Risk prediction of 30-day mortality after stroke using machine learning: a nationwide registry-based cohort study.

BACKGROUNDS: We aimed to develop and validate machine learning (ML) models for 30-day stroke mortality for mortality risk stratification and as benchmarking models for quality improvement in stroke care. METHODS: Data from the UK Sentinel Stroke National Audit Program between 2013 to 2019 were used. Models were developed using XGBoost, Logistic Regression (LR), LR with elastic net with/without interaction terms using 80% randomly selected admissions from 2013 to 2018, validated on the 20% remaining admissions, and temporally validated on 2019 admissions. The models were developed with 30 variables. A reference model was developed using LR and 4 variables. Performances of all models was evaluated in terms of discrimination, calibration, reclassification, Brier scores and Decision-curves. RESULTS: In total, 488,497 stroke patients with a 12.3% 30-day mortality rate were included in the analysis. In 2019 temporal validation set, XGBoost model obtained the lowest Brier score (0.069 (95% CI: 0.068-0.071)) and the highest area under the ROC curve (AUC) (0.895 (95% CI: 0.891-0.900)) which outperformed LR reference model by 0.04 AUC (p < 0.001) and LR with elastic net and interaction term model by 0.003 AUC (p < 0.001). All models were perfectly calibrated for low (< 5%) and moderate risk groups (5-15%) and ≈1% underestimation for high-risk groups (> 15%). The XGBoost model reclassified 1648 (8.1%) low-risk cases by the LR reference model as being moderate or high-risk and gained the most net benefit in decision curve analysis. CONCLUSIONS: All models with 30 variables are potentially useful as benchmarking models in stroke-care quality improvement with ML slightly outperforming others.

Stroke Care in the United Kingdom During the COVID-19 Pandemic.

BACKGROUND AND PURPOSE: The coronavirus disease 2019 (COVID-19) pandemic has potentially caused indirect harm to patients with other conditions via reduced access to health care services. We aimed to describe the impact of the initial wave of the pandemic on admissions, care quality, and outcomes in patients with acute stroke in the United Kingdom. METHODS: Registry-based cohort study of patients with acute stroke admitted to hospital in England, Wales, and Northern Ireland between October 1, 2019, and April 30, 2020, and equivalent periods in the 3 prior years. RESULTS: One hundred fourteen hospitals provided data for a study cohort of 184 017 patients. During the lockdown period (March 23 to April 30), there was a 12% reduction (6923 versus 7902) in the number of admissions compared with the same period in the 3 previous years. Admissions fell more for ischemic than hemorrhagic stroke, for older patients, and for patients with less severe strokes. Quality of care was preserved for all measures and in some domains improved during lockdown (direct access to stroke unit care, 1-hour brain imaging, and swallow screening). Although there was no change in the proportion of patients discharged with good outcome (modified Rankin Scale score, ≤2; 48% versus 48%), 7-day inpatient case fatality increased from 6.9% to 9.4% (P<0.001) and was 22.0% in patients with confirmed or suspected COVID-19 (adjusted rate ratio, 1.41 [1.11-1.80]). CONCLUSIONS: Assuming that the true incidence of acute stroke did not change markedly during the pandemic, hospital avoidance may have created a cohort of untreated stroke patients at risk of poorer outcomes or recurrent events. Unanticipated improvements in stroke care quality should be used as an opportunity for quality improvement and to learn about how to develop resilient health care systems.

Trends in prevalence of acute stroke impairments: A population-based cohort study using the South London Stroke Register.

BACKGROUND: Acute stroke impairments often result in poor long-term outcome for stroke survivors. The aim of this study was to estimate the trends over time in the prevalence of these acute stroke impairments. METHODS AND FINDINGS: All first-ever stroke patients recorded in the South London Stroke Register (SLSR) between 2001 and 2018 were included in this cohort study. Multivariable Poisson regression models with robust error variance were used to estimate the adjusted prevalence of 8 acute impairments, across six 3-year time cohorts. Prevalence ratios comparing impairments over time were also calculated, stratified by age, sex, ethnicity, and aetiological classification (Trial of Org 10172 in Acute Stroke Treatment [TOAST]). A total of 4,683 patients had a stroke between 2001 and 2018. Mean age was 68.9 years, 48% were female, and 64% were White. After adjustment for demographic factors, pre-stroke risk factors, and stroke subtype, the prevalence of 3 out of the 8 acute impairments declined during the 18-year period, including limb motor deficit (from 77% [95% CI 74%-81%] to 62% [56%-68%], p < 0.001), dysphagia (37% [33%-41%] to 15% [12%-20%], p < 0.001), and urinary incontinence (43% [39%-47%) to 29% [24%-35%], p < 0.001). Declines in limb impairment over time were 2 times greater in men than women (prevalence ratio 0.73 [95% CI 0.64-0.84] and 0.87 [95% CI 0.77-0.98], respectively). Declines also tended to be greater in younger patients. Stratified by TOAST classification, the prevalence of all impairments was high for large artery atherosclerosis (LAA), cardioembolism (CE), and stroke of undetermined aetiology. Conversely, small vessel occlusions (SVOs) had low levels of all impairments except for limb motor impairment and dysarthria. While we have assessed 8 key acute stroke impairments, this study is limited by a focus on physical impairments, although cognitive impairments are equally important to understand. In addition, this is an inner-city cohort, which has unique characteristics compared to other populations. CONCLUSIONS: In this study, we found that stroke patients in the SLSR had a complexity of acute impairments, of which limb motor deficit, dysphagia, and incontinence have declined between 2001 and 2018. These reductions have not been uniform across all patient groups, with women and the older population, in particular, seeing fewer reductions.

The Impact of Pre- and Post-Stroke Statin Use on Stroke Severity and Long-Term Outcomes: A Population-Based Cohort Study.

BACKGROUND: The benefit of statins on stroke incidence is well known. However, data on the relationship between pre- and post-stroke statin use, recurrence, and survival outcomes are limited. We aim to investigate the short- and long-term relationships between statin prescription, stroke recurrence, and survival in patients with first-ever ischemic stroke. METHODS: Data were collected from the population-based South London Stroke Register for the years 1995-2015. Patients were assessed at the time of first ever stroke, 3 months, and annually thereafter. Data on vascular risk factors, treatments prescribed, sociodemographic characteristics, stroke subtype, survival, and stroke recurrence were collected. Cox proportional hazard analyses were used to assess the relationship of statin prescriptions pre- and post-stroke on stroke severity, long-term recurrence and survival. RESULTS: Patients prescribed statins both pre- and post-stroke showed a 24% reduction in mortality (adjusted Hazard Ratio [aHR] 0.76, 0.60-0.97), those who were prescribed statins pre-stroke and then stopped post-stroke showed greater risk of mortality (aHR 1.85, 1.10-3.12) and stroke recurrence (aHR 3.25, 1.35-7.84) compared to those that were not prescribed statins at any time. No associations were observed between pre-stroke statin and severity of the initial stroke overall, though a protective effect against moderate/severe stroke (Glasgow Coma Scale ≤12) was observed in those aged 75+ years (aOR 0.70, 0.52-0.95). CONCLUSIONS: Statins play a significant role in improving the survival rates after a stroke. Adherence to the National Guidelines that promote statin treatment, primary and secondary prevention of stroke should be monitored and a focus for quality improvement programs.

Associations Between 30-Day Mortality, Specialist Nursing, and Daily Physician Ward Rounds in a National Stroke Registry.

Background and Purpose- Well-organized stroke care is associated with better patient outcomes, but the most important organizational factors are unknown. Methods- Data were extracted from the Sentinel Stroke National Audit Programme of adults with acute stroke treated in stroke hospitals in England and Wales between April 2013 and March 2015. Multilevel models with random intercepts for hospitals were used to estimate the association of each variable with 30-day mortality to estimate the impact of admission to differently organized hospitals. Results- Of the 143 578 patients with acute stroke admitted to 154 hospitals, 14.4% died within 30 days of admission. In adjusted analyses, admission to hospitals with higher ratios of nurses trained in swallow screening was associated with reduced odds of death ( P=0.004), and admission to hospitals with daily physician ward rounds was associated with 10% lower odds of mortality compared with less-frequent ward rounds (95% CI, 0.82-0.98; P=0.013). Number of stroke admissions and overall ratio of registered nurses on duty at weekends were not found to be independently associated with mortality after adjustment for other factors. Conclusions- If these associations are causal, an extra 1332 deaths annually in England and Wales could be saved by hospitals providing care associated with a ratio of nurses trained in swallow screening of at least 3 per 10 beds and daily stroke physician ward rounds.

20 years of researching stroke through audit.

Over the last 20 years, England, Wales and Northern Ireland have developed an audit programme that now encompasses nearly all patients admitted to hospital with a stroke. This article records and reviews some questions that have been answered using data from the audit: Is the rate of institutional care after rehabilitation a possible measure of outcome? Does stroke unit care in routine practice give the benefits shown in randomized controlled trials? How is the quality of stroke care affected by a patient's age and the time of their stroke? Do patient-reported measures match those obtained from the professionals recording of processes of care? How do the processes of care after stroke affect mortality? Is thrombolysis safe to use in patients over the age of 80? Do staffing levels matter? Does assessing the safety of swallowing really make a difference? Do clinicians make rational decisions about end-of-life care in patients with haemorrhage? Does socioeconomic status influence the risk of stroke, outcome after stroke and the quality of stroke care? How much does stroke really cost in England, Wales and Northern Ireland? The article concludes that this national audit has improved stroke care across the United Kingdom, has given answers to important questions that could not be answered in any other way and has shown that benefits found in research do generalize into real clinical benefits in day-to-day practice.

The potential role of cost-utility analysis in the decision to implement major system change in acute stroke services in metropolitan areas in England.

BACKGROUND: The economic implications of major system change are an important component of the decision to implement health service reconfigurations. Little is known about how best to report the results of economic evaluations of major system change to inform decision-makers. Reconfiguration of acute stroke care in two metropolitan areas in England, namely London and Greater Manchester (GM), was used to analyse the economic implications of two different implementation strategies for major system change. METHODS: A decision analytic model was used to calculate difference-in-differences in costs and outcomes before and after the implementation of two major system change strategies in stroke care in London and GM. Values in the model were based on patient level data from Hospital Episode Statistics, linked mortality data from the Office of National Statistics and data from two national stroke audits. Results were presented as net monetary benefit (NMB) and using Programme Budgeting and Marginal Analysis (PBMA) to assess the costs and benefits of a hypothetical typical region in England with approximately 4000 strokes a year. RESULTS: In London, after 90 days, there were nine fewer deaths per 1000 patients compared to the rest of England (95% CI -24 to 6) at an additional cost of £770,027 per 1000 stroke patients admitted. There were two additional deaths (95% CI -19 to 23) in GM, with a total costs saving of £156,118 per 1000 patients compared to the rest of England. At a £30,000 willingness to pay the NMB was higher in London and GM than the rest of England over the same time period. The results of the PBMA suggest that a GM style reconfiguration could result in a total greater health benefit to a region. Implementation costs were £136 per patient in London and £75 in GM. CONCLUSIONS: The implementation of major system change in acute stroke care may result in a net health benefit to a region, even one functioning within a fixed budget. The choice of what model of stroke reconfiguration to implement may depend on the relative importance of clinical versus cost outcomes.

Weekly variation in health-care quality by day and time of admission: a nationwide, registry-based, prospective cohort study of acute stroke care.

BACKGROUND: Studies in many health systems have shown evidence of poorer quality health care for patients admitted on weekends or overnight than for those admitted during the week (the so-called weekend effect). We postulated that variation in quality was dependent on not only day, but also time, of admission, and aimed to describe the pattern and magnitude of variation in the quality of acute stroke care across the entire week. METHODS: We did this nationwide, registry-based, prospective cohort study using data from the Sentinel Stroke National Audit Programme. We included all adult patients (aged >16 years) admitted to hospital with acute stroke (ischaemic or primary intracerebral haemorrhage) in England and Wales between April 1, 2013, and March 31, 2014. Our outcome measure was 30 day post-admission survival. We estimated adjusted odds ratios for 13 indicators of acute stroke-care quality by fitting multilevel multivariable regression models across 42 4-h time periods per week. FINDINGS: The study cohort comprised 74,307 patients with acute stroke admitted to 199 hospitals. Care quality varied across the entire week, not only between weekends and weekdays, with different quality measures showing different patterns and magnitudes of temporal variation. We identified four patterns of variation: a diurnal pattern (thrombolysis, brain scan within 12 h, brain scan within 1 h, dysphagia screening), a day of the week pattern (stroke physician assessment, nurse assessment, physiotherapy, occupational therapy, and assessment of communication and swallowing by a speech and language therapist), an off-hours pattern (door-to-needle time for thrombolysis), and a flow pattern whereby quality changed sequentially across days (stroke-unit admission within 4 h). The largest magnitude of variation was for door-to-needle time within 60 min (range in quality 35-66% [16/46-232/350]; coefficient of variation 18·2). There was no difference in 30 day survival between weekends and weekdays (adjusted odds ratio 1·03, 95% CI 0·95-1·13), but patients admitted overnight on weekdays had lower odds of survival (0·90, 0·82-0·99). INTERPRETATION: The weekend effect is a simplification, and just one of several patterns of weekly variation occurring in the quality of stroke care. Weekly variation should be further investigated in other health-care settings, and quality improvement should focus on reducing temporal variation in quality and not only the weekend effect. FUNDING: None.

The association between delays in screening for and assessing dysphagia after acute stroke, and the risk of stroke-associated pneumonia.

BACKGROUND: There is no robust evidence that screening patients with acute stroke for dysphagia reduces the risk of stroke-associated pneumonia (SAP), or of how quickly it should be done after admission. We aimed to identify if delays in bedside dysphagia screening and comprehensive dysphagia assessments by a speech and language therapist (SALT) were associated with patients' risk of SAP. METHODS: Nationwide, registry-based, prospective cohort study of patients admitted with acute stroke in England and Wales. Multilevel multivariable logistic regression models were fitted, adjusting for patient variables and stroke severity. The exposures were time from (1) admission to bedside dysphagia screen, and (2) admission to comprehensive dysphagia assessment. RESULTS: Of 63 650 patients admitted with acute stroke, 55 838 (88%) had a dysphagia screen, and 24 542 (39%) a comprehensive dysphagia assessment. Patients with the longest delays in dysphagia screening (4th quartile adjusted OR 1.14, 1.03 to 1.24) and SALT dysphagia assessment (4th quartile adjusted OR 2.01, 1.76 to 2.30) had a higher risk of SAP. The risk of SAP increased in a dose-response manner with delays in SALT dysphagia assessment, with an absolute increase of pneumonia incidence of 1% per day of delay. CONCLUSIONS: Delays in screening for and assessing dysphagia after stroke, are associated with higher risk of SAP. Since SAP is one of the main causes of mortality after acute stroke, early dysphagia assessment may contribute to preventing deaths from acute stroke and could be implemented even in settings without access to high-technology specialist stroke care.

Differences in the distribution of stroke subtypes in a UK black stroke population - final results from the South London Ethnicity and Stroke Study.

BACKGROUND: Stroke incidence is increased in Black individuals but the reasons for this are poorly understood. Exploring the differences in aetiological stroke subtypes, and the extent to which they are explained by conventional and novel risk factors, is an important step in elucidating the underlying mechanisms for this increased stroke risk. METHODS: Between 1999 and 2010, 1200 black and 1200 white stroke patients were prospectively recruited from a contiguous geographical area in South London in the UK. The Trial of Org 10172 (TOAST) classification was used to classify stroke subtype. Age- and sex-adjusted comparisons of socio-demographics, traditional vascular risk factors and stroke subtypes were performed between black and white stroke patients and between Black Caribbean and Black African stroke patients using age-, sex-, and social deprivation-adjusted univariable and multivariable logistic regression analyses. RESULTS: Black stroke patients were younger than white stroke patients (mean (SD) 65.1 (13.7) vs. 74.8 (13.7) years). There were significant differences in the distribution of stroke subtypes. Small vessel disease stroke was increased in black patients versus white patients (27 % vs. 12 %; OR, 2.74; 95 % CI, 2.19-3.44), whereas large vessel and cardioembolic stroke was less frequent in black patients (OR, 0.59; 95 % CI, 0.45-0.78 and OR, 0.61; 95 % CI, 0.50-0.74, respectively). These associations remained after controlling for traditional vascular risk factors and socio-demographics. Black Caribbean patients appeared to have an intermediate risk factor and stroke subtype profile between that found in Black African and white stroke patients. Cardioembolic stroke was more strongly associated with Black Caribbean ethnicity versus Black African ethnicity (OR, 1.48; 95 % CI, 1.04-2.10), whereas intracranial large vessel disease was less frequent in Black Caribbean patients versus Black African subjects (OR, 0.44; 95 % CI, 0.24-0.83). CONCLUSIONS: Clear differences exist in stroke subtype distribution between black and white stroke patients, with a marked increase in small vessel stroke. These could not be explained by differences in the assessed traditional risk factors. Possible explanations for these differences might include variations in genetic susceptibility, differing rates of control of vascular risk factors, or as yet undetermined environmental risk factors.

Patient outcomes up to 15†years after stroke: survival, disability, quality of life, cognition and mental health.

BACKGROUND: The global epidemiological shift of disease burden towards long-term conditions means understanding long-term outcomes of cardiovascular disease is increasingly important. More people are surviving stroke to experience its long-term consequences, but outcomes in people living more >10 years after stroke have not been described in detail. METHODS: Data were collected for the population-based South London Stroke Register, with participants followed up annually until death. Outcomes were survival, disability, activity, cognitive impairment, quality of life, depression and anxiety. FINDINGS: Of 2625 people having first-ever stroke, 262 (21%) survived to 15 years. By 15 years, 61% (95% CI 55% to 67%) of the survivors were male, with a median age of stroke onset of 58 years (IQR 48-66). 87% of the 15-year survivors were living at home and 33.8% (26.2% to 42.4%) had mild disability, 14.3% (9.2% to 21.4%) moderate disability and 15.0% (9.9% to 22.3%) severe disability. The prevalence of disability increased with time but 1 in 10 of the 15-year survivors had lived with moderate-severe disability since their stroke. At 15 years, the prevalence of cognitive impairment was 30.0% (19.5% to 43.1%), depression 39.1% (30.9% to 47.9%) and anxiety 34.9% (27.0% to 43.8%), and survivors reported greater loss of physical than mental quality of life. CONCLUSIONS: One in five people live at least 15 years after a stroke and poor functional, cognitive and psychological outcomes affect a substantial proportion of these long-term survivors. As the global population of individuals with cardiovascular long-term conditions grows, research and health services will need to increasingly focus on preventing and managing the long-term consequences of stroke.

Association between socioeconomic deprivation and functional impairment after stroke: the South London Stroke Register.

BACKGROUND AND PURPOSE: Previous findings of the association between socioeconomic deprivation and functional impairment after stroke are inconsistent. There is a lack of data on long-term association. We assessed the association and differences by age, sex, prestroke comorbidities, and stroke phenotypes. METHODS: We examined data from the South London Stroke Register cohort of 1995 to 2011, recording all first-ever strokes in patients of all ages in South London. A total of 2104 patients were alive at 3 months after stroke. Socioeconomic deprivation was measured using the index of multiple deprivation based on patient postcodes, and functional impairment after stroke was defined as a Barthel index of <15. RESULTS: At 3 months after stroke, 643 patients had functional impairment (30.6%; 95% confidence interval, 28.6%-32.5%). Compared with the first quartile of index of multiple deprivation (the least deprived), multivariate-adjusted odds ratios for functional impairment in patients with the second, third, and fourth quartiles were 1.29 (95% confidence interval, 0.94-1.76), 1.33 (0.97-1.82), and 1.78 (1.31-2.43), overall P=0.004. The association was significant in patients aged ≥65 years (corresponding odds ratios were 1.49 [1.02-2.17], 1.21 [0.83-1.75], and 1.94 [1.34-2.81]; P=0.003); in women, P=0.008, in patients who do not have prestroke comorbidities, P=0.009, and in patients with ischemic stroke, P<0.001, but not significant in their counterpart patients. There were similar patterns of the associations of socioeconomic deprivation with impairment at 3 years after stroke. CONCLUSIONS: There are significant inequalities in short- and long-term functional recovery after stroke. General socioeconomic improvement, targeting groups at high risk of functional impairment is likely to reduce inequality in functional recovery after stroke.

Cross-National Key Performance Measures of the Quality of Acute Stroke Care in Western Europe.

BACKGROUND AND PURPOSE: There are no agreed measures of stroke care quality that enable the standardized comparison of stroke care between countries. We aimed to develop a set of measures of quality of acute stroke care involving stroke quality registers in Western Europe. METHODS: A multinational working group identified 6 regional or national stroke quality registers in Europe and reviewed their data sets, performance measures, and the method by which these had been developed. Measures used in the registers were presented for discussion to a consensus group of representatives from the quality registers identified, as well as other stroke experts, and the final set of common performance measures was agreed through majority consensus. RESULTS: Thirty final performance measures were agreed by the European consensus group, encompassing the domains of coordination of care (stroke unit-based care), diagnosis (brain imaging, vascular imaging, cardiac arrhythmia detection, and therapy assessment), preservation of neural tissue (thrombolytic therapy and door-to-needle time), prevention of complications (dysphagia screening), initiation of secondary prevention (antiplatelet, anticoagulation, lipid lowering, blood pressure lowering, carotid surgery, time from vascular imaging to carotid surgery, and smoking cessation), survival (90-day poststroke mortality), and functional outcomes (90-day modified Rankin Scale). CONCLUSIONS: On the basis of experience of quality registers in Europe, we have proposed a common set of performance measures that will facilitate the international comparison of acute stroke care quality.

Variations in acute hospital stroke care and factors influencing adherence to quality indicators in 6 European audits.

BACKGROUND AND PURPOSE: We compared compliance with standards of acute stroke care between 6 European audits and identified factors associated with delivery of appropriate care. METHODS: Data were derived from stroke audits in Germany, Poland, Scotland, Catalonia, Sweden, and England/Wales/Northern-Ireland participating within the European Implementation Score (EIS) collaboration. Associations between demographic and clinical characteristics with adherence to predefined quality indicators were investigated by hierarchical logistic regression analyses. RESULTS: In 2007/2008 data from 329 122 patients with stroke were documented. Substantial variations in adherence to quality indicators were found; older age was associated with a lower probability of receiving thrombolytic therapy, anticoagulant therapy, or stroke unit treatment and a higher probability of being tested for dysphagia. Women were less likely to receive anticoagulant or antiplatelet therapy or stroke unit treatment. No major weekend effect was found. CONCLUSIONS: Detected variations in performance of acute stroke services were found. Differences in adherence to quality indicators might indicate population subgroups with specific needs for improving care delivery.

Diagnosis of Stroke-Associated Pneumonia: Recommendations From the Pneumonia in Stroke Consensus Group.

BACKGROUND AND PURPOSE: Lower respiratory tract infections frequently complicate stroke and adversely affect outcome. There is currently no agreed terminology or gold-standard diagnostic criteria for the spectrum of lower respiratory tract infections complicating stroke, which has implications for clinical practice and research. The aim of this consensus was to propose standardized terminology and operational diagnostic criteria for lower respiratory tract infections complicating acute stroke. METHODS: Systematic literature searches of multiple electronic databases were undertaken. An evidence review and 2 rounds of consensus consultation were completed before a final consensus meeting in September 2014, held in Manchester, United Kingdom. Consensus was defined a priori as ≥75% agreement between the consensus group members. RESULTS: Consensus was reached for the following: (1) stroke-associated pneumonia (SAP) is the recommended terminology for the spectrum of lower respiratory tract infections within the first 7 days after stroke onset; (2) modified Centers for Disease Control and Prevention (CDC) criteria are proposed for SAP as follows-probable SAP: CDC criteria met, but typical chest x-ray changes absent even after repeat or serial chest x-ray; definite SAP: CDC criteria met, including typical chest x-ray changes; (3) there is limited evidence for a diagnostic role of white blood cell count or C-reactive protein in SAP; and (4) there is insufficient evidence for the use of other biomarkers (eg, procalcitonin). CONCLUSIONS: Consensus operational criteria for the terminology and diagnosis of SAP are proposed based on the CDC criteria. These require prospective evaluation in patients with stroke to determine their reliability, validity, impact on clinician behaviors (including antibiotic prescribing), and clinical outcomes.

Effects of Centralizing Acute Stroke Services on Stroke Care Provision in Two Large Metropolitan Areas in England.

BACKGROUND AND PURPOSE: In 2010, Greater Manchester and London centralized acute stroke care into hyperacute units (Greater Manchester=3, London=8), with additional units providing ongoing specialist stroke care nearer patients' homes. Greater Manchester patients presenting within 4 hours of symptom onset were eligible for hyperacute unit admission; all London patients were eligible. Research indicates that postcentralization, only London's stroke mortality fell significantly more than elsewhere in England. This article attempts to explain this difference by analyzing how centralization affects provision of evidence-based clinical interventions. METHODS: Controlled before and after analysis was conducted, using national audit data covering Greater Manchester, London, and a noncentralized urban comparator (38 623 adult stroke patients, April 2008 to December 2012). Likelihood of receiving all interventions measured reliably in pre- and postcentralization audits (brain scan; stroke unit admission; receiving antiplatelet; physiotherapist, nutrition, and swallow assessments) was calculated, adjusting for age, sex, stroke-type, consciousness, and whether stroke occurred in-hospital. RESULTS: Postcentralization, likelihood of receiving interventions increased in all areas. London patients were overall significantly more likely to receive interventions, for example, brain scan within 3 hours: Greater Manchester=65.2% (95% confidence interval=64.3-66.2); London=72.1% (71.4-72.8); comparator=55.5% (54.8-56.3). Hyperacute units were significantly more likely to provide interventions, but fewer Greater Manchester patients were admitted to these (Greater Manchester=39%; London=93%). Differences resulted from contrasting hyperacute unit referral criteria and how reliably they were followed. CONCLUSIONS: Centralized systems admitting all stroke patients to hyperacute units, as in London, are significantly more likely to provide evidence-based clinical interventions. This may help explain previous research showing better outcomes associated with fully centralized models.

How is pneumonia diagnosed in clinical stroke research? A systematic review and meta-analysis.

BACKGROUND AND PURPOSE: Diagnosis of pneumonia complicating stroke is challenging, and there are currently no consensus diagnostic criteria. As a first step in developing such consensus-based diagnostic criteria, we undertook a systematic review to identify the existing diagnostic approaches to pneumonia in recent clinical stroke research to establish the variation in diagnosis and terminology. METHODS: Studies of ischemic stroke, intracerebral hemorrhage, or both, which reported occurrence of pneumonia from January 2009 to March 2014, were considered and independently screened for inclusion by 2 reviewers after multiple searches using electronic databases. The primary analysis was to identify existing diagnostic approaches for pneumonia. Secondary analyses explored potential reasons for any heterogeneity where standard criteria for pneumonia had been applied. RESULTS: Sixty-four studies (56% ischemic stroke, 6% intracerebral hemorrhage, 38% both) of 639 953 patients were included. Six studies (9%) reported no information on the diagnostic approach, whereas 12 (19%) used unspecified clinician-reported diagnosis or initiation of antibiotics. The majority used objective diagnostic criteria: 20 studies (31%) used respiratory or other published standard criteria; 26 studies (41%) used previously unpublished ad hoc criteria. The overall occurrence of pneumonia was 14.3% (95% confidence interval 13.2%-15.4%; I(2)=98.9%). Occurrence was highest in studies applying standard criteria (19.1%; 95% confidence interval 15.1%-23.4%; I(2)=98.5%). The substantial heterogeneity observed was not explained by stratifying for other potential confounders. CONCLUSIONS: We found considerable variation in terminology and the diagnostic approach to pneumonia. Our review supports the need for consensus development of operational diagnostic criteria for pneumonia complicating stroke.

Explanatory factors for the association between depression and long-term physical disability after stroke.

OBJECTIVES: To identify explanatory factors for the association between depression at 3 months after stroke and physical disability at 3 years. METHODS: Data from the South London Stroke Register (1998-2013) were used. Patients (n = 3,612) were assessed at stroke onset. Follow-up at 3 months included assessment for depression with the Hospital Anxiety and Depression scale (scores ≥ 7 = depression), physical disability (Barthel index) cognitive function, smoking habit, selective serotonin reuptake inhibitors (SSRIs) use, perception of recovery and social support. Physical disability was reassessed at 3 years. The associations between depression at 3 months and physical disability at 3 years were estimated with multinomial regression adjusting for age, gender, ethnicity, stroke severity and possible explanatory factors for the association (introduced in the models first individually and then sequentially): pre-stroke medical history and physical disability, cognitive function, smoking, SSRIs, perception of recovery and social support at 3 months. RESULTS: One thousand three hundred and seven survivors were assessed at 3 months, of which 418 (32.0%) had depression. Survivors with depression had a higher physical disability rate at 3 years. These associations remained significant after adjustment for individual explanatory factors but were not significant after adjustment for combined explanatory factors. Physical disability at 3 months was a relevant explanatory factor for this association. SSRIs were associated with severe, relative risk: 6.62 (2.92-15.02) P < 0.001, and moderate physical disability, relative risk: 3.45 (1.58-7.52) P = 0.002, at 3 years. CONCLUSION: The association between depression and physical disability appears to be multifactorial. The use of SSRIs after stroke requires further research.

Long-term effects of secondary prevention on cognitive function in stroke patients.

BACKGROUND: Limited long-term follow-up data exist on the impact of appropriate secondary prevention therapies on cognitive function in patients after first-ever stroke. The aim of this study is to determine the effect of secondary prevention of vascular events on cognitive function after stroke. METHODS AND RESULTS: Data were collected between 1995 and 2011 (n=4413) from the community-based South London Stroke Register covering an inner-city multiethnic source population of 271 817 inhabitants. Modified Poisson regression models were constructed to adjust for cognitive function status at 3 months, demographic and socioeconomic characteristics, case mix, stroke subtype, vascular risk factors, disability, and stroke recurrence. In patients with ischemic strokes without a history of atrial fibrillation (AF), there was a reduced risk of cognitive impairment associated with the use of different prevention treatments: (1) antihypertensives (relative risk, 0.7 [95% confidence interval, 0.57-0.82] for diuretics; relative risk, 0.8 [95% confidence interval, 0.64-0.98] for angiotensin-converting enzyme inhibitors; and relative risk, 0.7 [95% confidence interval, 0.55-0.81] for their combination), (2) a combination of aspirin and dipyridamole (relative risk, 0.8 [95% confidence interval, 0.68-1.01]), and (3) statin (relative risk, 0.9 [95% confidence interval, 0.76-1.06]) when clinically indicated. Protective effects against cognitive impairment were also observed in patients on the combination of antihypertensives, antithrombotic agents, and lipid-lowering drugs (relative risk, 0.55 [95% confidence interval, 0.40-0.77]). No significant associations were noted between poststroke cognitive impairment and antihypertensives among hemorrhagic stroke patients. CONCLUSIONS: Appropriate vascular risk management was associated with a long-term reduced risk of cognitive impairment. Focus on optimal preventive drug therapy of vascular risk factors and management should be supported.