Machine Learning-Based Prediction of Stroke in Emergency Departments.

Background: Stroke misdiagnosis, associated with poor outcomes, is estimated to occur in 9% of all stroke patients. Objectives: We hypothesized that machine learning (ML) could assist in the diagnosis of ischemic stroke in emergency departments (EDs). Design: The study was conducted and reported according to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis guidelines. We performed model development and prospective temporal validation, using data from pre- and post-COVID periods; we also performed a case study on a small cohort of previously misdiagnosed stroke patients. Methods: We used structured and unstructured electronic health records (EHRs) of 56,452 patient encounters from 13 hospitals in Pennsylvania, from September 2003 to January 2021. ML pipelines, including natural language processing, were created using pre-event clinical data and provider notes in the EDs. Results: Using pre-event information, our model's area under the receiver operating characteristics curve (AUROC) ranged from 0.88 to 0.92 with a similar range accuracy (0.87-0.90). Using provider notes, we identified five models that reached a balanced performance in terms of AUROC, sensitivity, and specificity. Model AUROC ranged from 0.93 to 0.99. Model sensitivity and specificity reached 0.90 and 0.99, respectively. Four of the top five performing models were based on the post-COVID provider notes; however, no performance difference between models tested on pre- and post-COVID was observed. Conclusion: This study leveraged pre-event and at-encounter level EHR for stroke prediction. The results indicate that available clinical information can be used for building EHR-based stroke prediction models and ED stroke alert systems.

An integrated pipeline for prediction of Clostridioides difficile infection.

With the expansion of electronic health records(EHR)-linked genomic data comes the development of machine learning-enable models. There is a pressing need to develop robust pipelines to evaluate the performance of integrated models and minimize systemic bias. We developed a prediction model of symptomatic Clostridioides difficile infection(CDI) by integrating common EHR-based and genetic risk factors(rs2227306/IL8). Our pipeline includes (1) leveraging phenotyping algorithm to minimize temporal bias, (2) performing simulation studies to determine the predictive power in samples without genetic information, (3) propensity score matching to control for the confoundings, (4) selecting machine learning algorithms to capture complex feature interactions, (5) performing oversampling to address data imbalance, and (6) optimizing models and ensuring proper bias-variance trade-off. We evaluate the performance of prediction models of CDI when including common clinical risk factors and the benefit of incorporating genetic feature(s) into the models. We emphasize the importance of building a robust integrated pipeline to avoid systemic bias and thoroughly evaluating genetic features when integrated into the prediction models in the general population and subgroups.

The Phenotypic Spectrum of COL4A3 Heterozygotes.

Introduction: The penetrance and phenotypic spectrum of autosomal dominant Alport Syndrome (ADAS), affecting 1 in 106, remains understudied. Methods: Using data from 174,418 participants in the Geisinger MyCode/DiscovEHR study, an unselected health system-based cohort with whole exome sequencing, we identified 403 participants who were heterozygous for likely pathogenic COL4A3 variants. Phenotypic data was evaluated using International Classification of Diseases (ICD) codes, laboratory data, and chart review. To evaluate the phenotypic spectrum of genetically-determined ADAS, we matched COL4A3 heterozygotes 1:5 to nonheterozygotes using propensity scores by demographics, hypertension, diabetes, and nephrolithiasis. Results: COL4A3 heterozygotes were at significantly increased risks of hematuria, decreased estimated glomerular filtration rate (eGFR), albuminuria, and kidney failure (P < 0.05 for all comparisons) but not bilateral sensorineural hearing loss (P = 0.9). Phenotypic severity was more severe for collagenous domain glycine missense variants than protein truncating variants (PTVs). For example, patients with Gly695Arg (n = 161) had markedly increased risk of dipstick hematuria (odds ratio [OR] 9.50; 95% confidence interval [CI]: 6.32, 14.28) and kidney failure (OR 7.02; 95% CI: 3.48, 14.16) whereas those with PTVs (n = 119) had moderately increased risks of dipstick hematuria (OR 1.64; 95% CI: 1.03, 2.59) and kidney failure (OR 3.44; 95% CI: 1.28, 9.22). Less than a third of patients had albuminuria screening completed, and fewer than 1 of 3 were taking inhibitors of the renin-angiotensin-aldosterone system. Conclusion: This study demonstrates a wide spectrum of phenotypic severity in ADAS due to COL4A3 with phenotypic variability by genotype. Future studies are needed to evaluate the impact of earlier diagnosis, appropriate evaluation, and treatment of ADAS.

The Phenotypic Spectrum of COL4A3 Heterozygotes.

Most data on Alport Syndrome (AS) due to COL4A3 are limited to families with autosomal recessive AS or severe manifestations such as focal segmental glomerulosclerosis (FSGS). Using data from 174,418 participants in the Geisinger MyCode/DiscovEHR study, an unselected health system-based cohort with whole exome sequencing, we identified 403 participants (0.2%) who were heterozygous for likely pathogenic COL4A3 variants. Phenotypic data was evaluated using International Classification of Diseases (ICD) codes, laboratory data, and chart review. To evaluate the phenotypic spectrum of genetically-determined autosomal dominant AS, we matched COL4A3 heterozygotes 1:5 to non-heterozygotes using propensity scores by demographics, hypertension, diabetes, and nephrolithiasis. COL4A3 heterozygotes were at significantly increased risks of hematuria, decreased estimated glomerular filtration rate (eGFR), albuminuria, and end-stage kidney disease (ESKD) (p<0.05 for all comparisons) but not bilateral sensorineural hearing loss (p=0.9). Phenotypic severity tended to be more severe among patients with glycine missense variants located within the collagenous domain. For example, patients with Gly695Arg (n=161) had markedly increased risk of dipstick hematuria (OR 9.47, 95% CI: 6.30, 14.22) and ESKD diagnosis (OR 7.01, 95% CI: 3.48, 14.12) whereas those with PTVs (n=119) had moderately increased risks of dipstick hematuria (OR 1.63, 95% CI: 1.03, 2.58) and ESKD diagnosis (OR 3.43, 95% CI: 1.28, 9.19). Less than a third of patients had albuminuria screening completed, and fewer than 1/3 were taking inhibitors of the renin-angiotensin-aldosterone system (RAASi). Future studies are needed to evaluate the impact of earlier diagnosis, appropriate evaluation, and treatment of ADAS.

Defining the Age of Young Ischemic Stroke Using Data-Driven Approaches.

Introduction: The cut-point for defining the age of young ischemic stroke (IS) is clinically and epidemiologically important, yet it is arbitrary and differs across studies. In this study, we leveraged electronic health records (EHRs) and data science techniques to estimate an optimal cut-point for defining the age of young IS. Methods: Patient-level EHRs were extracted from 13 hospitals in Pennsylvania, and used in two parallel approaches. The first approach included ICD9/10, from IS patients to group comorbidities, and computed similarity scores between every patient pair. We determined the optimal age of young IS by analyzing the trend of patient similarity with respect to their clinical profile for different ages of index IS. The second approach used the IS cohort and control (without IS), and built three sets of machine-learning models-generalized linear regression (GLM), random forest (RF), and XGBoost (XGB)-to classify patients for seventeen age groups. After extracting feature importance from the models, we determined the optimal age of young IS by analyzing the pattern of comorbidity with respect to the age of index IS. Both approaches were completed separately for male and female patients. Results: The stroke cohort contained 7555 ISs, and the control included 31,067 patients. In the first approach, the optimal age of young stroke was 53.7 and 51.0 years in female and male patients, respectively. In the second approach, we created 102 models, based on three algorithms, 17 age brackets, and two sexes. The optimal age was 53 (GLM), 52 (RF), and 54 (XGB) for female, and 52 (GLM and RF) and 53 (RF) for male patients. Different age and sex groups exhibited different comorbidity patterns. Discussion: Using a data-driven approach, we determined the age of young stroke to be 54 years for women and 52 years for men in our mainly rural population, in central Pennsylvania. Future validation studies should include more diverse populations.

Social Determinants of Stroke Hospitalization and Mortality in United States' Counties.

(1) Background: Stroke incidence and outcomes are influenced by socioeconomic status. There is a paucity of reported population-level studies regarding these determinants. The goal of this ecological analysis was to determine the county-level associations of social determinants of stroke hospitalization and death rates in the United States. (2) Methods: Publicly available data as of 9 April 2021, for the socioeconomic factors and outcomes, was extracted from the Centers for Disease Control and Prevention. The outcomes of interest were "all stroke hospitalization rates per 1000 Medicare beneficiaries" (SHR) and "all stroke death rates per 100,000 population" (SDR). We used a multivariate binomial generalized linear mixed model after converting the outcomes to binary based on their median values. (3) Results: A total of 3226 counties/county-equivalents of the states and territories in the US were analyzed. Heart disease prevalence (odds ratio, OR = 2.03, p < 0.001), blood pressure medication nonadherence (OR = 2.02, p < 0.001), age-adjusted obesity (OR = 1.24, p = 0.006), presence of hospitals with neurological services (OR = 1.9, p < 0.001), and female head of household (OR = 1.32, p = 0.021) were associated with high SHR while cost of care per capita for Medicare patients with heart disease (OR = 0.5, p < 0.01) and presence of hospitals (OR = 0.69, p < 0.025) were associated with low SHR. Median household income (OR = 0.6, p < 0.001) and park access (OR = 0.84, p = 0.016) were associated with low SDR while no college degree (OR = 1.21, p = 0.049) was associated with high SDR. (4) Conclusions: Several socioeconomic factors (e.g., education, income, female head of household) were found to be associated with stroke outcomes. Additional research is needed to investigate intermediate and potentially modifiable factors that can serve as targeted interventions.

Artificial Intelligence: A Shifting Paradigm in Cardio-Cerebrovascular Medicine.

The future of healthcare is an organic blend of technology, innovation, and human connection. As artificial intelligence (AI) is gradually becoming a go-to technology in healthcare to improve efficiency and outcomes, we must understand our limitations. We should realize that our goal is not only to provide faster and more efficient care, but also to deliver an integrated solution to ensure that the care is fair and not biased to a group of sub-population. In this context, the field of cardio-cerebrovascular diseases, which encompasses a wide range of conditions-from heart failure to stroke-has made some advances to provide assistive tools to care providers. This article aimed to provide an overall thematic review of recent development focusing on various AI applications in cardio-cerebrovascular diseases to identify gaps and potential areas of improvement. If well designed, technological engines have the potential to improve healthcare access and equitability while reducing overall costs, diagnostic errors, and disparity in a system that affects patients and providers and strives for efficiency.

Imputation of missing values for electronic health record laboratory data.

Laboratory data from Electronic Health Records (EHR) are often used in prediction models where estimation bias and model performance from missingness can be mitigated using imputation methods. We demonstrate the utility of imputation in two real-world EHR-derived cohorts of ischemic stroke from Geisinger and of heart failure from Sutter Health to: (1) characterize the patterns of missingness in laboratory variables; (2) simulate two missing mechanisms, arbitrary and monotone; (3) compare cross-sectional and multi-level multivariate missing imputation algorithms applied to laboratory data; (4) assess whether incorporation of latent information, derived from comorbidity data, can improve the performance of the algorithms. The latter was based on a case study of hemoglobin A1c under a univariate missing imputation framework. Overall, the pattern of missingness in EHR laboratory variables was not at random and was highly associated with patients' comorbidity data; and the multi-level imputation algorithm showed smaller imputation error than the cross-sectional method.

Predicting short and long-term mortality after acute ischemic stroke using EHR.

OBJECTIVE: Despite improvements in treatment, stroke remains a leading cause of mortality and long-term disability. In this study, we leveraged administrative data to build predictive models of short- and long-term post-stroke all-cause-mortality. METHODS: The study was conducted and reported according to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guideline. We used patient-level data from electronic health records, three algorithms, and six prediction windows to develop models for post-stroke mortality. RESULTS: We included 7144 patients from which 5347 had survived their ischemic stroke after two years. The proportion of mortality was between 8%(605/7144) within 1-month, to 25%(1797/7144) for the 2-years window. The three most common comorbidities were hypertension, dyslipidemia, and diabetes. The best Area Under the ROC curve(AUROC) was reached with the Random Forest model at 0.82 for the 1-month prediction window. The negative predictive value (NPV) was highest for the shorter prediction windows - 0.91 for the 1-month - and the best positive predictive value (PPV) was reached for the 6-months prediction window at 0.92. Age, hemoglobin levels, and body mass index were the top associated factors. Laboratory variables had higher importance when compared to past medical history and comorbidities. Hypercoagulation state, smoking, and end-stage renal disease were more strongly associated with long-term mortality. CONCLUSION: All the selected algorithms could be trained to predict the short and long-term mortality after stroke. The factors associated with mortality differed depending on the prediction window. Our classifier highlighted the importance of controlling risk factors, as indicated by laboratory measures.

SARS-CoV-2 Is a Culprit for Some, but Not All Acute Ischemic Strokes: A Report from the Multinational COVID-19 Stroke Study Group.

BACKGROUND: SARS-CoV-2 infected patients are suggested to have a higher incidence of thrombotic events such as acute ischemic strokes (AIS). This study aimed at exploring vascular comorbidity patterns among SARS-CoV-2 infected patients with subsequent stroke. We also investigated whether the comorbidities and their frequencies under each subclass of TOAST criteria were similar to the AIS population studies prior to the pandemic. METHODS: This is a report from the Multinational COVID-19 Stroke Study Group. We present an original dataset of SASR-CoV-2 infected patients who had a subsequent stroke recorded through our multicenter prospective study. In addition, we built a dataset of previously reported patients by conducting a systematic literature review. We demonstrated distinct subgroups by clinical risk scoring models and unsupervised machine learning algorithms, including hierarchical K-Means (ML-K) and Spectral clustering (ML-S). RESULTS: This study included 323 AIS patients from 71 centers in 17 countries from the original dataset and 145 patients reported in the literature. The unsupervised clustering methods suggest a distinct cohort of patients (ML-K: 36% and ML-S: 42%) with no or few comorbidities. These patients were more than 6 years younger than other subgroups and more likely were men (ML-K: 59% and ML-S: 60%). The majority of patients in this subgroup suffered from an embolic-appearing stroke on imaging (ML-K: 83% and ML-S: 85%) and had about 50% risk of large vessel occlusions (ML-K: 50% and ML-S: 53%). In addition, there were two cohorts of patients with large-artery atherosclerosis (ML-K: 30% and ML-S: 43% of patients) and cardioembolic strokes (ML-K: 34% and ML-S: 15%) with consistent comorbidity and imaging patterns. Binominal logistic regression demonstrated that ischemic heart disease (odds ratio (OR), 4.9; 95% confidence interval (CI), 1.6-14.7), atrial fibrillation (OR, 14.0; 95% CI, 4.8-40.8), and active neoplasm (OR, 7.1; 95% CI, 1.4-36.2) were associated with cardioembolic stroke. CONCLUSIONS: Although a cohort of young and healthy men with cardioembolic and large vessel occlusions can be distinguished using both clinical sub-grouping and unsupervised clustering, stroke in other patients may be explained based on the existing comorbidities.

Prediction of Long-Term Stroke Recurrence Using Machine Learning Models.

BACKGROUND: The long-term risk of recurrent ischemic stroke, estimated to be between 17% and 30%, cannot be reliably assessed at an individual level. Our goal was to study whether machine-learning can be trained to predict stroke recurrence and identify key clinical variables and assess whether performance metrics can be optimized. METHODS: We used patient-level data from electronic health records, six interpretable algorithms (Logistic Regression, Extreme Gradient Boosting, Gradient Boosting Machine, Random Forest, Support Vector Machine, Decision Tree), four feature selection strategies, five prediction windows, and two sampling strategies to develop 288 models for up to 5-year stroke recurrence prediction. We further identified important clinical features and different optimization strategies. RESULTS: We included 2091 ischemic stroke patients. Model area under the receiver operating characteristic (AUROC) curve was stable for prediction windows of 1, 2, 3, 4, and 5 years, with the highest score for the 1-year (0.79) and the lowest score for the 5-year prediction window (0.69). A total of 21 (7%) models reached an AUROC above 0.73 while 110 (38%) models reached an AUROC greater than 0.7. Among the 53 features analyzed, age, body mass index, and laboratory-based features (such as high-density lipoprotein, hemoglobin A1c, and creatinine) had the highest overall importance scores. The balance between specificity and sensitivity improved through sampling strategies. CONCLUSION: All of the selected six algorithms could be trained to predict the long-term stroke recurrence and laboratory-based variables were highly associated with stroke recurrence. The latter could be targeted for personalized interventions. Model performance metrics could be optimized, and models can be implemented in the same healthcare system as intelligent decision support for targeted intervention.

Trends in ischemic stroke outcomes in a rural population in the United States.

INTRODUCTION: The stroke mortality rate has gradually declined due to improved interventions and controlled risk factors. We investigated the associated factors and trends in recurrence and all-cause mortality in ischemic stroke patients from a rural population in the United States between 2004 and 2018. METHODS: This was a retrospective cohort study based on electronic health records (EHR) data. A comprehensive stroke database called "Geisinger NeuroScience Ischemic Stroke (GNSIS)" was built for this study. Clinical data were extracted from multiple sources, including EHR and quality data. RESULTS: The cohort included in the study comprised of 8561 consecutive ischemic stroke patients (mean age: 70.1 ± 13.9 years, men: 51.6%, 95.1% Caucasian). Hypertension was the most prevalent risk factor (75.2%). The one-year recurrence and all-cause mortality rates were 6.3% and 16.1%, respectively. Although the one-year stroke recurrence increased during the study period, the one-year stroke mortality rate decreased significantly. Age > 65 years, atrial fibrillation or flutter, heart failure, and prior ischemic stroke were independently associated with one-year all-cause mortality in stratified Cox proportional hazards model. In the Cause-specific hazard model, diabetes, chronic kidney disease and age < 65 years were found to be associated with one-year ischemic stroke recurrence. CONCLUSION: Although all-cause mortality after stroke has decreased, stroke recurrence has significantly increased in stroke patients from rural population between 2004 and 2018. Older age, atrial fibrillation or flutter, heart failure, and prior ischemic stroke were independently associated with one-year all-cause mortality while diabetes, chronic kidney disease and age less than 65 years were predictors of ischemic stroke recurrence.

Early Detection of Septic Shock Onset Using Interpretable Machine Learners.

BACKGROUND: Developing a decision support system based on advances in machine learning is one area for strategic innovation in healthcare. Predicting a patient's progression to septic shock is an active field of translational research. The goal of this study was to develop a working model of a clinical decision support system for predicting septic shock in an acute care setting for up to 6 h from the time of admission in an integrated healthcare setting. METHOD: Clinical data from Electronic Health Record (EHR), at encounter level, were used to build a predictive model for progression from sepsis to septic shock up to 6 h from the time of admission; that is, T = 1, 3, and 6 h from admission. Eight different machine learning algorithms (Random Forest, XGBoost, C5.0, Decision Trees, Boosted Logistic Regression, Support Vector Machine, Logistic Regression, Regularized Logistic, and Bayes Generalized Linear Model) were used for model development. Two adaptive sampling strategies were used to address the class imbalance. Data from two sources (clinical and billing codes) were used to define the case definition (septic shock) using the Centers for Medicare & Medicaid Services (CMS) Sepsis criteria. The model assessment was performed using Area under Receiving Operator Characteristics (AUROC), sensitivity, and specificity. Model predictions for each feature window (1, 3 and 6 h from admission) were consolidated. RESULTS: Retrospective data from April 2005 to September 2018 were extracted from the EHR, Insurance Claims, Billing, and Laboratory Systems to create a dataset for septic shock detection. The clinical criteria and billing information were used to label patients into two classes-septic shock patients and sepsis patients at three different time points from admission, creating two different case-control cohorts. Data from 45,425 unique in-patient visits were used to build 96 prediction models comparing clinical-based definition versus billing-based information as the gold standard. Of the 24 consolidated models (based on eight machine learning algorithms and three feature windows), four models reached an AUROC greater than 0.9. Overall, all the consolidated models reached an AUROC of at least 0.8820 or higher. Based on the AUROC of 0.9483, the best model was based on Random Forest, with a sensitivity of 83.9% and specificity of 88.1%. The sepsis detection window at 6 h outperformed the 1 and 3-h windows. The sepsis definition based on clinical variables had improved performance when compared to the sepsis definition based on only billing information. CONCLUSION: This study corroborated that machine learning models can be developed to predict septic shock using clinical and administrative data. However, the use of clinical information to define septic shock outperformed models developed based on only administrative data. Intelligent decision support tools can be developed and integrated into the EHR and improve clinical outcomes and facilitate the optimization of resources in real-time.

Racial, Economic, and Health Inequality and COVID-19 Infection in the United States.

OBJECTIVES: There is preliminary evidence of racial and social economic disparities in the population infected by and dying from COVID-19. The goal of this study is to report the associations of COVID-19 with respect to race, health, and economic inequality in the United States. METHODS: We performed an ecological study of the associations between infection and mortality rate of COVID-19 and demographic, socioeconomic, and mobility variables from 369 counties (total population, 102,178,117 [median, 73,447; IQR, 30,761-256,098]) from the seven most affected states (Michigan, New York, New Jersey, Pennsylvania, California, Louisiana, Massachusetts). RESULTS: The risk factors for infection and mortality are different. Our analysis shows that counties with more diverse demographics, higher population, education, income levels, and lower disability rates were at a higher risk of COVID-19 infection. However, counties with higher proportion with disability and poverty rates had a higher death rate. African Americans were more vulnerable to COVID-19 than other ethnic groups (1981 African American infected cases versus 658 Whites per million). Data on mobility changes corroborate the impact of social distancing. CONCLUSION: Our study provides evidence of racial, economic, and health inequality in the population infected by and dying from COVID-19. These observations might be due to the workforce of essential services, poverty, and access to care. Counties in more urban areas are probably better equipped at providing care. The lower rate of infection, but a higher death rate in counties with higher poverty and disability could be due to lower levels of mobility, but a higher rate of comorbidities and health care access.

Using artificial intelligence for improving stroke diagnosis in emergency departments: a practical framework.

Stroke is the fifth leading cause of death in the United States and a major cause of severe disability worldwide. Yet, recognizing the signs of stroke in an acute setting is still challenging and leads to loss of opportunity to intervene, given the narrow therapeutic window. A decision support system using artificial intelligence (AI) and clinical data from electronic health records combined with patients' presenting symptoms can be designed to support emergency department providers in stroke diagnosis and subsequently reduce the treatment delay. In this article, we present a practical framework to develop a decision support system using AI by reflecting on the various stages, which could eventually improve patient care and outcome. We also discuss the technical, operational, and ethical challenges of the process.

Risk of stroke in hospitalized SARS-CoV-2 infected patients: A multinational study.

BACKGROUND: There is an increased attention to stroke following SARS-CoV-2. The goal of this study was to better depict the short-term risk of stroke and its associated factors among SARS-CoV-2 hospitalized patients. METHODS: This multicentre, multinational observational study includes hospitalized SARS-CoV-2 patients from North and South America (United States, Canada, and Brazil), Europe (Greece, Italy, Finland, and Turkey), Asia (Lebanon, Iran, and India), and Oceania (New Zealand). The outcome was the risk of subsequent stroke. Centres were included by non-probability sampling. The counts and clinical characteristics including laboratory findings and imaging of the patients with and without a subsequent stroke were recorded according to a predefined protocol. Quality, risk of bias, and heterogeneity assessments were conducted according to ROBINS-E and Cochrane Q-test. The risk of subsequent stroke was estimated through meta-analyses with random effect models. Bivariate logistic regression was used to determine the parameters with predictive outcome value. The study was reported according to the STROBE, MOOSE, and EQUATOR guidelines. FINDINGS: We received data from 26,175 hospitalized SARS-CoV-2 patients from 99 tertiary centres in 65 regions of 11 countries until May 1st, 2020. A total of 17,799 patients were included in meta-analyses. Among them, 156(0.9%) patients had a stroke-123(79%) ischaemic stroke, 27(17%) intracerebral/subarachnoid hemorrhage, and 6(4%) cerebral sinus thrombosis. Subsequent stroke risks calculated with meta-analyses, under low to moderate heterogeneity, were 0.5% among all centres in all countries, and 0.7% among countries with higher health expenditures. The need for mechanical ventilation (OR: 1.9, 95% CI:1.1-3.5, p = 0.03) and the presence of ischaemic heart disease (OR: 2.5, 95% CI:1.4-4.7, p = 0.006) were predictive of stroke. INTERPRETATION: The results of this multi-national study on hospitalized patients with SARS-CoV-2 infection indicated an overall stroke risk of 0.5%(pooled risk: 0.9%). The need for mechanical ventilation and the history of ischaemic heart disease are the independent predictors of stroke among SARS-CoV-2 patients. FUNDING: None.

Regional Association of Disability and SARS-CoV-2 Infection in 369 Counties of the United States.

Background: There have been outbreaks of SARS-CoV-2 in long term care facilities and recent reports of disproportionate death rates among the vulnerable population. The goal of this study was to better understand the impact of SARS-CoV-2 infection on the non-institutionalized disabled population in the United States using data from the most affected states as of April 9th, 2020. Methods: This was an ecological study of county-level factors associated with the infection and mortality rate of SARS-CoV-2 in the non-institutionalized disabled population. We analyzed data from 369 counties from the most affected states (Michigan, New York, New Jersey, Pennsylvania, California, Louisiana, Massachusetts) in the United States using data available by April 9th, 2020. The variables include changes in mobility reported by Google, race/ethnicity, median income, education level, health insurance, and disability information from the United States Census Bureau. Bivariate regression analysis adjusted for state and median income was used to analyze the association between death rate and infection rate. Results: The independent sample t-test of two groups (group 1: Death rate≥ 3.4% [median] and group 2: Death rate < 3.4%) indicates that counties with a higher total population, a lower percentage of Black males and females, higher median income, higher education, and lower percentage of disabled population have a lower rate (< 3.4%) of SARS-CoV-2 related mortality (all p-values<4.3E-02). The results of the bivariate regression when controlled for median income and state show counties with a higher White disabled population (est: 0.19, 95% CI: 0.01-0.37; p-value:3.7E-02), and higher population with independent living difficulty (est: 0.15, 95% CI: -0.01-0.30; p-value: 6.0E-02) have a higher rate of SARS-CoV-2 related mortality. Also, the regression analysis indicates that counties with higher White disabled population (est: -0.22, 95% CI: -0.43-(-0.02); p-value: 3.3E-02), higher population with hearing disability (est: -0.26, 95% CI: -0.42-(-0.11); p-value:1.2E-03), and higher population with disability in the 18-34 years age group (est: -0.25, 95% CI: -0.41-(-0.09); p-value:2.4E-03) show a lower rate of SARS-CoV-2 infection. Conclusion: Our results indicate that while counties with a higher percentage of non-institutionalized disabled population, especially White disabled population, show a lower infection rate, they have a higher rate of SARS-CoV-2 related mortality.

Racial, Economic and Health Inequality and COVID-19 Infection in the United States.

BACKGROUND: There is preliminary evidence of racial and social-economic disparities in the population infected by and dying from COVID-19. The goal of this study is to report the associations of COVID-19 with respect to race, health and economic inequality in the United States. METHODS: We performed a cross-sectional study of the associations between infection and mortality rate of COVID-19 and demographic, socioeconomic and mobility variables from 369 counties (total population: 102,178,117 [median: 73,447, IQR: 30,761-256,098]) from the seven most affected states (Michigan, New York, New Jersey, Pennsylvania, California, Louisiana, Massachusetts). FINDINGS: The risk factors for infection and mortality are different. Our analysis shows that counties with more diverse demographics, higher population, education, income levels, and lower disability rates were at a higher risk of COVID-19 infection. However, counties with higher disability and poverty rates had a higher death rate. African Americans were more vulnerable to COVID-19 than other ethnic groups (1,981 African American infected cases versus 658 Whites per million). Data on mobility changes corroborate the impact of social distancing. INTERPRETATION: The observed inequality might be due to the workforce of essential services, poverty, and access to care. Counties in more urban areas are probably better equipped at providing care. The lower rate of infection, but a higher death rate in counties with higher poverty and disability could be due to lower levels of mobility, but a higher rate of comorbidities and health care access.

Sex disparity in long-term stroke recurrence and mortality in a rural population in the United States.

BACKGROUND: Several studies suggest women may be disproportionately affected by poorer stroke outcomes than men. This study aims to investigate whether women have a higher risk of all-cause mortality and recurrence after an ischemic stroke than men in a rural population in central Pennsylvania, United States. METHODS: We analyzed consecutive ischemic stroke patients captured in the Geisinger NeuroScience Ischemic Stroke research database from 2004 to 2019. Kaplan-Meier (KM) estimator curves stratified by gender and age were used to plot survival probabilities and Cox Proportional Hazards Ratios were used to analyze outcomes of all-cause mortality and the composite outcome of ischemic stroke recurrence or death. Fine-Gray Competing Risk models were used for the outcome of recurrent ischemic stroke, with death as the competing risk. Two models were generated; Model 1 was adjusted by data-driven associated health factors, and Model 2 was adjusted by traditional vascular risk factors. RESULTS: Among 8900 adult ischemic stroke patients [median age of 71.6 (interquartile range: 61.1-81.2) years and 48% women], women had a higher crude all-cause mortality. The KM curves demonstrated a 63.3% survival in women compared with a 65.7% survival in men (p = 0.003) at 5 years; however, the survival difference was not present after controlling for covariates, including age, atrial fibrillation or flutter, myocardial infarction, diabetes mellitus, dyslipidemia, heart failure, chronic lung diseases, rheumatic disease, chronic kidney disease, neoplasm, peripheral vascular disease, past ischemic stroke, past hemorrhagic stroke, and depression. There was no adjusted or unadjusted sex difference in terms of recurrent ischemic stroke or composite outcome. CONCLUSION: Sex was not an independent risk factor for all-cause mortality and ischemic stroke recurrence in the rural population in central Pennsylvania.

Increasing the Density of Laboratory Measures for Machine Learning Applications.

BACKGROUND: The imputation of missingness is a key step in Electronic Health Records (EHR) mining, as it can significantly affect the conclusions derived from the downstream analysis in translational medicine. The missingness of laboratory values in EHR is not at random, yet imputation techniques tend to disregard this key distinction. Consequently, the development of an adaptive imputation strategy designed specifically for EHR is an important step in improving the data imbalance and enhancing the predictive power of modeling tools for healthcare applications. METHOD: We analyzed the laboratory measures derived from Geisinger's EHR on patients in three distinct cohorts-patients tested for Clostridioides difficile (Cdiff) infection, patients with a diagnosis of inflammatory bowel disease (IBD), and patients with a diagnosis of hip or knee osteoarthritis (OA). We extracted Logical Observation Identifiers Names and Codes (LOINC) from which we excluded those with 75% or more missingness. The comorbidities, primary or secondary diagnosis, as well as active problem lists, were also extracted. The adaptive imputation strategy was designed based on a hybrid approach. The comorbidity patterns of patients were transformed into latent patterns and then clustered. Imputation was performed on a cluster of patients for each cohort independently to show the generalizability of the method. The results were compared with imputation applied to the complete dataset without incorporating the information from comorbidity patterns. RESULTS: We analyzed a total of 67,445 patients (11,230 IBD patients, 10,000 OA patients, and 46,215 patients tested for C. difficile infection). We extracted 495 LOINC and 11,230 diagnosis codes for the IBD cohort, 8160 diagnosis codes for the Cdiff cohort, and 2042 diagnosis codes for the OA cohort based on the primary/secondary diagnosis and active problem list in the EHR. Overall, the most improvement from this strategy was observed when the laboratory measures had a higher level of missingness. The best root mean square error (RMSE) difference for each dataset was recorded as -35.5 for the Cdiff, -8.3 for the IBD, and -11.3 for the OA dataset. CONCLUSIONS: An adaptive imputation strategy designed specifically for EHR that uses complementary information from the clinical profile of the patient can be used to improve the imputation of missing laboratory values, especially when laboratory codes with high levels of missingness are included in the analysis.