Functional Disability and Cognitive Impairment after Hospitalization for Myocardial Infarction and Stroke

Introduction

Myocardial infarction (MI) and stroke are the leading vascular contributors to illness and death in the United States,^{1, 2} with hundreds of thousands of Americans experiencing one of these events annually.² In 2010, 7.6 million US adults were MI survivors and 6.8 million were stroke survivors,² and these prevalence rates are predicted to increase by up to 25% over the next two decades³ owing to treatment advances and growth of the population aged 65 or older.^4-6 The chronic care of MI and stroke survivors will place huge burdens on families, health care systems, and public programs such as Medicare and Medicaid.

Disability accounts for a substantial amount of the health burden and costs due to MI and stroke.¹ Although both MI and stroke experienced major improvements in acute care (e.g., thrombolysis, organized inpatient care) and secondary cardiovascular disease (CVD) prevention⁷ between 1990 and 2010, disability rates for MI fell nearly 15% but disability rates for stroke increased 40% during this period.¹ Yet, few studies have compared acute and long-term changes in disability due to functional or cognitive problems after MI and stroke in a contemporary cohort of survivors.⁸

Importantly, declines in pre-morbid functioning may precede acute medical events and these declines may confound or even modify the effect of these events on functional and cognitive outcomes.⁹ While MI and stroke have been associated with acute increases in functional and cognitive disability,^10-12 studies conflict on whether acute MI or stroke increase the rate of disability after controlling for individuals’ longitudinal changes in disability over time before the vascular event.^13-18 Many studies suggest that average yearly changes in disability after MI or stroke are not significantly greater than those before MI or stroke.^13-18 It is also not established whether pre-morbid functioning modifies the effect of acute MI or stroke on functional outcomes. If MI and stroke contribute to acute functional and cognitive disability and also new impairments in physical and cognitive functioning over the long-term, then clinicians and payers will need to consider interventions that identify and treat these problems beyond the immediate post-event period.

The present study used a longitudinal, nationally representative cohort of older Americans to compare the acute and long-term changes in functional disability, cognitive impairment, and depression after MI and stroke in a contemporary cohort of survivors. Our analyses focused on the pivotal role of pre-morbid functioning by controlling for and also stratifying by individuals’ longitudinal changes in function and cognition over time before the vascular event.

Methods

Results

There were 391 individuals who experienced 450 hospitalizations for MI but not for stroke and 370 individuals who experienced 432 hospitalizations for stroke though not for MI in which individuals survived to complete ≥1 follow-up outcome assessment. Patients had up to 5 outcome measurements (range: 7.7-9.8 years) before hospitalization and up to 6 outcome measurements (range: 9.9-12.7 years) after hospitalization.

Characteristics of the patients and the hospitalizations have been reported previously.³³ MI survivors were younger and more likely to be male, white, married, and former smokers compared with survivors of stroke hospitalization. Functional limitations, cognitive impairments, and depressive symptoms before the hospitalization were less common in survivors of MI hospitalization than in survivors of stroke hospitalization. Hospitalizations for MI more frequently involved major surgery and stays in ICUs or CCUs compared with hospitalizations for stroke. Survivors of stroke hospitalization had a greater likelihood of switching to proxy-respondent status than MI hospitalization survivors (28% vs. 8%).

Discussion

In this nationally-representative cohort of older Americans, MI hospitalization was associated with significant acute and long-term increases in functional disability. We were able to demonstrate that MI and stroke each increase the rate of additional impairments in the following decade. This emphasizes the extent to which these acute vascular events are potential inflection points in patients’ lives. Clinically, this suggests that clinicians should have an increased index of suspicion that patients may need additional help with ADLs and IADLs over the years after MI and stroke. Scientifically, it suggests an urgent need to disentangle the extent to which this acute and also accelerated disability are the result of incomplete rehabilitation from the initial hospitalization, subsequent vascular events, behavioral changes, or other mechanisms.

Most studies have lacked information on individuals’ changes in disability over time before the MI or stroke hospitalization and consequently have controlled for baseline functioning using measures obtained during or soon after the hospitalization.¹⁰ However, many adults experience significant declines in functional disability over time before MI or stroke.^13-17 Studies assessing changes in functional disability before and after MI and stroke in older adults have yielded mixed results.^13-17 Some studies observed that MI increased the risk of a decline in physical functioning but that the effect was explained by comorbidity¹³ or limited to subjects with recent MI (<1 year)¹⁵. Other studies have found that the average rate of yearly increase in functional disability after MI or stroke did not change or lessened compared to changes in disability before the vascular event.^{14, 16, 17} We observed that acute MI is associated with increases in functional impairments years after the event, even after accounting for individuals’ changes in functional disability before the MI and acutely after the event. We likely detected a significant increase in rates of disability after MI and stroke because we had larger sample size, more outcome data before and after the vascular events, and included IADL changes.^{14, 16, 17}

We also observed that stroke significantly increased the odds of survivors’ developing moderate-to-severe cognitive impairment acutely at the time of the event with a non-significant trend for MI to do so. After accounting for changes in cognitive impairment before and acutely after the vascular events, we did not detect a significant increase in survivors’ subsequent risk of cognitive impairment over the long-term, consistent with previous stroke research.^{18, 35} Our results may differ from others^{8, 36, 37} because of cohort differences and we studied incident MI and stroke events, used a different measure of cognitive impairment, and controlled for cognitive changes before the event and acutely at the time of the event. It is also possible that our measure of cognitive impairment was not sensitive to more subtle long-term changes in executive functioning, attention, or processing speed present after acute MI or stroke.

We did not detect a significant increase in the presence of substantial depressive symptoms after MI hospitalization³⁸ when controlling for depressive symptoms prior to the acute event, similar to previous MI research.¹⁵ Still, we observed that having a higher number of functional impairments after acute MI or stroke was independently associated with substantial depressive symptoms, consistent with studies of other acute medical events like pneumonia and sepsis.^{32, 33} Taken together, these studies suggest that it is not the acute medical event itself but how disabling the event is that incrementally increases the risk of developing depression.

The key finding of our study—that functional disability mounts after an MI and stroke—has clinical implications. Clinical practice guidelines and quality improvement programs recommend assessments of function, cognition, and depression in stroke patients prior to discharge and also in the post-acute settings.^{39, 40} Our results suggest that survivors of MI merit functional assessment prior to hospital discharge because they may have acutely acquired functional impairments. Moreover, our results suggest that survivors of MI and stroke also warrant monitoring for mounting functional disability over the years after the event. Clinicians can screen survivors for difficulty performing ADLs or IADLs as these were our measures. Those survivors who acquire new functional impairments further warrant screening for depression because, in our study, the odds of developing substantial depressive symptoms increased 20% with each new functional limitation gained after MI and 34% greater per each new functional limitation gained after stroke.

We observed that progression of functional disability is relatively worse in some patients than in others. Survivors with fewer baseline limitations were most likely to gain functional limitations. These intriguing results should be interpreted with caution. We may not have detected significant changes in physical functioning after MI or stroke in patients with more baseline limitations because acute clinical events have less impact⁹ or our measures are relatively insensitive in patients with moderate-to-severe pre-existing functional disability. The patients with moderate-to-severe disability at baseline had steep functional declines over the years before MI or stroke and these slopes did not appear to change over the long-term after the events. Those with substantial functional decline before the MI or stroke event likely had CVD or other disabling clinical events before the study period while those with no disability at baseline may have experienced their first event. This vulnerable population warrants further study as to the reasons for, and prevention of, these steep functional declines.

The World Health Organization's International Classification of Functioning, Disability, and Health describes the complex interplay between a health condition like MI or stroke, functioning, and disability with contributions by pathophysiological processes, the impact of the health condition on the survivor, and contextual factors such as each survivor's personal and environmental resources.^{39, 41, 42} Survivors of MI or stroke may experience continued functional decline over the years after the event because of exacerbation of underlying neurodegenerative disease, subclinical vascular events (e.g, cerebral infarcts), or ongoing vascular and neuronal damage.^43-45 Although we did not aim to identify the activities, behaviors, or contextual factors that predict long-term functional declines after MI or stroke, studies have shown that having Medicaid or no health insurance, lack of cardiac rehabilitation referral, and moderate-to-high perceived stress increase the risk for functional decline after MI.^{10, 16, 46} Conversely, vascular risk factor modification may reduce the long-term risk of cognitive impairment after stroke.⁴⁷ Research is needed to better elucidate the mechanisms and risk factors for functional decline over the long-term after MI or stroke.

Since more adults are elderly and surviving MI and stroke, we need to better understand how to best care for an older population with CVD and functional or cognitive impairments. While much is known about the benefits and risks of rehabilitation, transitions of care, and pharmacological management in patients during the early months after MI or stroke, less is known about the benefits and risks of these interventions in functionally or cognitively disabled patients. How do we best apply or tailor effective CVD interventions to an older survivor population with functional or cognitive impairments? Research to identify efficacious, patient-centered clinical interventions and systems of care for the large and growing population of older adults with CVD and functional or cognitive impairments is urgently needed.

Our study has some potential limitations. Our results are generalizable to older, community-dwelling survivors of MI and stroke including those who move to nursing homes. Although survivors were sampled from a nationally-representative cohort, the small numbers, particularly within functional strata, may limit generalizability. Still, our study was larger than other studies of this topic. Although the HRS-Medicare data do not include the results of testing for cardiac biomarkers, the principal and secondary diagnoses of MI and dates for a hospitalization are highly accurate in CMS records. Our analysis does not include individuals with out-of-hospital or severe events that do not survive hospitalization. The use of proxies by survivors during longitudinal follow-up is a potential limitation. Although 28% of stroke survivors and 8% of MI survivors converted to proxy-respondents at some point during follow-up, sensitivity analyses with non-response propensity score adjustment found similar results for the depression outcome.

Our estimates of declines in physical and cognitive functioning may be underestimated because of survivor bias and outmigration due to death (i.e., death of patients with greater disability at baseline or patients with severe deficits after MI or stroke). We were unable to examine the role of clinical variables such as disease severity, treatments, atrial fibrillation, or revascularization surgery. However, among patients with coronary artery disease, long-term cognitive decline may occur similarly in those who do and do not undergo surgery¹¹ and physical disability after one year may be less in those undergoing surgery compared to percutaneous coronary intervention.⁴⁸ We did not identify personal and environmental factors associated with long-term functional declines after MI or stroke. Although we did not study disease-specific functional measures, we studied functional and cognitive impairments using standardized, valid instruments that are commonly used in older adults with vascular and non-vascular conditions. Our outcome measures represent clinically meaningful changes in patient-centered outcomes: physical and cognitive functioning.

Conclusion

The population of MI and stroke survivors is substantial and growing. Given improved event survival, the increasing number of older adults, and recent disability trends, the chronic care of MI and stroke survivors will prove costly. In this nationally-representative cohort of older Americans, we found that hospitalizations for MI and stroke were associated with significant acute and long-term increases in functional disability. Our results suggest that survivors of MI merit functional assessment prior to hospital discharge because they may have acutely acquired functional impairments. Moreover, our results suggest that survivors of MI and stroke also warrant monitoring for mounting functional disability over the years after the event.