Race/Ethnic and Stroke Subtype Differences in Poststroke Functional Recovery After Acute Rehabilitation

Explainable machine learning for predicting neurological outcome in hemorrhagic and ischemic stroke patients in critical care

Aim

The objective of this study is to develop accurate machine learning (ML) models for predicting the neurological status at hospital discharge of critically ill patients with hemorrhagic and ischemic stroke and identify the risk factors associated with the neurological outcome of stroke, thereby providing healthcare professionals with enhanced clinical decision-making guidance.

Materials and methods

Data of stroke patients were extracted from the eICU Collaborative Research Database (eICU-CRD) for training and testing sets and the Medical Information Mart for Intensive Care IV (MIMIC IV) database for external validation. Four machine learning models, namely gradient boosting classifier (GBC), logistic regression (LR), multi-layer perceptron (MLP), and random forest (RF), were used for prediction of neurological outcome. Furthermore, shapley additive explanations (SHAP) algorithm was applied to explain models visually.

Results

A total of 1,216 hemorrhagic stroke patients and 954 ischemic stroke patients from eICU-CRD and 921 hemorrhagic stroke patients 902 ischemic stroke patients from MIMIC IV were included in this study. In the hemorrhagic stroke cohort, the LR model achieved the highest area under curve (AUC) of 0.887 in the test cohort, while in the ischemic stroke cohort, the RF model demonstrated the best performance with an AUC of 0.867 in the test cohort. Further analysis of risk factors was conducted using SHAP analysis and the results of this study were converted into an online prediction tool.

Conclusion

ML models are reliable tools for predicting hemorrhagic and ischemic stroke neurological outcome and have the potential to improve critical care of stroke patients. The summarized risk factors obtained from SHAP enable a more nuanced understanding of the reasoning behind prediction outcomes and the optimization of the treatment strategy.

Health Care Disparities in Stroke Rehabilitation

Stroke outcomes are influenced by factors such as education, lifestyle, and access to care, which determine the extent of functional recovery. Disparities in stroke rehabilitation research have traditionally included age, race/ethnicity, and sex, but other areas make up a gap in the literature. This article conducted a literature review of original research articles published between 2008 and 2022. The article also expands on research that highlights stroke disparities in risk factors, rehabilitative stroke care, language barriers, outcomes for stroke survivors, and interventions focused on rehabilitative stroke disparities.

Racial and Ethnic Disparities in the Medical Management of Poststroke Complications Among Patients With Acute Stroke

BACKGROUND

To inform clinical practice, we sought to identify racial and ethnic differences in the medical management of common poststroke complications.

METHODS AND RESULTS

A cohort of acutely hospitalized, first-time non-Hispanic White (NHW), non-Hispanic Black, and Hispanic patients with stroke was identified from electronic medical records of 51 large health care organizations (January 1, 2003 to December 5, 2022). Matched propensity scores were used to account for baseline differences. Primary outcomes included receipt of medication(s) associated with the management of the following poststroke complications: arousal/fatigue, spasticity, mood, sleep, neurogenic bladder, neurogenic bowel, and seizure. Differences were measured at 14, 90, and 365 days. Subgroup analyses included differences restricted to patients with ischemic stroke, younger age (<65 years), and stratified by decade (2003-2012 and 2013-2022). Before matching, the final cohort consisted of 348 286 patients with first-time stroke. Matching resulted in 63 722 non-Hispanic Black-NHW pairs and 24 009 Hispanic-NHW pairs. Non-Hispanic Black (versus NHW) patients were significantly less likely to be treated for all poststroke complications, with differences largest for arousal/fatigue (relative risk (RR), 0.58 [95% CI, 0.54-0.62]), spasticity (RR, 0.64 [95% CI, 0.0.62-0.67]), and mood disorders (RR, 0.72 [95% CI, 0.70-0.74]) at 14 days. Hispanic-NHW differences were similar, albeit with smaller magnitudes, with the largest differences present for spasticity (RR, 0.67 [95% CI, 0.63-0.72]), arousal/fatigue (RR, 0.77 [95% CI, 0.70-0.85]), and mood disorders (RR, 0.79 [95% CI, 0.77-0.82]). Subgroup analyses revealed similar patterns for ischemic stroke and patients aged <65 years. Disparities for the current decade remained significant but with smaller magnitudes compared with the prior decade.

CONCLUSIONS

There are significant racial and ethnic disparities in the treatment of poststroke complications. The differences were greatest at 14 days, outlining the importance of early identification and management.

Association Between Social Determinants of Health and Communication Difficulties in Poststroke U.S. Hispanic and Non-Hispanic White Populations

PURPOSE

The relationship among ethnicity, social determinants of health (SDOH), and disparities in poststroke outcomes is complex, and the impact on communication difficulties is unclear. This study investigated the presence and nature of communication difficulties in poststroke non-Hispanic White (PsnHw) and Hispanic U.S. populations using population-level data.

METHOD

We performed a cross-sectional analysis of 2,861 non-Hispanic White and 353 Hispanic poststroke respondents included in the 2014-2018 National Health Interview Survey. Respondents self-reported difficulties communicating in their usual language, in addition to providing information relating to demographics and lifestyle, health care access and utilization, health status, and SDOH. We used univariate statistics, generalized linear models, and an exploratory mediation analysis, to characterize the pattern of differences between these cohorts, examine associations between variables and communication difficulties, and determine the potential intermediate role of cumulative SDOH on the likelihood of reporting communication difficulties.

RESULTS

Findings indicated a more challenging life context for the poststroke Hispanic population due to SDOH disparities. Poverty and Internet use were associated with greater and lower odds of communication difficulties for PsnHw, respectively. The mediation analysis showed that ethnicity significantly affected communication difficulties, but only when mediated by SDOH. SDOH accounted for approximately two thirds of the total effect on reporting communication difficulties.

CONCLUSIONS

This study underscores the need for uniform measures of SDOH in prospective research and for interventions aimed at mitigating health disparities through addressing disparities in SDOH. Future research should focus on evaluating the effectiveness of such strategies in diverse ethnic and socioeconomic poststroke populations.

SUPPLEMENTAL MATERIAL

https://doi.org/10.23641/asha.24521419.

Racial Disparities in Utilization of Emergency Medical Services and Related Impact on Poststroke Disability

BACKGROUND

Prompt seeking of emergency medical services (EMS) assistance at stroke onset is critical to minimize poststroke disability.

OBJECTIVE

The aim was to study how racial differences in EMS decision-relevant factors and EMS use impact stroke care and disability outcomes.

DESIGN

A prospective observational study.

PARTICIPANTS

A total of 1168 acute ischemic stroke patients discharged from April 2016 to October 2017 at a safety net hospital were included; 108 patients were surveyed before discharge.

MEASURES

(1) Prehospital delay: EMS use, timely hospital arrival; (2) Stroke care: alteplase receipt and inpatient rehab; (3) Outcomes: Functional improvement at discharge (admission minus discharge scores on National Institutes of Health Stroke Scale), 90-day modified Rankin Scale; (4) EMS decision-relevant factors: Stroke symptom knowledge, source of knowledge, unfavorable past EMS/care experiences, and financial barriers to EMS use.

RESULTS

Despite more Black patients using EMS than Whites/Asians (56% vs. 48%, P =0.003), their timely hospital arrival was 30% less likely. Adjusted for stroke severity, receipt of alteplase, and inpatient rehab were similar, but Black patients fared worse on functional improvement at discharge (among severe strokes, 2.4 National Institutes of Health Stroke Scale points less improvement, P <0.01), and on functional normalcy at 90 days (modified Rankin Scale score 0-1 being 60% less likely across severity categories) ( P <0.01). Fewer Black patients knew any stroke symptoms before the stroke (72% vs. 87%, P =0.03), and fewer learned about stroke from providers ( P =0.01). Financial barriers and provider mistrust were similar.

CONCLUSIONS

Black patients had less knowledge of stroke symptoms, more care-seeking delay, and poorer outcomes. Including stroke education as a standard of chronic disease care may mitigate stroke outcome disparities.

Essential information for neurorecovery clinical trial design: trajectory of global disability in first 90 days post-stroke in patients discharged to acute rehabilitation facilities

BACKGROUND

Many stroke recovery interventions are most beneficial when started 2-14d post-stroke, a time when patients become eligible for inpatient rehabilitation facilities (IRF) and neuroplasticity is often at its peak. Clinical trials focused on recovery need to expand the time from this plasticity to later outcome timepoints.

METHODS

The disability course of patients with acute ischemic stroke (AIS) and intracranial hemorrhage (ICH) enrolled in Field Administration of Stroke Therapy Magnesium (FAST-MAG) Trial with moderate-severe disability (modified Rankin Scale [mRS] 3-5) on post-stroke day4 who were discharged to IRF 2-14d post-stroke were analyzed.

RESULTS

Among 1422 patients, 446 (31.4%) were discharged to IRFs, including 23.6% within 2-14d and 7.8% beyond 14d. Patients with mRS 3-5 on day4 discharged to IRFs between 2-14d accounted for 21.7% (226/1041) of AIS patients and 28.9% (110/381) of ICH patients, (p < 0.001). Among these AIS patients, age was 69.8 (± 12.7), initial NIHSS median 8 (IQR 4-12), and day4 mRS = 3 in 16.4%, mRS = 4 in 50.0%, and mRS = 5 in 33.6%. Among these ICH patients, age was 62.4 (± 11.7), initial NIHSS median 9 (IQR 5-13), day 4 mRS = 3 in 9.4%, mRS = 4 in 45.3%, and mRS = 5 in 45.3% (p < 0.01 for AIS vs ICH). Between day4 to day90, mRS improved ≥ 1 levels in 72.6% of AIS patients vs 77.3% of ICH patients, p = 0.3. For AIS, mRS improved from mean 4.17 (± 0.7) to 2.84 (± 1.5); for ICH, mRS improved from mean 4.35 (± 0.7) to 2.75 (± 1.3). Patients discharged to IRF beyond day14 had less improvement on day90 mRS compared with patients discharged between 2-14d.

CONCLUSIONS

In this acute stroke cohort, nearly 1 in 4 patients with moderate-severe disability on post-stroke day4 were transferred to IRF within 2-14d post-stroke. ICH patients had nominally greater mean improvement on mRS day90 than AIS patients. This course delineation provides a roadmap for future rehabilitation intervention studies.

Precision aiming performance with the paretic upper limb is associated with center of pressure patterns in individuals with chronic stroke

BACKGROUND

Individuals with stroke commonly demonstrate upper-limb sensorimotor impairments. Upper-limb tasks occur against a background level of postural control and thus require a flexible postural control system to facilitate performance. Anterior precision aiming tasks, for example, benefit from lower medial-lateral (ML) center of pressure (COP) fluctuations (where increased fluctuations erode performance) relative to anterior-posterior (AP) fluctuations (where increased fluctuations do not strongly influence performance). After stroke, individuals may compensate for upper-limb impairments by increasing trunk movement which increases overall COP fluctuations and thus may make it more difficult to modulate COP in a task-sensitive manner.

RESEARCH QUESTION

Do upper-limb task demands modulate COP movement patterns after stroke?

METHODS

In this cross-sectional study, adults with chronic stroke (n = 23) and unilateral upper-limb impairments were immersed in a virtual environment displaying an anterior target. Participants aimed to maintain the position of a virtual laser pointer (via handheld controller) in the target with each hand. COP was concurrently recorded. Mixed effects models and correlations were used to detect differences in COP patterns between limbs and movement planes and evaluate associations between task performance and COP patterns, respectively.

RESULTS

Participants showed greater COP standard deviation and regularity in the AP compared to the ML direction. The magnitude of difference between AP and ML COP metrics was greater using the nonparetic limb. Task performance was moderately and positively associated with task-sensitive COP patterns (i.e., higher AP:ML ratios of COP metrics) using the paretic upper limb. Participants consistently demonstrated high levels of task performance and task-sensitive COP movement patterns using the nonparetic limb.

SIGNIFICANCE

Impairments in postural control after stroke may be related to the upper limb used. It is important to recognize the role of directional COP variability and regularity in the context of a task goal after stroke.

Recovery of Patients With Upper Limb Paralysis Due to Stroke Who Underwent Intervention Using Low-Frequency Repetitive Transcranial Magnetic Stimulation Combined With Occupational Therapy: A Retrospective Cohort Study

OBJECTIVES

The combination of repetitive transcranial magnetic stimulation (rTMS) and motor practice is based on the theory of neuromodulation and use-dependent plasticity. Predictive planning of occupational therapy (OT) is important for patients with rTMS conditioning. Recovery characteristics based on the severity of pretreatment upper extremity paralysis can guide the patient's practice plan for using the paretic hand. Therefore, we evaluated the recovery of patients with upper limb paralysis due to stroke who underwent a novel intervention of rTMS combined with OT (NEURO) according to the severity of upper limb paralysis based on the scores of the Fugl-Meyer assessment for upper extremity (FMA-UE) with recovery in proximal upper extremity, wrist, hand, and coordination.

MATERIALS AND METHODS

In this multicenter retrospective cohort study, the recovery of 1397 patients with upper limb paralysis was analyzed by severity at six hospitals that were accredited by the Japanese Stimulation Therapy Society for treatment. The delta values of the FMA-UE scores before and after NEURO were compared among the groups with severe, moderate, and mild paralysis using the generalized linear model.

RESULTS

NEURO significantly improved the FMA-UE total score according to the severity of paralysis (severe = 5.3, moderate = 6.0, and mild = 2.9). However, when the FMA-UE subscores were analyzed separately, the results indicated specific improvements in shoulder/elbow, wrist, fingers, and coordination movements, depending on the severity.

CONCLUSIONS

This study had enough patients who were divided according to severity and stratified by lesion location and handedness parameters. Our results suggest that independently of these factors, the extent of recovery of upper limb motor parts after NEURO varies according to the severity of paralysis.

Upper Limb Function Recovery by Combined Repetitive Transcranial Magnetic Stimulation and Occupational Therapy in Patients with Chronic Stroke According to Paralysis Severity

Repetitive transcranial magnetic stimulation (rTMS) with intensive occupational therapy improves upper limb motor paralysis and activities of daily living after stroke; however, the degree of improvement according to paralysis severity remains unverified. Target activities of daily living using upper limb functions can be established by predicting the amount of change after treatment for each paralysis severity level to further aid practice planning. We estimated post-treatment score changes for each severity level of motor paralysis (no, poor, limited, notable, and full), stratified according to Action Research Arm Test (ARAT) scores before combined rTMS and intensive occupational therapy. Motor paralysis severity was the fixed factor for the analysis of covariance; the delta (post-pre) of the scores was the dependent variable. Ordinal logistic regression analysis was used to compare changes in ARAT subscores according to paralysis severity before treatment. We implemented a longitudinal, prospective, interventional, uncontrolled, and multicenter cohort design and analyzed a dataset of 907 patients with stroke hemiplegia. The largest treatment-related changes were observed in the Limited recovery group for upper limb motor paralysis and the Full recovery group for quality-of-life activities using the paralyzed upper limb. These results will help predict treatment effects and determine exercises and goal movements for occupational therapy after rTMS.

Stroke Rehabilitation: Recent Progress and Future Promise

Significant advancements in acute stroke medical management have changed stroke rehabilitation. In addition, an ever-changing health care ecosystem and heightened awareness of continued and new challenges requires that the occupational therapy profession consider new, innovative, and pragmatic approaches to measurement, intervention, and health services research, and clinical practice. The profession must elevate the focus and rigor of research examining occupation and participation after stroke, and their associations with health. Intervention research must progress beyond early phase pilot studies to a robust collection of meaningful large multisite studies that demonstrate the effectiveness of our interventions and the effectiveness of wide-scale implementation to ensure quality and consistent delivery of evidence-based practices in occupational therapy. These studies must address the accessibility of these practices for all people who have sustained stroke, and particularly those people who are most vulnerable to inaccessible stroke rehabilitation service delivery systems.

The differential impact of the COVID-19 pandemic on healthcare utilization disruption for community-dwelling individuals with and without acquired brain injury

•

All groups experienced notable disruption in health care utilization.

•

Disruption in care for traumatic brain injury or mental health was associated with decreased health-related quality of life.

•

Telehealth was a viable alternative to in-person visits.

•

Telehealth is not a panacea and should be adopted using a nuanced approach.

Objective

To delineate health care disruption for individuals with acquired brain injury (ABI) during the peak of the pandemic and to understand the impact of health care disruption on health-related quality of life (HRQoL).

Design

Cross-sectional survey.

Setting

General community.

Participants

Volunteer sample of adults with traumatic brain injury (TBI; n=33), adults with stroke (n=66), and adults without TBI or stroke (n=108) with access to the internet and personal technology (N=207).

Interventions

Not applicable.

Main Outcome Measures

Not applicable.

Results

Participants with TBI and stroke reported high rates of disruption in care specific to their diagnosis (53%-54.5%), while participants across all groups reported disruption for major medical care (range, 68.2%-80%), general health care (range, 60.3%-72.4%), and mental health care (range, 31.8%-83.3%). During the pandemic, participants with TBI and stroke used telehealth for care specific to their diagnosis (40.9%-42.4%), whereas all participants used telehealth for major medical care (range, 50%-86.7%), general health care (range, 31.2%-53.3%), and mental health care (range, 53.8%-72.7%). Disruption in TBI or stroke care and type of ABI explained 27.1% of the variance in HRQoL scores (F_2,95=16.82, P<.001, R²=0.262), and disruption in mental health care explained 14.8% of the variance (F_1,51=8.86, P=.004, R²=0.148).

Conclusions

Individuals with and without ABI experienced pronounced disruption in health care utilization overall. However, individuals who experienced a disruption in care specific to TBI or mental health care were most vulnerable to decreased HRQoL. Telehealth was a viable alternative to in-person visits for individuals with and without ABI, but limitations included difficulty with technology, difficulty with comprehensive examination, and decreased rapport with providers.

Social Determinants of Health in Physiatry: Challenges and Opportunities for Clinical Decision Making and Improving Treatment Precision

Physiatry is a medical specialty focused on improving functional outcomes in patients with a variety of medical conditions that affect the brain, spinal cord, peripheral nerves, muscles, bones, joints, ligaments, and tendons. Social determinants of health (SDH) play a key role in determining therapeutic process and patient functional outcomes. Big data and precision medicine have been used in other fields and to some extent in physiatry to predict patient outcomes, however many challenges remain. The interplay between SDH and physiatry outcomes is highly variable depending on different phases of care, and more favorable patient profiles in acute care may be less favorable in the outpatient setting. Furthermore, SDH influence which treatments or interventional procedures are accessible to the patient and thus determine outcomes. This opinion paper describes utility of existing datasets in combination with novel data such as movement, gait patterning and patient perceived outcomes could be analyzed with artificial intelligence methods to determine the best treatment plan for individual patients in order to achieve maximal functional capacity.

Rationale for a Clinical Trial That Compares Acute Stroke Rehabilitation at Inpatient Rehabilitation Facilities to Skilled Nursing Facilities: Challenges and Opportunities

In the United States, approximately 400,000 patients with acute stroke are discharged annually to inpatient rehabilitation facilities (IRFs) or skilled nursing facilities (SNFs). Typically, IRFs provide time-intensive therapy for an average of 2-3 weeks, whereas SNFs provide more moderately intensive therapy for 4-5 weeks. The factors that influence discharge to an IRF or SNF are multifactorial and poorly understood. The complexity of these factors in combination with subjective clinical indications contributes to large variations in the use of IRFs and SNFs. This has significant financial implications for health care expenditure, given that stroke rehabilitation at IRFs costs approximately double that at SNFs. To control health care spending without compromising outcomes, the Institute of Medicine has stated that policy reforms that promote more efficient use of IRFs and SNFs are critically needed. A major barrier to the formulation of such policies is the highly variable and low-quality evidence for the comparative effectiveness of IRF- vs SNF-based stroke rehabilitation. The current evidence is limited by the inability of observational data to control for residual confounding, which contributes to substantial uncertainty around any magnitude of benefit for IRF- vs SNF-based care. Furthermore, it is unclear which specific patients would receive the most benefit from each setting. A randomized controlled trial addresses these issues, because random treatment allocation facilitates an equitable distribution of measured and unmeasured confounders. We discuss several measurement, practical, and ethical issues of a trial and provide our rationale for design suggestions that overcome some of these issues.