Association Between Use of Rehabilitation in the Acute Care Hospital and Hospital Readmission or Mortality in Patients With Stroke.

OBJECTIVE: To examine the association between the number of physical and occupational therapist visits received in the acute care hospital and the risk of hospital readmission or death. DESIGN: Retrospective cohort study of electronic health records and administrative claims data collected for 2.25 years (January 1, 2016-March 30, 2018). SETTING: Twelve acute care hospitals in a large health care system in western Pennsylvania. PARTICIPANTS: Adults (N=8279) discharged with a primary stroke diagnosis. INTERVENTIONS: The exposure was number of physical and occupational therapist visits during the acute care stay. MAIN OUTCOME MEASURES: Generalized linear mixed models were estimated to examine the relationship between therapy use and 30- and 7-day hospital readmission or death (outcome), controlling for patient demographic and clinical characteristics. RESULTS: The 30- and 7-day readmission or death rates were 16.0% and 5.7%, respectively. The number of therapist visits was inversely related to the risk of 30-day readmission or death. Relative to no therapist visits, the odds of readmission or death were 0.70 (95% confidence interval [CI], 0.54-0.90) for individuals who received 1-2 visits, 0.59 (95% CI, 0.43-0.81) for 3-5 visits, and 0.57 (95% CI, 0.39-0.81) for >5 visits. A similar pattern was seen for the 7-day outcome, with slightly larger effect sizes. Effects were also greater in individuals with more mobility limitations on admission and for those discharged to a postacute care facility vs home. CONCLUSIONS: There was an inverse relationship between the number of therapist visits and risk for readmission or death for patients with stroke discharged from an acute care hospital. Effects differed by time to the event (30d vs 7d), discharge location, and mobility limitations on admission.

Sedentary time and physical activity patterns of stroke survivors during the inpatient rehabilitation week.

Physical activity is recommended after stroke. However, the rehabilitation day is largely spent sedentary. Understanding patterns of physical activity across the rehabilitation week may help identify opportunities to improve participation. We aimed to examine: (1) differences between weekday and weekend sedentary time and physical activity, (2) the pattern of 24-h rehabilitation activity. Participants with stroke (n = 29) wore an activity monitor continuously during the final 7-days of inpatient rehabilitation. Linear mixed models (adjusted for waking hours) were performed with activity (sedentary, steps per day, walking time) as the dependent variable, and day type (weekday or weekend) as the independent variable. Patterns of upright time during the 24-h period were determined by averaging daily activity in 60-min intervals and generating a heat map of activity levels as a function of time. Participant mean age was 69 (SD 13) years (52% male) and mean National Institutes of Health Stroke Scale score was 7.0 (SD, 5.5). There was no significant difference in sedentary time between weekdays and weekends. At the weekend, participants spent 8.4 min less time walking (95% CI, -12.1 to -4.6) taking 624 fewer steps/day (95% CI, -951 to -296) than during the week. Activity patterns showed greatest upright time in the morning during the week. Afternoon and evening activities were low on all days. Sedentary time did not change across the 7-day rehabilitation week, but less walking activity occurred on the weekend. There are opportunities for stroke survivors to increase physical activity during afternoons and evenings and on weekend mornings during rehabilitation.

Machine learning to predict mortality after rehabilitation among patients with severe stroke.

Stroke is among the leading causes of death and disability worldwide. Approximately 20-25% of stroke survivors present severe disability, which is associated with increased mortality risk. Prognostication is inherent in the process of clinical decision-making. Machine learning (ML) methods have gained increasing popularity in the setting of biomedical research. The aim of this study was twofold: assessing the performance of ML tree-based algorithms for predicting three-year mortality model in 1207 stroke patients with severe disability who completed rehabilitation and comparing the performance of ML algorithms to that of a standard logistic regression. The logistic regression model achieved an area under the Receiver Operating Characteristics curve (AUC) of 0.745 and was well calibrated. At the optimal risk threshold, the model had an accuracy of 75.7%, a positive predictive value (PPV) of 33.9%, and a negative predictive value (NPV) of 91.0%. The ML algorithm outperformed the logistic regression model through the implementation of synthetic minority oversampling technique and the Random Forests, achieving an AUC of 0.928 and an accuracy of 86.3%. The PPV was 84.6% and the NPV 87.5%. This study introduced a step forward in the creation of standardisable tools for predicting health outcomes in individuals affected by stroke.

Factors related to interrupted inpatient stroke rehabilitation due to acute care transfer or death.

OBJECTIVES: To identify factors that are independently related to interrupted stroke rehabilitation due to acute care transfer or death. METHODS: Medical records of stroke inpatients admitted from 2012 to 2017 were reviewed. Stroke inpatients with interrupted stroke rehabilitation due to acute care transfer or death were enrolled into the case group. Those without interruption admitted in the same month were randomly selected into the control group (case to control ratio of 1 : 5). Ten clinical factors were studied. RESULTS: Among stroke inpatients, 3.2% were transferred to acute care facilities and 0.2% died. The most common causes of acute care transfer were respiratory tract infection, intracranial hemorrhage, recurrent ischemic stroke, ischemic heart disease, and seizure. Three factors were found to be significantly associated with interrupted stroke rehabilitation, i.e. presence of feeding tube, presence of anemia and age. Our results also revealed significant association between presence of feeding tube and respiratory tract infection (p = 0.005). CONCLUSION: Feeding tube, anemia and old age were identified as independent predictors of interrupted stroke rehabilitation due to acute care transfer or death. Interventions to reduce severe complications should be implemented in order to prevent interruption of rehabilitation process and to reduce the patient transfer rate.

Reliability and validity of the long-distance corridor walk among stroke survivors.

OBJECTIVE: To identify the psychometric properties of the Long-Distance Corridor Walk (LDCW) among community-dwelling stroke survivors. DESIGN: Cross-sectional. SUBJECTS: Twenty-five stroke survivors and 25 healthy older adults. METHODS: The LDCW was administered to the 25 stroke survivors on 2 separate days with a 7-day interval. Fugl-Meyer Assessment for the Lower Extremities (FMA-LE), measurement of lower limb muscle strength, Berg Balance Scale (BBS), limit of stability (LOS), Narrow-Corridor Walk Test (NCWT), Timed Up and Go (TUG) test, and the Community Integration Measure-Cantonese version (CIM) were performed on either day. The healthy older adults completed the LDCW once, and the results were recorded by a random rater. RESULTS: The LDCW showed excellent inter-rater reliability and test-retest reliability, and significant correlations with FMA-LE, BBS, TUG, and NCWT. A cut-off score of 127.5 m for the 2-min walk and 426.69 s for the 400-m walk distinguished stroke survivors from healthy older adults. The MDC in the LDCW in the 2-min walk and 400-m walk were 18.69 m and 121.43 s, respectively. CONCLUSION: The LDCW is a reliable clinical measurement tool for the assessment of advanced walking capacity in stroke survivors.

Long-Term Trends in Stroke Survivors Discharged to Care Homes: The South London Stroke Register.

Background and Purpose- Care homes provide care to many stroke survivors, yet little is known about changes in care home use over time. We aim to determine trends in discharge to care homes, to explore the characteristics of stroke survivors over time (1995-2018), and to identify the associations between these characteristics and discharge to care homes poststroke. Methods- Using data from the South London Stroke Register between 1995 and 2018, we estimated the proportions discharged to care homes and their characteristics over time, assessed by tests for trends. Multivariable logistic regression models were built to assess the associations between their characteristics and discharge destination. Results- Of 4172 stroke survivors, 484 (12%) were discharged to care homes. This proportion has decreased from 24% in 1995 to 2000 to 5% in 2013 to 2018. The mean age of those discharged to care homes has increased over time, from 73 to 75 (P<0.001). Among stroke survivors discharged to a care home, the proportion with a prestroke Barthel Index <15 has also increased over time from 7% to 21% (P=0.027), while the proportion with a 7-day poststroke Barthel Index <15 remains largely unchanged over time (93% in 1995-2000, 90% in 2013-2018). The characteristics most strongly associated with discharge to care homes were (odds ratio [95% CI]) age (1.05 [1.04-1.07] per year), stroke subtype (hemorrhagic; 0.64 [0.43-0.95]), stroke severity (Glasgow Coma Scale score, <13; 1.67 [1.19-2.35]), failed swallow test at admission (1.65 [1.20-2.25]), 7-day poststroke Barthel Index <15 (3.58 [2.20-6.03]), and a longer hospital stay (1.02 [1.02-1.03] per day). Conclusions- Over >20 years, there has been an 80% reduction in the proportion of stroke survivors discharged to care homes, influenced by changes in the demographics, disability, and stroke care in the underlying stroke population. In those moving to care homes, the level of poststroke disability remains high, requiring continued attention and investment.

Trends in Outcomes of Patients With Ischemic Stroke Treated Between 2002 and 2016: Insights From a Chinese Cohort.

BACKGROUND: Little is known about long-term trends in outcomes of patients with ischemic stroke in China. We aimed to assess longitudinal trends in these outcomes over the past 15 years in China and explore possible factors behind the trends. METHODS AND RESULTS: Patients with ischemic stroke admitted to the Department of Neurology at West China Hospital were prospectively and consecutively enrolled in a central registry since 2002, and the present study analyzed data from those admitted to hospital within 7 days of stroke during the period 2002 to 2016. Patients were binned into three 5-year intervals for temporal analysis. Death, disability, and death/disability at 3 and 12 months after stroke were compared among the time intervals across the entire sample and in subsets stratified by age (<65 or ≥65 years). To explore the possible factors related to the trends in outcomes, interaction between the factors and time on outcomes was entered separately into the multivariable logistic regression model. Of 6462 patients with ischemic stroke in the final analysis, 3837 (59.4%) were men, and mean age was 64.2 years (SD, 13.7). Mean age at stroke onset and National Institutes of Health Stroke Scale score at admission decreased significantly during the 15-year period (P<0.001). Between 2002 to 2006 and 2012 to 2016, cumulative incidences declined significantly for death at 3 months (from 9.6% to 6.4%), disability at 3 months (from 36.8% to 28.7%), and death/disability at 3 months (from 42.9% to 33.3%), as well as for death at 12 months (from 15.9% to 10.7%), disability at 12 months (from 23.2% to 17.6%), and death/disability at 12 months (from 35.4% to 26.4%; all P<0.001). The decreases in disability and death/disability at 3 and 12 months between 2002 to 2006 and 2012 to 2016 remained significant after adjusting for confounders, and the results were similar for the entire cohort and for subgroups of patients <65 or ≥65 years. Only interactions of National Institutes of Health Stroke Scale score on admission and time period (2012-2016) were found to significantly correlate with disability and death/disability at 3 and 12 months (all P≤0.03). CONCLUSIONS: Our study from a large medical center in southwest China suggests that since 2002, risks of disability and death/disability at 3 and 12 months after ischemic stroke have declined. This appears to be due, at least in part, to a significant decline in National Institutes of Health Stroke Scale score on admission, which may reflect greater public awareness of stroke detection, willingness to seek medical attention, and ease of access to healthcare infrastructure. The factors behind this apparent improvement require further study.

Life Expectancy after Stroke Based On Age, Sex, and Rankin Grade of Disability: A Synthesis.

BACKGROUND: Stroke is a leading cause of death and disability in the developed world. The major factor affecting long term survival (other than age) is known to be the severity of disability. Yet to our knowledge there are no studies reporting life expectancies stratified by both age and severity. Remaining life expectancy is a key measure of health. METHODS: We identified 11 long-term follow-up studies of stroke patients that reported the multivariate effects of age, sex, the modified Rankin Scale (mRS) grade of disability, and other factors. From these we computed the composite effects of these factors on survival, then used these to calculate age-, sex-, and mRS-specific mortality rates. Finally we used the rates to construct life tables, and hence obtain life expectancies. RESULTS: Life expectancy varies by age, sex, and mRS. The life expectancies of males age 70, for example, were 13, 13, 11, 8, 6, and 5 years for Rankin Grades 0-5, respectively, representing reductions of 1, 1, 3, 6, 8, and 9 years from the corresponding general population figure. CONCLUSIONS: These figures demonstrate the importance of rehabilitation following stroke, and can be used in discussion of public policy and benchmarking of future results.

Timing of Direct Enteral Tube Placement and Outcomes after Acute Stroke.

BACKGROUND: Direct enteral feeding tube (DET) placement for dysphagia after stroke is associated with poor outcomes. However, the relationship between timing of DET placement and poststroke mortality and disability is unknown. We sought to determine the risk of mortality and severe disability in patients who receive DET at different times after stroke. METHODS: We used the Ontario Stroke Registry and linked administrative databases to identify patients with acute ischemic stroke or intracerebral hemorrhage between 2003 and 2013 who received DET (gastrostomy or jejunostomy) during their hospital admission. We grouped patients by week of DET placement and evaluated mortality at 30 days and 6 months after DET insertion, and disability at discharge. We used Cox proportional hazard models and multiple logistic regression to determine the association between time from admission to DET placement and outcomes, adjusting for patient and hospital factors. RESULTS: In the study sample of 1367 patients, the median time from admission to DET placement was 17 days. After adjustment, each week of delay to DET placement was associated with lower mortality at 30 days (adjusted hazard ratio [aHR] .88, 95% confidence interval [CI] .79-.98), but not at 6 months (aHR .98, 95% CI .91- 1.05), and a higher likelihood of severe disability at discharge (adjusted odds ratio 1.35, 95% CI 1.13- 1.60). CONCLUSIONS: Later DET placement after stroke was associated with lower 30-day mortality but higher severe disability at discharge. Further research is needed to understand the reasons for these observations and to optimize patient selection and timing of DET.

Evidence-Based Performance Measures and Outcomes in Patients With Acute Ischemic Stroke.

BACKGROUND: Stroke is the leading cause of death in China. Despite the wide dissemination of evidence-based guidelines, data about adherence to these in routine clinical practice are scarce. We conducted a study using a nationwide registry to evaluate the implementation of evidence-based stroke performance indicators and associated guidelines, for patients with an ischemic stroke in China. METHODS AND RESULTS: The China National Stroke Registry is a prospective cohort study, including 12 416 patients diagnosed with acute ischemic stroke from 132 hospitals across China, for 1 year beginning September 2007. Twelve performance indicators were selected to evaluate the quality of stroke care. Multivariable Cox models were used to determine the association between optimal compliance and clinical outcomes. Conformity with performance measures ranged from a median of 6.5% for the use of intravenous tPA (tissue-type plasminogen activator) to 81.8% for early use of antithrombotics. The optimal compliance with all in-hospital measures was associated with 1-year death after admission (hazard ratio, 0.66; 95% CI, 0.55-0.79). The optimal compliance with all discharge measures was associated with the 1-year death after discharge (hazard ratio, 0.55; 95% CI, 0.44-0.69), 1-year stroke recurrence (hazard ratio, 0.81; 95% CI, 0.70-0.93), and favorable functional outcomes (defined as modified Rankin Scale score of ≤2) (hazard ratio, 1.10; 95% CI, 1.04-1.16). CONCLUSIONS: Adherence to evidence-based ischemic stroke care measures in China revealed substantial gaps, and select measures were associated with improved outcomes. These findings support the need for ongoing quality measurement and improvement in stroke care in China.

Very early versus delayed mobilisation after stroke.

BACKGROUND: Very early mobilisation (VEM) is performed in some stroke units and recommended in some acute stroke clinical guidelines. However, it is unclear whether very early mobilisation independently improves outcome after stroke. OBJECTIVES: To determine whether very early mobilisation (started as soon as possible, and no later than 48 hours after onset of symptoms) in people with acute stroke improves recovery (primarily the proportion of independent survivors) compared with usual care. SEARCH METHODS: We searched the Cochrane Stroke Group Trials Register (last searched 31 July 2017). We also systematically searched 19 electronic databases including; CENTRAL; 2017, Issue 7 in the Cochrane Library (searched July 2017), MEDLINE Ovid (1950 to August 2017), Embase Ovid (1980 to August 2017), CINAHL EBSCO (Cumulative Index to Nursing and Allied Health Literature; 1937 to August 2017) , PsycINFO Ovid (1806 to August 2017), AMED Ovid (Allied and Complementary Medicine Database), SPORTDiscus EBSCO (1830 to August 2017). We searched relevant ongoing trials and research registers (searched December 2016), the Chinese medical database, Wanfangdata (searched to November 2016), and reference lists, and contacted researchers in the field. SELECTION CRITERIA: Randomised controlled trials (RCTs) of people with acute stroke, comparing an intervention group that started out-of-bed mobilisation within 48 hours of stroke, and aimed to reduce time-to-first mobilisation, with or without an increase in the amount or frequency (or both) of mobilisation activities, with usual care, where time-to-first mobilisation was commenced later. DATA COLLECTION AND ANALYSIS: Two review authors independently selected trials, extracted data, assessed risk of bias, and applied the GRADE approach to assess the quality of the evidence. The primary outcome was death or poor outcome (dependency or institutionalisation) at the end of scheduled follow-up. Secondary outcomes included death, dependency, institutionalisation, activities of daily living (ADL), extended ADL, quality of life, walking ability, complications (e.g. deep vein thrombosis), patient mood, and length of hospital stay. We also analysed outcomes at three-month follow-up. MAIN RESULTS: We included nine RCTs with 2958 participants; one trial provided most of the information (2104 participants). The median (range) delay to starting mobilisation after stroke onset was 18.5 (13.1 to 43) hours in the VEM group and 33.3 (22.5 to 71.5) hours in the usual care group. The median difference within trials was 12.7 (4 to 45.6) hours. Other differences in intervention varied between trials; in five trials, the VEM group were also reported to have received more time in therapy, or more mobilisation activity.Primary outcome data were available for 2542 of 2618 (97.1%) participants randomized and followed up for a median of three months. VEM probably led to similar or slightly more deaths and participants who had a poor outcome, compared with delayed mobilisation (51% versus 49%; odds ratio (OR) 1.08, 95% confidence interval (CI) 0.92 to 1.26; P = 0.36; 8 trials; moderate-quality evidence). Death occurred in 7% of participants who received delayed mobilisation, and 8.5% of participants who received VEM (OR 1.27, 95% CI 0.95 to 1.70; P = 0.11; 8 trials, 2570 participants; moderate-quality evidence), and the effects on experiencing any complication were unclear (OR 0.88; 95% CI 0.73 to 1.06; P = 0.18; 7 trials, 2778 participants; low-quality evidence). Analysis using outcomes collected only at three-month follow-up did not alter the conclusions.The mean ADL score (measured at end of follow-up, with the 20-point Barthel Index) was higher in those who received VEM compared with the usual care group (mean difference (MD) 1.94, 95% CI 0.75 to 3.13, P = 0.001; 8 trials, 9 comparisons, 2630/2904 participants (90.6%); low-quality evidence), but there was substantial heterogeneity (93%). Effect sizes were smaller for outcomes collected at three-month follow-up, rather than later.The mean length of stay was shorter in those who received VEM compared with the usual care group (MD -1.44, 95% CI -2.28 to -0.60, P = 0.0008; 8 trials, 2532/2618 participants (96.7%); low-quality evidence). Confidence in the answer was limited by the variable definitions of length of stay. The other secondary outcome analyses (institutionalisation, extended activities of daily living, quality of life, walking ability, patient mood) were limited by lack of data.Sensitivity analyses by trial quality: none of the outcome conclusions were altered if we restricted analyses to trials with the lowest risk of bias (based on method of randomization, allocation concealment, completeness of follow-up, and blinding of final assessment), or information about the amount of mobilisation.Sensitivity analysis by intervention characteristics: analyses restricted to trials where the mean VEM time-to-first mobilisation was less than 24 hours, showed an odds of death of 1.35 (95% CI 0.99 to 1.83; P = 0.06; I² = 25%; 5 trials). Analyses restricted to the trials that clearly reported a more prolonged out-of-bed activity showed a similar primary outcome (OR 1.14; 0.96 to 1.35; P = 0.13; I² = 28%; 5 trials), and odds of death (OR 1.27; 0.93 to 1.73; P = 0.13; I² = 0%; 4 trials) to the main analysis.Exploratory network meta-analysis (NMA): we were unable to analyze by the amount of therapy, but low-quality evidence indicated that time-to-first mobilisation at around 24 hours was associated with the lowest odds of death or poor outcome, compared with earlier or later mobilisation. AUTHORS' CONCLUSIONS: VEM, which usually involved first mobilisation within 24 hours of stroke onset, did not increase the number of people who survived or made a good recovery after their stroke. VEM may have reduced the length of stay in hospital by about one day, but this was based on low-quality evidence. Based on the potential hazards reported in the single largest RCT, the sensitivity analysis of trials commencing mobilisation within 24 hours, and the NMA, there was concern that VEM commencing within 24 hours may carry an increased risk, at least in some people with stroke. Given the uncertainty around these effect estimates, more detailed research is still required.

Stroke Self-Management Support Improves Survivors' Self-Efficacy and Outcome Expectation of Self-Management Behaviors.

BACKGROUND AND PURPOSE: Evidence shows self-management programs are associated with improved recovery outcomes. This article reports on the effectiveness of a new nurse-led self-efficacy-based stroke self-management program. METHODS: A randomized controlled trial of participants recruited from 3 acute stroke units was conducted. The intervention group received the 4-week stroke self-management program. The control group received usual care. All participants were assessed at baseline and 8 weeks after randomization. Data were analyzed using generalized estimating equations. Outcomes included self-efficacy, outcome expectation, and satisfaction with performance of self-management behaviors. RESULTS: One hundred twenty-eight participants were randomized with mean age, 67.46 years (SD, 11.95); 59% men; and mean duration poststroke, 45 days (SD, 26.16). At 8 weeks of follow-up in the intention-to-treat population, the intervention group improved significantly in self-efficacy (95% confidence interval, 2.55-12.45; P<0.01), outcome expectation (95% confidence interval, 5.47-14.01; P<0.01), and satisfaction with performance of self-management behaviors (95% confidence interval, 3.38-13.87; P<0.01) compared with the control. Similar results were obtained at 8 weeks of follow-up in the per-protocol population. CONCLUSIONS: The stroke self-management program improved survivors' self-efficacy, outcome expectation, and satisfaction with performance of self-management behaviors. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT02112955.

Effects of Early Mobilization after Acute Stroke: A Meta-Analysis of Randomized Control Trials.

BACKGROUND: Early mobilization is inconsistently associated with the recovery of stroke. We aim to examine the effect of early mobilization on patients with acute stroke. METHODS: PubMed, EMBASE, and the Cochrane library were searched up to April 2017. Randomized controlled trials that reported risk estimates or mean with standard deviation were included. Primary outcomes were defined as modified Rankin scale score 0-2 and mortality, and secondary outcomes were Barthel Index, length of stay, and incidence of complications. Summary relative risk, standardized mean difference (SMD), and weighted mean difference (WMD) were calculated as needed. Sensitivity analyses were also conducted to test stability of results. RESULTS: Six studies (8 publications) were included to analyze the effects of early mobilization after stroke. No differences between groups were observed for modified Rankin scale 0-2 (relative risk [RR]: .80; 95% confidence interval [CI]: .58-1.02; I2=45%) and the risk of death (RR: 1.21, 95% CI: .76-1.65; I2=0%). Compared with conventional practice, early mobilization was superior in Barthel Index (SMD: .66; 95% CI: .00-1.31; I2=85.9%), and shorter hospital stay for stroke patients (WMD: -1.97; 95% CI: -2.63 to -1.32; I2=15.3%). We found no significant difference between groups on the incidence of complications. CONCLUSIONS: Current evidence revealed that no statistical significant difference between early mobilization and non-early mobilization was observed on modified Rankin scale score 0-2 and mortality. Interestingly, early mobilization is associated with an increased Barthel Index and shorter hospital stay for patients. Further research is necessary to verify the effect of early mobilization on patients with acute stroke.

Malnutrition Increases the Incidence of Death, Cardiovascular Events, and Infections in Patients with Stroke after Rehabilitation.

BACKGROUND: Although the impact of malnutrition in patients with acute stroke has been reported, its significance after rehabilitation is not well understood. The geriatric nutritional risk index (GNRI) is a simple and well-established nutritional screening tool that predicts poor prognosis in elderly patients and in those with a high risk of cardiovascular events. We investigated the associations between GNRI and all-cause mortality, cardiovascular events, and infectious diseases in patients with stroke after rehabilitation. METHODS: This study included 138 patients aged 80 years or below who were discharged between 2010 and 2013 in a single center, and followed up for more than 1 year. Malnutrition was defined as a GNRI of 96 or lower. RESULTS: The mean age was 63.9 ± 11.0 years, the mean GNRI at discharge was 98.8 ± 6.5, and the mean total functional independence measure (FIM) score at discharge was 91.8 ± 25.8. Among the patients, 37 (27%) had malnutrition. During the follow-up period, all-cause mortality, cardiovascular events, and infectious diseases were recorded in 11 (8%), 21 (15%), and 20 (15%) patients, respectively. Kaplan-Meier curves showed a significantly higher incidence of each outcome in patients with a GNRI of 96 or lower. In the Cox proportional analysis, GNRI was an independent determinant of all-cause mortality (hazard ratio [HR], .71; 95% confidence interval [CI], .61-.83), cardiovascular events (HR, .87; 95% CI, .80-.95), and infectious diseases (HR, .80; 95% CI, .74-.87) after adjusting for age, gender, and total FIM score. CONCLUSIONS: Malnutrition has a negative impact on prognosis in patients with stroke even after rehabilitation.

Effect of Rehabilitation Intensity on Mortality Risk After Stroke.

OBJECTIVE: To determine the relation between rehabilitation intensity and poststroke mortality. DESIGN: Retrospective cohort study. SETTING: Nationwide claims data. PARTICIPANTS: From Taiwan's National Health Insurance claims databases, patients (N=6737; mean age, 66.9y; 40.3% women) hospitalized between 2001 and 2013 for a first-ever stroke who had mild to moderate stroke and survived the first 90 days of stroke were enrolled. INTERVENTIONS: The intensity of rehabilitation therapy within 90 days after stroke was categorized into low, medium, or high based on the tertile distribution of the number of rehabilitation sessions. MAIN OUTCOME MEASURES: Long-term all-cause mortality. The Cox proportional hazard models with Bonferroni correction were used to assess the association between rehabilitation intensity and mortality, adjusting for age, comorbidities, stroke severity, and other covariates. RESULTS: Patients in the high-intensity group were younger but had a higher burden of comorbidities and greater stroke severity. During follow-up, the high-intensity group was associated with a significantly lower adjusted risk (hazard ratio [HR], .73; 95% confidence interval [CI], .63-.84) of mortality than the low-intensity group, whereas the medium-intensity group carried a similar risk of mortality (HR, 0.94; 95% CI, 0.84-1.06) compared with the low-intensity group. This association was not modified by stroke severity. CONCLUSIONS: Among patients with mild to moderate stroke severity, high-intensity rehabilitation therapy within the first 90 days was associated with a lower mortality risk than low-intensity therapy. Efforts to promote high-intensity rehabilitation therapy for this group of patients with stroke should be encouraged.

Postacute Care Setting, Facility Characteristics, and Poststroke Outcomes: A Systematic Review.

OBJECTIVES: To synthesize research comparing poststroke health outcomes between patients rehabilitated in skilled nursing facilities (SNFs) and those in inpatient rehabilitation facilities (IRFs) as well as to evaluate relations between facility characteristics and outcomes. DATA SOURCES: PubMed and CINAHL searches spanned January 1, 1998, to October 6, 2016, and encompassed MeSH and free-text keywords for stroke, IRF/SNF, and study outcomes. Searches were restricted to peer-reviewed research in humans published in English. STUDY SELECTION: Observational and experimental studies examining outcomes of adult patients with stroke rehabilitated in an IRF or SNF were eligible. Studies had to provide site of care comparisons and/or analyses incorporating facility-level characteristics and had to report ≥1 primary outcome (discharge setting, functional status, readmission, quality of life, all-cause mortality). Unpublished, single-center, descriptive, and non-US studies were excluded. Articles were reviewed by 1 author, and when uncertain, discussion with study coauthors achieved consensus. Fourteen titles (0.3%) were included. DATA EXTRACTION: The types of data, time period, size, design, and primary outcomes were extracted. We also extracted 2 secondary outcomes (length of IRF/SNF stay, cost) when reported by included studies. Effect measures, modeling approaches, methods for confounding adjustment, and potential confounders were extracted. Data were abstracted by 1 author, and the accuracy was verified by a second reviewer. DATA SYNTHESIS: Two studies evaluating community discharge, 1 study evaluating the predicted probability of readmission, and 3 studies evaluating all-cause mortality favored IRFs over SNFs. Functional status comparisons were inconsistent. No studies evaluated quality of life. Two studies confirmed increased costs in the IRF versus SNF setting. Although substantial facility variation was described, few studies characterized sources of variation. CONCLUSIONS: The few studies comparing poststroke outcomes indicated better outcomes (with higher costs) for patients in IRFs versus those in SNFs. Contemporary research on the role of the postacute care setting and its attributes in determining health outcomes should be prioritized to inform reimbursement system reform.

A Pilot Trial of a Lifestyle Intervention for Stroke Survivors: Design of Healthy Eating and Lifestyle after Stroke (HEALS).

BACKGROUND: Stroke survivors have high rates of subsequent cardiovascular and recurrent cerebrovascular events, and mortality. While healthy lifestyle practices - including a diet rich in fruits and vegetables, limited alcohol intake, and regular physical activity - can mitigate these outcomes, few stroke survivors adhere to them. Minorities from socioeconomically disadvantaged communities who obtain care in safety-net health systems experience the most barriers to implementing healthy lifestyle changes after stroke. PURPOSE: To report the design of Healthy Eating and Lifestyle After Stroke (HEALS), a randomized controlled trial (RCT) was designed to test the feasibility of using a manualized, lifestyle management intervention in a safety-net setting to improve lifestyle practices among ethnically diverse individuals with stroke or transient ischemic attack (TIA). METHODS: Design: Pilot RCT. PARTICIPANTS: Inclusion criteria: 1) Adults (≥40 years) with ischemic stroke or TIA (≥ 90 days prior); 2) English- or Spanish-speaking. SETTING: Outpatient clinic, safety-net setting. INTERVENTION: Weekly two-hour small group sessions led by an occupational therapist for six weeks. The sessions focused on implementing nutrition, physical activity, and self-management strategies tailored to each participant's goals. MAIN OUTCOME MEASURES: Body mass index, diet, and physical activity. CONCLUSIONS: Recruitment for this study is complete. If the HEALS intervention study is feasible and effective, it will serve as a platform for a large-scale RCT that will investigate the efficacy and cost-effectiveness of life management interventions for racially and ethnically diverse, low-income individuals with a history of stroke or TIA who seek healthcare in the safety-net system.

The Impact of Rehabilitation Frequencies in the First Year after Stroke on the Risk of Recurrent Stroke and Mortality.

BACKGROUND: Rehabilitation is essential for all poststroke patients to improve self-care ability. However, whether an increased frequency of rehabilitation reduces poststroke adverse events remains undetermined. METHODS: We recruited 4899 patients with newly diagnosed ischemic stroke between January 1, 2000, and December 31, 2008, from our database and divided them into 3 groups according to their Charlson Comorbidity Index, and they were further categorized into 3 groups of different rehabilitation frequencies during their first year after stroke. Clinical adverse events including recurrent stroke, hip fracture, pneumonia, and all-cause mortality were analyzed by Cox regression analysis to investigate the protective effects of aggressive rehabilitation. RESULTS: We discovered that aggressive rehabilitation in the first year after stroke was significantly associated with a lower incidence of recurrent stroke and all-cause mortality despite the severity of patients' comorbidities. Further Cox regression analysis revealed decreased hazard ratios to develop recurrent stroke and all-cause mortality in patients with more intensive rehabilitation (P for trend <.05). However, no significant associations between rehabilitation frequency and pneumonia and hip fracture were identified in our study. CONCLUSION: Intensive rehabilitation during the first year after stroke should be recommended to prevent detrimental adverse events for stroke survivors.

Functional Outcomes, Subsequent Healthcare Utilization, and Mortality of Stroke Postacute Care Patients in Taiwan: A Nationwide Propensity Score-matched Study.

OBJECTIVE: To evaluate the benefits of the national stroke postacute care (PAC) program on clinical outcomes and subsequent healthcare utilization. DESIGN: Propensity score-matched case-control study using the National Health Insurance data. PARTICIPANTS: A total of 1480 stroke cases receiving PAC services and 3159 matched controls with similar stroke severity but without PAC services. MEASUREMENTS: Demographic characteristics, functional outcomes (modified Rankin Scale, Barthel Index, Lawton-Brody Instrumental Activities of Daily Living, Functional Oral Intake Scale, Mini-Nutritional Assessment, Berg Balance Test, Usual Gait Speed Test, 6-Minute Walk Test, Fugl-Meyer Assessment (modified sensation and motor), Mini-Mental State Examination, Motor Activity Log, and the Concise Chinese Aphasia Test), subsequent healthcare utilization (90-day stroke re-admission and emergency department visits), and 90-day mortality. RESULTS: After propensity score matching, baseline characteristics, stroke severity, and status of healthcare utilization before index stroke admission were similar between cases and controls. After PAC services, the case group obtained significant improvement in all functional domains and may have reduced subsequent disability. Among all functional assessments, balance was the most significantly improved domain and was suggestive for the reduction of subsequent falls risk and related injuries. Compared with controls, patients receiving PAC services had significantly lower 90-day hospital re-admissions [11.1% vs 21.0%, adjusted odds ratio (aOR) 0.47 with 95% confidence interval (CI) 0.34-0.64], stroke-related re-admissions (2.1% vs 8.8%, aOR 0.22, 95% CI 0.12-0.41), and emergency department visits (13.5% vs 24.0%, aOR 0.49, 95% CI 0.37-0.65), but the 90-day mortality rate remained similar between groups (1.4% case group vs 2.0% control group, aOR 0.68, 95% CI 0.29-1.62). CONCLUSIONS: PAC significantly improved the recovery of stroke patients in all functional domains through the program, with universal interorganizational staff training, periodic functional assessment, and high-intensity rehabilitation. Further longitudinal research is needed to evaluate the long-term survival benefits and healthcare utilization.