Abstract
Background
Although studies to date have broadly shown that cardiovascular disease (CVD) increases cognitive and physical impairment risk, there is still limited understanding of the magnitude of this risk among relevant CVD subtypes or age cohorts.
Methods
We analyzed longitudinal data from 16 679 U.S. Health and Retirement Study participants who were aged ≥65 years at study entry. Primary endpoints were physical impairment (activities of daily living impairment) or cognitive impairment (Langa–Weir Classification of dementia). We compared these endpoints among participants who developed incident CVD versus those who were CVD free, both in the short term (<2-year postdiagnosis) and long term (>5 years), controlling for sociodemographic and health characteristics. We then analyzed the effects by CVD subtype (atrial fibrillation, congestive heart failure, ischemic heart disease, and stroke) and age-at-diagnosis (65–74, 75–84, and ≥85).
Results
Over a median follow-up of 10 years, 8 750 participants (52%) developed incident CVD. Incident CVD was associated with significantly higher adjusted odds (aOR) of short-term and long-term physical and cognitive impairment. The oldest (≥85) age-at-diagnosis subgroup had the highest risk of short-term physical (aOR 3.01, 95% confidence interval [CI]: 2.40–3.77) and cognitive impairment (aOR 1.96, 95% CI: 1.55–2.48), as well as long-term impairment. All CVD subtypes were associated with higher odds of physical and cognitive impairment, with the highest risk for patients with incident stroke.
Conclusions
Incident CVD was associated with an increased risk of physical and cognitive impairment across CVD subtypes. Impairment risk after CVD was highest among the oldest patients (≥85 years) who should therefore remain a target for prevention efforts.
Keywords: Cardiovascular disease, Cognitive impairment, Physical impairment
By 2035, the number of people over age 65 in the United States will outnumber children below the age of 18 (1). As age confers increased cardiovascular risk, it is expected that among this growing older adult population nearly 4 out of 5 individuals will have 1 or more forms of cardiovascular disease (CVD), and given changing demographics the absolute number of people with CVD will markedly increase (2,3).
In this context, a key area of research has been to understand the effects of CVD on “healthspan,” defined as aging without major physical or cognitive impairment (4,5). Plausibly, CVD represents a phenotype of accelerated vascular aging that impairs these domains through mechanisms including chronic inflammation (6) and oxidative stress (7). Although studies to date have broadly suggested an association between CVD and impairment (8–11), they have generally failed to comparatively examine the effects of clinically distinct CVD subtypes (eg, stroke, ischemic heart disease, heart failure, atrial fibrillation) and age at CVD diagnosis. In addition, the timing of impairment onset after incident CVD (eg, immediate- or late onset) is not clear given the longitudinal limitations of many data sets.
Given these knowledge gaps, we used data from the Health and Retirement Study (HRS) to better understand the relationship between incident CVD (including clinically relevant subtypes) and the odds of physical and cognitive impairment. The HRS is a large, well-characterized nationally representative cohort study in the United States that currently includes over 42 000 participants enrolled between 1992 and 2018, with over 2 decades of regular interval assessments for almost 10 000 individuals (12,13). This long duration of follow-up provides an opportunity to measure the long-term impacts of slow-acting biological processes such as cognitive decline. Furthermore, HRS was explicitly designed to examine the demographic trends of an aging population and accordingly includes measures of both cognitive and physical impairment (13). Elucidating the relationship between incident CVD and subsequent impairment, while adjusting for important sociodemographic and clinical characteristics, can inform targeted and tailored screening (eg, measures for early cognitive decline) and prevention efforts (eg, aggressive management of hypertension and dyslipidemia) among growing population of older adults.
Methods
Study Participants
We performed a secondary analysis of data from waves 1992–2018 of the HRS core survey and its linked Medicare claims. The HRS is a biennial longitudinal health survey of U.S. older-age adults, nearly all of whom are aged ≥50 years (12,13). The HRS is a nationally representative cohort, with a slight oversampling of Black, Hispanic, and Floridian households (12,13). Survey questions, as described previously, address physical, mental, and financial well-being, as well as employment history and family circumstances (12,13). Our study included respondents from 6 HRS cohorts (encompassing those born prior to 1924 to those born in 1953), where a new cohort of individuals aged 51–56 is added every 6 years (eg, in 1998, 2004, 2010, etc.), for a total of 42 361 individuals in the overall study sample. From the overall study sample, we included HRS participants with incident CVD at age-at-diagnosis ≥65 and HRS participants with no CVD throughout their HRS follow-ups for the main analysis.
Variables and Their Measurements
Incident cardiovascular disease
Participants were classified as having incident CVD using International Classification of Diseases-10 (ICD-10) diagnosis claims from linked Medicare records (Supplementary Table 1). The year/age of CVD incidence was defined as the earliest date/age with a CVD diagnosis claim of atrial fibrillation, congestive heart failure, ischemic heart disease, or stroke in the Medicare files. We then classified all participants with incident CVD into 3 age-at-diagnosis groups: 65–74, 75–84, and ≥85, as well as into 4 common types of CVD using their diagnostic codes: ischemic heart disease (including the subgroup hospitalized for acute myocardial infarction), congestive heart failure, atrial fibrillation, and stroke (Figure 1). Because there is extensive literature documenting the association of stroke with physical and cognitive impairment (14–18), the subgroup with stroke effectively served as a positive control.
Figure 1.
Flow chart of the study design. The Health and Retirement Study (HRS) included 42 361 participants between 1992 and 2018. After excluding 1 344 participants with missing information on age, 14 511 younger than 65, 4 767 without linked Medicare data, and 5 060 participants with prevalent cardiovascular disease (CVD) before or at the time of HRS entry, our final analysis cohort was 16 679 HRS participants aged 65 and older who were CVD free at HRS entry. In this cohort, 8 750 participants developed incident CVD and 7 929 are CVD free.
When assessing how the incident CVD subtype affects physical and cognitive impairment, we considered in an interaction term only the subtype for the first CVD event. Given that nearly 60% of participants had >1 diagnosis of CVD subtype, the majority of participants would go on to experience other CVD subtypes. A full overview of the breakdown of incident CVD types by the participants is shown in Figure 2.
Figure 2.
Count of incident cardiovascular disease (CVD) subtypes, demonstrating that many participants had multiple CVD diagnoses after entering the Health and Retirement Study (HRS). AFIB = atrial fibrillation; CHF = congestive heart failure; count = number of HRS participants with the indicated combination of CVD subtype diagnoses; Ischemic = ischemic heart disease.
Outcomes
The primary outcomes were physical impairment and cognitive impairment using biennial self-responded outcomes in the HRS core survey. Physical impairment was measured using any dependency in activities of daily living (ADLs), which included walking across a room, dressing, bathing, feeding, transferring, or toileting. Cognitive impairment was measured using Langa–Weir classification (LWC), which uses the Telephone Interview for Cognitive Status (range 0–27; higher scores indicate better cognitive function) for self-respondents and a validated 11-point scale for proxy respondents (19). The LWC maps participants into 3 categories: normal, cognitively impaired without dementia (CIND), and with dementia (19). For our analysis, we treated both physical and cognitive disabilities as binary outcomes (physical impairment: ADL >0 and cognitive impairment: LWC demented).
Covariates
We controlled for demographic and clinical characteristics at baseline based on prior literature (9,11,20–22) as well as the clinical judgment of study investigators. Demographic variables included age, gender, race (non-Hispanic White, non-Hispanic Black, Hispanic, and Other), HRS cohort (a decade when a participant was born), whether the participant was self-respondent (vs proxy), marital status, educational attainment, and net worth. Clinical variables included body mass index, smoking status, history of depression, history of drinking, and comorbid chronic conditions based on self-report: hypertension, cancer, chronic lung disease, and diabetes. We additionally controlled for ADL impairment and LWC status at the HRS baseline.
Statistical Analysis
We performed descriptive analyses to compare participants with and without incident CVD. For continuous measures, we used the Wilcoxon rank-sum test, and for categorical measures, we used the Chi-square test.
For the primary analysis of our binary outcome variables, which were linked to participant change over time, we fit linear mixed-effects logistic regression models to measure changes in the odds of physical and cognitive impairment over time after adjusting for all covariates. Time was expressed as years from the first HRS baseline interview. Incident CVD changes from 0 to 1 on the year of CVD diagnosis and remains 1 in all years after diagnosis, and a duration covariate is included as years since CVD diagnosis.
In this model, the parameters of interest are (a), the adjusted odds ratio (aOR) of incident CVD representing the short-term (within 2 years) increase in the outcome odds after CVD diagnosis; and (b), the aOR of duration representing the relative change in slope of odds of impairment after CVD compared with the slope before CVD and the slope of participants who never were diagnosed with CVD (ie, acceleration). We then calculated a 5-year aOR, using the short-term and acceleration coefficient estimates, for patients with incident CVD and patients without incident CVD, controlling for covariates.
The adjusted covariates, including demographics, clinical characteristics, and outcomes at the HRS baseline, were retained in all models regardless of statistical significance. Interaction terms between age cohort and time were included to adjust for changes across generations and in CVD management. Random effects were included to accommodate for within-participant outcome correlations.
Additional analyses were performed to examine the association of age and CVD subtype on outcomes. Specifically, we implemented the above-described analysis in subgroups of patients with incident CVD—including age at CVD diagnosis (65–74, 75–84, and ≥85), and CVD subtype (ischemic heart disease, congestive heart failure, atrial fibrillation, and stroke). For both analyses, the aORs were calculated from the above main model with the addition of interaction terms of CVD age-at-diagnosis and CVD subtype to the existing incident CVD and the duration of CVD covariates.
To illustrate the differential aging trajectories by CVD age-at-diagnosis, we calculated the predicted trajectory for each outcome of physical impairment and cognitive impairment for a model participant. Our participant was defined as a non-Hispanic White woman born in the 1930s entering HRS at age 65 with the median values of all covariates at baseline (eg, high school education, net worth <$40 000, married/partnered, never smoker, no diabetes, lung disease, or cancer, normal LWC status, and ADL score of 0). We then predicted the probability of physical and cognitive impairment for the patient from ages 65–100 based on her never experiencing a CVD event or experiencing one at different ages of diagnosis (65–74, 75–84, or ≥85).
Statistical significance for all analyses was a 2-sided p value less than .05. All analyses used R software, version 2021.09.1 + 372 “Ghost Orchid” Release.
Sensitivity Analysis
In our first set of sensitivity analyses we removed patients who already had physical (ADL >0) and cognitive (LWC demented) impairment at or prior to incident CVD diagnosis and repeated the primary analysis previously described. We kept these patients in the main model to be reflective of the overall patient population.
In our second set of sensitivity analyses, our goal was to account for the high overlap of participant CVD diagnoses. To do this, we ran 2 sensitivity analyses (1) including patients in a CVD subtype if they ever received a diagnosis of that CVD subtype during their time in HRS, regardless of whether it was their first CVD diagnosis (indicating that patients can appear across multiple subgroups) and (2) only looking at patients who had a single CVD diagnosis throughout their entire HRS observation period. We then compared outcomes from both of these CVD subtype groupings to outcomes prior to that particular CVD diagnosis and to outcomes in participants who never had a CVD diagnosis.
In our final set of sensitivity analyses, we assessed how the definition of cognitive impairment influenced results. Our primary outcome for cognitive impairment was dementia, as defined by LWC, and we compared these results to a broader definition of cognitive impairment that also includes CIND.
Results
Baseline Characteristics
There were 42 361 individuals in the overall HRS sample from waves 1992 to 2018. After excluding those with missing age information (1 344), age <65 years (14 511), and those without linked Medicare data (4 767) there were 21 739 participants. We then further excluded respondents who had a prior CVD diagnosis at their baseline HRS interview (5 060), which left a final sample of 16 679 participants. Over the observation period for our study (median follow-up of 10 years, IQR: 6–16 years), 8 750 participants experienced incident CVD and 7 929 were free of CVD (Figure 1). Among those with incident CVD, 4 620 (53%) developed CVD between the ages of 65–74, 3 053 (35%) between the ages of 75–84, and 1 077 (12%) at ≥85.
Participant characteristics are shown in Table 1. Participants with incident CVD were more likely to be older, male, non-Hispanic White, with less education, and with lower income at the HRS baseline. Baseline physical impairment and cognitive state were worse in participants with incident CVD than those who never experienced CVD. Of note, participants with physical impairment (defined as ADL > 0) had dependency on bathing and dressing as the most frequent measures of physical impairment (Supplementary Table 2). Additionally, nearly 60% of participants had greater than 1 type of CVD diagnosis over the course of their time within HRS, with ischemic heart disease being the most frequent with 7 641 participants (753 of whom also had an acute myocardial infarction). A summary of the types of incident CVD experienced by participants is shown in Figure 2. Additionally, a summary of participant demographics who joined HRS over the age of 65, but that had a prior CVD diagnosis, is shown in Supplementary Table 3; this subset was slightly older and with greater physical and cognitive disability at the HRS baseline.
Table 1.
Demographic Characteristics of the Health and Retirement Study Respondents at Baseline Overall and by Cardiovascular Disease Status
| Overall | CVD-Free Participants | Participants With Incident CVD | p Value* | |
|---|---|---|---|---|
| N | 16 679 | 7 929 | 8 750 | |
| Gender (N [%]) | .035 | |||
| Male | 7 054 (42.3) | 3 277 (41.3) | 3 777 (43.2) | |
| Female | 9 624 (57.7) | 4 652 (58.7) | 4 972 (56.8) | |
| NA | 1 (0.0) | 0 (0.0) | 1 (0.0) | |
| Race/ethnicity (N [%]) | <.001 | |||
| Hispanic | 818 (4.9) | 442 (5.6) | 376 (4.3) | |
| Non-Hispanic Black | 2 367 (14.2) | 1 132 (14.3) | 1 235 (14.1) | |
| Non-Hispanic White | 13 034 (78.1) | 6 089 (76.8) | 6 945 (79.4) | |
| Other | 453 (2.7) | 260 (3.3) | 193 (2.2) | |
| NA | 7 (0.0) | 6 (0.1) | 1 (0.0) | |
| Cohort (N [%])† | <.001 | |||
| AHEAD | 4 620 (27.7) | 1 544 (19.5) | 3 076 (35.2) | |
| CODA | 2 767 (16.6) | 1 050 (13.2) | 1 717 (19.6) | |
| HRS | 6 344 (38.0) | 3 033 (38.3) | 3 311 (37.8) | |
| WB | 1 919 (11.5) | 1 385 (17.5) | 534 (6.1) | |
| EBB | 1 029 (6.2) | 917 (11.6) | 112 (1.3) | |
| Education (N [%]) | <.001 | |||
| <12 | 4 940 (29.6) | 2 062 (26.0) | 2 878 (32.9) | |
| =12 | 5 499 (33.0) | 2 597 (32.8) | 2 902 (33.2) | |
| >12 | 6 224 (37.3) | 3 259 (41.1) | 2 965 (33.9) | |
| NA | 16 (0.1) | 11 (0.1) | 5 (0.1) | |
| Characteristics at HRS baseline | ||||
| Age (mean [SD]) | 69.20 (5.73) | 68.35 (5.29) | 69.98 (6.00) | <.001 |
| Marriage (N [%])‡ | Overall | CVD-free participants | Participants with incident CVD | .490 |
| Married/partnered | 11 316 (67.8) | 5 383 (67.9) | 5 933 (67.8) | |
| Not married | 5 333 (32.0) | 2 535 (32.0) | 2 798 (32.0) | |
| NA | 30 (0.2) | 11 (0.1) | 19 (0.2) | |
| Net worth (N [%])‡ | <.001 | |||
| <$40 000 | 10 375 (62.2) | 4 560 (57.5) | 5 815 (66.5) | |
| $40 000–150 000 | 5 442 (32.6) | 2 861 (36.1) | 2 581 (29.5) | |
| $150 000–300 000 | 634 (3.8) | 382 (4.8) | 252 (2.9) | |
| ±$300 000 | 206 (1.2) | 120 (1.5) | 86 (1.0) | |
| NA | 22 (0.1) | 6 (0.1) | 16 (0.2) | |
| BMI (median [IQR]) | 26.50 [23.60, 29.80] | 26.40 [23.40, 29.80] | 26.50 [23.74, 29.80] | .057 |
| Smoker (%) | <.001 | |||
| No | 10 879 (65.2) | 4 641 (58.5) | 6 238 (71.3) | |
| Yes | 2 475 (14.8) | 1 219 (15.4) | 1 256 (14.4) | |
| NA | 3 325 (19.9) | 2 069 (26.1) | 1 256 (14.4) | |
| Ever drinker (N [%]) | <.001 | |||
| No | 7 764 (46.5) | 3 547 (44.7) | 4 217 (48.2) | |
| Yes | 8 257 (49.5) | 4 180 (52.7) | 4 077 (46.6) | |
| NA | 658 (3.9) | 202 (2.5) | 456 (5.2) | |
| Hypertension (N [%]) | <.001 | |||
| No | 8 287 (49.7) | 4 129 (52.1) | 4 158 (47.5) | |
| Yes | 8 383 (50.3) | 3 795 (47.9) | 4 588 (52.4) | |
| NA | 9 (0.1) | 5 (0.1) | 4 (0.0) | |
| Cancer (N [%]) | .207 | |||
| No | 14 722 (88.3) | 6 972 (87.9) | 7 750 (88.6) | |
| Yes | 1 957 (11.7) | 957 (12.1) | 1 000 (11.4) | |
| Ever depression (N [%]) | .702 | |||
| No | 15 969 (95.7) | 7 586 (95.7) | 8 383 (95.8) | |
| Yes | 710 (4.3) | 343 (4.3) | 367 (4.2) | |
| Chronic lung disease (N [%]) | .001 | |||
| No | 15 190 (91.1) | 7 290 (91.9) | 7 900 (90.3) | |
| Yes | 1 482 (8.9) | 636 (8.0) | 846 (9.7) | |
| NA | 7 (0.0) | 3 (0.0) | 4 (0.0) | |
| Diabetes (N [%]) | .024 | |||
| No | 14 143 (84.8) | 6 778 (85.5) | 7 365 (84.2) | |
| Yes | 2 529 (15.2) | 1 146 (14.5) | 1 383 (15.8) | |
| NA | 7 (0.0) | 5 (0.1) | 2 (0.0) | |
| Physical and cognitive impairment at HRS baseline | ||||
| Physical state (ADL, N [%])‡ | .002 | |||
| No | 15 988 (95.9) | 7 642 (96.4) | 8 346 (95.4) | |
| Yes | 668 (4.0) | 281 (3.5) | 387 (4.4) | |
| NA | 23 (0.1) | 6 (0.1) | 17 (0.2) | |
| Cognitive state (N [%])‡ | <.001 | |||
| Normal | 12 673 (76.0) | 6 141 (77.4) | 6 532 (74.7) | |
| CIND | 2 604 (15.6) | 1 120 (14.1) | 1 484 (17.0) | |
| Dementia | 967 (5.8) | 367 (4.6) | 600 (6.9) | |
| NA | 435 (2.6) | 301 (3.8) | 134 (1.5) | |
Notes: ADL = activities of daily living; AHEAD = study of assets and health dynamics among the oldest old cohort; BMI = body mass index; CIND = cognitive impairment without dementia; CODA = children of the depression cohort; CVD = cardiovascular disease; EBB = early baby boomer cohort; HRS = Health and Retirement Study cohort; IQR = interquartile range; SD = standard deviation; WB = war baby cohort.
* p Values are from the 2-sample t test or Wilcoxon rank-sum test for comparing continuous covariates and the Chi-square test for comparing categorical covariates.
†HRS included 6 cohorts: AHEAD cohort, born prior to 1924; CODA cohort, born 1924–1930; the original HRS cohort, born 1931–1941; the WB cohort, born 1942–1947; and the EBB cohort, born 1948–1953.
‡ADL, marriage and net worth were first measured for HRS participants in year 1996. For participants with baseline year earlier than 1996, their ADL, marriage and net worth on 1996 was used as baseline. Cognitive state was first measured in 1995 or 1996; thus, the baseline cognitive state was defined as the measurements in either 1995 or 1996. Cognitive state is measured using the Langa–Weir classification (LWC).
Physical Impairment After CVD
Among participants with incident CVD, the prevalence (N) of physical impairment increased from 8.6% (8 611) at the last interview before CVD, to 16.4% (8 317; p < .001 compared with before CVD) at the immediate post-CVD interview, to 25.6% (6 502; p < .001 compared with before CVD) on average in interviews of 2+ years post-CVD. Further details are shown in Supplementary Table 4.
The short-term adjusted odds of developing physical impairment were increased among participants with incident CVD (aOR 2.03, 95% confidence interval [CI]: 1.84–2.23; Table 2 and Supplementary Figure 1). Controlling for the short-term increase, the adjusted acceleration odds of developing physical impairment in the 2+ years following CVD diagnosis increased significantly faster (aOR 1.04 times faster, 95% CI: 1.02–1.06; Table 2 and Supplementary Figure 1) than the increase in adjusted odds of physical impairment for those who never experienced incident CVD (aOR 1.35 per year, 95% CI: 1.33–1.37). Combining the short-term increase and acceleration, participants with incident CVD at year 5 experienced greater adjusted odds of physical impairment (aOR 2.46, 95% CI: 2.21–2.73; Table 2 and Supplementary Figure 1) than participants without incident CVD.
Table 2.
The Estimated Short-Term, Acceleration, and 5-Year aOR for Physical and Cognitive Impairment for All Participants With Incident Cardiovascular Disease, Including by Age-At-Diagnosis and Cardiovascular Disease Subtype
| Short-Term Impairment | Acceleration Impairment | 5-Year Impairment | ||||||
|---|---|---|---|---|---|---|---|---|
| N | Physical (aOR, 95% CI) | Cognitive (aOR, 95% CI) | Physical (aOR, 95% CI) | Cognitive (aOR, 95% CI) | Physical (aOR, 95% CI) | Cognitive (aOR, 95% CI) | ||
| Overall | 8 750 | 2.03 (1.84, 2.23) | 1.30 (1.18, 1.44) | 1.04 (1.02, 1.06) | 1.05 (1.03, 1.06) | 2.46 (2.21, 2.73) | 1.63 (1.47, 1.82) | |
| By age at diagnosis | 65–74 | 4 620 | 1.76 (1.53, 2.02) | 1.35 (1.17, 1.56) | 1.03 (1.01, 1.05) | 1.01 (0.99, 1.04) | 2.06 (1.81, 2.34) | 1.45 (1.27, 1.65) |
| 75–84 | 3 053 | 1.97 (1.70, 2.28) | 1.04 (0.89, 1.2) | 1.09 (1.06, 1.11) | 1.13 (1.1, 1.16) | 2.97 (2.52, 3.50) | 1.88 (1.60, 2.21) | |
| ≥85 | 1 077 | 3.01 (2.40, 3.77) | 1.96 (1.55, 2.48) | 1.15 (1.09, 1.22) | 1.12 (1.06, 1.18) | 6.16 (4.58, 8.29) | 3.40 (2.55, 4.54) | |
| By CVD subtype (first diagnosis) | Atrial fibrillation | 753 | 2.11 (1.53, 2.9) | 1.71 (1.25, 2.35) | 1.1 (1.05, 1.15) | 1.08 (1.02, 1.13) | 3.37 (2.52, 4.52) | 2.47 (1.86, 3.29) |
| Congestive heart failure | 1 975 | 3.49 (2.9, 4.21) | 1.75 (1.46, 2.1) | 1.03 (1.00, 1.06) | 1.05 (1.02, 1.09) | 4.05 (3.30, 4.98) | 2.28 (1.88, 2.78) | |
| Ischemic heart disease | 4 867 | 1.75 (1.52, 2.02) | 1.14 (1, 1.31) | 1.06 (1.04, 1.09) | 1.07 (1.05, 1.1) | 2.40 (2.03, 2.83) | 1.63 (1.40, 1.91) | |
| Subset of ischemic heart disease w/acute myocardial infarction | 107 | 3.26 (1.38, 7.71) | 0.88 (0.35, 2.22) | 1.04 (0.9, 1.19) | 1.11 (0.95, 1.3) | 3.90 (1.80, 8.46) | 1.48 (0.67, 3.24) | |
| Stroke | 1 155 | 5.28 (4.19, 6.65) | 2.92 (2.34, 3.66) | 1.05 (1.01, 1.09) | 1.06 (1.02, 1.1) | 6.63 (5.22, 8.43) | 3.84 (3.07, 4.81) | |
Notes: aOR = adjusted odds ratio; 95% CI = 95% confidence interval; CVD = cardiovascular disease.
Cognitive Impairment (Dementia) After CVD
Among participants with incident CVD, the prevalence (N) of cognitive impairment increased from 9.2% (7 710) before CVD to 14.7% (8 229) in the short term after CVD diagnosis and 21.6% (6 161) on average in interviews of 2+ years after CVD (both p < .001 compared with before CVD). Further details are shown in Supplementary Table 4.
The short-term odds of developing cognitive impairment were increased among participants with incident CVD (aOR 1.30, 95% CI: 1.18–1.44; Table 2 and Supplementary Figure 1). In the years following CVD diagnosis, the acceleration, or annual change in slope, of cognitive impairment was significantly faster (aOR 1.05 times faster, 95% CI: 1.03–1.06; Table 2 and Supplementary Figure 1) than the annual increase in adjusted odds of cognitive impairment for those who never experienced incident CVD. Thus, the aOR of cognitive impairment increased by 1.23 (95% CI: 1.21–1.25) per year for participants without incident CVD, those with incident CVD increased 1.05 times faster, at an accelerated rate of 1.29 per year (1.05 × 1.23). Combining the short-term increase and acceleration over time, participants with incident CVD had significantly greater odds of cognitive impairment at 5 years (aOR 1.63, 95% CI: 1.47–1.82; Table 2 and Supplementary Figure 1) compared with participants without incident CVD.
Variability of Impact on Healthspan by Age at CVD Diagnosis
The increases in short-term and acceleration odds of physical and cognitive impairment were greater with increasing age at CVD diagnosis. Although participants with incident CVD at age 65–74 years experienced significant short-term increases in physical impairment (aOR 1.76, 95% CI: 1.53–2.02) and acceleration (aOR 1.03 times faster, 95% CI: 1.01–1.05) compared with those without incident CVD, those at older ages had more pronounced increases: age 75–84 and ≥85 had significant short-term (75–84: aOR 1.97, 95% CI: 1.70–2.28; ≥85: aOR 3.01, 95% CI: 2.40–3.77) and acceleration (75–84: aOR 1.09 times faster, 95% CI: 1.06–1.11; ≥85: aOR 1.15 times faster, 95% CI: 1.09–1.22) odds of physical impairment (Table 2). With each increase in a decade, participants with incident CVD experienced a greater increase in odds of physical decline that manifested both in short- and long-term outcomes, and culminated in the largest increased 5-year adjusted odds of physical impairment for the ≥85 cohort (aOR 6.16, 955 CI: 4.58–8.29). The 75–84 cohort had lower 5-year adjusted odds of physical impairment (aOR 2.97, 95% CI: 2.52–3.50) and the 65–74 cohort had even lower (aOR 2.06, 95% CI: 1.81–2.34) (Table 2).
For cognitive impairment, only participants with CVD age-at-diagnosis in the age 65–74 and age ≥85 ranges had significant short-term impairment development (65–74: aOR 1.35, 95% CI: 1.17–1.56; ≥85: aOR 1.96, 95% CI: 1.55–2.48). Concerning acceleration in the odds of cognitive impairment, the 65–74 age group did not have a significant association, but the 75–84 and ≥85 age groups did (75–84: aOR 1.13 times faster, 95% CI: 1.10–1.16; ≥85: aOR 1.12 times faster, 95% CI: 1.06–1.18). Similar to physical impairment, the oldest participants (≥85) had the largest 5-year adjusted odds of cognitive impairment (aOR 3.40, 95% CI: 2.55–4.54; Table 2).
To illustrate the implications of our findings, we generated a simulated participant in Figure 3. This sample participant is described in Method section and the figure shows the predicted probability of physical and cognitive impairment if she experienced incident CVD at ages 65–74, 75–84, and ≥85, respectively, comparing it to her predicted trajectory in the absence of CVD. As shown, at age-at-diagnosis ≥85, the increase in short-term and acceleration odds of physical and cognitive impairment is greater than at younger ages-at-diagnosis.
Figure 3.
Predicted mean change in the risk of physical and cognitive impairment before and after cardiovascular disease by age-at-diagnosis: Participant-specific (conditional) predicted values were calculated for a non-Hispanic White woman born in the 1930s entering Health and Retirement Study cohort (HRS) at age 65 with the median values of all covariates at baseline (eg, high school education, net worth <$40 000, married, never smoker, no diabetes, normal cognitive state, and activities of daily living [ADL] score of 0). The black line shows the trajectory of the outcome risk if she was CVD free. The green, red, and purple lines show the trajectories of outcome risks if she develops incident CVD at the ages of 65–74, 75–84, and ≥85, respectively. The dash lines show the pre-CVD rate of increases of the risks for comparison.
Variability of Impact on Healthspan by Subtype of CVD
Among the 4 subtypes of CVD, all were associated with a significant short-term increase in the adjusted odds of impairment (atrial fibrillation: aOR 2.11, 95% CI: 1.53–2.90; congestive heart failure: aOR 3.49, 95% CI: 2.90–4.21; ischemic heart disease: aOR 1.75, 95% CI: 1.52–2.02; stroke: aOR 5.28, 95% CI: 4.19–6.65), as well as significant adjusted acceleration odds in physical impairment (atrial fibrillation: aOR: 1.10, 95% CI: 1.05–1.15; congestive heart failure: aOR 1.03, 95% CI: 1.00–1.06; ischemic heart disease: aOR 1.06, 95% CI: 1.04–1.09; stroke: aOR 1.05, 95% CI: 1.01–1.09; Table 2). The subset of patients with ischemic heart disease that also experienced an acute myocardial infarction had a higher increase in the odds of short-term physical impairment (aOR: 3.26, 95% CI: 1.38–7.71) as compared with patients with ischemic heart disease overall, but had insignificant increases in the acceleration odds of physical impairment and in short-term and acceleration odds of cognitive impairment.
With regards to cognitive impairment, all 4 CVD subtypes were significantly associated with both an increase in short-term odds (atrial fibrillation: aOR 1.71, 95% CI: 1.25–2.35; congestive heart failure: aOR 1.75, 95% CI: 1.46–2.10; ischemic heart disease: aOR 1.14, 95% CI: 1.00–1.31; stroke: aOR 2.92, 95% CI: 2.34–3.66) and acceleration odds (atrial fibrillation: aOR 1.08, 95% CI: 1.02–1.13; congestive heart failure: aOR 1.05, 95% CI: 1.02–1.09; ischemic heart disease: aOR 1.07, 95% CI: 1.05–1.10; stroke: aOR 1.06, 95% CI: 1.02–1.10; Table 2). At 5 years post-CVD, participants with any of the 4 CVD types had increased odds of physical and cognitive impairment, with the subset of ischemic heart disease patients with acute myocardial infarction having only significantly increased odds of physical impairment.
Sensitivity Analysis
In the first sensitivity analysis, we assessed how the results would differ by excluding patients who already had physical impairment (ADL > 0; N = 936 removed) and cognitive impairment (LWC = demented; N = 885 removed) prior to or at the time of experiencing incident CVD. We found that there were significant short-term increases in odds for both physical impairment (aOR 4.49, 95% CI: 3.97–5.09) and cognitive impairment (aOR 2.98, 95% CI: 2.61–3.41). The estimated accelerations of the rates of incident physical and cognitive impairment were similar to the main results; but were not statistically significant for cognitive impairment (p = .601, Supplementary Figure 2).
In the second sensitivity analysis, we observed how differences in CVD subtype definition would affect estimates. When including patients in a CVD subtype if they ever received a diagnosis of that CVD subtype during their time in HRS, regardless of whether it was their first CVD diagnosis (indicating that patients can appear across multiple subgroups), the odds of physical and cognitive impairment estimated very similar patterns (Supplementary Table 5). When looking only at patients who had a single CVD diagnosis throughout their entire HRS observation period we did observe slight differences in outcome patterns. However, this subset represented less than 50% of all study participants. In this sensitivity analysis, incident stroke and congestive heart failure continued to lead to significant short-term increases in the odds of physical and cognitive impairment, and no incident CVD subtypes led to increases in the acceleration odds of impairment (Supplementary Table 6).
In the final sensitivity analysis comparing variable definitions of cognitive impairment, we found that results for a broader definition of cognitive impairment that includes CIND in addition to dementia were very similar to results just considering dementia (Supplementary Table 7).
Discussion
Consistent with prior literature, our study found that patients with incident CVD experienced an increase in odds of physical and cognitive impairment both in the short and long terms, indicating that they have a shortened healthspan compared with those who remained free from CVD. Furthermore, we found differential effects on the odds of physical and cognitive impairment by CVD subtype and by age. Specifically, the increases in odds of impairment were most pronounced at very advanced age-at-diagnosis (≥85); and within the considerable variability across CVD subtypes the diagnosis of stroke consistently, and expectedly (14–17), demonstrated the greatest odds of subsequent physical and cognitive impairment. Finally, when assessing which participants were more likely to have incident CVD to begin with, our results mirror prior findings in that we also found that participants with less education (23–25) and lower income (23,25,26) were more likely to have incident CVD. Incorporating these factors into global assessments of risk is therefore an important consideration for future prevention efforts.
We undertook our study in the context of an aging U.S. population where CVD is increasing in prevalence. Our study supports prior findings by Yazdanyar et al. (27) and more recent studies (8,10,11,20–22) demonstrating that incident and cumulative CVD contributes to both physical and cognitive decline. We were able to further these findings by assessing the variability by age at CVD diagnosis, the differential impacts of each CVD subtype (stroke, ischemic heart disease, heart failure, atrial fibrillation), and the differences in manifestation between short- and long-term outcomes.
Age has been shown to be a risk factor for both impairment onset and a faster rate of impairment progression (11). However, there have been limited data on the mediating role of age in impairment with incident CVD. We found that the “oldest old” patients (age ≥85 years) had the highest odds of impairment following incident CVD. This finding may help inform practice guidelines for the oldest adults. Using statin therapy as a primary prevention tool as an example, the 2018 American College of Cardiology/American Heart Association guidelines weigh heavily on patient–physician discussions over direct guidelines (28) given the limitations of many clinical trials (28). The findings from our study can help to inform these discussions by showing that CVD events have a major impact on functional status, highlighting the additional benefits of prevention over more traditional outcomes (eg, recurrent myocardial infarction). We acknowledge that until findings from definitive statin prevention trials such as PREVENTABLE (NCT04262206) are published, which is evaluating whether statin therapy for primary prevention prevents incident dementia and persistent impairment in patients aged ≥75 years, this area remains one of clinical uncertainty (29). For blood pressure, the SPRINT trial recently demonstrated that intensive control of blood pressure can preserve function and cognition (30), and we believe that aggressive antihypertensive therapy is an actionable step for older patients provided it aligns with their own goals. Our findings underscore that CVD events lead to major functional declines in this population, which again can help to inform patient discussions.
Last, our study evaluated patterns of impairment by CVD subtype. A prior study found that among patients with heart failure, after incident diagnosis there was a faster rate of decline in global cognitive ability (20). Our study assessed this impact among a broader CVD population and found that across subtypes, CVD led to increased odds of physical and cognitive function decline and that short-term patterns of impairment vary by CVD subtype. As expected, the highest increases in adjusted odds were seen among patients experiencing stroke given an existing large body of evidence showing that poststroke cognitive and physical deficits lead to dementia in one third of stroke survivors and impairments in ADLs in over half of survivors (14–17). Thus, the positive results seen in participants with incident stroke effectively served as a positive control in our data set.
However, even excluding stroke from the equation, other CVD conditions had significant impacts on impairment. Some explanatory mechanisms for our findings include the high proinflammatory state seen after an acute myocardial infarction (31) and with ischemic heart disease (32), which can affect muscle function and lead to physical impairment. Additionally, reduced cerebral perfusion in the context of atrial fibrillation and congestive heart failure (33) has been linked with cognitive decline. Furthermore, there has been increasing research looking at the link between frailty and health outcomes following incident CVD with findings demonstrating that frailty is a risk factor for cognitive and physical impairment (34). In addition to frailty, dyspnea and fatigue are prominent in patients with CVD, particularly in those who are older, and may lead to exercise intolerance followed by physical impairment (35). Last, nonelective hospitalizations in older adults, for which CVD is a risk factor (36), lead to accelerations in cognitive decline as compared to elective hospitalizations (37). Vascular aging along with the mechanisms of a high inflammatory state, hypoperfusion, frailty, and dyspnea as well as fatigue are risk factors for both CVD as well as impairment following CVD, indicating that after any incident CVD event, particularly one that required a nonelective hospitalization, focused monitoring on physical and cognitive decline may be warranted.
Strengths of our study include a large, well-characterized, and representative cohort that has serial assessments with nearly a decade of follow-up for each participant. Additionally, incident CVD diagnosis was validated with linked Medicare data, which, along with the large sample size of the subgroups, allowed for a targeted analysis that focused on the heterogeneity of outcomes by CVD subtype and age at CVD diagnosis. Furthermore, the study was able to adjust for not only demographic factors but also social and economic factors known to influence CVD (38). Last, the study incorporated multiple sensitivity analyses to account for the heterogeneity of underlying functional impairment and heterogeneity of CVD diagnoses across participants. Overall, we found that patterns of short-term (<2 years) increases in the adjusted odds of physical and cognitive impairment remained significant across the sensitivity analyses. The greatest variability was seen in the acceleration odds of impairment. For example, the adjusted odds of cognitive impairment acceleration were not significant in participants that did not have either underlying physical or cognitive impairment, indicating that careful monitoring of physical and cognitive function is most critical in the first few years following incident CVD diagnosis.
There are several limitations in our analysis that also deserve consideration. First, we were unable to control for the clinical severity of participants’ CVD (eg, heart failure symptom burden) and its association with physical and cognitive impairment. Second, our use of the LWC for the dementia outcome meant that we were unable to distinguish the dementia subtype, which would have required clinical adjudication. We also considered cognitive impairment a binary variable and, in the future, can consider the spectrum of cognitive impairment that includes cognitive impairment without dementia. Third, we did not have details on treatments received for CVD and are unable to determine whether preventive therapies slowed declines over the long term. Last, we did not use a competing risk analysis in our study, and findings may have been influenced by the competing risk of death. Our inference is with respect to the impact of incident CVD on subsequent functional outcomes of survivors only. However, higher mortality is expected in the CVD group. Thus, the actual impact of CVD on subsequent functional outcomes in all individuals would be even higher if one could observe the functional outcomes of the subjects who died, as they typically would have worse functional outcomes.
In conclusion, we found that incident CVD was associated with short-term and long-term increases in the odds of physical and cognitive impairment. Furthermore, patients of advanced age-at-diagnosis had the most marked increases in impairment. This has implications for primary preventive therapies in the “oldest old” as well as careful clinical surveillance for functional declines after an incident CVD event.
Supplementary Material
Contributor Information
Katherine L Stone, Division of Geriatric Medicine and Palliative Care, Department of Medicine, New York University Langone Medical Center, New York, New York, USA.
Judy Zhong, Division of Biostatistics, Department of Population Health, New York University Langone Medical Center, New York, New York, USA.
Chen Lyu, Division of Biostatistics, Department of Population Health, New York University Langone Medical Center, New York, New York, USA.
Joshua Chodosh, Division of Geriatric Medicine and Palliative Care, Department of Medicine, New York University Langone Medical Center, New York, New York, USA.
Nina L Blachman, Division of Geriatric Medicine and Palliative Care, Department of Medicine, New York University Langone Medical Center, New York, New York, USA.
John A Dodson, Division of Biostatistics, Department of Population Health, New York University Langone Medical Center, New York, New York, USA; Leon H. Charney Division of Cardiology, Department of Medicine, New York University Langone Medical Center, New York, New York, USA.
Funding
This work was supported by the Summer Research Training in Aging for Medical Students (MSTAR) program (T35AG050998), which is sponsored by the National Institute on Aging to encourage medical students to consider a career in academic geriatrics. The project used data from the Health and Retirement Study, which is supported by the National Institute on Aging (U01 AG009740) and the Social Security Administration. J.A.D. is supported by a mid-career mentoring award from the NIH/NIA (K24AG080025).
Conflict of Interest
None declared.
Author Contributions
All authors have read and approved the manuscript. All authors have made substantial contributions to all of the following: (a) the conception and design of the study, or acquisition of data, or analysis and interpretation of data, (b) drafting the article or revising it critically for important intellectual content, (c) final approval of the version to be submitted. K.L.S. and C.L. were also involved in data curation, formal analysis, and investigation. K.L.S. lead the validation and visualization. K.L.S. and J.D. drafted the article. This article has not been previously published and is not being considered for publication elsewhere, in whole or in part.
Data Availability
The HRS core survey data is available from HRS website. HRS-linked Medicare data is obtained under DUA RDA# 2018-045.
References
- 1. National Center for Chronic Disease Prevention and Health Promotion. Heart Disease Facts. Centers for Disease Control and Prevention. [Google Scholar]
- 2. Nelson S, Whitsel L.. Projections of Cardiovascular Disease Prevalence and Costs: 2015–2035 (Technical Report). RTI International; 2016. [Google Scholar]
- 3. Heidenreich PA, Trogdon JG, Khavjou OA, et al. Forecasting the future of cardiovascular disease in the United States. Circulation. 2011;123(8):933–944. doi: 10.1161/CIR.0b013e31820a55f5 [DOI] [PubMed] [Google Scholar]
- 4. Queen NJ, Hassan QN, Cao L.. Improvements to healthspan through environmental enrichment and lifestyle interventions: where are we now? Front Neurosci. 2020;14. doi: 10.3389/fnins.2020.00605 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Garmany A, Yamada S, Terzic A.. Longevity leap: mind the healthspan gap. NPJ Regen Med. 2021;6(1):57. doi: 10.1038/s41536-021-00169-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ferrucci L, Fabbri EI.. Chronic inflammation in ageing, cardiovascular disease, and frailty. Nat Rev Cardiol. 2018;15(9):505–522. doi: 10.1038/s41569-018-0064-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Liguori I, Russo G, Curcio F, et al. Oxidative stress, aging, and diseases. Clin Interv Aging. 2018;13:757–772. doi: 10.2147/cia.s158513 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jin Y, Tanaka T, Ma Y, Bandinelli S, Ferrucci L, Talegawkar SA.. Cardiovascular health is associated with physical function among older community dwelling men and women. J Gerontol A Biol Sci Med Sci. 2017;72(12):1710–1716. doi: 10.1093/gerona/glw329 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Covello AL, Horwitz LI, Singhal S, Blaum CS, Li Y, Dodson JA.. Cardiovascular disease and cumulative incidence of cognitive impairment in the Health and Retirement Study. BMC Geriatr. 2021;21(1):274. doi: 10.1186/s12877-021-02191-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Li CL, Chiu YC, Chang HY, Hsu KH, Bai YB, Wang HH.. Association of geriatric conditions and cardiovascular diseases with disability in older adults with diabetes: findings from a nationally representative survey. Geriatr Gerontol Int. 2013;13(3):563–570. doi: 10.1111/j.1447-0594.2012.00935.x [DOI] [PubMed] [Google Scholar]
- 11. Cui K, Song R, Xu H, et al. Association of cardiovascular risk burden with risk and progression of disability: mediating role of cardiovascular disease and cognitive decline. J Am Heart Assoc. 2020;9(18). doi: 10.1161/JAHA.120.017346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Health and Retirement Study. Published online September 15, 2021. https://hrsonline.isr.umich.edu/
- 13. Fisher GG, Ryan LH.. Overview of the Health and Retirement Study and introduction to the special issue. Work Aging Retire. 2018;4(1):1–9. doi: 10.1093/workar/wax032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Levine DA, Wadley VG, Langa KM, et al. Risk factors for poststroke cognitive decline. Stroke. 2018;49(4):987–994. doi: 10.1161/STROKEAHA.117.018529 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Heshmatollah A, Mutlu U, Koudstaal PJ, Ikram MA, Ikram MK.. Cognitive and physical impairment and the risk of stroke—a prospective cohort study. Sci Rep. 2020;10(1):6274. doi: 10.1038/s41598-020-63295-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kelly-Hayes M, Beiser A, Kase CS, Scaramucci A, D’Agostino RB, Wolf PA.. The influence of gender and age on disability following ischemic stroke: the Framingham study. J Stroke Cerebrovasc Dis. 2003;12(3):119–126. doi: 10.1016/s1052-3057(03)00042-9 [DOI] [PubMed] [Google Scholar]
- 17. Capistrant BD, Wang Q, Liu SY, Glymour MM.. Stroke-associated differences in rates of activity of daily living loss emerge years before stroke onset. J Am Geriatr Soc. 2013;61(6):931–938. doi: 10.1111/jgs.12270 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Centers for Disease Control and Prevention. Prevalence and most common causes of disability among adults—United States. MMWR Morb Mortal Wkly Rep. 2009;58:421–426. [PubMed] [Google Scholar]
- 19. Crimmins EM, Kim JK, Langa KM, Weir DR.. Assessment of cognition using surveys and neuropsychological assessment: the Health and Retirement Study and the aging, demographics, and memory study. J Gerontol B Psychol Sci Soc Sci. 2011;66B(Supple 1):i162–i171. doi: 10.1093/geronb/gbr048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hammond CA, Blades NJ, Chaudhry SI, et al. Long-term cognitive decline after newly diagnosed heart failure. Circ Heart Fail. 2018;11(3). doi: 10.1161/CIRCHEARTFAILURE.117.004476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Schievink SHJ, van Boxtel MPJ, Deckers K, van Oostenbrugge RJ, Verhey FRJ, Köhler S.. Cognitive changes in prevalent and incident cardiovascular disease: a 12-year follow-up in the Maastricht Aging Study (MAAS). Eur Heart J. 2022;43(7):e2–e9. doi: 10.1093/eurheartj/ehx365 [DOI] [PubMed] [Google Scholar]
- 22. Xie W, Zheng F, Yan L, Zhong B.. Cognitive decline before and after incident coronary events. J Am Coll Cardiol. 2019;73(24):3041–3050. doi: 10.1016/j.jacc.2019.04.019 [DOI] [PubMed] [Google Scholar]
- 23. Havranek EP, Mujahid MS, Barr DA, et al. Social determinants of risk and outcomes for cardiovascular disease. Circulation. 2015;132(9):873–898. doi: 10.1161/CIR.0000000000000228 [DOI] [PubMed] [Google Scholar]
- 24. Kaplan GA, Keil JE.. Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation. 1993;88(4):1973–1998. doi: 10.1161/01.cir.88.4.1973 [DOI] [PubMed] [Google Scholar]
- 25. Mensah GA, Mokdad AH, Ford ES, Greenlund KJ, Croft JB.. State of disparities in cardiovascular health in the United States. Circulation. 2005;111(10):1233–1241. doi: 10.1161/01.CIR.0000158136.76824.04 [DOI] [PubMed] [Google Scholar]
- 26. Wang SY, Tan ASL, Claggett B, et al. Longitudinal associations between income changes and incident cardiovascular disease. JAMA Cardiol. 2019;4(12):1203–1212. doi: 10.1001/jamacardio.2019.3788 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Yazdanyar A, Newman AB.. The burden of cardiovascular disease in the elderly: morbidity, mortality, and costs. Clin Geriatr Med. 2009;25(4):563–577, vii. doi: 10.1016/j.cger.2009.07.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Anum Saeed M, Laxmi S, Mehta MF.. Statin Therapy in Older Adults for Primary Prevention of Atherosclerotic Cardiovascular Disease: The Balancing Act. American College of Cardiology; 2020. [Google Scholar]
- 29. Alexander K, Ambrosius W, Hernandez A, Williamson J.. Pragmatic Evaluation of Events and Benefits of Lipid-Lowering in Older Adults (PREVENTABLE). clinicaltrials.gov; 2022. [Google Scholar]
- 30. Williamson JD, Pajewski NM, Auchus AP, et al. Effect of intensive vs standard blood pressure control on probable dementia. JAMA. 2019;321(6):553. doi: 10.1001/jama.2018.21442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Ong SB, Hernández-Reséndiz S, Crespo-Avilan GE, et al. Inflammation following acute myocardial infarction: multiple players, dynamic roles, and novel therapeutic opportunities. Pharmacol Ther. 2018;186:73–87. doi: 10.1016/j.pharmthera.2018.01.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Hodzic E. Potential anti-inflammatory treatment of ischaemic heart disease. Med Arch. 2018;72(2):94. doi: 10.5455/medarh.2018.72.94-98 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Alonso A, Arenas de Larriva AP.. Atrial fibrillation, cognitive decline and dementia. Eur Cardiol Rev. 2016;11(1):49. doi: 10.15420/ecr.2016:13:2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Damluji AA, Chung SE, Xue QL, et al. Physical frailty phenotype and the development of geriatric syndromes in older adults with coronary heart disease. Am J Med. 2021;134(5):662–671. doi: 10.1016/j.amjmed.2020.09.057 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Lu J, Wang M, Zhang Y, et al. The prevalence of activities of daily living impairment in patients with heart failure: a systematic review and meta-analysis. Front Cardiovasc Med. 2022;9. doi: 10.3389/fcvm.2022.952926 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Damush TM, Smith DM, Perkins AJ, Dexter PR, Smith F.. Risk factors for nonelective hospitalization in frail and older adult, inner-city outpatients. Gerontologist. 2004;44(1):68–75. doi: 10.1093/geront/44.1.68 [DOI] [PubMed] [Google Scholar]
- 37. James BD, Wilson RS, Capuano AW, et al. Cognitive decline after elective and nonelective hospitalizations in older adults. Neurology. 2019;92(7):e690–e699. doi: 10.1212/wnl.0000000000006918 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Schultz WM, Kelli HM, Lisko JC, et al. Socioeconomic status and cardiovascular outcomes. Circulation. 2018;137(20):2166–2178. doi: 10.1161/CIRCULATIONAHA.117.029652 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The HRS core survey data is available from HRS website. HRS-linked Medicare data is obtained under DUA RDA# 2018-045.