|---|
| Variable | OR | 95% CI | p-value | OR | 95% CI | p-value |
| Cognitive Impairment | ||||||
| No Cognitive Impairment | — | — | — | — | ||
| Cognitive Impairment | 3.10 | 1.89, 5.08 | <0.001 | 1.94 | 1.15,3.27 | 0.013 |
| Latino Ethnicity | ||||||
| Mexican | — | — | ||||
| Dominican | 2.59 | 1.35,4.95 | 0.004 | |||
| Central American | 1.17 | 0.74,1.85 | 0.5 | |||
| Cuban | 1.12 | 0.76,1.66 | 0.6 | |||
| Puerto Rican | 1.61 | 0.80,3.27 | 0.2 | |||
| South American | 1.01 | 0.59,1.74 | >0.9 | |||
| More than One/Other | 1.54 | 0.71,3.35 | 0.3 | |||
| Age (years) | 1.06 | 1.03,1.08 | <0.001 | |||
| Language Preference | ||||||
| English | — | — | ||||
| Spanish | 0.89 | 0.51,1.53 | 0.7 | |||
| Blood Pressure Status | ||||||
| Normotensive | — | — | ||||
| Hypertensive | 1.58 | 1.18,2.13 | 0.002 | |||
| Education | ||||||
| Less than High School | — | — | ||||
| High School or Equivalent | 1.44 | 0.97,2.13 | 0.067 | |||
| Greater than High School | 1.00 | 0.73,1.38 | >0.9 | |||
| Current Smoker Status | ||||||
| Never/Former Smoker | — | — | ||||
| Current Smoker | 0.77 | 0.55,1.08 | 0.13 | |||
| Income | ||||||
| More than $30,000 | — | — | — | |||
| Less than $30,000 | 0.96 | 0.73,1.26 | 0.7 | |||
| Nativity | ||||||
| Born in US or Territory | — | — | ||||
| Not Born in US or Territory | 0.97 | 0.50,1.89 | >0.9 | |||
| Diabetes Status | ||||||
| Does Not Have Diabetes | — | — | ||||
| Has Diabetes | 2.78 | 1.55,5.01 | <0.001 | |||
| Gender | ||||||
| Female | — | — | — | |||
| Male | 0.50 | 0.38,0.66 | <0.001 | |||
| Marital Status | ||||||
| Married | — | — | — | |||
| Single | 0.83 | 0.57,1.19 | 0.3 | |||
| Separated/Divorced | 0.91 | 0.64,1.29 | 0.6 | |||
| Widowed | 1.31 | 0.60,2.85 | 0.5 | |||
| Living with a Partner | 0.76 | 0.47,1.22 | 0.3 | |||
| Years in the US | ||||||
| >10 Years in US | — | — | — | |||
| <10 Years in US | 1.11 | 0.66,1.87 | 0.7 | |||
| Health Insurance Status | ||||||
| Currently Have Health Insurance | — | — | — | |||
| No Current Health Insurance | 4.35 | 3.19,5.93 | <0.001 | |||
Note. — = Reference Category; CI = Confidence Intervals; MCI = Mild Cognitive Impairment; OR = Odds Ratio; US = United States
Associations Between Avoidable Healthcare Utilization and Cognitive Impairment
We analyzed the subpopulation of participants who reported overall healthcare utilization, and the reason for seeking healthcare was avoidable healthcare utilization (ER or hospitalizations). In our crude Model 1, we observed a statistical association between avoidable healthcare utilization and cognitive impairment (OR=1.63, p= <0.001, 95% CI [1.24, 2.13]) (see Table 4). This association remained significant in our fully adjusted Model 2 (OR=1.51, p=0.004, 95% CI [1.14, 1.98]).
Table 4.
Crude and Adjusted Models for Avoidable Healthcare Utilization and Cognitive Impairment
| Model 1 | Model 2 | |||||
|---|---|---|---|---|---|---|
| Variable | OR | 95% CI | p-value | OR | 95% CI | p-value |
| Cognitive Impairment | ||||||
| No Cognitive Impairment | — | — | — | — | ||
| Cognitive Impairment | 1.63 | 1.24, 2.13 | <0.001 | 1.51 | 1.14,1.98 | 0.004 |
| Latino Ethnicity | ||||||
| Mexican | — | — | ||||
| Dominican | 1.27 | 0.90,1.80 | 0.2 | |||
| Central American | 0.97 | 0.66,1.43 | 0.9 | |||
| Cuban | 0.96 | 0.71, 1.30 | 0.8 | |||
| Puerto Rican | 1.24 | 0.76,2.01 | 0.4 | |||
| South American | 0.86 | 0.55,1.34 | 0.5 | |||
| More than One/Other | 1.34 | 0.81,2.23 | 0.3 | |||
| Age (years) | 1.00 | 0.98,1.01 | 0.9 | |||
| Language Preference | ||||||
| English | — | — | — | |||
| Spanish | 0.88 | 0.63,1.23 | 0.5 | |||
| Blood Pressure Status | ||||||
| Normotensive | — | — | — | |||
| Hypertensive | 1.08 | 0.86,1.35 | 0.5 | |||
| Education | ||||||
| Greater than High School | — | — | — | |||
| Less Than High School | 0.95 | 0.75,1.20 | 0.7 | |||
| High School or Equivalent | 0.99 | 0.76,1.29 | >0.9 | |||
| Current Smoker Status | ||||||
| Never/Former Smoker | — | — | — | |||
| Current Smoker | 1.29 | 0.99,1.70 | 0.063 | |||
| Income | ||||||
| More than $30,000 | — | — | — | |||
| Less than $30,000 | 1.07 | 0.86,1.33 | 0.5 | |||
| Nativity | ||||||
| Born in US or Territory | — | — | — | |||
| Not Born in US or Territory | 0.84 | 0.51, 1.39 | 0.5 | |||
| Diabetes Status | ||||||
| Does Not Have Diabetes | — | — | — | |||
| Has Diabetes | 1.38 | 1.10,1.74 | 0.006 | |||
| Gender | ||||||
| Female | — | — | — | |||
| Male | 0.83 | 0.69, 1.01 | 0.060 | |||
| Marital Status | ||||||
| Married | — | — | — | |||
| Single | 1.44 | 1.09,1.92 | 0.011 | |||
| Separated/Divorced | 1.18 | 0.91,1.54 | 0.2 | |||
| Widowed | 1.37 | 0.89,2.10 | 0.2 | |||
| Living with a Partner | 1.14 | 0.75,1.75 | 0.5 | |||
| Years in the US | ||||||
| >10 Years in US | — | — | — | |||
| <10 Years in US | 1.14 | 0.75,1.72 | 0.5 | |||
| Health Insurance Status | ||||||
| Currently Have Health Insurance | — | — | — | |||
| No Current Health Insurance | 0.83 | 0.60,1.16 | 0.3 | |||
Note. — = Reference Category; CI = Confidence Intervals; MCI = Mild Cognitive Impairment; OR = Odds Ratio; US = United States
Moderation Analysis
Given the statistically significant associations observed in the main effects between avoidable healthcare utilization and cognitive impairment, we proceeded to examine the moderating effects of Latino ethnicity. Although the overall interaction between Latino ethnicity and cognitive impairment was not statistically significant (p=0.33), the objective of this study was to report specific estimates across Latino ethnic groups. Thus, we proceeded with the examination of these interactions. Mexican participants were chosen as the reference group for practical considerations, including their status as the largest ethnic group in our sample size [23].
By examining the interactions directly, we visualize significant differences between certain Latino ethnic groups. This suggests that the effect of cognitive impairment on avoidable healthcare utilization differs by certain reported types of Latino ethnicity in this subpopulation. Specifically, our analyses indicate that only one ethnic group, Puerto Ricans with cognitive impairment (OR=2.51, p=0.012, 95% CI [1.23, 5.15]), were more likely to have avoidable healthcare utilization than Mexican respondents without cognitive impairment, which remained significant after controlling for potential confounders (OR=2.61, p =0.014, 95% CI [1.22, 5.59]) (see Table 5). Moreover, although not statistically significant, Central Americans’ effect sizes were high (OR = 2.24, p-value = 0.093, 95% CI = 0.87, 5.76). Figure 1 displays the predicted probabilities of participants reporting avoidable healthcare utilization by Latino ethnic group and cognitive impairment status estimated from the logistic model.
Table 5.
Interaction Between Latino Ethnicity and Cognitive Impairment on Avoidable Healthcare Utilization
| Model 1 | Model 2 | |||||
|---|---|---|---|---|---|---|
| Variable | OR | 95% CI | p-value | OR | 95% CI | p-value |
| Cognitive Impairment | ||||||
| No Cognitive Impairment | — | — | — | — | — | — |
| Cognitive Impairment | 1.03 | 0.61,1.74 | >0.9 | 0.95 | 0.56,1.62 | 0.9 |
| Latino Ethnicity | ||||||
| Mexican | — | — | — | — | — | — |
| Dominican | 1.24 | 0.89,1.73 | 0.2 | 1.16 | 0.81,1.66 | 0.4 |
| Central American | 0.85 | 0.57,1.27 | 0.4 | 0.87 | 0.58,1.31 | 0.5 |
| Cuban | 1.01 | 0.76,1,.34 | >0.9 | 0.93 | 0.68,1.27 | 0.6 |
| Puerto Rican | 1.63 | 1.24,2.15 | <0.001 | 1.11 | 0.68,1.80 | 0.7 |
| South American | 0.82 | 0.52,1.28 | 0.4 | 0.79 | 0.49,1.28 | 0.3 |
| More than One/Other | 1.53 | 0.88,2.63 | 0.13 | 1.29 | 0.74,2.24 | 0.4 |
| MCI * Latino Ethnicity | ||||||
| Cognitive Impairment * Dominican | 1.87 | 0.70,4.97 | 0.2 | 2.12 | 0.80,5.66 | 0.13 |
| Cognitive Impairment *Central American | 2.33 | 0.88,6.21 | 0.090 | 2.24 | 0.87,5.76 | 0.093 |
| Cognitive Impairment * Cuban | 1.35 | 0.62,2.94 | 0.4 | 1.31 | 0.60,2.85 | 0.5 |
| Cognitive Impairment * Puerto Rican | 2.51 | 1.23,5.15 | 0.012 | 2.61 | 1.22,5.59 | 0.014 |
| Cognitive Impairment * South American | 1.74 | 0.55,5.52 | 0.3 | 2.09 | 0.59,7.30 | 0.2 |
| Cognitive Impairment * More Than One/Other | 1.21 | 0.30,4.84 | 0.8 | 1.39 | 0.29,7.35 | 0.7 |
| Age (years) | 1.00 | 0.98,1.01 | >0.9 | |||
| Language Preference | ||||||
| English | — | — | — | |||
| Spanish | 0.85 | 0.61,1.18 | 0.3 | |||
| Blood Pressure Status | ||||||
| Normotensive | — | — | — | |||
| Hypertensive | 1.08 | 0.86,1.36 | 0.5 | |||
| Education | ||||||
| Greater Than High School | — | — | — | |||
| Less Than High School | 0.95 | 0.75,1.20 | 0.6 | |||
| High School or Equivalent | 0.99 | 0.76,1.29 | >0.9 | |||
| Smoking Status | ||||||
| Never/Former Smoker | — | — | — | |||
| Current Smoker | 1.29 | 0.98,1.70 | 0.065 | |||
| Income | ||||||
| More Than $30,000 | — | — | — | |||
| Less than $30,000 | 1.07 | 0.86,1.33 | 0.5 | |||
| Nativity | ||||||
| Born in US or Territory | — | — | — | |||
| Not Born in US or Territory | 0.87 | 0.52,1.43 | 0.6 | |||
| Diabetes Status | ||||||
| Does Not Have Diabetes | — | — | — | |||
| Has Diabetes | 1.38 | 1.10,1.74 | 0.006 | |||
| Gender | ||||||
| Female | — | — | — | |||
| Male | 0.84 | 0.69,1.01 | 0.066 | |||
| Marital Status | ||||||
| Married | — | — | — | |||
| Single | 1.45 | 1.09,1.92 | 0.010 | |||
| Separated/Divorced | 1.18 | 0.91,1.55 | 0.2 | |||
| Widowed | 1.37 | 0.89,2.11 | 0.2 | |||
| Living with a Partner | 1.14 | 0.74,1.75 | 0.6 | |||
| Years in the US | ||||||
| >10 Years in US | — | — | — | |||
| <10 Years in US | 1.14 | 0.76,1.73 | 0.5 | |||
| Health Insurance Status | ||||||
| Currently Have Health Insurance | — | — | — | |||
| No Current Health Insurance | 0.82 | 0.59,1.15 | 0.3 | |||
Abbreviations: CI = Confidence Interval, OR = Odds Ratio
Figure 1. Interaction Between Cognitive Impairment and Latino Ethnicity for Avoidable Healthcare Utilization Plot.
Note. Error bars represent 95% confidence intervals.
Discussion
This cross-sectional study identified increased odds of both overall and avoidable healthcare utilization among Latinos with cognitive impairment. Additionally, a significant interaction indicated that patterns of association between cognitive impairment and avoidable healthcare utilization varied across specific Latino ethnicities. These findings underscore the importance of disaggregating Latino ethnic data to identify heterogeneity in healthcare experiences and outcomes.
While our study did not examine direct measures of access to preventive care, prior research has cited disparities in accessing preventive care as drivers of avoidable healthcare utilization among Latino adults [24,25]. Therefore, while it is plausible that limited preventive care use contributes to higher avoidable healthcare utilization, this interpretation should be considered hypothesis-generating rather than causal. Future research should consider using longitudinal or claims-based data to test whether preventive care use mediates the relationship between cognitive impairment and avoidable healthcare utilization among Latino adults.
Expanding preventive care represents a cost-effective strategy by reducing avoidable healthcare utilization and associated expenditures. Prior research has found that the removal of health insurance mandates increased the probabilities of higher ER visits and delaying care due to costs in Latino populations, which has important implications for expanding Medicaid and bolstering the healthcare safety net to provide community-based services [26]. However, Latino adults remain disproportionately underrepresented in programs designed to promote community-based care access, including Medicaid-funded initiatives such as the Program of All-Inclusive Care for the Elderly (PACE) and the newly introduced Guiding an Improved Dementia Experience (GUIDE) Model [27,28]. The PACE and GUIDE Models may be less accessible for diverse Latino communities, particularly for individuals with non-English language preference and non-citizen status, underscoring the critical need for inclusive models of care [29]. Future research is needed to evaluate whether these programs are effectively reaching and accessible to diverse Latino populations.
The elevated odds of avoidable healthcare utilization among Puerto Rican participants warrant nuanced interpretation. Our findings align with prior research that Puerto Rican participants exhibit higher rates of ER utilization compared to other Latino ethnic groups, which only became evident through data disaggregation [30]. In addition, the literature suggests that more acculturated Latinos may be more likely to pursue avoidable healthcare utilization at rates almost identical to non-Hispanic Whites [31]. However, Puerto Ricans are U.S citizens by birth and thus may experience different structural and legal contexts than other Latino groups [32]. Attributing our findings of higher avoidable healthcare utilization in Puerto Rican participants solely to acculturation is inappropriate. Alternative explanations may include place-based factors (e.g., resource-limited communities) [33] or differences in healthcare-seeking behaviors shaped by local health system characteristics across Latino ethnic groups [34]. Future research should examine in more detail key social and structural determinants such as neighborhood disadvantage, health insurance coverage continuity, or healthcare provider density that may account for these ethnic-specific patterns observed in our study.
Overall, this study highlights a gap in the literature regarding increased overall and avoidable healthcare utilization among disaggregated Latino groups with cognitive impairment. More awareness and training are needed within healthcare systems about intra-ethnic differences among Latino populations and how that may influence healthcare utilization. With a better understanding of the needs of diverse Latino populations, health systems may be better able to deliver more efficient and effective care.
Limitations and Future Directions
Several limitations must be considered. First, the present study is exploratory; the cross-sectional design of this quantitative study precludes the ability to make inferences regarding causal relationships between healthcare utilization and cognitive impairment. Second, due to the privacy and data protection protocols in place in the SOL-INCA dataset, we are unable to assess or adjust for citizenship status in our statistical analysis. Future studies that integrate immigration and citizenship data may illuminate how policy exclusions and legal precarity shape healthcare utilization among different Latino ethnic groups. In addition, data regarding the specific HCHS/SOL field centers from which participants were recruited were not available. Consequently, we were unable to account for potential statistical dependency arising from correlated outcomes among respondents from the same center. Lastly, although the SOL-INCA cohort is relatively large (n=6,377), a smaller subset of SOL-INCA participant who did not give consent for future analyses were excluded by BioLINCC [15]. As a result, the analytic sample was reduced (n=4,825), leading to smaller cell sizes, particularly following disaggregation by Latino ethnicity and cognitive impairment status. The resulting estimates were accompanied by wide confidence intervals, which reflect a degree of uncertainty that should be considered when interpreting the results. Future large secondary datasets that collect and provide disaggregated data are essential to adequately power analyses involving data disaggregation. The SOL-INCA dataset is establishing a foundation for conducting such large-scale studies.
Conclusions
Our findings indicate that Latino individuals with cognitive impairment experience significantly higher odds of both overall and avoidable healthcare utilization. Moreover, we observed evidence suggesting that Latino ethnicity may moderate the association between avoidable healthcare utilization and cognitive impairment. These results underscore the critical importance of disaggregating health data within Latino populations, as heterogeneous ethnic subgroups exhibit distinct patterns of healthcare utilization and may require culturally and contextually tailored interventions. Consequently, targeted strategies aimed at reducing avoidable healthcare utilization among Latino ethnic groups may contribute to decreased healthcare expenditures and enhanced access to timely, culturally appropriate care for this rapidly growing demographic in the US.
Acknowledgments
The authors thank the staff and participants of HCHS/SOL and SOL-INCA for their important contributions.
Funding
This study is partially supported by the National Institutes of Health (NIH)/National Institute of Minority Health and Health Disparities (NIMHD) U54MD000538 and NIH/NIMHD R01MD018204. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Footnotes
Competing Interests: The authors have no relevant financial or non-financial interests to disclose.
Conflicts of Interest
The authors have no conflicts of interest to disclose.
Ethics Approval: The New York University Institutional Review Board determined this study as Not Human Subjects Research (IRB-FY2024-8548).
References
- 1.Matthews KA, Xu W, Gaglioti AH, Holt JB, Croft JB, Mack D, & McGuire LC (2019). Racial and ethnic estimates of Alzheimer’s disease and related dementias in the United States (2015–2060) in adults aged ≥65 years. Alzheimer’s & Dementia, 15(1), 17–24. 10.1016/j.jalz.2018.06.3063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chung S-D, Liu S-P, Sheu J-J, Lin C-C, Lin H-C, & Chen C-H (2014). Increased Healthcare Service Utilizations for Patients with Dementia: A Population-Based Study. PLoS ONE, 9(8), e105789. 10.1371/journal.pone.0105789 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Anderson TS, Marcantonio ER, McCarthy EP, & Herzig SJ (2020). National Trends in Potentially Preventable Hospitalizations of Older Adults with Dementia. Journal of the American Geriatrics Society, 68(10), 2240. 10.1111/jgs.16636 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.McDermott KW, & Jiang J (2020). Characteristics and costs of potentially preventable inpatient stays, 2017. HCUP Statistical Brief #259. Agency for Healthcare Research and Quality. https://hcup-us.ahrq.gov/reports/statbriefs/sb259-Potentially-Preventable-Hospitalizations-2017.pdf. [PubMed] [Google Scholar]
- 5.Hunt LJ, Coombs LA, & Stephens CE (2018). Emergency Department Use by Community-Dwelling Individuals With Dementia in the United States. Journal of Gerontological Nursing, 44(3), 23–30. 10.3928/00989134-20171206-01 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rahemi Z, Bacsu J-DR, Shalhout SZ, Sadafipoor MS, Smith ML, & Adams SA (2025). Exploring social determinants of healthcare and cognition levels among diverse older adults. Geriatric Nursing, 61, 614–621. 10.1016/j.gerinurse.2024.12.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fernandez Cajavilca M, Lee M, & Ðoàn LN (2025). Reviewing and Advocating for Data Disaggregation in Aging-Related Health Research: NIA Grants and Recommendations for Equity. The Gerontologist, gnaf013. 10.1093/geront/gnaf013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chin MK, Đoàn LN, Russo RG, Roberts T, Persaud S, Huang E, Fu L, Kui KY, Kwon SC, & Yi SS (2023a). Methods for retrospectively improving race/ethnicity data quality: A scoping review. Epidemiologic Reviews, 45(1), 127–139. 10.1093/epirev/mxad002 [DOI] [PubMed] [Google Scholar]
- 9.Kumar RG, Lund EB, Ornstein KA, Li J, Covinsky KE, & Kelley AS (2023). Health care utilization and costs in the years preceding dementia identification. Alzheimer’s & Dementia, 19(12), 5852–5859. 10.1002/alz.13476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhu CW, Cosentino S, Ornstein K, Gu Y, Scarmeas N, Andrews H, & Stern Y (2015). Medicare Utilization and Expenditures Around Incident Dementia in a Multiethnic Cohort. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 70(11), 1448–1453. 10.1093/gerona/glv124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Callahan KE, Lovato JF, Miller ME, Easterling D, Snitz B, & Williamson JD (2015). Associations Between Mild Cognitive Impairment and Hospitalization and Readmission. Journal of the American Geriatrics Society, 63(9), 1880–1885. 10.1111/jgs.13593 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Lin P-J, Fillit HM, Cohen JT, & Neumann PJ (2013). Potentially avoidable hospitalizations among Medicare beneficiaries with Alzheimer’s disease and related disorders. Alzheimer’s & Dementia, 9(1), 30–38. 10.1016/j.jalz.2012.11.002 [DOI] [PubMed] [Google Scholar]
- 13.Downer B, Al Snih S, Raji M, Chou L, Kuo Y, Markides K, & Ottenbacher K (2020). Healthcare utilization of Mexican-American Medicare beneficiaries with and without Alzheimer’s disease and related dementias. PLOS ONE, 15(1). 10.1371/journal.pone.0227681 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Green RK, Shah MN, Clark LR, Batt RJ, Chin NA, & Patterson BW (2022). Comparing emergency department use among individuals with varying levels of cognitive impairment. BMC Geriatrics, 22, 382. 10.1186/s12877-022-03093-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gonzalez H, Tarraf W, Fornage M, Gonzalez K, Chai A, Youngblood M, Abreu M, Zeng D, Thomas S, Talavera G, Gallo L, Kaplan R, Daviglus M, & Schneiderman N (2019). A research framework for cognitive aging and Alzheimer’s disease among diverse US Latinos: Design and implementation of the Hispanic Community Health Study/Study of Latinos-Investigation of Neurocognitive Aging (SOL-INCA). ALZHEIMERS & DEMENTIA, 15(12), 1624–1632. 10.1016/j.jalz.2019.08.192 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sorlie PD, Avilés-Santa LM, Wassertheil-Smoller S, Kaplan RC, Daviglus ML, Giachello AL, Schneiderman N, Raij L, Talavera G, Allison M, LaVange L, Chambless LE, & Heiss G (2010). Design and Implementation of the Hispanic Community Health Study/Study of Latinos. Annals of Epidemiology, 20(8), 629–641. 10.1016/j.annepidem.2010.03.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.American Diabetes Association. (2024). Diagnosis and Classification of Diabetes. Standards of Care in Diabetes-2024. Diabetes Care, 47(Suppl. 1). 10.2337/dc24-S002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gomez S, Blumer V, & Rodriguez F (2022). Unique Cardiovascular Disease Risk Factors in Hispanic Individuals. Current Cardiovascular Risk Reports, 16(7), 53–61. 10.1007/s12170-022-00692-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Whelton PK, Carey RM, Aronow WS, Casey DE, Collins KJ, Dennison Himmelfarb C, DePalma SM, Gidding S, Jamerson KA, Jones DW, MacLaughlin EJ, Muntner P, Ovbiagele B, Smith SC, Spencer CC, Stafford RS, Taler SJ, Thomas RJ, Williams KA, … Wright JT (2018). 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APHA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension, 71(6), e13–e115. 10.1161/HYP.0000000000000065 [DOI] [PubMed] [Google Scholar]
- 20.Zhang Q, Zhang M, Chen Y, Zhu S, Zhou W, Zhang L, Dong G, & Cao Y (2022). Smoking Status and Cognitive Function in a National Sample of Older Adults. Frontiers in Psychiatry, 13, 926708. 10.3389/fpsyt.2022.926708 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Estrella ML, Tarraf W, Kuwayama S, Gallo LC, Wu B, Marquine MJ, Perreira KM, Vasquez PM, Isasi CR, Lipton RB, Mattei J, González HM, Daviglus ML, & Lamar M (2024). Psychosocial factors associated with 7-year change in cognition among middle-aged and older Hispanics/Latinos: The Hispanic Community Health Study/Study of Latinos-Investigation of Neurocognitive Aging (SOL-INCA) and Sociocultural ancillary studies. Alzheimer’s & Dementia, 20(2), 1137–1148. 10.1002/alz.13527 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Márquez F, Tarraf W, Stickel AM, González KA, Testai FD, Cai J, Gallo LC, Talavera GA, Daviglus ML, Wassertheil-Smoller S, DeCarli C, Schneiderman N, & González HM (2024). Hypertension, Cognitive Decline, and Mild Cognitive Impairment Among Diverse Hispanics/Latinos: Study of Latinos-Investigation of Neurocognitive Aging Results (SOL-INCA). Journal of Alzheimer’s Disease, 97(3), 1449–1461. 10.3233/JAD-230424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Johfre SS, & Freese J (2021). Reconsidering the reference category. Sociological Methodology, 51(2), 253–269. 10.1177/0081175020982632. [DOI] [Google Scholar]
- 24.Cobb S, Bazargan M, Assari S, Barkley L, & Bazargan-Hejazi S (2023). Emergency Department Utilization, Hospital Admissions, and Office-Based Physician Visits Among Under-Resourced African American and Latino Older Adults. Journal of Racial and Ethnic Health Disparities, 10(1), 205–218. 10.1007/s40615-021-01211-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Parast L, Mathews M, Martino S, Lehrman WG, Stark D, & Elliott MN (2022). Racial/Ethnic Differences in Emergency Department Utilization and Experience. Journal of General Internal Medicine, 37(1), 49–56. 10.1007/s11606-021-06738-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ortega AN, Chen J, Roby DH, Mortensen K, Rivera-González AC, & Bustamante AV (2022). Changes in Coverage and Cost-Related Delays in Care for Latino Individuals After Elimination of the Affordable Care Act’s Individual Mandate. JAMA Network Open, 5(3), e221476. 10.1001/jamanetworkopen.2022.1476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Eng C, Pedulla J, Eleazer GP, McCann R, & Fox N (1997). Program of All-inclusive Care for the Elderly (PACE): An Innovative Model of Integrated Geriatric Care and Financing. Journal of the American Geriatrics Society, 45(2), 223–232. 10.1111/j.1532-5415.1997.tb04513.x [DOI] [PubMed] [Google Scholar]
- 28.Haggerty KL, Epstein-Lubow G, Stoeckle RJ, Carmody J, Maglich A, Johnson M, Super N, Spragens L, Lee DR, & Snyder RE (2023). Applying An Evidence-Based Approach To Comprehensive Dementia Care Under The New GUIDE Model. Health Affairs Forefront. 10.1377/forefront.20231120.826884 [DOI] [Google Scholar]
- 29.Travers JL, D’Arpino S, Bradway C, Kim SJ, & Naylor MD (2022). Minority Older Adults’ Access to and Use of Programs of All-Inclusive Care for the Elderly. Journal of Aging & Social Policy, 34(6), 976–1002. 10.1080/08959420.2021.2024411 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Graves TD, Garcia MT, & Lee W-C (2025). Cross-Sectional Study of Emergency Room Visits Among Hispanic Immigrants by Country of Origin. Journal of Racial and Ethnic Health Disparities. 10.1007/s40615-025-02449-y [DOI] [PubMed] [Google Scholar]
- 31.Allen L, & Cummings J (2016). Emergency Department Use Among Hispanic Adults: The Role of Acculturation. Medical Care, 54(5), 449–456. 10.1097/MLR.0000000000000511 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Cohn D, Patten E, & Lopez MH (2014). Chapter 1: Puerto Ricans on the U.S. Mainland. https://www.pewresearch.org/race-and-ethnicity/2014/08/11/chapter-1-puerto-ricans-on-the-u-s-mainland/. Accessed May 27th, 2025.
- 33.Caraballo C, Ndumele CD, Roy B, Lu Y, Riley C, Herrin J, & Krumholz HM (2022). Trends in Racial and Ethnic Disparities in Barriers to Timely Medical Care Among Adults in the US, 1999 to 2018. JAMA Health Forum, 3(10), e223856. 10.1001/jamahealthforum.2022.3856 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Vargas Bustamante A, Fang H, Rizzo JA, & Ortega AN (2009). Understanding Observed and Unobserved Health Care Access and Utilization Disparities Among U.S. Latino Adults. Medical Care Research and Review, 66(5), 561–577. 10.1177/1077558709338487 [DOI] [PubMed] [Google Scholar]