Assessing stroke severity using electronic health record data: a machine learning approach

Characterisation of young stroke presentations, pathways of care, and support for 'invisible' difficulties: a retrospective clinical audit study

Background Young stroke survivors are likely to be discharged home from acute hospital care without rehabilitation more quickly than older survivors, but it is not clear why. File-audit studies capturing real-world clinical practice are lacking for this cohort. We aimed to compare characteristics and care pathways of young and older survivors and describe stroke presentations and predictors of pathways of care in young survivors (≤45years), including a focus on care received for 'invisible' (cognitive, psychological) difficulties. Methods A retrospective audit of 847 medical records (67 young stroke survivors, mean age=36years; 780 older patients, mean age=70years) was completed for stroke survivors admitted to an Australian tertiary hospital. Stroke characteristics and presence of cognitive difficulties (identified through clinician opinion or cognitive screening) were used to predict length of stay and discharge destination in young stroke survivors. Results There were no differences in length of stay between young and older survivors, however, young stroke survivors were more likely to be discharged home without rehabilitation (though this may be due to milder strokes observed in young stroke survivors). For young stroke survivors, stroke severity and age predicted discharge destination, while cognitive difficulties predicted longer length of stay. While almost all young survivors were offered occupational therapy and physiotherapy, none received psychological input (clinical, health or neuropsychology). Conclusions Cognitive and psychological needs of young stroke survivors may remain largely unmet by a service model designed for older people. Findings can inform service development or models of care, such as the new Australian Young Stroke Service designed to better meet the needs of young survivors.

Automatic quantitative stroke severity assessment based on Chinese clinical named entity recognition with domain-adaptive pre-trained large language model

BACKGROUND

Stroke is a prevalent disease with a significant global impact. Effective assessment of stroke severity is vital for an accurate diagnosis, appropriate treatment, and optimal clinical outcomes. The National Institutes of Health Stroke Scale (NIHSS) is a widely used scale for quantitatively assessing stroke severity. However, the current manual scoring of NIHSS is labor-intensive, time-consuming, and sometimes unreliable. Applying artificial intelligence (AI) techniques to automate the quantitative assessment of stroke on vast amounts of electronic health records (EHRs) has attracted much interest.

OBJECTIVE

This study aims to develop an automatic, quantitative stroke severity assessment framework through automating the entire NIHSS scoring process on Chinese clinical EHRs.

METHODS

Our approach consists of two major parts: Chinese clinical named entity recognition (CNER) with a domain-adaptive pre-trained large language model (LLM) and automated NIHSS scoring. To build a high-performing CNER model, we first construct a stroke-specific, densely annotated dataset "Chinese Stroke Clinical Records" (CSCR) from EHRs provided by our partner hospital, based on a stroke ontology that defines semantically related entities for stroke assessment. We then pre-train a Chinese clinical LLM coined "CliRoberta" through domain-adaptive transfer learning and construct a deep learning-based CNER model that can accurately extract entities directly from Chinese EHRs. Finally, an automated, end-to-end NIHSS scoring pipeline is proposed by mapping the extracted entities to relevant NIHSS items and values, to quantitatively assess the stroke severity.

RESULTS

Results obtained on a benchmark dataset CCKS2019 and our newly created CSCR dataset demonstrate the superior performance of our domain-adaptive pre-trained LLM and the CNER model, compared with the existing benchmark LLMs and CNER models. The high F1 score of 0.990 ensures the reliability of our model in accurately extracting the entities for the subsequent automatic NIHSS scoring. Subsequently, our automated, end-to-end NIHSS scoring approach achieved excellent inter-rater agreement (0.823) and intraclass consistency (0.986) with the ground truth and significantly reduced the processing time from minutes to a few seconds.

CONCLUSION

Our proposed automatic and quantitative framework for assessing stroke severity demonstrates exceptional performance and reliability through directly scoring the NIHSS from diagnostic notes in Chinese clinical EHRs. Moreover, this study also contributes a new clinical dataset, a pre-trained clinical LLM, and an effective deep learning-based CNER model. The deployment of these advanced algorithms can improve the accuracy and efficiency of clinical assessment, and help improve the quality, affordability and productivity of healthcare services.

Automated Extraction of Stroke Severity from Unstructured Electronic Health Records using Natural Language Processing

BACKGROUND

Multi-center electronic health records (EHR) can support quality improvement initiatives and comparative effectiveness research in stroke care. However, limitations of EHR-based research include challenges in abstracting key clinical variables from non-structured data at scale. This is further compounded by missing data. Here we develop a natural language processing (NLP) model that automatically reads EHR notes to determine the NIH stroke scale (NIHSS) score of patients with acute stroke.

METHODS

The study included notes from acute stroke patients (>= 18 years) admitted to the Massachusetts General Hospital (MGH) (2015-2022). The MGH data were divided into training (70%) and hold-out test (30%) sets. A two-stage model was developed to predict the admission NIHSS. A linear model with the least absolute shrinkage and selection operator (LASSO) was trained within the training set. For notes in the test set where the NIHSS was documented, the scores were extracted using regular expressions (stage 1), for notes where NIHSS was not documented, LASSO was used for prediction (stage 2). The reference standard for NIHSS was obtained from Get With The Guidelines Stroke Registry. The two-stage model was tested on the hold-out test set and validated in the MIMIC-III dataset (Medical Information Mart for Intensive Care-MIMIC III 2001-2012) v1.4, using root mean squared error (RMSE) and Spearman correlation (SC).

RESULTS

We included 4,163 patients (MGH = 3,876; MIMIC = 287); average age of 69 [SD 15] years; 53% male, and 72% white. 90% patients had ischemic stroke and 10% hemorrhagic stroke. The two-stage model achieved a RMSE [95% CI] of 3.13 [2.86-3.41] (SC = 0.90 [0.88-0. 91]) in the MGH hold-out test set and 2.01 [1.58-2.38] (SC = 0.96 [0.94-0.97]) in the MIMIC validation set.

CONCLUSIONS

The automatic NLP-based model can enable large-scale stroke severity phenotyping from EHR and therefore support real-world quality improvement and comparative effectiveness studies in stroke.

Subacute stroke: new-onset poststroke bladder and bowel dysfunctions and possible associated factors

PURPOSE

Bladder and bowel poststroke dysfunctions negatively impact patients' health. Stroke-related characteristics associated to these dysfunctions are poorly known. This study aims to estimate the prevalence of new-onset poststroke bladder and bowel dysfunctions, characterize their associated factors, and describe the dysfunctions' clinical approach.

MATERIALS AND METHODS

Cross-sectional study including 157 patients admitted to a single hospital's stroke unit with a first-ever stroke, during 3 months. An 18-item questionnaire was applied to assess dysfunctions pre and poststroke. The McNemar test was used to compare pre and poststroke prevalence. A logistic regression was used to estimate associations (OR, 95% CI) between individual characteristics and new-onset dysfunctions.

RESULTS

We had 113 (72%) respondents. There was a significant increase in the prevalence of bladder and bowel dysfunctions poststroke (p < 0.001). Higher stroke severity was significantly associated with both new-onset poststroke bladder and bowel dysfunctions (OR = 15.00, 95% CI [4.92,45.76] and OR = 5.87,95%CI [2.14,16.12], respectively). Total anterior circulation strokes, cardioembolic strokes, and lower functionality at discharge were also significantly associated with both dysfunctions. Thirteen patients (11.5%) reported that health professionals addressed these dysfunctions.

CONCLUSIONS

Poststroke bladder and bowel dysfunctions are highly prevalent. Being aware of their epidemiology helps draw attention to patients at higher risk of developing these dysfunctions, enhancing the rehabilitation process.IMPLICATIONS FOR REHABILITATIONPoststroke bladder and bowel dysfunctions are highly prevalent and under-recognised consequences of stroke.Being aware of their epidemiology and associated factors may help identify patients at higher risk of developing these dysfunctions.It is necessary to raise clinical awareness to ensure a more efficient diagnostic and therapeutic approach, enhancing patients' rehabilitation process, quality of life and lowering collateral societal burden.

Beyond diagnosis: Leveraging routine blood and urine biomarkers to predict severity and functional outcome in acute ischemic stroke

Background

The initial severity of acute ischemic stroke (AIS) is a crucial predictor of the disease outcome. In this study, blood and urine biomarkers from patients with AIS were measured to estimate stroke severity and predict long-term stroke outcomes.

Methods

The medical records of patients with AIS between October 2016 and May 2020 were retrospectively analyzed. The relationships of blood and urine biomarkers with stroke severity at admission were evaluated in patients with AIS. Predictive models for initial stroke severity and long-term prognosis were then developed using a panel of identified biomarkers.

Results

A total of 2229 patients were enrolled. Univariate analysis revealed 12 biomarkers associated with the National Institutes of Health Stroke Scale scores at admission. The area under the curve values for predicting initial stroke severity and long-term prognosis on the basis of these biomarkers were 0.7465, 0.7470, and 0.8061, respectively. Among multiple tested machine-learning, eXtreme gradient boosting exhibited the highest effectiveness in predicting 90-day modified Rankin Scale scores. SHapley Additive exPlanations revealed fasting glucose, albumin, hemoglobin, prothrombin time, and urine-specific gravity to be the top five most crucial biomarkers.

Conclusion

These findings demonstrate that clinically available blood and urine biomarkers can effectively estimate initial stroke severity and predict long-term prognosis in patients with AIS. Our results provide a scientific basis for developing tailored clinical treatment and management strategies for AIS, through incorporating liquid biomarkers into stroke risk assessment and patient care protocols for patients with AIS.

Acute Dysphagia Following Reperfusion Therapies: A Prospective Pilot Cohort Study

Dysphagia is a well-documented sequela of stroke. Recent advancements in medical treatments for stroke include reperfusion therapies (endovascular thrombectomy (EVT) and thrombolysis). As outcomes following reperfusion therapies are typically measured via general functional scales, the pattern and progression of acute dysphagia following reperfusion therapies is less known. To determine the progression of acute dysphagia (0-72 h) following reperfusion therapies and relationships between various stroke parameters and dysphagia, twenty-six patients were prospectively recruited across two EVT and thrombolysis centres in Brisbane, Australia. Dysphagia was screened via the Gugging Swallowing Screen (GUSS) at the bedside at three timepoints: 0-24 h, 24-48 h, and 48-72 h post-reperfusion therapies. Across three groups (EVT only, thrombolysis only, or both), the incidence of any dysphagia within the first 24 h of reperfusion therapy was 92.31% (n = 24/26), 91.30% (n = 21/23) by 48 h, and 90.91% (n = 20/22) by 72 h. Fifteen patients presented with severe dysphagia at 0-24 h, 10 at 24-48 h, and 10 at 48-72 h. Whilst dysphagia was not significantly correlated to infarct penumbra/core size, dysphagia severity was significantly related to the number of passes required during EVT (p = 0.009).Dysphagia continues to persist in the acute stroke population despite recent advancements in technology aimed to reduce morbidity and mortality post-stroke. Further research is required to establish protocols for management of dysphagia post-reperfusion therapies.

Barriers and facilitators to access post-stroke rehabilitation services in the first six months of recovery in Brazil

PURPOSE

To identify barriers and facilitators to accessing post-stroke rehabilitation services six months after discharge from the stroke unit of a Brazilian public hospital.

MATERIALS AND METHODS

This cross-sectional and descriptive study collected sociodemographic and clinical-functional data during hospitalization. Then, barriers and facilitators for accessing the post-stroke rehabilitation services were collected six months after discharge. We considered economic conditions and displacement, the quality and organization of post-stroke rehabilitation services, and personal conditions.

RESULTS

A total of 174 patients were included. Among the 20 aspects analyzed, 17 (85.0%) were reported as facilitators, while three (15.0%) were as barriers. The identified barriers included financial income available for healthcare (49.4%), waiting time to schedule or to be seen (47.0%), and process to scheduling (45.4%). The main facilitators (> 79.0%) were the expectation of the patient with the treatment and assistance from family and friends. Moreover, most patients indicated as facilitators all aspects related to the quality of post-stroke rehabilitation services.

CONCLUSION

Access to post-stroke rehabilitation services presented more facilitators than barriers. Public policies to subsidize health costs, optimize waiting time, and process for scheduling post-stroke rehabilitation services should be considered to reduce barriers. Likewise, human and financial resources must promote the facilitators.

Association between Vitamin D and Short-Term Functional Outcomes in Acute Ischemic Stroke

Vitamin D (Vit D) affects musculoskeletal performance and central nervous system neuroprotection. We aimed to investigate the association between serum Vit D levels and short-term functional outcomes in patients with acute ischemic stroke. This study involved patients with acute ischemic stroke confirmed on brain MRI. The National Institutes of Health Stroke Scale (NIHSS) was used to assess initial stroke severity upon admission. We evaluated the functional outcomes using the Berg Balance Scale (BBS), Manual Function Test (MFT), Korean Mini-Mental State Examination (K-MMSE), Korean version of the modified Barthel Index (K-MBI) within three weeks from the onset of stroke, and modified Rankin Scale (mRS) score at discharge. Overall, 192 patients were finally included and divided into three groups: Vit D sufficient (n = 28), insufficient (n = 49), and deficient (n = 115). Multivariate analysis showed that the Vit D deficient group presented with a higher risk of initially severe stroke (p = 0.025) and poor functional outcomes on the BBS (p = 0.048), MFT (p = 0.017), K-MMSE (p = 0.001), K-MBI (p = 0.003), and mRS (p = 0.032) compared to the Vit D sufficient group. Vit D deficiency may be associated with severe initial stroke and poor short-term post-stroke functional outcomes.

Applications of Natural Language Processing for the Management of Stroke Disorders: Scoping Review

BACKGROUND

Recent advances in natural language processing (NLP) have heightened the interest of the medical community in its application to health care in general, in particular to stroke, a medical emergency of great impact. In this rapidly evolving context, it is necessary to learn and understand the experience already accumulated by the medical and scientific community.

OBJECTIVE

The aim of this scoping review was to explore the studies conducted in the last 10 years using NLP to assist the management of stroke emergencies so as to gain insight on the state of the art, its main contexts of application, and the software tools that are used.

METHODS

Data were extracted from Scopus and Medline through PubMed, using the keywords "natural language processing" and "stroke." Primary research questions were related to the phases, contexts, and types of textual data used in the studies. Secondary research questions were related to the numerical and statistical methods and the software used to process the data. The extracted data were structured in tables and their relative frequencies were calculated. The relationships between categories were analyzed through multiple correspondence analysis.

RESULTS

Twenty-nine papers were included in the review, with the majority being cohort studies of ischemic stroke published in the last 2 years. The majority of papers focused on the use of NLP to assist in the diagnostic phase, followed by the outcome prognosis, using text data from diagnostic reports and in many cases annotations on medical images. The most frequent approach was based on general machine learning techniques applied to the results of relatively simple NLP methods with the support of ontologies and standard vocabularies. Although smaller in number, there has been an increasing body of studies using deep learning techniques on numerical and vectorized representations of the texts obtained with more sophisticated NLP tools.

CONCLUSIONS

Studies focused on NLP applied to stroke show specific trends that can be compared to the more general application of artificial intelligence to stroke. The purpose of using NLP is often to improve processes in a clinical context rather than to assist in the rehabilitation process. The state of the art in NLP is represented by deep learning architectures, among which Bidirectional Encoder Representations from Transformers has been found to be especially widely used in the medical field in general, and for stroke in particular, with an increasing focus on the processing of annotations on medical images.

Access to rehabilitation professionals by individuals with stroke one month after hospital discharge from a stroke unit in Brazil is insufficient regardless of the pandemic

OBJECTIVE

To compare access to rehabilitation professionals by individuals with stroke one month after hospital discharge from a stroke unit in Brazil, before and during the COVID-19 pandemic.

MATERIALS AND METHODS

This longitudinal and prospective study included individuals aged 20 years or older without previous disabilities admitted into a stroke unit due to a first stroke. Individuals were divided into two groups: before (G1) and during (G2) the COVID-19 pandemic. Groups were matched for age, sex, education level, socioeconomic status, and stroke severity. One month after hospital discharge, individuals were contacted via telephone to collect data regarding their access to rehabilitation services based on the number of referred rehabilitation professionals. Then, between-group comparisons were conducted (α = 5%).

RESULTS

The access to rehabilitation professionals was similar between groups. Rehabilitation professionals accessed included medical doctors, occupational therapists, physical therapists, and speech therapists. The first consultation after hospital discharge was mainly provided by public services. Despite the pandemic, telehealth was not frequent in any period evaluated. In both groups, the number of accessed professionals (G1 = 110 and G2 = 90) was significantly lower than the number of referrals (G1 = 212 and G2 = 194; p < 0.001).

CONCLUSIONS

Access to rehabilitation professionals was similar between groups. However, the number of accessed rehabilitation professionals was lower than that of referred ones during both periods. This finding indicates a compromised comprehensiveness of care for individuals with stroke, regardless of the pandemic.

Study of the obstructive sleep apnea syndrome in cerebral infarction patients

Introduction

Obstructive Sleep Apnea Syndrome (OSAS) is the most common respiratory disorder during sleep. Many studies have shown an association between obstructive sleep apnea syndrome and stroke, and OSAS has not been adequately considered in Vietnam compared to the actual clinical dangers. This study aims to assess the prevalence and general characteristics of obstructive sleep apnea syndrome in patients with cerebral infarction and investigate the relationship between obstructive sleep apnea syndrome and the severity of cerebral infarction.

Methods

Descriptive cross-sectional study. We identified 56 participants from August 2018 to July 2019. Subacute infarcts were identified by neuroradiologists. For each participant, vascular risk factors, medications, clinical symptoms, and neurological examination were abstracted from the medical record. Patients were taken for history and clinical examination. The patients were divided into two groups according to their AHI (Apnea-Hypopnea Index) (<5 and ≥5).

Results

A total of 56 patients were registered for the study. The mean age is 67.70 ± 11.07. The proportion of men is 53.6%. AHI has a positive correlation with neck circumference (r = 0.4), BMI (r = 0.38), the Epworth Sleepiness Scale (r = 0.61), LDL cholesterol (r = 0.38), the Modified Rankin Scale (r = 0.49), NIHSS (National Institutes of Health Stroke Scale) (r = 0.53), and an inverse correlation with SpO₂ (r = 0.61).

Conclusion

Obstructive sleep apnea Syndrome is a factor in the prognosis of cerebral infarction as well as cardiovascular diseases such as hypertension. Thus, understanding the risk of stroke in people with sleep apnea is necessary and working with a doctor to diagnose and treat sleep apnea is important.

Identifying stroke-related quantified evidence from electronic health records in real-world studies

BACKGROUND

Stroke is one of the leading causes of death and disability worldwide. The National Institutes of Health Stroke Scale (NIHSS) scores in electronic health records (EHRs), which quantitatively describe patients' neurological deficits in evidence-based treatment, are crucial in stroke-related clinical investigations. However, the free-text format and lack of standardization inhibit their effective use. Automatically extracting the scale scores from the clinical free text so that its potential value in real-world studies is realized has become an important goal.

OBJECTIVE

This study aims to develop an automated method to extract scale scores from the free text of EHRs.

METHODS

We propose a two-step pipeline method to identify NIHSS items and numerical scores and validate its feasibility using a freely accessible critical care database: MIMIC-III (Medical Information Mart for Intensive Care III). First, we utilize MIMIC-III to create an annotated corpus. Then, we investigate possible machine learning methods for two subtasks, NIHSS item and score recognition and item-score relation extraction. In the evaluation, we conduct both task-specific and end-to-end evaluations and compare our method with the rule-based method using precision, recall and F1 scores as evaluation metrics.

RESULTS

We use all available discharge summaries of stroke cases in MIMIC-III. The annotated NIHSS corpus contains 312 cases, 2929 scale items, 2774 scores and 2733 relations. The results show that the best F1-score of our method was 0.9006, which was attained by combining BERT-BiLSTM-CRF and Random Forest, and it outperformed the rule-based method (F1-score = 0.8098). In the end-to-end task, our method could successfully recognize the item "1b level of consciousness questions", the score "1" and their relation "('1b level of consciousness questions', '1', 'has value')" from the sentence "1b level of consciousness questions: said name = 1", while the rule-based method could not.

CONCLUSIONS

The two-step pipeline method we propose is an effective approach to identify NIHSS items, scores and their relations. With its help, clinical investigators can easily retrieve and access structured scale data, thereby supporting stroke-related real-world studies.

Patient Embeddings From Diagnosis Codes for Health Care Prediction Tasks: Pat2Vec Machine Learning Framework

BACKGROUND

In health care, diagnosis codes in claims data and electronic health records (EHRs) play an important role in data-driven decision making. Any analysis that uses a patient's diagnosis codes to predict future outcomes or describe morbidity requires a numerical representation of this diagnosis profile made up of string-based diagnosis codes. These numerical representations are especially important for machine learning models. Most commonly, binary-encoded representations have been used, usually for a subset of diagnoses. In real-world health care applications, several issues arise: patient profiles show high variability even when the underlying diseases are the same, they may have gaps and not contain all available information, and a large number of appropriate diagnoses must be considered.

OBJECTIVE

We herein present Pat2Vec, a self-supervised machine learning framework inspired by neural network-based natural language processing that embeds complete diagnosis profiles into a small real-valued numerical vector.

METHODS

Based on German outpatient claims data with diagnosis codes according to the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10), we discovered an optimal vectorization embedding model for patient diagnosis profiles with Bayesian optimization for the hyperparameters. The calibration process ensured a robust embedding model for health care-relevant tasks by aggregating the metrics of different regression and classification tasks using different machine learning algorithms (linear and logistic regression as well as gradient-boosted trees). The models were tested against a baseline model that binary encodes the most common diagnoses. The study used diagnosis profiles and supplementary data from more than 10 million patients from 2016 to 2019 and was based on the largest German ambulatory claims data set. To describe subpopulations in health care, we identified clusters (via density-based clustering) and visualized patient vectors in 2D (via dimensionality reduction with uniform manifold approximation). Furthermore, we applied our vectorization model to predict prospective drug prescription costs based on patients' diagnoses.

RESULTS

Our final models outperform the baseline model (binary encoding) with equal dimensions. They are more robust to missing data and show large performance gains, particularly in lower dimensions, demonstrating the embedding model's compression of nonlinear information. In the future, other sources of health care data can be integrated into the current diagnosis-based framework. Other researchers can apply our publicly shared embedding model to their own diagnosis data.

CONCLUSIONS

We envision a wide range of applications for Pat2Vec that will improve health care quality, including personalized prevention and signal detection in patient surveillance as well as health care resource planning based on subcohorts identified by our data-driven machine learning framework.

Natural Language Processing in Electronic Health Records in relation to healthcare decision-making: A systematic review

BACKGROUND

Natural Language Processing (NLP) is widely used to extract clinical insights from Electronic Health Records (EHRs). However, the lack of annotated data, automated tools, and other challenges hinder the full utilisation of NLP for EHRs. Various Machine Learning (ML), Deep Learning (DL) and NLP techniques are studied and compared to understand the limitations and opportunities in this space comprehensively.

METHODOLOGY

After screening 261 articles from 11 databases, we included 127 papers for full-text review covering seven categories of articles: (1) medical note classification, (2) clinical entity recognition, (3) text summarisation, (4) deep learning (DL) and transfer learning architecture, (5) information extraction, (6) Medical language translation and (7) other NLP applications. This study follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.

RESULT AND DISCUSSION

EHR was the most commonly used data type among the selected articles, and the datasets were primarily unstructured. Various ML and DL methods were used, with prediction or classification being the most common application of ML or DL. The most common use cases were: the International Classification of Diseases, Ninth Revision (ICD-9) classification, clinical note analysis, and named entity recognition (NER) for clinical descriptions and research on psychiatric disorders.

CONCLUSION

We find that the adopted ML models were not adequately assessed. In addition, the data imbalance problem is quite important, yet we must find techniques to address this underlining problem. Future studies should address key limitations in studies, primarily identifying Lupus Nephritis, Suicide Attempts, perinatal self-harmed and ICD-9 classification.

Minimal Clinically Important Differences and Changes in Stroke Disease-Specific Quality of Life in Stroke Survivors: A Prospective Cohort Study

The aims of the present study were to investigate changes in QOL in post-stroke patients during the first 3 months of rehabilitation treatment. We estimate minimal detectable changes (MDCs) and minimal clinically important differences (MCIDs) of the eight dimensions of QOL and assess the proportion of patients' change scores that exceed MDCs and MCIDs in stroke survivors who receive rehabilitation in a hospital ward. This prospective cohort study enrolled 40 stroke survivors (57.5% male; M_age = 58.3 years) who received in-hospital rehabilitation for a total of 3 months. The Stroke Impact Scale 3.0, which has eight subscales-strength, activities of daily living (ADLs)/instrumental ADLs (IADLs), mobility, hand function, communication, memory and thinking, emotion, and social participation-was used for assessment on the third day of rehabilitation (Time 1), 1 month later (Time 2), and 3 months later (Time 3). Our findings indicated that the MDC₉₅ and MCID proportions varied from 7.5% to 30% and 7.5% to 65%, respectively, of individuals who exhibited change based on individual change scores. The findings show compliance with MDC and MCID values in physical function, with the lowest proportion in hand function.

Antiplatelet treatment patterns and outcomes of secondary stroke prevention in the United States

Objective

Patients who have an ischemic stroke (IS) or transient ischemic attack (TIA) are at risk of having a secondary stroke. Single antiplatelet therapy (SAPT) or dual antiplatelet therapy (DAPT) may be recommended for secondary stroke prevention (SSP), depending on severity and etiology. This study evaluated outpatient antiplatelet treatment patterns for SSP and outcomes after first hospitalization for IS/TIA among adults without atrial fibrillation in the United States.

Materials and methods

This retrospective observational study utilized data from an adjudicated administrative health claims database. Eligible patients had an imputed National Institutes of Health Stroke Scale index event score ≤7. Over-the-counter medication use (eg, aspirin) was not captured.

Results

Of 154,273 patients, 41,622 (27%) were prescribed antiplatelet therapy within 90 days of the event; 93.8% received SAPT, 6.1% received DAPT. The first line of antiplatelet therapy after discharge was started a mean of 17.0 days after the event; mean treatment duration was 61.9 days. The incidence rate for secondary IS was 5.53, 2.03, and 1.17 per person-year 90-days, 1-year, and 3-years following treatment initiation, respectively. Among patients matched for demographic and clinical characteristics, the risk of secondary IS was increased with DAPT versus SAPT (hazard ratio [95% CI]: 1.27 [1.20-1.34]; p < 0.0001).

Conclusions

Many patients were not prescribed or discontinued antiplatelet therapy within 90 days of hospitalization for IS/TIA and, in most cases, prescriptions were not compliant with SSP consensus guidelines. Patients remained at risk for IS, which was highest within 90 days. More effective strategies for SSP are needed to improve outcomes in this patient population.

Comparative Effectiveness and Safety of Seizure Prophylaxis Among Adults After Acute Ischemic Stroke

BACKGROUND

Older adults occasionally receive seizure prophylaxis in an acute ischemic stroke (AIS) setting, despite safety concerns. There are no trial data available about the net impact of early seizure prophylaxis on post-AIS survival.

METHODS

Using a stroke registry (American Heart Association's Get With The Guidelines) individually linked to electronic health records, we examined the effect of initiating seizure prophylaxis (ie, epilepsy-specific antiseizure drugs) within 7 days of an AIS admission versus not initiating in patients ≥65 years admitted for a new, nonsevere AIS (National Institutes of Health Stroke Severity score ≤20) between 2014 and 2021 with no recorded use of epilepsy-specific antiseizure drugs in the previous 3 months. We addressed confounding by using inverse-probability weights. We performed standardization accounting for pertinent clinical and health care factors (eg, National Institutes of Health Stroke Severity scale, prescription counts, seizure-like events).

RESULTS

The study sample included 151 patients who received antiseizure drugs and 3020 who did not. The crude 30-day mortality risks were 219 deaths per 1000 patients among epilepsy-specific antiseizure drugs initiators and 120 deaths per 1000 among noninitiators. After standardization, the estimated mortality was 251 (95% CI, 190-307) deaths per 1000 among initiators and 120 (95% CI, 86-144) deaths per 1000 among noninitiators, corresponding to a risk difference of 131 (95% CI, 65-200) excess deaths per 1000 patients. In the prespecified subgroup analyses, the risk difference was 52 (95% CI, 11-72) among patients with minor AIS and 138 (95% CI, 52-222) among moderate-to-severe AIS patients. Similarly, the risk differences were 86 (95% CI, 18-118) and 157 (95% CI, 57-219) among patients aged 65 to 74 years and ≥75 years, respectively.

CONCLUSIONS

There was a higher risk of 30-day mortality associated with initiating versus not initiating seizure prophylaxis within 7 days post-AIS. This study does not support the role of seizure prophylaxis in reducing 30-day poststroke mortality.

Neutrophil-to-Lymphocyte Ratio as a Predictor of Short-Term Functional Outcomes in Acute Ischemic Stroke Patients

Background: Neutrophil-to-lymphocyte ratio (NLR), a systemic inflammatory biomarker, has been associated with poorer outcomes in acute ischemic stroke patients. The present study was designed to expand these findings by investigating the association between NLR and short-term functional outcomes in acute ischemic stroke patients. Methods: This retrospective study evaluated patients within 7 days after the onset of acute ischemic stroke. Stroke severity on admission was measured using the National Institutes of Health Stroke Scale (NIHSS). The functional outcomes were assessed using the Berg Balance Scale (BBS), Manual Function Test (MFT), the Korean version of the modified Barthel Index (K-MBI), and the Korean Mini-Mental State Examination (K-MMSE) within 2 weeks of stroke onset. The modified Rankin Scale (mRS) was evaluated at discharge. Results: This study included 201 patients, who were grouped into three NLR tertiles (<1.84, 1.84−2.71, and >2.71) on admission. A multivariate analysis showed that the top tertile group (NLR > 2.71) had significantly higher risks of unfavorable outcomes on the K-MBI (p = 0.010) and K-MMSE (p = 0.029) than the bottom tertile group (NLR < 1.84). Based on the optimal cut-off values from a receiver operating characteristic curve analysis, a higher NLR was significantly associated with higher NIHSS scores (p = 0.011) and unfavorable outcomes on the K-MBI (p = 0.002) and K-MMSE (p = 0.001). Conclusions: A higher NLR is associated with poorer short-term functional outcomes in acute ischemic stroke patients.

Drivers of Ischemic Stroke Hospital Cost Trends Among Older Adults in the United States

PURPOSE

The increased use of neuroimaging and innovations in ischemic stroke (IS) treatment have improved outcomes, but the impact on median hospital costs is not well understood.

METHODS

A retrospective study was conducted using Medicare 5% claims data for 75,525 consecutive index IS hospitalizations for patients aged ≥65 years from 2012 to 2019 (values in 2019 dollars). IS episode cost was calculated in each year for trend analysis and stratified by cost components, including neuroimaging (CT angiography [CTA], CT perfusion [CTP], MRI, and MR angiography [MRA]), treatment (endovascular thrombectomy [EVT] and/or intravenous thrombolysis), and patient sociodemographic factors. Logistic regression was performed to analyze the drivers of high-cost episodes and median regression to assess drivers of median costs.

RESULTS

The median IS episode cost increased by 4.9% from $9,509 in 2012 to $9,973 in 2019 (P = .0021). Treatment with EVT resulted in the greatest odds of having a high-cost (>$20,000) hospitalization (odds ratio [OR], 71.86; 95% confidence interval [CI], 54.62-94.55), as did intravenous thrombolysis treatment (OR, 3.19; 95% CI, 2.90-3.52). Controlling for other factors, neuroimaging with CTA (OR, 1.72; 95% CI, 1.58-1.87), CTP (OR, 1.32; 95% CI, 1.14-1.52), and/or MRA (OR, 1.26; 95% CI, 1.15-1.38) had greater odds of having high-cost episodes than those without CTA, CTP, and MRA. Length of stay > 4 days (OR, 4.34; 95% CI, 3.99-4.72) and in-hospital mortality (OR, 1.85; 95% CI, 1.63-2.10) were also associated with high-cost episodes.

CONCLUSIONS

From 2012 to 2019, the median IS episode cost increased by 4.9%, with EVT as the main cost driver. However, the increasing treatment cost trends have been partially offset by decreases in median length of stay and in-hospital mortality.

Machine learning models to detect social distress, spiritual pain, and severe physical psychological symptoms in terminally ill patients with cancer from unstructured text data in electronic medical records

BACKGROUND

Few studies have developed automatic systems for identifying social distress, spiritual pain, and severe physical and phycological symptoms from text data in electronic medical records.

AIM

To develop models to detect social distress, spiritual pain, and severe physical and psychological symptoms in terminally ill patients with cancer from unstructured text data contained in electronic medical records.

DESIGN

A retrospective study of 1,554,736 narrative clinical records was analyzed 1 month before patients died. Supervised machine learning models were trained to detect comprehensive symptoms, and the performance of the models was tested using the area under the receiver operating characteristic curve (AUROC) and precision recall curve (AUPRC).

SETTING/PARTICIPANTS

A total of 808 patients was included in the study using records obtained from a university hospital in Japan between January 1, 2018 and December 31, 2019. As training data, we used medical records labeled for detecting social distress (n = 10,000) and spiritual pain (n = 10,000), and records that could be combined with the Support Team Assessment Schedule (based on date) for detecting severe physical/psychological symptoms (n = 5409).

RESULTS

Machine learning models for detecting social distress had AUROC and AUPRC values of 0.98 and 0.61, respectively; values for spiritual pain, were 0.90 and 0.58, respectively. The machine learning models accurately identified severe symptoms (pain, dyspnea, nausea, insomnia, and anxiety) with a high level of discrimination (AUROC > 0.8).

CONCLUSION

The machine learning models could detect social distress, spiritual pain, and severe symptoms in terminally ill patients with cancer from text data contained in electronic medical records.

The number of risk factors increases the recurrence events in ischemic stroke

PURPOSE

Stroke is a significant cause of disability worldwide and is considered a disease caused by long-term exposure to lifestyle-related risk factors. These risk factors influence the first event of stroke and recurrent stroke events, which carry more significant risks for more severe disabilities. This study specifically compared the risk factors and neurological outcome of patients with recurrent ischemic stroke to those who had just experienced their first stroke among patients admitted to the Hospital.

PATIENTS AND METHODS

We observed and analyzed 300 patients' data who met the inclusion and exclusion criteria. This retrospective observational study was conducted on consecutive acute ischemic stroke patients admitted to the top referral hospital, West Java, Indonesia. The data displayed are epidemiological characteristics, NIHSS score at admission and discharge, and the type and number of risk factors. Data were then analyzed using appropriate statistical tests.

RESULTS

Most patients had more than one risk factor with hypertension as the most frequent (268 subjects or 89.3%). In patients who experienced ischemic stroke for the first time, the average National Institutes of Health Stroke Scale (NIHSS) score was lower (6.52 ± 3.55), and the alteration of NIHSS score was higher (1.22 ± 2.26) than those with recurrent stroke (6.96 ± 3.55) for NIHSS score and 1.21 ± 1.73 for alteration of NIHSS score). We processed the data with statistical analysis and showed a positive correlation between age (P < 0.05) and the number of risk factors (P < 0.001) in the recurrent ischemic stroke group.

CONCLUSIONS

Age and the number of risk factors correlate with recurrent ischemic strokes.

Understanding and Predicting Cognitive Improvement of Young Adults in Ischemic Stroke Rehabilitation Therapy

Accurate early predictions of a patient's likely cognitive improvement as a result of a stroke rehabilitation programme can assist clinicians in assembling more effective therapeutic programs. In addition, sufficient levels of explainability, which can justify these predictions, are a crucial requirement, as reported by clinicians. This article presents a machine learning (ML) prediction model targeting cognitive improvement after therapy for stroke surviving patients. The prediction model relies on electronic health records from 201 ischemic stroke surviving patients containing demographic information, cognitive assessments at admission from 24 different standardized neuropsychology tests (e.g., TMT, WAIS-III, Stroop, RAVLT, etc.), and therapy information collected during rehabilitation (72,002 entries collected between March 2007 and September 2019). The study population covered young-adult patients with a mean age of 49.51 years and only 4.47% above 65 years of age at the stroke event (no age filter applied). Twenty different classification algorithms (from Python's Scikit-learn library) are trained and evaluated, varying their hyper-parameters and the number of features received as input. Best-performing models reported Recall scores around 0.7 and F1 scores of 0.6, showing the model's ability to identify patients with poor cognitive improvement. The study includes a detailed feature importance report that helps interpret the model's inner decision workings and exposes the most influential factors in the cognitive improvement prediction. The study showed that certain therapy variables (e.g., the proportion of memory and orientation executed tasks) had an important influence on the final prediction of the cognitive improvement of patients at individual and population levels. This type of evidence can serve clinicians in adjusting the therapeutic settings (e.g., type and load of therapy activities) and selecting the one that maximizes cognitive improvement.

Clinical and Computerized Volumetric Analysis of Posterior Fossa Decompression for Space-Occupying Cerebellar Infarction

Background and Purpose

Surgical decompression of the posterior fossa is often performed in cases with a space-occupying cerebellar infarction to prevent coma and death. In this study, we analyzed our institutional experience with this condition. We specifically attempted to address timing issues and investigated the role of cerebellar necrosectomy using imaging data and conducting volumetric analyses.

Methods

We retrospectively studied pertinent clinical and imaging data, including computerized volumetric analyses (preoperative/postoperative infarction volume, necrosectomy volume, and posterior fossa volume), from all 49 patients who underwent posterior fossa decompression surgery for cerebellar infarction in our department from January 2012 to January 2021.

Results

Thirty-five (71%) patients had a Glasgow Coma Scale (GCS) of 14–15 at admission vs. only 14 (29%) before vs. 41 (84%) following surgery. Seven (14%) patients had preventive surgery (initial GCS 14–15, preoperative GCS change ≤ 1). Only 18 (37%) patients had an mRS score of 0–3 at discharge. Estimated overall survival was 70.5% at 1 year. Interestingly, 18/20 (90%) surviving cases had a modified Rankin Scale (mRS) outcome of 0–3 (mRS 0–2: 12/20 [60%]) 1 year after surgery. Surgical timing, including preventive surgery and mass effect of the infarct, in the posterior fossa assessed semi-quantitatively (Kirollos grade) and with volumetric parameters that were not predictive of the patients' (functional) outcomes.

Conclusion

Posterior fossa decompression for cerebellar infarction is a life-saving procedure, but rapid recovery of the GCS after surgery does not necessarily translate into good functional outcome. Many patients died during follow-up, but long-term mRS outcomes of 4–5 are rare. Surgery should probably aim primarily at pressure relief, and our clinical as well as volumetric data suggest that the impact of removing an infarcted tissue may be limited. It is presumably relatively safe to initially withhold surgery in cases with a GCS of 14–15.

An overview of real-world data sources for oncology and considerations for research

Generating evidence on the use, effectiveness, and safety of new cancer therapies is a priority for researchers, health care providers, payers, and regulators given the rapid pace of change in cancer diagnosis and treatments. The use of real-world data (RWD) is integral to understanding the utilization patterns and outcomes of these new treatments among patients with cancer who are treated in clinical practice and community settings. An initial step in the use of RWD is careful study design to assess the suitability of an RWD source. This pivotal process can be guided by using a conceptual model that encourages predesign conceptualization. The primary types of RWD included are electronic health records, administrative claims data, cancer registries, and specialty data providers and networks. Careful consideration of each data type is necessary because they are collected for a specific purpose, capturing a set of data elements within a certain population for that purpose, and they vary by population coverage and longitudinality. In this review, the authors provide a high-level assessment of the strengths and limitations of each data category to inform data source selection appropriate to the study question. Overall, the development and accessibility of RWD sources for cancer research are rapidly increasing, and the use of these data requires careful consideration of composition and utility to assess important questions in understanding the use and effectiveness of new therapies.

The Allure of Big Data to Improve Stroke Outcomes: Review of Current Literature

PURPOSE OF REVIEW

To critically appraise literature on recent advances and methods using "big data" to evaluate stroke outcomes and associated factors.

RECENT FINDINGS

Recent big data studies provided new evidence on the incidence of stroke outcomes, and important emerging predictors of these outcomes. Main highlights included the identification of COVID-19 infection and exposure to a low-dose particulate matter as emerging predictors of mortality post-stroke. Demographic (age, sex) and geographical (rural vs. urban) disparities in outcomes were also identified. There was a surge in methodological (e.g., machine learning and validation) studies aimed at maximizing the efficiency of big data for improving the prediction of stroke outcomes. However, considerable delays remain between data generation and publication. Big data are driving rapid innovations in research of stroke outcomes, generating novel evidence for bridging practice gaps. Opportunity exists to harness big data to drive real-time improvements in stroke outcomes.

Diagnostic Trends and Prescription Patterns in Disruptive Mood Dysregulation Disorder and Bipolar Disorder

OBJECTIVE

Disruptive mood dysregulation disorder (DMDD) was introduced in DSM-5 to distinguish a subset of chronically irritable youth who may be incorrectly diagnosed and/or treated for pediatric bipolar disorder (BPD). This study characterized the rate of new treatment episodes and treated prevalence of BPD and DMDD from a longitudinal electronic health record database and examined the impact of DMDD on prescription trends.

METHOD

A retrospective cohort study using 2008-2018 Optum electronic health record data was conducted. Youth aged 10 to < 18 years with ≥ 183 days of database enrollment before the study cohort entry were included. Annual new treatment episode rates per 1,000 patient-years and treated prevalence (%) were estimated. Prescriptions for medications, concomitant diagnoses, and acute mental health service use for 2016-2018 were evaluated.

RESULTS

There were 7,677 youths with DMDD and 6,480 youths with BPD identified. Mean age (13-15 years) and ethnicity were similar for both groups. A rise in new treatment episode rates (0.87-1.75 per 1,000 patient-years, p < .0001) and treated prevalence (0.08%-0.35%, p < .0001) of DMDD diagnoses (2016-2018) following diagnosis inception was paralleled by decreasing new treatment episode rates (1.22-1.14 per 1,000 patient-years, p < .01) and treated prevalence (0.42%-0.36%, p < .0001) of BPD diagnoses (2015-2018). More youth in the DMDD group were prescribed medications compared with the BPD group (81.9% vs 69.4%), including antipsychotics (58.9% vs 51.0%). Higher proportions of youth with DMDD vs youth with BPD had disruptive behavior disorders (eg, 35.9% vs 20.5% had oppositional defiant disorder), and required inpatient hospitalization related to their mental health disorder (45.0% vs 33.0%).

CONCLUSION

Diagnosis of DMDD has had rapid uptake in clinical practice but is associated with increased antipsychotic and polypharmacy prescriptions and higher rates of comorbidity and inpatient hospitalization in youth with a DMDD diagnosis compared with a BPD diagnosis.

Data Science Trends Relevant to Nursing Practice: A Rapid Review of the 2020 Literature

BACKGROUND

The term "data science" encompasses several methods, many of which are considered cutting edge and are being used to influence care processes across the world. Nursing is an applied science and a key discipline in health care systems in both clinical and administrative areas, making the profession increasingly influenced by the latest advances in data science. The greater informatics community should be aware of current trends regarding the intersection of nursing and data science, as developments in nursing practice have cross-professional implications.

OBJECTIVES

This study aimed to summarize the latest (calendar year 2020) research and applications of nursing-relevant patient outcomes and clinical processes in the data science literature.

METHODS

We conducted a rapid review of the literature to identify relevant research published during the year 2020. We explored the following 16 topics: (1) artificial intelligence/machine learning credibility and acceptance, (2) burnout, (3) complex care (outpatient), (4) emergency department visits, (5) falls, (6) health care-acquired infections, (7) health care utilization and costs, (8) hospitalization, (9) in-hospital mortality, (10) length of stay, (11) pain, (12) patient safety, (13) pressure injuries, (14) readmissions, (15) staffing, and (16) unit culture.

RESULTS

Of 16,589 articles, 244 were included in the review. All topics were represented by literature published in 2020, ranging from 1 article to 59 articles. Numerous contemporary data science methods were represented in the literature including the use of machine learning, neural networks, and natural language processing.

CONCLUSION

This review provides an overview of the data science trends that were relevant to nursing practice in 2020. Examinations of such literature are important to monitor the status of data science's influence in nursing practice.

Machine Learning in Action: Stroke Diagnosis and Outcome Prediction

The application of machine learning has rapidly evolved in medicine over the past decade. In stroke, commercially available machine learning algorithms have already been incorporated into clinical application for rapid diagnosis. The creation and advancement of deep learning techniques have greatly improved clinical utilization of machine learning tools and new algorithms continue to emerge with improved accuracy in stroke diagnosis and outcome prediction. Although imaging-based feature recognition and segmentation have significantly facilitated rapid stroke diagnosis and triaging, stroke prognostication is dependent on a multitude of patient specific as well as clinical factors and hence accurate outcome prediction remains challenging. Despite its vital role in stroke diagnosis and prognostication, it is important to recognize that machine learning output is only as good as the input data and the appropriateness of algorithm applied to any specific data set. Additionally, many studies on machine learning tend to be limited by small sample size and hence concerted efforts to collate data could improve evaluation of future machine learning tools in stroke. In the present state, machine learning technology serves as a helpful and efficient tool for rapid clinical decision making while oversight from clinical experts is still required to address specific aspects not accounted for in an automated algorithm. This article provides an overview of machine learning technology and a tabulated review of pertinent machine learning studies related to stroke diagnosis and outcome prediction.

Prehospital Notification Using a Mobile Application Can Improve Regional Stroke Care System in a Metropolitan Area

BACKGROUND

Acute ischemic stroke is a time-sensitive disease. Emergency medical service (EMS) prehospital notification of potential patients with stroke could play an important role in improving the in-hospital medical response and timely treatment of patients with acute ischemic stroke. We analyzed the effects of FASTroke, a mobile app that EMS can use to notify hospitals of patients with suspected acute ischemic stroke at the prehospital stage.

METHODS

We conducted a retrospective observational study of patients diagnosed with acute ischemic stroke at 5 major hospitals in metropolitan Daegu City, Korea, from February 2020 to January 2021. The clinical conditions and time required for managing patients were compared according to whether the EMS employed FASTroke app and further compared the factors by dividing the patients into subgroups according to the preregistration received by the hospitals when using FASTroke app.

RESULTS

Of the 563 patients diagnosed with acute ischemic stroke, FASTroke was activated for 200; of these, 93 were preregistered. The FASTroke prenotification showed faster door-to-computed-tomography times (19 minutes vs. 25 minutes, P < 0.001), faster door-to-intravenous-thrombolysis times (37 minutes vs. 48 minutes, P < 0.001), and faster door-to-endovascular-thrombectomy times (82 minutes vs. 119 minutes, P < 0.001). The time was further shortened when the preregistration was conducted simultaneously by the receiving hospital.

CONCLUSION

The FASTroke app is an easy and useful tool for prenotification as a regional stroke care system in the metropolitan area, leading to reduced transport and acute ischemic stroke management time and more reperfusion treatment. The effect was more significant when the preregistration was performed jointly.

Stroke Outcome Measurements From Electronic Medical Records: Cross-sectional Study on the Effectiveness of Neural and Nonneural Classifiers

BACKGROUND

With the rapid adoption of electronic medical records (EMRs), there is an ever-increasing opportunity to collect data and extract knowledge from EMRs to support patient-centered stroke management.

OBJECTIVE

This study aims to compare the effectiveness of state-of-the-art automatic text classification methods in classifying data to support the prediction of clinical patient outcomes and the extraction of patient characteristics from EMRs.

METHODS

Our study addressed the computational problems of information extraction and automatic text classification. We identified essential tasks to be considered in an ischemic stroke value-based program. The 30 selected tasks were classified (manually labeled by specialists) according to the following value agenda: tier 1 (achieved health care status), tier 2 (recovery process), care related (clinical management and risk scores), and baseline characteristics. The analyzed data set was retrospectively extracted from the EMRs of patients with stroke from a private Brazilian hospital between 2018 and 2019. A total of 44,206 sentences from free-text medical records in Portuguese were used to train and develop 10 supervised computational machine learning methods, including state-of-the-art neural and nonneural methods, along with ontological rules. As an experimental protocol, we used a 5-fold cross-validation procedure repeated 6 times, along with subject-wise sampling. A heatmap was used to display comparative result analyses according to the best algorithmic effectiveness (F1 score), supported by statistical significance tests. A feature importance analysis was conducted to provide insights into the results.

RESULTS

The top-performing models were support vector machines trained with lexical and semantic textual features, showing the importance of dealing with noise in EMR textual representations. The support vector machine models produced statistically superior results in 71% (17/24) of tasks, with an F1 score >80% regarding care-related tasks (patient treatment location, fall risk, thrombolytic therapy, and pressure ulcer risk), the process of recovery (ability to feed orally or ambulate and communicate), health care status achieved (mortality), and baseline characteristics (diabetes, obesity, dyslipidemia, and smoking status). Neural methods were largely outperformed by more traditional nonneural methods, given the characteristics of the data set. Ontological rules were also effective in tasks such as baseline characteristics (alcoholism, atrial fibrillation, and coronary artery disease) and the Rankin scale. The complementarity in effectiveness among models suggests that a combination of models could enhance the results and cover more tasks in the future.

CONCLUSIONS

Advances in information technology capacity are essential for scalability and agility in measuring health status outcomes. This study allowed us to measure effectiveness and identify opportunities for automating the classification of outcomes of specific tasks related to clinical conditions of stroke victims, and thus ultimately assess the possibility of proactively using these machine learning techniques in real-world situations.

Use of real-world evidence in translational pharmacology research

Real-world evidence (RWE) refers to observational data gathered outside the formalism of randomized controlled trials, in real life situations, on marketed drugs. While clinical trials are the gold standards to demonstrate the efficacy and tolerability of a medicinal product, the generalizability of their results to actual use in real-life is limited by the biases induced by the very nature of clinical trials; indeed, the patients included in the trials may differ from actual users because of their concomitant diseases or treatments, or other factors excluding them from the trials. Clinical researchers and pharmaceutical industries have hence become increasingly interested in expanding and integrating RWE into clinical research, by capitalizing on the exponential growth in access to data from electronic health records, claims databases, electronic devices, software or mobile applications, registries embedded in clinical practice and social media. Meanwhile, applications of RWE may also be used for drug discovery and repurposing, for clinical developments and post-marketing studies. The aim of this review is to provide our opinion regarding the use of RWE in translational research, including non-clinical and clinical pharmacology research, at the different step of drugs development use.

Association of Social Behaviors With Community Discharge in Patients with Total Hip and Knee Replacement

OBJECTIVES

Understand the association between social determinants of health and community discharge after elective total joint arthroplasty.

DESIGN

Retrospective cohort design using Optum de-identified electronic health record dataset.

SETTING AND PARTICIPANTS

A total of 38 hospital networks and 18 non-network hospitals in the United States; 79,725 patients with total hip arthroplasty and 136,070 patients with total knee arthroplasty between 2011 and 2018.

METHODS

Logistic regression models were used to examine the association among pain, weight status, smoking status, alcohol use, substance disorder, and postsurgical community discharge, adjusted for patient demographics.

RESULTS

Mean ages for patients with hip and knee arthroplasty were 64.5 (SD 11.3) and 65.9 (SD 9.6) years; most patients were women (53.6%, 60.2%), respectively. The unadjusted community discharge rate was 82.8% after hip and 81.1% after knee arthroplasty. After adjusting for demographics, clinical factors, and behavioral factors, we found obesity [hip: odds ratio (OR) 0.81, 95% confidence interval (CI) 0.76-0.85; knee: OR 0.73, 95% CI 0.69-0.77], current smoking (hip: OR 0.82, 95% CI 0.77-0.88; knee: OR 0.90, 95% CI 0.85-0.95), and history of substance use disorder (hip: OR 0.55, 95% CI 0.50-0.60; knee: OR 0.57, 95% CI 0.53-0.62) were associated with lower odds of community discharge after hip and knee arthroplasty, respectively.

CONCLUSIONS AND IMPLICATIONS

Social determinants of health are associated with odds of community discharge after total hip and knee joint arthroplasty. Our findings demonstrate the value of using electronic health record data to analyze more granular patient factors associated with patient discharge location after total joint arthroplasty. Although bundled payment is increasing community discharge rates, post-acute care facilities must be prepared to manage more complex patients because odds of community discharge are diminished in those who are obese, smoking, or have a history of substance use disorder.

Discrepancies in Stroke Distribution and Dataset Origin in Machine Learning for Stroke

BACKGROUND

Machine learning algorithms depend on accurate and representative datasets for training in order to become valuable clinical tools that are widely generalizable to a varied population. We aim to conduct a review of machine learning uses in stroke literature to assess the geographic distribution of datasets and patient cohorts used to train these models and compare them to stroke distribution to evaluate for disparities.

AIMS

582 studies were identified on initial searching of the PubMed database. Of these studies, 106 full texts were assessed after title and abstract screening which resulted in 489 papers excluded. Of these 106 studies, 79 were excluded due to using cohorts from outside the United States or being review articles or editorials. 27 studies were thus included in this analysis.

SUMMARY OF REVIEW

Of the 27 studies included, 7 (25.9%) used patient data from California, 6 (22.2%) were multicenter, 3 (11.1%) were in Massachusetts, 2 (7.4%) each in Illinois, Missouri, and New York, and 1 (3.7%) each from South Carolina, Washington, West Virginia, and Wisconsin. 1 (3.7%) study used data from Utah and Texas. These were qualitatively compared to a CDC study showing the highest distribution of stroke in Mississippi (4.3%) followed by Oklahoma (3.4%), Washington D.C. (3.4%), Louisiana (3.3%), and Alabama (3.2%) while the prevalence in California was 2.6%.

CONCLUSIONS

It is clear that a strong disconnect exists between the datasets and patient cohorts used in training machine learning algorithms in clinical research and the stroke distribution in which clinical tools using these algorithms will be implemented. In order to ensure a lack of bias and increase generalizability and accuracy in future machine learning studies, datasets using a varied patient population that reflects the unequal distribution of stroke risk factors would greatly benefit the usability of these tools and ensure accuracy on a nationwide scale.

Risk of Stroke Outcomes in Atrial Fibrillation Patients Treated with Rivaroxaban and Warfarin

OBJECTIVES

In a previous real-world study, rivaroxaban reduced the risk of stroke overall and severe stroke compared with warfarin in patients with nonvalvular atrial fibrillation (NVAF). The aim of this study was to assess the reproducibility in a different database of our previously observed results (Alberts M, et al. Stroke. 2020;51:549-555) on the risk of severe stroke among NVAF patients in a different population treated with rivaroxaban or warfarin.

MATERIAL AND METHODS

This retrospective cohort study included patients from the IBM® MarketScan® Commercial and Medicare databases (2011-2019) who initiated rivaroxaban or warfarin after a diagnosis of NVAF, had ≥6 months of continuous health plan enrollment, had a CHA₂DS₂-VASc score ≥2, and had no history of stroke or anticoagulant use. Patient data were assessed until the earliest occurrence of a primary inpatient diagnosis of stroke, death, end of health plan enrollment, or end of study. Stroke severity was defined by National Institutes of Health Stroke Scale (NIHSS) score, imputed by random forest model. Cox proportional hazard regression was used to compare risk of stroke between cohorts, balanced by inverse probability of treatment weighting.

RESULTS

The mean observation period from index date to either stroke, or end of eligibility or end of data was 28 months. Data from 13,599 rivaroxaban and 39,861 warfarin patients were included. Stroke occurred in 272 rivaroxaban-treated patients (0.97/100 person-years [PY]) and 1,303 warfarin-treated patients (1.32/100 PY). Rivaroxaban patients had lower risk for stroke overall (hazard ratio [HR], 0.82; 95% confidence interval [CI], 0.76-0.88) and for minor (NIHSS 1 to <5; HR, 0.83; 95% CI, 0.74-0.93), moderate (NIHSS 5 to <16; HR, 0.88; 95% CI, 0.78-0.99), and severe stroke (NIHSS 16 to 42; HR, 0.44; 95% CI, 0.22-0.91).

CONCLUSIONS

The results of this study in a larger population of NVAF patients align with previous real-world findings and the ROCKET-AF trial by showing improved stroke prevention with rivaroxaban versus warfarin across all stroke severities.

Electronic Medical Record-Based Case Phenotyping for the Charlson Conditions: Scoping Review

Background

Electronic medical records (EMRs) contain large amounts of rich clinical information. Developing EMR-based case definitions, also known as EMR phenotyping, is an active area of research that has implications for epidemiology, clinical care, and health services research.

Objective

This review aims to describe and assess the present landscape of EMR-based case phenotyping for the Charlson conditions.

Methods

A scoping review of EMR-based algorithms for defining the Charlson comorbidity index conditions was completed. This study covered articles published between January 2000 and April 2020, both inclusive. Embase (Excerpta Medica database) and MEDLINE (Medical Literature Analysis and Retrieval System Online) were searched using keywords developed in the following 3 domains: terms related to EMR, terms related to case finding, and disease-specific terms. The manuscript follows the Preferred Reporting Items for Systematic reviews and Meta-analyses extension for Scoping Reviews (PRISMA) guidelines.

Results

A total of 274 articles representing 299 algorithms were assessed and summarized. Most studies were undertaken in the United States (181/299, 60.5%), followed by the United Kingdom (42/299, 14.0%) and Canada (15/299, 5.0%). These algorithms were mostly developed either in primary care (103/299, 34.4%) or inpatient (168/299, 56.2%) settings. Diabetes, congestive heart failure, myocardial infarction, and rheumatology had the highest number of developed algorithms. Data-driven and clinical rule–based approaches have been identified. EMR-based phenotype and algorithm development reflect the data access allowed by respective health systems, and algorithms vary in their performance.

Conclusions

Recognizing similarities and differences in health systems, data collection strategies, extraction, data release protocols, and existing clinical pathways is critical to algorithm development strategies. Several strategies to assist with phenotype-based case definitions have been proposed.

Healthcare costs before and after stroke in patients with non-valvular atrial fibrillation who initiated treatment with rivaroxaban or warfarin

AIMS

Rivaroxaban reduces stroke compared with warfarin in patients with non-valvular atrial fibrillation (NVAF). This study compared healthcare costs before and after stroke in NVAF patients treated with rivaroxaban or warfarin.

MATERIALS AND METHODS

Using de-identified IBM MarketScan Commercial and Medicare databases, this retrospective cohort study (from 2011 to 2019) included patients with NVAF who initiated rivaroxaban or warfarin within 30 days after initial NVAF diagnosis. Patients who developed stroke were identified, and stroke severity was determined by the National Institutes of Health Stroke Scale (NIHSS) score, imputed by a random forest method. Total all-cause per-patient per-year (PPPY) costs of care were determined for patients: (1) who developed stroke during the pre- and post-stroke periods and (2) who remained stroke-free during the follow-up period. Treatment groups were balanced using inverse probability of treatment weighting.

RESULTS

A total of 13,599 patients initiated rivaroxaban and 39,861 initiated warfarin, of which 272 (2.0%) and 1,303 (3.3%), respectively, developed stroke during a mean follow-up of 28 months. Among patients who developed stroke, PPPY costs increased from the pre-stroke to post-stroke period, with greater increases in the warfarin cohort relative to the rivaroxaban cohort. Overall, the costs increased by 1.78-fold for rivaroxaban vs 3.07-fold for warfarin; for less severe strokes (NIHSS < 5), costs increased 0.88-fold and 1.05-fold, respectively. Cost increases for more severe strokes (NIHSS ≥ 5) among rivaroxaban patients were half those for warfarin patients (3.19-fold vs 6.37-fold, respectively). Among patients without stroke, costs were similar during the follow-up period between the two treatment groups.

CONCLUSIONS

Total all-cause costs of care increased in the post-stroke period, and particularly in the patients treated with warfarin relative to those treated with rivaroxaban. The lower rate of stroke in the rivaroxaban cohort suggests that greater pre- to post-stroke cost increases result from more strokes occurring in the warfarin cohort.

Association of Pain on Hospital Discharge with the Risk of 30-Day Readmission in Patients with Total Hip and Knee Replacement

BACKGROUND

It is not clear if there is a risk of 30-day readmissions following total hip and knee arthroplasty in patients reporting high levels of pain at hospital discharge. We examined the relationship between post-surgical pain on the day of discharge and 30-day readmission in patients who received total knee and hip arthroplasty.

METHODS

Retrospective cohort study was conducted of patients who received total knee (n = 155,284) or hip arthroplasty (n = 89,283) from 2011 to 2018 using electronic health records from the Optum database. Four categories of pain at discharge were created, from none to severe. Multivariate logistic regression models to predict 30-day all-cause readmission were adjusted for patient and clinical characteristics and built separately for knee and hip arthroplasty patients.

RESULTS

Mean ages for hip and knee patients were 64.4 (standard deviation 11.3) and 65.7 (standard deviation 9.7) years, respectively. The majority of patients were female (hip: 54.4%; knee: 61.5%). The unadjusted rate of 30-day readmission was 3.54% for hip replacement and 3.66% for knee replacement. In models adjusted for patient and clinical characteristics, for patients with total hip replacement, the odds of 30-day readmission for those with severe pain score at discharge vs those with no pain at discharge were 1.60 (95% confidence interval 1.33-1.92). Similarly, readmission likelihood increased as pain at discharge increased (severe pain vs no pain) for patients with total knee arthroplasty (odds ratio 1.38, 95% confidence interval 1.19-1.59).

CONCLUSION

Our findings demonstrated that the pain scores on the day of discharge are associated with 30-day hospital readmission.

Semantic text mining in early drug discovery for type 2 diabetes

Background

Surveying the scientific literature is an important part of early drug discovery; and with the ever-increasing amount of biomedical publications it is imperative to focus on the most interesting articles. Here we present a project that highlights new understanding (e.g. recently discovered modes of action) and identifies potential drug targets, via a novel, data-driven text mining approach to score type 2 diabetes (T2D) relevance. We focused on monitoring trends and jumps in T2D relevance to help us be timely informed of important breakthroughs.

Methods

We extracted over 7 million n-grams from PubMed abstracts and then clustered around 240,000 linked to T2D into almost 50,000 T2D relevant ‘semantic concepts’. To score papers, we weighted the concepts based on co-mentioning with core T2D proteins. A protein’s T2D relevance was determined by combining the scores of the papers mentioning it in the five preceding years. Each week all proteins were ranked according to their T2D relevance. Furthermore, the historical distribution of changes in rank from one week to the next was used to calculate the significance of a change in rank by T2D relevance for each protein.

Results

We show that T2D relevant papers, even those not mentioning T2D explicitly, were prioritised by relevant semantic concepts. Well known T2D proteins were therefore enriched among the top scoring proteins. Our ‘high jumpers’ identified important past developments in the apprehension of how certain key proteins relate to T2D, indicating that our method will make us aware of future breakthroughs. In summary, this project facilitated keeping up with current T2D research by repeatedly providing short lists of potential novel targets into our early drug discovery pipeline.