Search

ObjectiveWe aimed to investigate the impact of social circumstances on cancer therapy selection using natural language processing to derive insights from social worker documentation.Materials and methodsWe developed and employed a Bidirectional Encoder Representations from Transformers (BERT) based approach, using a hierarchical multi-step BERT model (BERT-MS), to predict the prescription of targeted cancer therapy to patients based solely on documentation by clinical social workers. Our corpus included free-text clinical social work notes, combined with medication prescription information, for all patients treated for breast cancer at UCSF between 2012 and 2021. We conducted a feature importance analysis to identify the specific social circumstances that impact cancer therapy regimen.ResultsUsing only social work notes, we consistently predicted the administration of targeted therapies, suggesting systematic differences in treatment selection exist due to non-clinical factors. The findings were confirmed by several language models, with GatorTron achieving the best performance with an area under the receiver operating characteristic curve (AUROC) of 0.721 and a Macro F1 score of 0.616. The UCSF BERT-MS model, capable of leveraging multiple pieces of notes, surpassed the UCSF-BERT model in both AUROC and Macro-F1. Our feature importance analysis identified several clinically intuitive social determinants of health that potentially contribute to disparities in treatment.DiscussionLeveraging social work notes can be instrumental in identifying disparities in clinical decision-making. Hypotheses generated in an automated way could be used to guide patient-specific quality improvement interventions. Further validation with diverse clinical outcomes and prospective studies is essential.ConclusionsOur findings indicate that significant disparities exist among breast cancer patients receiving different types of therapies based on social determinants of health. Social work reports play a crucial role in understanding these disparities in clinical decision-making.

Digital therapeutics (DTx) are a somewhat novel class of US Food and Drug Administration-regulated software that help patients prevent, manage, or treat disease. Here, we use natural language processing to characterise registered DTx clinical trials and provide insights into the clinical development landscape for these novel therapeutics. We identified 449 DTx clinical trials, initiated or expected to be initiated between 2010 and 2030, from ClinicalTrials.gov using 27 search terms, and available data were analysed, including trial durations, locations, MeSH categories, enrolment, and sponsor types. Topic modelling of eligibility criteria, done with BERTopic, showed that DTx trials frequently exclude patients on the basis of age, comorbidities, pregnancy, language barriers, and digital determinants of health, including smartphone or data plan access. Our comprehensive overview of the DTx development landscape highlights challenges in designing inclusive DTx clinical trials and presents opportunities for clinicians and researchers to address these challenges. Finally, we provide an interactive dashboard for readers to conduct their own analyses.

ObjectiveExisting research on social determinants of health (SDoH) predominantly focuses on physician notes and structured data within electronic medical records. This study posits that social work notes are an untapped, potentially rich source for SDoH information. We hypothesize that clinical notes recorded by social workers, whose role is to ameliorate social and economic factors, might provide a complementary information source of data on SDoH compared to physician notes, which primarily concentrate on medical diagnoses and treatments. We aimed to use word frequency analysis and topic modeling to identify prevalent terms and robust topics of discussion within a large cohort of social work notes including both outpatient and in-patient consultations.Materials and methodsWe retrieved a diverse, deidentified corpus of 0.95 million clinical social work notes from 181 644 patients at the University of California, San Francisco. We conducted word frequency analysis related to ICD-10 chapters to identify prevalent terms within the notes. We then applied Latent Dirichlet Allocation (LDA) topic modeling analysis to characterize this corpus and identify potential topics of discussion, which was further stratified by note types and disease groups.ResultsWord frequency analysis primarily identified medical-related terms associated with specific ICD10 chapters, though it also detected some subtle SDoH terms. In contrast, the LDA topic modeling analysis extracted 11 topics explicitly related to social determinants of health risk factors, such as financial status, abuse history, social support, risk of death, and mental health. The topic modeling approach effectively demonstrated variations between different types of social work notes and across patients with different types of diseases or conditions.DiscussionOur findings highlight LDA topic modeling's effectiveness in extracting SDoH-related themes and capturing variations in social work notes, demonstrating its potential for informing targeted interventions for at-risk populations.ConclusionSocial work notes offer a wealth of unique and valuable information on an individual's SDoH. These notes present consistent and meaningful topics of discussion that can be effectively analyzed and utilized to improve patient care and inform targeted interventions for at-risk populations.

Rheumatoid arthritis (RA) management leans toward achieving remission or low disease activity. In this study, we conducted single-cell RNA sequencing (scRNA-Seq) of peripheral blood mononuclear cells (PBMCs) from 36 individuals (18 patients with RA and 18 matched controls, accounting for age, sex, race, and ethnicity), to identify disease-relevant cell subsets and cell type-specific signatures associated with disease activity. Our analysis revealed 18 distinct PBMC subsets, including an IFN-induced transmembrane 3-overexpressing (IFITM3-overexpressing) IFN-activated monocyte subset. We observed an increase in CD4+ T effector memory cells in patients with moderate-high disease activity (DAS28-CRP ≥ 3.2) and a decrease in nonclassical monocytes in patients with low disease activity or remission (DAS28-CRP < 3.2). Pseudobulk analysis by cell type identified 168 differentially expressed genes between RA and matched controls, with a downregulation of proinflammatory genes in the γδ T cell subset, alteration of genes associated with RA predisposition in the IFN-activated subset, and nonclassical monocytes. Additionally, we identified a gene signature associated with moderate-high disease activity, characterized by upregulation of proinflammatory genes such as TNF, JUN, EGR1, IFIT2, MAFB, and G0S2 and downregulation of genes including HLA-DQB1, HLA-DRB5, and TNFSF13B. Notably, cell-cell communication analysis revealed an upregulation of signaling pathways, including VISTA, in both moderate-high and remission-low disease activity contexts. Our findings provide valuable insights into the systemic cellular and molecular mechanisms underlying RA disease activity.

ObjectivesAs the real-world electronic health record (EHR) data continue to grow exponentially, novel methodologies involving artificial intelligence (AI) are becoming increasingly applied to enable efficient data-driven learning and, ultimately, to advance healthcare. Our objective is to provide readers with an understanding of evolving computational methods and help in deciding on methods to pursue.Target audienceThe sheer diversity of existing methods presents a challenge for health scientists who are beginning to apply computational methods to their research. Therefore, this tutorial is aimed at scientists working with EHR data who are early entrants into the field of applying AI methodologies.ScopeThis manuscript describes the diverse and growing AI research approaches in healthcare data science and categorizes them into 2 distinct paradigms, the bottom-up and top-down paradigms to provide health scientists venturing into artificial intelligent research with an understanding of the evolving computational methods and help in deciding on methods to pursue through the lens of real-world healthcare data.

Differential host responses in coronavirus disease 2019 (COVID-19) and multisystem inflammatory syndrome in children (MIS-C) remain poorly characterized. Here, we use next-generation sequencing to longitudinally analyze blood samples from pediatric patients with COVID-19 or MIS-C across three hospitals. Profiling of plasma cell-free nucleic acids uncovers distinct signatures of cell injury and death between COVID-19 and MIS-C, with increased multiorgan involvement in MIS-C encompassing diverse cell types, including endothelial and neuronal cells, and an enrichment of pyroptosis-related genes. Whole-blood RNA profiling reveals upregulation of similar pro-inflammatory pathways in COVID-19 and MIS-C but also MIS-C-specific downregulation of T cell-associated pathways. Profiling of plasma cell-free RNA and whole-blood RNA in paired samples yields different but complementary signatures for each disease state. Our work provides a systems-level view of immune responses and tissue damage in COVID-19 and MIS-C and informs future development of new disease biomarkers.

BackgroundAdverse events of special interest (AESIs) were pre-specified to be monitored for the COVID-19 vaccines. Some AESIs are not only associated with the vaccines, but with COVID-19. Our aim was to characterise the incidence rates of AESIs following SARS-CoV-2 infection in patients and compare these to historical rates in the general population.MethodsA multi-national cohort study with data from primary care, electronic health records, and insurance claims mapped to a common data model. This study's evidence was collected between Jan 1, 2017 and the conclusion of each database (which ranged from Jul 2020 to May 2022). The 16 pre-specified prevalent AESIs were: acute myocardial infarction, anaphylaxis, appendicitis, Bell's palsy, deep vein thrombosis, disseminated intravascular coagulation, encephalomyelitis, Guillain- Barré syndrome, haemorrhagic stroke, non-haemorrhagic stroke, immune thrombocytopenia, myocarditis/pericarditis, narcolepsy, pulmonary embolism, transverse myelitis, and thrombosis with thrombocytopenia. Age-sex standardised incidence rate ratios (SIR) were estimated to compare post-COVID-19 to pre-pandemic rates in each of the databases.FindingsSubstantial heterogeneity by age was seen for AESI rates, with some clearly increasing with age but others following the opposite trend. Similarly, differences were also observed across databases for same health outcome and age-sex strata. All studied AESIs appeared consistently more common in the post-COVID-19 compared to the historical cohorts, with related meta-analytic SIRs ranging from 1.32 (1.05 to 1.66) for narcolepsy to 11.70 (10.10 to 13.70) for pulmonary embolism.InterpretationOur findings suggest all AESIs are more common after COVID-19 than in the general population. Thromboembolic events were particularly common, and over 10-fold more so. More research is needed to contextualise post-COVID-19 complications in the longer term.FundingNone.

There is an increased interest in utilizing real-world data (RWD) for pharmaceutical research and regulatory decision-making. The development and use of pediatric medicines could benefit greatly from real-world data studies given nearly half of drugs prescribed to children are "off-label", meaning there is a lack of pediatric-specific evidence from controlled trials, while there is an abundance of data from routine clinical practice. Currently, the use of real-world data, such as data from electronic health records, is lacking in pediatric research, especially within infectious diseases. Here, we discuss opportunities and challenges for real-world data to generate evidence on the optimal treatment and management of infectious diseases in children.

Chronic low back pain (cLBP) afflicts 8. 2% of adults in the United States, and is the leading global cause of disability. Neuropsychiatric co-morbidities including anxiety, depression, and substance abuse- are common in cLBP patients. In particular, cLBP is a risk factor for opioid addiction, as more than 50% of opioid prescriptions in the United States are for cLBP. Misuse of these prescriptions is a common precursor to addiction. While associations between cLBP and neuropsychiatric disorders are well established, causal relationships for the most part are unknown. Developing effective treatments for cLBP, and associated co-morbidities, requires identifying and understanding causal relationships. Rigorous methods for causal inference, a process for quantifying causal effects from observational data, have been developed over the past 30 years. In this review we first discuss the conceptual model of cLBP that current treatments are based on, and how gaps in causal knowledge contribute to poor clinical outcomes. We then present cLBP as a "Big Data" problem and identify how advanced analytic techniques may close knowledge gaps and improve clinical outcomes. We will focus on causal discovery, which is a data-driven method that uses artificial intelligence (AI) and high dimensional datasets to identify causal structures, discussing both constraint-based (PC and Fast Causal Inference) and score-based (Fast Greedy Equivalent Search) algorithms.

BackgroundRandomized trials are the gold-standard for clinical evidence generation, but they can sometimes be limited by infeasibility and unclear generalizability to real-world practice. External control arm (ECA) studies may help address this evidence gaps by constructing retrospective cohorts that closely emulate prospective ones. Experience in constructing these outside the context of rare diseases or cancer is limited. We piloted an approach for developing an ECA in Crohn's disease using electronic health records (EHR) data.MethodsWe queried EHR databases and manually screened records at the University of California, San Francisco to identify patients meeting the eligibility criteria of TRIDENT, a recently completed interventional trial involving an ustekinumab reference arm. We defined timepoints to balance missing data and bias. We compared imputation models by their impacts on cohort membership and outcomes. We assessed the accuracy of algorithmic data curation against manual review. Lastly, we assessed disease activity following treatment with ustekinumab.ResultsScreening identified 183 patients. 30% of the cohort had missing baseline data. Nonetheless, cohort membership and outcomes were robust to the method of imputation. Algorithms for ascertaining non-symptom-based elements of disease activity using structured data were accurate against manual review. The cohort consisted of 56 patients, exceeding planned enrollment in TRIDENT. 34% of the cohort was in steroid-free remission at week 24.ConclusionWe piloted an approach for creating an ECA in Crohn's disease from EHR data by using a combination of informatics and manual methods. However, our study reveals significant missing data when standard-of-care clinical data are repurposed. More work will be needed to improve the alignment of trial design with typical patterns of clinical practice, and thereby enable a future of more robust ECAs in chronic diseases like Crohn's disease.

BackgroundReal-world performance of COVID-19 diagnostic tests under Emergency Use Authorization (EUA) must be assessed. We describe overall trends in the performance of serology tests in the context of real-world implementation.MethodsSix health systems estimated the odds of seropositivity and positive percent agreement (PPA) of serology test among people with confirmed SARS-CoV-2 infection by molecular test. In each dataset, we present the odds ratio and PPA, overall and by key clinical, demographic, and practice parameters.ResultsA total of 15,615 people were observed to have at least one serology test 14-90 days after a positive molecular test for SARS-CoV-2. We observed higher PPA in Hispanic (PPA range: 79-96%) compared to non-Hispanic (60-89%) patients; in those presenting with at least one COVID-19 related symptom (69-93%) as compared to no such symptoms (63-91%); and in inpatient (70-97%) and emergency department (93-99%) compared to outpatient (63-92%) settings across datasets. PPA was highest in those with diabetes (75-94%) and kidney disease (83-95%); and lowest in those with auto-immune conditions or who are immunocompromised (56-93%). The odds ratios (OR) for seropositivity were higher in Hispanics compared to non-Hispanics (OR range: 2.59-3.86), patients with diabetes (1.49-1.56), and obesity (1.63-2.23); and lower in those with immunocompromised or autoimmune conditions (0.25-0.70), as compared to those without those comorbidities. In a subset of three datasets with robust information on serology test name, seven tests were used, two of which were used in multiple settings and met the EUA requirement of PPA ≥87%. Tests performed similarly across datasets.ConclusionAlthough the EUA requirement was not consistently met, more investigation is needed to understand how serology and molecular tests are used, including indication and protocol fidelity. Improved data interoperability of test and clinical/demographic data are needed to enable rapid assessment of the real-world performance of in vitro diagnostic tests.

BackgroundAs diagnostic tests for COVID-19 were broadly deployed under Emergency Use Authorization, there emerged a need to understand the real-world utilization and performance of serological testing across the United States.MethodsSix health systems contributed electronic health records and/or claims data, jointly developed a master protocol, and used it to execute the analysis in parallel. We used descriptive statistics to examine demographic, clinical, and geographic characteristics of serology testing among patients with RNA positive for SARS-CoV-2.ResultsAcross datasets, we observed 930,669 individuals with positive RNA for SARS-CoV-2. Of these, 35,806 (4%) were serotested within 90 days; 15% of which occurred 30%) in some datasets-limiting our ability to examine differences in serological testing by race. In datasets where race/ethnicity information was available, we observed a greater distribution of White individuals among those serotested; however, the time between RNA and serology tests appeared shorter in Black compared to White individuals. Test manufacturer data was available in half of the datasets contributing to the analysis.ConclusionOur results inform the underlying context of serotesting during the first year of the COVID-19 pandemic and differences observed between claims and EHR data sources-a critical first step to understanding the real-world accuracy of serological tests. Incomplete reporting of race/ethnicity data and a limited ability to link test manufacturer data, lab results, and clinical data challenge the ability to assess the real-world performance of SARS-CoV-2 tests in different contexts and the overall U.S. response to current and future disease pandemics.

Pathologists can label pathologies differently, making it challenging to yield consistent assessments in the absence of one ground truth. To address this problem, we present a deep learning (DL) approach that draws on a cohort of experts, weighs each contribution, and is robust to noisy labels. We collected 100,495 annotations on 20,099 candidate amyloid beta neuropathologies (cerebral amyloid angiopathy (CAA), and cored and diffuse plaques) from three institutions, independently annotated by five experts. DL methods trained on a consensus-of-two strategy yielded 12.6-26% improvements by area under the precision recall curve (AUPRC) when compared to those that learned individualized annotations. This strategy surpassed individual-expert models, even when unfairly assessed on benchmarks favoring them. Moreover, ensembling over individual models was robust to hidden random annotators. In blind prospective tests of 52,555 subsequent expert-annotated images, the models labeled pathologies like their human counterparts (consensus model AUPRC = 0.74 cored; 0.69 CAA). This study demonstrates a means to combine multiple ground truths into a common-ground DL model that yields consistent diagnoses informed by multiple and potentially variable expert opinions.

BackgroundAngiotensin-converting enzyme 2 (ACE2) is the cell-entry receptor for SARS-CoV-2. It plays critical roles in both the transmission and the pathogenesis of COVID-19. Comprehensive profiling of ACE2 expression patterns could reveal risk factors of severe COVID-19 illness. While the expression of ACE2 in healthy human tissues has been well characterized, it is not known which diseases and drugs might be associated with ACE2 expression.ResultsWe develop GENEVA (GENe Expression Variance Analysis), a semi-automated framework for exploring massive amounts of RNA-seq datasets. We apply GENEVA to 286,650 publicly available RNA-seq samples to identify any previously studied experimental conditions that could be directly or indirectly associated with ACE2 expression. We identify multiple drugs, genetic perturbations, and diseases that are associated with the expression of ACE2, including cardiomyopathy, HNF1A overexpression, and drug treatments with RAD140 and itraconazole. Our joint analysis of seven datasets confirms ACE2 upregulation in all cardiomyopathy categories. Using electronic health records data from 3936 COVID-19 patients, we demonstrate that patients with pre-existing cardiomyopathy have an increased mortality risk than age-matched patients with other cardiovascular conditions. GENEVA is applicable to any genes of interest and is freely accessible at http://genevatool.org .ConclusionsThis study identifies multiple diseases and drugs that are associated with the expression of ACE2. The effect of these conditions should be carefully studied in COVID-19 patients. In particular, our analysis identifies cardiomyopathy patients as a high-risk group, with increased ACE2 expression in the heart and increased mortality after SARS-COV-2 infection.

BackgroundAccurate, pragmatic risk stratification for postoperative delirium (POD) is necessary to target preventative resources toward high-risk patients. Machine learning (ML) offers a novel approach to leveraging electronic health record (EHR) data for POD prediction. We sought to develop and internally validate a ML-derived POD risk prediction model using preoperative risk features, and to compare its performance to models developed with traditional logistic regression.MethodsThis was a retrospective analysis of preoperative EHR data from 24,885 adults undergoing a procedure requiring anesthesia care, recovering in the main post-anesthesia care unit, and staying in the hospital at least overnight between December 2016 and December 2019 at either of two hospitals in a tertiary care health system. One hundred fifteen preoperative risk features including demographics, comorbidities, nursing assessments, surgery type, and other preoperative EHR data were used to predict postoperative delirium (POD), defined as any instance of Nursing Delirium Screening Scale ≥2 or positive Confusion Assessment Method for the Intensive Care Unit within the first 7 postoperative days. Two ML models (Neural Network and XGBoost), two traditional logistic regression models ("clinician-guided" and "ML hybrid"), and a previously described delirium risk stratification tool (AWOL-S) were evaluated using the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, positive likelihood ratio, and positive predictive value. Model calibration was assessed with a calibration curve. Patients with no POD assessments charted or at least 20% of input variables missing were excluded.ResultsPOD incidence was 5.3%. The AUC-ROC for Neural Net was 0.841 [95% CI 0. 816-0.863] and for XGBoost was 0.851 [95% CI 0.827-0.874], which was significantly better than the clinician-guided (AUC-ROC 0.763 [0.734-0.793], p

ObjectiveEarly identification of chronic diseases is a pillar of precision medicine as it can lead to improved outcomes, reduction of disease burden, and lower healthcare costs. Predictions of a patient's health trajectory have been improved through the application of machine learning approaches to electronic health records (EHRs). However, these methods have traditionally relied on "black box" algorithms that can process large amounts of data but are unable to incorporate domain knowledge, thus limiting their predictive and explanatory power. Here, we present a method for incorporating domain knowledge into clinical classifications by embedding individual patient data into a biomedical knowledge graph.Materials and methodsA modified version of the Page rank algorithm was implemented to embed millions of deidentified EHRs into a biomedical knowledge graph (SPOKE). This resulted in high-dimensional, knowledge-guided patient health signatures (ie, SPOKEsigs) that were subsequently used as features in a random forest environment to classify patients at risk of developing a chronic disease.ResultsOur model predicted disease status of 5752 subjects 3 years before being diagnosed with multiple sclerosis (MS) (AUC = 0.83). SPOKEsigs outperformed predictions using EHRs alone, and the biological drivers of the classifiers provided insight into the underpinnings of prodromal MS.ConclusionUsing data from EHR as input, SPOKEsigs describe patients at both the clinical and biological levels. We provide a clinical use case for detecting MS up to 5 years prior to their documented diagnosis in the clinic and illustrate the biological features that distinguish the prodromal MS state.

BackgroundThe great majority of severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) infections are mild and uncomplicated, but some individuals with initially mild COVID-19 progressively develop more severe symptoms. Furthermore, there is substantial heterogeneity in SARS-CoV-2-specific memory immune responses following infection. There remains a critical need to identify host immune biomarkers predictive of clinical and immunological outcomes in SARS-CoV-2-infected patients.MethodsLeveraging longitudinal samples and data from a clinical trial (N=108) in SARS-CoV-2-infected outpatients, we used host proteomics and transcriptomics to characterize the trajectory of the immune response in COVID-19 patients. We characterized the association between early immune markers and subsequent disease progression, control of viral shedding, and SARS-CoV-2-specific T cell and antibody responses measured up to 7 months after enrollment. We further compared associations between early immune markers and subsequent T cell and antibody responses following natural infection with those following mRNA vaccination. We developed machine-learning models to predict patient outcomes and validated the predictive model using data from 54 individuals enrolled in an independent clinical trial.ResultsWe identify early immune signatures, including plasma RIG-I levels, early IFN signaling, and related cytokines (CXCL10, MCP1, MCP-2, and MCP-3) associated with subsequent disease progression, control of viral shedding, and the SARS-CoV-2-specific T cell and antibody response measured up to 7 months after enrollment. We found that several biomarkers for immunological outcomes are shared between individuals receiving BNT162b2 (Pfizer-BioNTech) vaccine and COVID-19 patients. Finally, we demonstrate that machine-learning models using 2-7 plasma protein markers measured early within the course of infection are able to accurately predict disease progression, T cell memory, and the antibody response post-infection in a second, independent dataset.ConclusionsEarly immune signatures following infection can accurately predict clinical and immunological outcomes in outpatients with COVID-19 using validated machine-learning models.FundingSupport for the study was provided from National Institute of Health/National Institute of Allergy and Infectious Diseases (NIH/NIAID) (U01 AI150741-01S1 and T32-AI052073), the Stanford's Innovative Medicines Accelerator, National Institutes of Health/National Institute on Drug Abuse (NIH/NIDA) DP1DA046089, and anonymous donors to Stanford University. Peginterferon lambda provided by Eiger BioPharmaceuticals.

Much scientific work over the past few decades has linked health outcomes and disease risk to genomics, to derive a better understanding of disease mechanisms at the genetic and molecular level. However, genomics alone does not quite capture the full picture of one's overall health. Modern computational biomedical research is moving in the direction of including social/environmental factors that ultimately affect quality of life and health outcomes at both the population and individual level. The future of studying disease now lies at the hands of the social determinants of health (SDOH) to answer pressing clinical questions and address healthcare disparities across population groups through its integration into electronic health records (EHRs). In this perspective article, we argue that the SDOH are the future of disease risk and health outcomes studies due to their vast coverage of a patient's overall health. SDOH data availability in EHRs has improved tremendously over the years with EHR toolkits, diagnosis codes, wearable devices, and census tract information to study disease risk. We discuss the availability of SDOH data, challenges in SDOH implementation, its future in real-world evidence studies, and the next steps to report study outcomes in an equitable and actionable way.

Background: The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection causes coronavirus disease-2019 (COVID-19) in some individuals, while the majority remain asymptomatic. Natural killer (NK) cells play an essential role in antiviral defense. NK cell maturation and function are regulated mainly by highly polymorphic killer cell immunoglobulin-like receptors (KIR) and cognate HLA class I ligands. Herein, we tested our hypothesis that the individualized KIR and HLA class I ligand combinations that control NK cell function determine the outcome of SARS-CoV-2 infection. Methods: We characterized KIR and HLA genes in 200 patients hospitalized for COVID-19 and 195 healthy general population controls. Results: The KIR3DL1+HLA-Bw4+ [Odds ratio (OR) = 0.65, p = 0.03] and KIR3DL2+HLA-A3/11+ (OR = 0.6, p = 0.02) combinations were encountered at significantly lower frequency in COVID-19 patients than in the controls. Notably, 40% of the patients lacked both of these KIR+HLA+ combinations compared to 24.6% of the controls (OR = 2.04, p = 0.001). Additionally, activating receptors KIR2DS1+KIR2DS5+ are more frequent in patients with severe COVID-19 than patients with mild disease (OR = 1.8, p = 0.05). Individuals carrying KIR2DS1+KIR2DS5+ genes but missing either KIR3DL1+HLA-Bw4+ combination (OR = 1.73, p = 0.04) or KIR3DL2+HLA-A3/11+ combination (OR = 1.75, p = 0.02) or both KIR3DL1+HLA-Bw4+ and KIR2DL2+HLA-A3/11+ combinations (OR = 1.63, p = 0.03) were more frequent in the COVID-19 cohort compared to controls. Conclusions: The absence of KIR3DL1+HLA-Bw4+ and KIR3DL2+HLA-A3/11+ combinations presumably yields inadequate NK cell maturation and reduces anti-SARS-CoV-2 defense, causing COVID-19. An increased frequency of KIR2DS1+KIR2DS5+ in severe COVID-19 patients suggests vigorous NK cell response triggered via these activating receptors and subsequent production of exuberant inflammatory cytokines responsible for severe COVID-19. Our results demonstrate that specific KIR-HLA combinations that control NK cell maturation and function are underlying immunogenetic variables that determine the dual role of NK cells in mediating beneficial antiviral and detrimental pathologic action. These findings offer a framework for developing potential host genetic biomarkers to distinguish individuals prone to COVID-19.

ObjectivesWearable fitness devices are increasingly being used by the general population, with many new applications being proposed for healthy adults as well as for adults with chronic diseases. Fewer, if any, studies of these devices have been conducted in healthy adolescents and teenagers, especially over a long period of time. The goal of this work was to document the successes and challenges involved in 5 years of a wearable fitness device use in a pediatric case study.Materials and methodsComparison of 5 years of step counts and minutes asleep from a teenaged girl and her father.ResultsAt 60 months, this may be the longest reported pediatric study involving a wearable fitness device, and the first simultaneously involving a parent and a child. We find step counts to be significantly higher for both the adult and teen on school/work days, along with less sleep. The teen walked significantly less towards the end of the 5-year study. Surprisingly, many of the adult's and teen's sleeping and step counts were correlated, possibly due to coordinated behaviors.DiscussionWe end with several recommendations for pediatricians and device manufacturers, including the need for constant adjustments of stride length and calorie counts as teens are growing.ConclusionWith periodic adjustments for growth, this pilot study shows these devices can be used for more accurate and consistent measurements in adolescents and teenagers over longer periods of time, to potentially promote healthy behaviors.

ObjectivesElectronic health records (EHR) are receiving growing attention from regulators, biopharmaceuticals and payors as a potential source of real-world evidence. However, their suitability for the study of diseases with complex activity measures is unclear. We sought to evaluate the use of EHR data for estimating treatment effectiveness in inflammatory bowel disease (IBD), using tofacitinib as a use case.MethodsRecords from the University of California, San Francisco (6/2012 to 4/2019) were queried to identify tofacitinib-treated IBD patients. Disease activity variables at baseline and follow-up were manually abstracted according to a preregistered protocol. The proportion of patients meeting the endpoints of recent randomised trials in ulcerative colitis (UC) and Crohn's disease (CD) was assessed.Results86 patients initiated tofacitinib. Baseline characteristics of the real-world and trial cohorts were similar, except for universal failure of tumour necrosis factor inhibitors in the former. 54% (UC) and 62% (CD) of patients had complete capture of disease activity at baseline (month -6 to 0), while only 32% (UC) and 69% (CD) of patients had complete follow-up data (month 2 to 8). Using data imputation, we estimated the proportion achieving the trial primary endpoints as being similar to the published estimates for both UC (16%, p value=0.5) and CD (38%, p-value=0.8).Discussion/conclusionThis pilot study reproduced trial-based estimates of tofacitinib efficacy despite its use in a different cohort but revealed substantial missingness in routinely collected data. Future work is needed to strengthen EHR data and enable real-world evidence in complex diseases like IBD.

ObjectiveUsing the newly created University of California (UC) Health Data Warehouse, we present the first study to analyze antihyperglycemic treatment utilization across the five large UC academic health systems (Davis, Irvine, Los Angeles, San Diego, and San Francisco).Research design and methodsThis retrospective analysis used deidentified electronic health records (EHRs; 2014-2019) including 97,231 patients with type 2 diabetes from 1,003 UC-affiliated clinical settings. Significant differences between health systems and individual providers were identified using binomial probabilities with cohort matching.ResultsOur analysis reveals statistically different treatment utilization patterns not only between health systems but also among individual providers within health systems. We identified 21 differences among health systems and 29 differences among individual providers within these health systems, with respect to treatment intensifications within existing guidelines on top of either metformin monotherapy or dual therapy with metformin and a sulfonylurea. Next, we identified variation for medications within the same class (e.g., glipizide vs. glyburide among sulfonylureas), with 33 differences among health systems and 86 among individual providers. Finally, we identified 2 health systems and 55 individual providers who more frequently used medications with known cardioprotective benefits for patients with high cardiovascular disease risk, but also 1 health system and 8 providers who prescribed such medications less frequently for these patients.ConclusionsOur study used cohort-matching techniques to highlight real-world variation in care between health systems and individual providers. This demonstrates the power of EHRs to quantify differences in treatment utilization, a necessary step toward standardizing precision care for large populations.

BackgroundMachine learning (ML) has captured the attention of many clinicians who may not have formal training in this area but are otherwise increasingly exposed to ML literature that may be relevant to their clinical specialties. ML papers that follow an outcomes-based research format can be assessed using clinical research appraisal frameworks such as PICO (Population, Intervention, Comparison, Outcome). However, the PICO frameworks strain when applied to ML papers that create new ML models, which are akin to diagnostic tests. There is a need for a new framework to help assess such papers.ObjectiveWe propose a new framework to help clinicians systematically read and evaluate medical ML papers whose aim is to create a new ML model: ML-PICO (Machine Learning, Population, Identification, Crosscheck, Outcomes). We describe how the ML-PICO framework can be applied toward appraising literature describing ML models for health care.ConclusionThe relevance of ML to practitioners of clinical medicine is steadily increasing with a growing body of literature. Therefore, it is increasingly important for clinicians to be familiar with how to assess and best utilize these tools. In this paper we have described a practical framework on how to read ML papers that create a new ML model (or diagnostic test): ML-PICO. We hope that this can be used by clinicians to better evaluate the quality and utility of ML papers.

Our goal is to summarize the collective experience of 15 organizations in dealing with uncoordinated efforts that result in unnecessary delays in understanding, predicting, preparing for, containing, and mitigating the COVID-19 pandemic in the US. Response efforts involve the collection and analysis of data corresponding to healthcare organizations, public health departments, socioeconomic indicators, as well as additional signals collected directly from individuals and communities. We focused on electronic health record (EHR) data, since EHRs can be leveraged and scaled to improve clinical care, research, and to inform public health decision-making. We outline the current challenges in the data ecosystem and the technology infrastructure that are relevant to COVID-19, as witnessed in our 15 institutions. The infrastructure includes registries and clinical data networks to support population-level analyses. We propose a specific set of strategic next steps to increase interoperability, overall organization, and efficiencies.

There is an urgent need to identify therapies that prevent SARS-CoV-2 infection and improve the outcome of COVID-19 patients. Although repurposed drugs with favorable safety profiles could have significant benefit, widely available prevention or treatment options for COVID-19 have yet to be identified. Efforts to identify approved drugs with in vitro activity against SARS-CoV-2 resulted in identification of antiviral sigma-1 receptor ligands, including antihistamines in the histamine-1 receptor binding class. We identified antihistamine candidates for repurposing by mining electronic health records of usage in population of more than 219,000 subjects tested for SARS-CoV-2. Usage of diphenhydramine, hydroxyzine and azelastine was associated with reduced incidence of SARS-CoV-2 positivity in subjects greater than age 61. We found diphenhydramine, hydroxyzine and azelastine to exhibit direct antiviral activity against SARS-CoV-2 in vitro. Although mechanisms by which specific antihistamines exert antiviral effects is not clear, hydroxyzine, and possibly azelastine, bind Angiotensin Converting Enzyme-2 (ACE2) and the sigma-1 receptor as off-targets. Clinical studies are needed to measure the effectiveness of diphenhydramine, hydroxyzine and azelastine for disease prevention, for early intervention, or as adjuvant therapy for severe COVID-19.

The field of immunology is rapidly progressing toward a systems-level understanding of immunity to tackle complex infectious diseases, autoimmune conditions, cancer, and beyond. In the last couple of decades, advancements in data acquisition techniques have presented opportunities to explore untapped areas of immunological research. Broad initiatives are launched to disseminate the datasets siloed in the global, federated, or private repositories, facilitating interoperability across various research domains. Concurrently, the application of computational methods, such as network analysis, meta-analysis, and machine learning have propelled the field forward by providing insight into salient features that influence the immunological response, which was otherwise left unexplored. Here, we review the opportunities and challenges in democratizing datasets, repositories, and community-wide knowledge sharing tools. We present use cases for repurposing open-access immunology datasets with advanced machine learning applications and more.

To evaluate whether different approaches in note text preparation (known as preprocessing) can impact machine learning model performance in the case of mortality prediction ICU.DesignClinical note text was used to build machine learning models for adults admitted to the ICU. Preprocessing strategies studied were none (raw text), cleaning text, stemming, term frequency-inverse document frequency vectorization, and creation of n-grams. Model performance was assessed by the area under the receiver operating characteristic curve. Models were trained and internally validated on University of California San Francisco data using 10-fold cross validation. These models were then externally validated on Beth Israel Deaconess Medical Center data.SettingICUs at University of California San Francisco and Beth Israel Deaconess Medical Center.SubjectsTen thousand patients in the University of California San Francisco training and internal testing dataset and 27,058 patients in the external validation dataset, Beth Israel Deaconess Medical Center.InterventionsNone.Measurements and main resultsMortality rate at Beth Israel Deaconess Medical Center and University of California San Francisco was 10.9% and 7.4%, respectively. Data are presented as area under the receiver operating characteristic curve (95% CI) for models validated at University of California San Francisco and area under the receiver operating characteristic curve for models validated at Beth Israel Deaconess Medical Center. Models built and trained on University of California San Francisco data for the prediction of inhospital mortality improved from the raw note text model (AUROC, 0.84; CI, 0.80-0.89) to the term frequency-inverse document frequency model (AUROC, 0.89; CI, 0.85-0.94). When applying the models developed at University of California San Francisco to Beth Israel Deaconess Medical Center data, there was a similar increase in model performance from raw note text (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.72) to the term frequency-inverse document frequency model (area under the receiver operating characteristic curve at Beth Israel Deaconess Medical Center: 0.83).ConclusionsDifferences in preprocessing strategies for note text impacted model discrimination. Completing a preprocessing pathway including cleaning, stemming, and term frequency-inverse document frequency vectorization resulted in the preprocessing strategy with the greatest improvement in model performance. Further study is needed, with particular emphasis on how to manage author implicit bias present in note text, before natural language processing algorithms are implemented in the clinical setting.

There has long been an interest in understanding how the hazards from spaceflight may trigger or exacerbate human diseases. With the goal of advancing our knowledge on physiological changes during space travel, NASA GeneLab provides an open-source repository of multi-omics data from real and simulated spaceflight studies. Alone, this data enables identification of biological changes during spaceflight, but cannot infer how that may impact an astronaut at the phenotypic level. To bridge this gap, Scalable Precision Medicine Oriented Knowledge Engine (SPOKE), a heterogeneous knowledge graph connecting biological and clinical data from over 30 databases, was used in combination with GeneLab transcriptomic data from six studies. This integration identified critical symptoms and physiological changes incurred during spaceflight.

Acid suppressants are widely-used classes of medications linked to increased risks of aerodigestive infections. Prior studies of these medications as potentially reversible risk factors for COVID-19 have been conflicting. We aimed to determine the impact of chronic acid suppression use on COVID-19 infection risk while simultaneously evaluating the influence of social determinants of health to validate known and discover novel risk factors. We assessed the association of chronic acid suppression with incident COVID-19 in a 1:1 case-control study of 900 patients tested across three academic medical centers in California, USA. Medical comorbidities and history of chronic acid suppression use were manually extracted from health records by physicians following a pre-specified protocol. Socio-behavioral factors by geomapping publicly-available data to patient zip codes were incorporated. We identified no evidence to support an association between chronic acid suppression and COVID-19 (adjusted odds ratio 1.04, 95% CI 0.92-1.17, P = 0.515). However, several medical and social features were positive (Latinx ethnicity, BMI ≥ 30, dementia, public transportation use, month of the pandemic) and negative (female sex, concurrent solid tumor, alcohol use disorder) predictors of new infection. These findings demonstrate the value of integrating publicly-available databases with medical data to identify critical features of communicable diseases.

Cytometry technologies are essential tools for immunology research, providing high-throughput measurements of the immune cells at the single-cell level. Existing approaches in interpreting and using cytometry measurements include manual or automated gating to identify cell subsets from the cytometry data, providing highly intuitive results but may lead to significant information loss, in that additional details in measured or correlated cell signals might be missed. In this study, we propose and test a deep convolutional neural network for analyzing cytometry data in an end-to-end fashion, allowing a direct association between raw cytometry data and the clinical outcome of interest. Using nine large cytometry by time-of-flight mass spectrometry or mass cytometry (CyTOF) studies from the open-access ImmPort database, we demonstrated that the deep convolutional neural network model can accurately diagnose the latent cytomegalovirus (CMV) in healthy individuals, even when using highly heterogeneous data from different studies. In addition, we developed a permutation-based method for interpreting the deep convolutional neural network model. We were able to identify a CD27- CD94+ CD8+ T cell population significantly associated with latent CMV infection, confirming the findings in previous studies. Finally, we provide a tutorial for creating, training, and interpreting the tailored deep learning model for cytometry data using Keras and TensorFlow (https://github.com/hzc363/DeepLearningCyTOF).

In a general inpatient population, we predicted patient-specific medication orders based on structured information in the electronic health record (EHR). Data on over three million medication orders from an academic medical center were used to train two machine-learning models: A deep learning sequence model and a logistic regression model. Both were compared with a baseline that ranked the most frequently ordered medications based on a patient's discharge hospital service and amount of time since admission. Models were trained to predict from 990 possible medications at the time of order entry. Fifty-five percent of medications ordered by physicians were ranked in the sequence model's top-10 predictions (logistic model: 49%) and 75% ranked in the top-25 (logistic model: 69%). Ninety-three percent of the sequence model's top-10 prediction sets contained at least one medication that physicians ordered within the next day. These findings demonstrate that medication orders can be predicted from information present in the EHR.

There is a great and growing need to ascertain what exactly is the state of a patient, in terms of disease progression, actual care practices, pathology, adverse events, and much more, beyond the paucity of data available in structured medical record data. Ascertaining these harder-to-reach data elements is now critical for the accurate phenotyping of complex traits, detection of adverse outcomes, efficacy of off-label drug use, and longitudinal patient surveillance. Clinical notes often contain the most detailed and relevant digital information about individual patients, the nuances of their diseases, the treatment strategies selected by physicians, and the resulting outcomes. However, notes remain largely unused for research because they contain Protected Health Information (PHI), which is synonymous with individually identifying data. Previous clinical note de-identification approaches have been rigid and still too inaccurate to see any substantial real-world use, primarily because they have been trained with too small medical text corpora. To build a new de-identification tool, we created the largest manually annotated clinical note corpus for PHI and develop a customizable open-source de-identification software called Philter ("Protected Health Information filter"). Here we describe the design and evaluation of Philter, and show how it offers substantial real-world improvements over prior methods.

Real-world data (RWD) continue to emerge as a new source of clinical evidence. Although the best-known use case of RWD has been in drug regulation, RWD are being generated and used by many other parties, including biopharmaceutical companies, payors, clinical researchers, providers, and patients. In this Review, we describe 21 potential uses for RWD across the spectrum of health care. We also discuss important challenges and limitations relevant to the translation of these data into evidence.

The First International Conference in Systems and Network Medicine gathered together 200 global thought leaders, scientists, clinicians, academicians, industry and government experts, medical and graduate students, postdoctoral scholars and policymakers. Held at Georgetown University Conference Center in Washington D.C. on September 11-13, 2019, the event featured a day of pre-conference lectures and hands-on bioinformatic computational workshops followed by two days of deep and diverse scientific talks, panel discussions with eminent thought leaders, and scientific poster presentations. Topics ranged from: Systems and Network Medicine in Clinical Practice; the role of -omics technologies in Health Care; the role of Education and Ethics in Clinical Practice, Systems Thinking, and Rare Diseases; and the role of Artificial Intelligence in Medicine. The conference served as a unique nexus for interdisciplinary discovery and dialogue and fostered formation of new insights and possibilities for health care systems advances.

Preterm birth (PTB) is defined as the birth of an infant before 37 weeks of gestational age. It is the leading cause of perinatal morbidity and mortality worldwide. In this study, we present a comprehensive meta-analysis of vaginal microbiome in PTB. We integrated raw longitudinal 16S rRNA vaginal microbiome data from five independent studies across 3,201 samples and were able to gain new insights into the vaginal microbiome state in women who deliver preterm in comparison to those who deliver at term. We found that women who deliver prematurely show higher within-sample variance in vaginal microbiome abundance, with the most significant difference observed during the first trimester. Modeling the data longitudinally revealed a number of microbial genera as associated with PTB, including several that were previously known and two newly identified by this meta-analysis: Olsenella and Clostridium sensu stricto. New hypotheses emerging from this integrative analysis can lead to novel diagnostics to identify women who are at higher risk for PTB and potentially inform new therapeutic interventions.

In this Comment, we characterize the current pipeline of digital therapeutics and offer a clinical perspective into the advantages, challenges, and barriers to implementation of this treatment modality for patient care, which we hope will inform future regulatory policy, prescribing decisions, and scope of real-world evidence collection.

MotivationElectronic health records (EHRs) are quickly becoming omnipresent in healthcare, but interoperability issues and technical demands limit their use for biomedical and clinical research. Interactive and flexible software that interfaces directly with EHR data structured around a common data model (CDM) could accelerate more EHR-based research by making the data more accessible to researchers who lack computational expertise and/or domain knowledge.ResultsWe present PatientExploreR, an extensible application built on the R/Shiny framework that interfaces with a relational database of EHR data in the Observational Medical Outcomes Partnership CDM format. PatientExploreR produces patient-level interactive and dynamic reports and facilitates visualization of clinical data without any programming required. It allows researchers to easily construct and export patient cohorts from the EHR for analysis with other software. This application could enable easier exploration of patient-level data for physicians and researchers. PatientExploreR can incorporate EHR data from any institution that employs the CDM for users with approved access. The software code is free and open source under the MIT license, enabling institutions to install and users to expand and modify the application for their own purposes.Availability and implementationPatientExploreR can be freely obtained from GitHub: https://github.com/BenGlicksberg/PatientExploreR. We provide instructions for how researchers with approved access to their institutional EHR can use this package. We also release an open sandbox server of synthesized patient data for users without EHR access to explore: http://patientexplorer.ucsf.edu.Supplementary informationSupplementary data are available at Bioinformatics online.

Precision health relies on the ability to assess disease risk at an individual level, detect early preclinical conditions and initiate preventive strategies. Recent technological advances in omics and wearable monitoring enable deep molecular and physiological profiling and may provide important tools for precision health. We explored the ability of deep longitudinal profiling to make health-related discoveries, identify clinically relevant molecular pathways and affect behavior in a prospective longitudinal cohort (n = 109) enriched for risk of type 2 diabetes mellitus. The cohort underwent integrative personalized omics profiling from samples collected quarterly for up to 8 years (median, 2.8 years) using clinical measures and emerging technologies including genome, immunome, transcriptome, proteome, metabolome, microbiome and wearable monitoring. We discovered more than 67 clinically actionable health discoveries and identified multiple molecular pathways associated with metabolic, cardiovascular and oncologic pathophysiology. We developed prediction models for insulin resistance by using omics measurements, illustrating their potential to replace burdensome tests. Finally, study participation led the majority of participants to implement diet and exercise changes. Altogether, we conclude that deep longitudinal profiling can lead to actionable health discoveries and provide relevant information for precision health.

ProblemPreterm birth (PTB), or the delivery of an infant prior to 37 weeks of gestation, is a major health concern. Although a variety of social, environmental, and maternal factors have been implicated in PTB, causes of preterm labor have remained largely unknown. There is evidence of effectiveness and safety of influenza vaccination during pregnancy, however fewer studies have looked at vaccination response as an indicator of an innate host response that may be associated with adverse pregnancy outcomes. We carried out a pilot study to analyze the flu vaccine response during pregnancy of women who later deliver preterm or term.Method of studyWe performed a secondary analysis of the individual-level data from an influenza vaccination response study (openly available from ImmPort) measured by hemagglutination inhibition assay of 91 pregnant women with term deliveries and 11 women who went on to deliver preterm. Flu vaccination responses for H1N1 and H3N2 influenza strains were compared between term and preterm deliveries.ResultsWomen who went on to deliver preterm showed a significantly (P

Tissue fibrosis is a major cause of mortality that results from the deposition of matrix proteins by an activated mesenchyme. Macrophages accumulate in fibrosis, but the role of specific subgroups in supporting fibrogenesis has not been investigated in vivo. Here, we used single-cell RNA sequencing (scRNA-seq) to characterize the heterogeneity of macrophages in bleomycin-induced lung fibrosis in mice. A novel computational framework for the annotation of scRNA-seq by reference to bulk transcriptomes (SingleR) enabled the subclustering of macrophages and revealed a disease-associated subgroup with a transitional gene expression profile intermediate between monocyte-derived and alveolar macrophages. These CX3CR1+SiglecF+ transitional macrophages localized to the fibrotic niche and had a profibrotic effect in vivo. Human orthologs of genes expressed by the transitional macrophages were upregulated in samples from patients with idiopathic pulmonary fibrosis. Thus, we have identified a pathological subgroup of transitional macrophages that are required for the fibrotic response to injury.

Recent remarkable advances in genome sequencing have enabled detailed maps of identified and interpreted genomic variation, dubbed "mutanomes." The availability of thousands of exome/genome sequencing data has prompted the emergence of new challenges in the identification of novel druggable targets and therapeutic strategies. Typically, mutanomes are viewed as one- or two-dimensional. The three-dimensional protein structural view of personal mutanomes sheds light on the functional consequences of clinically actionable mutations revealed in tumor diagnosis and followed up in personalized treatments, in a mutanome-informed manner. In this review, we describe the protein structural landscape of personal mutanomes and provide expert opinions on rational strategies for more streamlined oncological drug discovery and molecularly targeted therapies for each individual and each tumor. We provide the structural mechanism of orthosteric versus allosteric drugs at the atom-level via targeting specific somatic alterations for combating drug resistance and the "undruggable" challenges in solid and hematologic neoplasias. We discuss computational biophysics strategies for innovative mutanome-informed cancer immunotherapies and combination immunotherapies. Finally, we highlight a personal mutanome infrastructure for the emerging development of personalized cancer medicine using a breast cancer case study.

The identification of novel disease associations using big-data for patient care has had limited success. In this study, we created a longitudinal disease network of traced readmissions (disease trajectories), merging data from over 10.4 million inpatients through the Healthcare Cost and Utilization Project, which allowed the representation of disease progression mapping over 300 diseases. From these disease trajectories, we discovered an interesting association between schizophrenia and rhabdomyolysis, a rare muscle disease (incidence

In order to advance precision medicine, detailed clinical features ought to be described in a way that leverages current knowledge. Although data collected from biomedical research is expanding at an almost exponential rate, our ability to transform that information into patient care has not kept at pace. A major barrier preventing this transformation is that multi-dimensional data collection and analysis is usually carried out without much understanding of the underlying knowledge structure. Here, in an effort to bridge this gap, Electronic Health Records (EHRs) of individual patients are connected to a heterogeneous knowledge network called Scalable Precision Medicine Oriented Knowledge Engine (SPOKE). Then an unsupervised machine-learning algorithm creates Propagated SPOKE Entry Vectors (PSEVs) that encode the importance of each SPOKE node for any code in the EHRs. We argue that these results, alongside the natural integration of PSEVs into any EHR machine-learning platform, provide a key step toward precision medicine.

Biomedical ‘big data’ has opened opportunities for data repurposing to reveal new insights into complex diseases. Public data on IBD have been repurposed for novel diagnostics and therapeutics, and these datasets continue to grow. Here, we discuss the practicalities and implications of open data informatics for IBD.

Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient's record. We propose a representation of patients' entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93-0.94), 30-day unplanned readmission (AUROC 0.75-0.76), prolonged length of stay (AUROC 0.85-0.86), and all of a patient's final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient's chart.

Machine Learning Special Issue Guest Editors Suchi Saria, Atul Butte, and Aziz Sheikh cut through the hyperbole with an accessible and accurate portrayal of the forefront of machine learning in clinical translation.

Preterm birth, or the delivery of an infant prior to 37 weeks of gestation, is a significant cause of infant morbidity and mortality. In the last decade, the advent and continued development of molecular profiling technologies has enabled researchers to generate vast amount of 'omics' data, which together with integrative computational approaches, can help refine the current knowledge about disease mechanisms, diagnostics, and therapeutics. Here we describe the March of Dimes' Database for Preterm Birth Research (http://www.immport.org/resources/mod), a unique resource that contains a variety of 'omics' datasets related to preterm birth. The database is open publicly, and as of January 2018, links 13 molecular studies with data across tens of thousands of patients from 6 measurement modalities. The data in the repository are highly diverse and include genomic, transcriptomic, immunological, and microbiome data. Relevant datasets are augmented with additional molecular characterizations of almost 25,000 biological samples from public databases. We believe our data-sharing efforts will lead to enhanced research collaborations and coordination accelerating the overall pace of discovery in preterm birth research.

Preterm birth (PTB), or the delivery prior to 37 weeks of gestation, is a significant cause of infant morbidity and mortality. Although twin studies estimate that maternal genetic contributions account for approximately 30% of the incidence of PTB, and other studies reported fetal gene polymorphism association, to date no consistent associations have been identified. In this study, we performed the largest reported genome-wide association study analysis on 1,349 cases of PTB and 12,595 ancestry-matched controls from the focusing on genomic fetal signals. We tested over 2 million single nucleotide polymorphisms (SNPs) for associations with PTB across five subpopulations: African (AFR), the Americas (AMR), European, South Asian, and East Asian. We identified only two intergenic loci associated with PTB at a genome-wide level of significance: rs17591250 (P = 4.55E-09) on chromosome 1 in the AFR population and rs1979081 (P = 3.72E-08) on chromosome 8 in the AMR group. We have queried several existing replication cohorts and found no support of these associations. We conclude that the fetal genetic contribution to PTB is unlikely due to single common genetic variant, but could be explained by interactions of multiple common variants, or of rare variants affected by environmental influences, all not detectable using a GWAS alone.

The original version of the Data Descriptor contained errors in the author list and affiliations. Rita Leite's first name was misspelled as "Rite" and affiliations 4 and 5 were incorrectly swapped. In addition, members of the March of Dimes Prematurity Research Center consortium were not listed in the agreed positions within the author list. These errors have now been corrected in the HTML and PDF versions.

BackgroundDespite substantial research and development investment in Alzheimer's disease (AD), effective therapeutics remain elusive. Significant emerging evidence has linked cholesterol, β-amyloid and AD, and several studies have shown a reduced risk for AD and dementia in populations treated with statins. However, while some clinical trials evaluating statins in general AD populations have been conducted, these resulted in no significant therapeutic benefit. By focusing on subgroups of the AD population, it may be possible to detect endotypes responsive to statin therapy.MethodsHere we investigate the possible protective and therapeutic effect of statins in AD through the analysis of datasets of integrated clinical trials, and prospective observational studies.ResultsRe-analysis of AD patient-level data from failed clinical trials suggested by trend that use of simvastatin may slow the progression of cognitive decline, and to a greater extent in ApoE4 homozygotes. Evaluation of continual long-term use of various statins, in participants from multiple studies at baseline, revealed better cognitive performance in statin users. These findings were supported in an additional, observational cohort where the incidence of AD was significantly lower in statin users, and ApoE4/ApoE4-genotyped AD patients treated with statins showed better cognitive function over the course of 10-year follow-up.ConclusionsThese results indicate that the use of statins may benefit all AD patients with potentially greater therapeutic efficacy in those homozygous for ApoE4.