ABSTRACT Background Ticagrelor and prasugrel are recommended P2Y₁₂ inhibitors for patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI), yet uncertainty persists regarding their direct comparative evidence and guideline recommendations differ. Methods We conducted a multinational retrospective new-user cohort study across 7 claims and electronic health record databases. Adults with ACS undergoing first PCI who initiated ticagrelor or prasugrel were included; patients with prior major ischemic or hemorrhagic events or oral anticoagulant use were excluded. The primary outcome was 1-year major adverse cardiovascular events (MACE: all-cause mortality, acute myocardial infarction, or stroke). Secondary outcomes included net adverse clinical events (NACE) and individual components. Propensity scores were estimated using large-scale L1-regularized logistic regression and applied through stratification. Prespecified diagnostics (covariate balance, empirical equipoise, and systematic error) determined eligibility of each database for inclusion in meta-analysis. Database-specific hazard ratios (HRs) were combined using Bayesian random-effects meta-analysis. Results Among 7 participating databases, 3 met prespecified diagnostic criteria and were included in the primary meta-analysis, comprising 133,718 patients from one nationwide Korean claims database and two U.S. commercial claims databases (ticagrelor, 109,639; prasugrel, 24,079). For 1-year MACE, the pooled HR for ticagrelor versus prasugrel was 1.28 (95% credible interval [CrI], 0.89–1.88), with substantial between-database heterogeneity. Sensitivity analyses across alternative time-at-risk definitions and propensity score matching were consistent. No statistically credible differences were observed for NACE (HR 1.23, CrI 0.88–1.75), all-cause mortality (HR 1.17, CrI 0.78–1.77), cardiovascular mortality (HR 1.23, CrI 0.81–1.87), ischemic events (HR 1.28, CrI 0.88–1.90), hemorrhagic events (HR 1.01, CrI 0.72–1.39), acute myocardial infarction (HR 1.30, CrI 0.88–1.94), stroke (HR 1.09, CrI 0.73– 1.58), or gastrointestinal bleeding (HR 1.04, CrI 0.77–1.41). In a post hoc meta-analysis restricted to the two U.S. databases, the pooled HR for 1-year MACE was 1.49 (95% CrI 1.05–2.10). Conclusions In this pre-specified multinational observational study, no statistically credible difference in 1-year MACE was observed between ticagrelor and prasugrel in patients with ACS undergoing PCI. However, substantial cross-database heterogeneity warrants further investigation into context-specific comparative effectiveness and safety. CLINICAL PERSPECTIVE What is new? In a multinational new-user active comparator study across one nationwide Korean claims database and two U.S. commercial claims databases, ticagrelor and prasugrel did not show a statistically credible overall difference in 1-year MACE after PCI for ACS. Despite a prespecified protocol, standardized execution, and empirical calibration, substantial cross-database heterogeneity was observed, with neutral estimates in Korea and prasugrel-favoring ischemic signals in U.S. databases. What are the clinical implications? These findings do not support superiority of ticagrelor or prasugrel for 1-year MACE across all health-care settings. Substantial cross-database heterogeneity argues against assuming that the comparative effectiveness and safety of ticagrelor and prasugrel are uniform across populations and care settings. Selection between ticagrelor and prasugrel should remain individualized, integrating individual patient risk profiles and local care considerations rather than presume a universally preferred agent.

De-identification of clinical text remains essential for secondary use of electronic health records (EHRs), yet public benchmarks such as i2b2 2006/2014 are over a decade old and lack the semantic and demographic diversity of modern narratives. While Large Language Models (LLMs) achieve state-of-the-art zero-shot extraction, enterprise deployment is hindered by compute costs and governance restricting Protected Health Information (PHI) from cloud APIs. We introduce SHIELD (Synthetic Human-annotated Identifier-replaced Entries for Learning and De-identification), a diverse dataset of 1,394 notes with 10,505 gold-standard PHI spans across 9 categories, built via set-cover diversity sampling with human-in-the-loop adjudication. We evaluate four LLMs (two proprietary, two open-weight) to establish a performance ceiling, then distill these capabilities into locally deployable Small Language Models (SLMs). Distributional analysis using Frechet Text Distance and Jensen-Shannon Divergence confirms SHIELD occupies a distinct region of biomedical embedding and vocabulary space versus legacy benchmarks. Our best distilled model matches its teacher on structured PHI categories (DATE, DOCTOR, ID, PATIENT, PHONE) and achieves micro-averaged span-level precision of 0.88 and recall of 0.86 on standard workstation hardware. Cross-dataset evaluation shows diversity-trained models generalize well on universal structured PHI, while institution-specific entities remain hard to transfer, suggesting optimal deployment combines broad-coverage models with specialized models for high-volume notes. We publicly release the SHIELD dataset and the distilled DeBERTa v3 model.

Foundation models hold promise for transforming AI in healthcare by providing modular components that are easily adaptable to downstream healthcare tasks, making AI development more scalable and cost-effective. Structured EHR foundation models, trained on coded medical records from millions of patients, demonstrated benefits including increased performance with fewer training labels, and improved robustness to distribution shifts. However, questions remain on the feasibility of sharing these models across different hospitals and their performance for local task adaptation. This multi-center study examined the adaptability of a recently released structured EHR foundation model ($FM_{SM}$), trained on longitudinal medical record data from 2.57M Stanford Medicine patients. Experiments were conducted using EHR data at The Hospital for Sick Children and MIMIC-IV. We assessed both adaptability via continued pretraining on local data, and task adaptability compared to baselines of training models from scratch at each site, including a local foundation model. We evaluated the performance of these models on 8 clinical prediction tasks. In both datasets, adapting the off-the-shelf $FM_{SM}$ matched the performance of GBM models locally trained on all data while providing a 13% improvement in settings with few task-specific training labels. With continued pretraining on local data, label efficiency substantially improved, such that $FM_{SM}$ required fewer than 1% of training examples to match the fully trained GBM's performance. Continued pretraining was also 60 to 90% more sample-efficient than training local foundation models from scratch. Our findings show that adapting shared EHR foundation models across hospitals provides improved prediction performance at less cost, underscoring the utility of base foundation models as modular components to streamline the development of healthcare AI.

Objective To measure pediatrician adherence to evidence-based guidelines in the treatment of young children with attention-deficit/hyperactivity disorder (ADHD) in a diverse healthcare system using natural language processing (NLP) techniques.Materials and Methods We extracted structured and free-text data from electronic health records (EHRs) of all office visits (2015-2019) of children aged 4-6 years in a community-based primary healthcare network in California, who had >= 1 visits with an ICD-10 diagnosis of ADHD. Two pediatricians annotated clinical notes of the first ADHD visit for 423 patients. Inter-annotator agreement (IAA) was assessed for the recommendation for the first-line behavioral treatment (F-measure = 0.89). Four pre-trained language models, including BioClinical Bidirectional Encoder Representations from Transformers (BioClinicalBERT), were used to identify behavioral treatment recommendations using a 70/30 train/test split. For temporal validation, we deployed BioClinicalBERT on 1,020 unannotated notes from other ADHD visits and well-care visits; all positively classified notes (n = 53) and 5% of negatively classified notes (n = 50) were manually reviewed.Results Of 423 patients, 313 (74%) were male; 298 (70%) were privately insured; 138 (33%) were White; 61 (14%) were Hispanic. The BioClinicalBERT model trained on the first ADHD visits achieved F1 = 0.76, precision = 0.81, recall = 0.72, and AUC = 0.81 [0.72-0.89]. Temporal validation achieved F1 = 0.77, precision = 0.68, and recall = 0.88. Fairness analysis revealed low model performance in publicly insured patients (F1 = 0.53).Conclusion Deploying pre-trained language models on a variable set of clinical notes accurately captured pediatrician adherence to guidelines in the treatment of children with ADHD. Validating this approach in other patient populations is needed to achieve equitable measurement of quality of care at scale and improve clinical care for mental health conditions.

Abstract Background Diagnostic codes are commonly used as inputs for clinical prediction models, to create labels for prediction tasks, and to identify cohorts for multicenter network studies. However, the coverage rates of diagnostic codes and their variability across institutions are underexplored. The primary objective was to describe lab- and diagnosis-based labels for 7 selected outcomes at three institutions. Secondary objectives were to describe agreement, sensitivity, and specificity of diagnosis-based labels against lab-based labels. Methods This study included three cohorts: SickKids from The Hospital for Sick Children, and StanfordPeds and StanfordAdults from Stanford Medicine. We included seven clinical outcomes with lab-based definitions: acute kidney injury, hyperkalemia, hypoglycemia, hyponatremia, anemia, neutropenia and thrombocytopenia. For each outcome, we created four lab-based labels (abnormal, mild, moderate and severe) based on test result and one diagnosis-based label. Proportion of admissions with a positive label were presented for each outcome stratified by cohort. Using lab-based labels as the gold standard, agreement using Cohen’s Kappa, sensitivity and specificity were calculated for each lab-based severity level. Results The number of admissions included were: SickKids (n = 59,298), StanfordPeds (n = 24,639) and StanfordAdults (n = 159,985). The proportion of admissions with a positive diagnosis-based label was significantly higher for StanfordPeds compared to SickKids across all outcomes, with odds ratio (99.9% confidence interval) for abnormal diagnosis-based label ranging from 2.2 (1.7–2.7) for neutropenia to 18.4 (10.1–33.4) for hyperkalemia. Lab-based labels were more similar by institution. When using lab-based labels as the gold standard, Cohen’s Kappa and sensitivity were lower at SickKids for all severity levels compared to StanfordPeds. Conclusions Across multiple outcomes, diagnosis codes were consistently different between the two pediatric institutions. This difference was not explained by differences in test results. These results may have implications for machine learning model development and deployment.

Objective To develop and validate a novel application of natural language processing (NLP) techniques to measure pediatrician adherence to evidence-based guidelines in the treatment of young children with attention-deficit/hyperactivity disorder (ADHD). Materials and Methods We extracted structured and free text data from electronic health records of all office visits (2015-2019) of children aged 4-6 years seen in a community-based primary healthcare network in California, who had ≥1 visits with an ICD-10 diagnosis of ADHD. Two pediatricians manually annotated clinical notes of the first ADHD visit for 423 patients. Inter-annotator agreement was assessed for recommendation for first-line behavioral treatment; Disagreements were reconciled. The BioClinical Bidirectional Encoder Representations from Transformers (BioClinical-BERT) was used to identify mentions of behavioral treatment recommendations using a 70/30 train/test split. Following an error analysis and threshold selection, we completed external (temporal) validation by deploying the model on 1,020 unannotated notes representing other ADHD visits and well-care visits; all positively classified notes and 5% of negatively classified notes were annotated. Results Of 423 included patients, 313 (74%) were male, 268 (63%) were privately insured; 138 (33%) were white; 61 (14 %) were Hispanic. The BERT model of first ADHD visits achieved F1=0.78, precision=0.84, and recall=0.72. Following threshold selection, temporal validation on notes from other visits achieved F1=0.80, recall=0.92 and precision=0.7. Conclusion Deploying a machine learning algorithm on a large and variable set of clinical notes accurately captured pediatrician adherence to guidelines in treatment of children with ADHD. This approach can be used to measure quality-of-care at scale and improve clinical care for various chronic medical conditions. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement National Institute of Mental Health of the National Institutes of Health under Award Number K23MH128455 (Dr. Bannett). ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: This study was approved by the Stanford University School of Medicine institutional review board. I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes All data produced in the present study are available upon reasonable request to the authors. The code used for this study is available via Github at .

ImportanceRanitidine, the most widely used histamine-2 receptor antagonist (H2RA), was withdrawn because of N-nitrosodimethylamine impurity in 2020. Given the worldwide exposure to this drug, the potential risk of cancer development associated with the intake of known carcinogens is an important epidemiological concern.ObjectiveTo examine the comparative risk of cancer associated with the use of ranitidine vs other H2RAs.Design, Setting, and ParticipantsThis new-user active comparator international network cohort study was conducted using 3 health claims and 9 electronic health record databases from the US, the United Kingdom, Germany, Spain, France, South Korea, and Taiwan. Large-scale propensity score (PS) matching was used to minimize confounding of the observed covariates with negative control outcomes. Empirical calibration was performed to account for unobserved confounding. All databases were mapped to a common data model. Database-specific estimates were combined using random-effects meta-analysis. Participants included individuals aged at least 20 years with no history of cancer who used H2RAs for more than 30 days from January 1986 to December 2020, with a 1-year washout period. Data were analyzed from April to September 2021.ExposureThe main exposure was use of ranitidine vs other H2RAs (famotidine, lafutidine, nizatidine, and roxatidine).Main Outcomes and MeasuresThe primary outcome was incidence of any cancer, except nonmelanoma skin cancer. Secondary outcomes included all cancer except thyroid cancer, 16 cancer subtypes, and all-cause mortality.ResultsAmong 1 183 999 individuals in 11 databases, 909 168 individuals (mean age, 56.1 years; 507 316 [55.8%] women) were identified as new users of ranitidine, and 274 831 individuals (mean age, 58.0 years; 145 935 [53.1%] women) were identified as new users of other H2RAs. Crude incidence rates of cancer were 14.30 events per 1000 person-years (PYs) in ranitidine users and 15.03 events per 1000 PYs among other H2RA users. After PS matching, cancer risk was similar in ranitidine compared with other H2RA users (incidence, 15.92 events per 1000 PYs vs 15.65 events per 1000 PYs; calibrated meta-analytic hazard ratio, 1.04; 95% CI, 0.97-1.12). No significant associations were found between ranitidine use and any secondary outcomes after calibration.Conclusions and RelevanceIn this cohort study, ranitidine use was not associated with an increased risk of cancer compared with the use of other H2RAs. Further research is needed on the long-term association of ranitidine with cancer development.

Abstract Objective To describe the infrastructure, tools, and services developed at Stanford Medicine to maintain its data science ecosystem and research patient data repository for clinical and translational research. Materials and Methods The data science ecosystem, dubbed the Stanford Data Science Resources (SDSR), includes infrastructure and tools to create, search, retrieve, and analyze patient data, as well as services for data deidentification, linkage, and processing to extract high-value information from healthcare IT systems. Data are made available via self-service and concierge access, on HIPAA compliant secure computing infrastructure supported by in-depth user training. Results The Stanford Medicine Research Data Repository (STARR) functions as the SDSR data integration point, and includes electronic medical records, clinical images, text, bedside monitoring data and HL7 messages. SDSR tools include tools for electronic phenotyping, cohort building, and a search engine for patient timelines. The SDSR supports patient data collection, reproducible research, and teaching using healthcare data, and facilitates industry collaborations and large-scale observational studies. Discussion Research patient data repositories and their underlying data science infrastructure are essential to realizing a learning health system and advancing the mission of academic medical centers. Challenges to maintaining the SDSR include ensuring sufficient financial support while providing researchers and clinicians with maximal access to data and digital infrastructure, balancing tool development with user training, and supporting the diverse needs of users. Conclusion Our experience maintaining the SDSR offers a case study for academic medical centers developing data science and research informatics infrastructure.

Background:Adverse 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. Methods:A 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. Findings:Substantial 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. Interpretation:Our 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. Funding:None.

Temporal distribution shift negatively impacts the performance of clinical prediction models over time. Pretraining foundation models using self-supervised learning on electronic health records (EHR) may be effective in acquiring informative global patterns that can improve the robustness of task-specific models. The objective was to evaluate the utility of EHR foundation models in improving the in-distribution (ID) and out-of-distribution (OOD) performance of clinical prediction models. Transformer- and gated recurrent unit-based foundation models were pretrained on EHR of up to 1.8 M patients (382 M coded events) collected within pre-determined year groups (e.g., 2009–2012) and were subsequently used to construct patient representations for patients admitted to inpatient units. These representations were used to train logistic regression models to predict hospital mortality, long length of stay, 30-day readmission, and ICU admission. We compared our EHR foundation models with baseline logistic regression models learned on count-based representations (count-LR) in ID and OOD year groups. Performance was measured using area-under-the-receiver-operating-characteristic curve (AUROC), area-under-the-precision-recall curve, and absolute calibration error. Both transformer and recurrent-based foundation models generally showed better ID and OOD discrimination relative to count-LR and often exhibited less decay in tasks where there is observable degradation of discrimination performance (average AUROC decay of 3% for transformer-based foundation model vs. 7% for count-LR after 5–9 years). In addition, the performance and robustness of transformer-based foundation models continued to improve as pretraining set size increased. These results suggest that pretraining EHR foundation models at scale is a useful approach for developing clinical prediction models that perform well in the presence of temporal distribution shift.

Objective:To assess the uptake of second line antihyperglycaemic drugs among patients with type 2 diabetes mellitus who are receiving metformin. Design:Federated pharmacoepidemiological evaluation in LEGEND-T2DM. Setting:10 US and seven non-US electronic health record and administrative claims databases in the Observational Health Data Sciences and Informatics network in eight countries from 2011 to the end of 2021. Participants:4.8 million patients (≥18 years) across US and non-US based databases with type 2 diabetes mellitus who had received metformin monotherapy and had initiated second line treatments. Exposure:The exposure used to evaluate each database was calendar year trends, with the years in the study that were specific to each cohort. Main outcomes measures:The outcome was the incidence of second line antihyperglycaemic drug use (ie, glucagon-like peptide-1 receptor agonists, sodium-glucose cotransporter-2 inhibitors, dipeptidyl peptidase-4 inhibitors, and sulfonylureas) among individuals who were already receiving treatment with metformin. The relative drug class level uptake across cardiovascular risk groups was also evaluated. Results:4.6 million patients were identified in US databases, 61 382 from Spain, 32 442 from Germany, 25 173 from the UK, 13 270 from France, 5580 from Scotland, 4614 from Hong Kong, and 2322 from Australia. During 2011-21, the combined proportional initiation of the cardioprotective antihyperglycaemic drugs (glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors) increased across all data sources, with the combined initiation of these drugs as second line drugs in 2021 ranging from 35.2% to 68.2% in the US databases, 15.4% in France, 34.7% in Spain, 50.1% in Germany, and 54.8% in Scotland. From 2016 to 2021, in some US and non-US databases, uptake of glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors increased more significantly among populations with no cardiovascular disease compared with patients with established cardiovascular disease. No data source provided evidence of a greater increase in the uptake of these two drug classes in populations with cardiovascular disease compared with no cardiovascular disease. Conclusions:Despite the increase in overall uptake of cardioprotective antihyperglycaemic drugs as second line treatments for type 2 diabetes mellitus, their uptake was lower in patients with cardiovascular disease than in people with no cardiovascular disease over the past decade. A strategy is needed to ensure that medication use is concordant with guideline recommendations to improve outcomes of patients with type 2 diabetes mellitus.

Background Temporal dataset shift can cause degradation in model performance as discrepancies between training and deployment data grow over time. The primary objective was to determine whether parsimonious models produced by specific feature selection methods are more robust to temporal dataset shift as measured by out-of-distribution (OOD) performance, while maintaining in-distribution (ID) performance.Methods Our dataset consisted of intensive care unit patients from MIMIC-IV categorized by year groups (2008-2010, 2011-2013, 2014-2016, and 2017-2019). We trained baseline models using L2-regularized logistic regression on 2008-2010 to predict in-hospital mortality, long length of stay (LOS), sepsis, and invasive ventilation in all year groups. We evaluated three feature selection methods: L1-regularized logistic regression (L1), Remove and Retrain (ROAR), and causal feature selection. We assessed whether a feature selection method could maintain ID performance (2008-2010) and improve OOD performance (2017-2019). We also assessed whether parsimonious models retrained on OOD data performed as well as oracle models trained on all features in the OOD year group.Results The baseline model showed significantly worse OOD performance with the long LOS and sepsis tasks when compared with the ID performance. L1 and ROAR retained 3.7 to 12.6% of all features, whereas causal feature selection generally retained fewer features. Models produced by L1 and ROAR exhibited similar ID and OOD performance as the baseline models. The retraining of these models on 2017-2019 data using features selected from training on 2008-2010 data generally reached parity with oracle models trained directly on 2017-2019 data using all available features. Causal feature selection led to heterogeneous results with the superset maintaining ID performance while improving OOD calibration only on the long LOS task.Conclusions While model retraining can mitigate the impact of temporal dataset shift on parsimonious models produced by L1 and ROAR, new methods are required to proactively improve temporal robustness.

Background: Characterization studies of COVID-19 patients with chronic obstructive pulmonary disease (COPD) are limited in size and scope. The aim of the study is to provide a large-scale characterization of COVID-19 patients with COPD. Methods: We included thirteen databases contributing data from January-June 2020 from North America (US), Europe and Asia. We defined two cohorts of patients with COVID-19 namely a ‘diagnosed’ and ‘hospitalized’ cohort. We followed patients from COVID-19 index date to 30 days or death. We performed descriptive analysis and reported the frequency of characteristics and outcomes among COPD patients with COVID-19. Results: The study included 934,778 patients in the diagnosed COVID-19 cohort and 177,201 in the hospitalized COVID-19 cohort. Observed COPD prevalence in the diagnosed cohort ranged from 3.8% (95%CI 3.5-4.1%) in French data to 22.7% (95%CI 22.4-23.0) in US data, and from 1.9% (95%CI 1.6-2.2) in South Korean to 44.0% (95%CI 43.1-45.0) in US data, in the hospitalized cohorts. COPD patients in the hospitalized cohort had greater comorbidity than those in the diagnosed cohort, including hypertension, heart disease, diabetes and obesity. Mortality was higher in COPD patients in the hospitalized cohort and ranged from 7.6% (95%CI 6.9-8.4) to 32.2% (95%CI 28.0-36.7) across databases. ARDS, acute renal failure, cardiac arrhythmia and sepsis were the most common outcomes among hospitalized COPD patients. Conclusion: COPD patients with COVID-19 have high levels of COVID-19-associated comorbidities and poor COVID-19 outcomes. Further research is required to identify patients with COPD at high risk of worse outcomes.

Patients diagnosed with systemic lupus erythematosus (SLE) suffer from a decreased quality of life, an increased risk of medical complications, and an increased risk of death. In particular, approximately 50% of SLE patients progress to develop lupus nephritis, which oftentimes leads to life-threatening end stage renal disease (ESRD) and requires dialysis or kidney transplant1. The challenge is that lupus nephritis is diagnosed via a kidney biopsy, which is typically performed only after noticeable decreased kidney function, leaving little room for proactive or preventative measures. The ability to predict which patients are most likely to develop lupus nephritis has the potential to shift lupus nephritis disease management from reactive to proactive. We present a clinically useful prediction model to predict which patients with newly diagnosed SLE will go on to develop lupus nephritis in the next five years.

OBJECTIVEDevelopment of electronic health records (EHR)-based machine learning models for pediatric inpatients is challenged by limited training data. Self-supervised learning using adult data may be a promising approach to creating robust pediatric prediction models. The primary objective was to determine whether a self-supervised model trained in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients, for pediatric inpatient clinical prediction tasks.MATERIALS AND METHODSThis retrospective cohort study used EHR data and included patients with at least one admission to an inpatient unit. One admission per patient was randomly selected. Adult inpatients were 18 years or older while pediatric inpatients were more than 28 days and less than 18 years. Admissions were temporally split into training (January 1, 2008 to December 31, 2019), validation (January 1, 2020 to December 31, 2020), and test (January 1, 2021 to August 1, 2022) sets. Primary comparison was a self-supervised model trained in adult inpatients versus count-based logistic regression models trained in pediatric inpatients. Primary outcome was mean area-under-the-receiver-operating-characteristic-curve (AUROC) for 11 distinct clinical outcomes. Models were evaluated in pediatric inpatients.RESULTSWhen evaluated in pediatric inpatients, mean AUROC of self-supervised model trained in adult inpatients (0.902) was noninferior to count-based logistic regression models trained in pediatric inpatients (0.868) (mean difference = 0.034, 95% CI=0.014-0.057; P < .001 for noninferiority and P = .006 for superiority).CONCLUSIONSSelf-supervised learning in adult inpatients was noninferior to logistic regression models trained in pediatric inpatients. This finding suggests transferability of self-supervised models trained in adult patients to pediatric patients, without requiring costly model retraining.

BACKGROUND:One key aspect of a learning health system (LHS) is utilizing data generated during care delivery to inform clinical care. However, institutional guidelines that utilize observational data are rare and require months to create, making current processes impractical for more urgent scenarios such as those posed by the COVID-19 pandemic. There exists a need to rapidly analyze institutional data to drive guideline creation where evidence from randomized control trials are unavailable. OBJECTIVES:This article provides a background on the current state of observational data generation in institutional guideline creation and details our institution's experience in creating a novel workflow to (1) demonstrate the value of such a workflow, (2) demonstrate a real-world example, and (3) discuss difficulties encountered and future directions. METHODS:Utilizing a multidisciplinary team of database specialists, clinicians, and informaticists, we created a workflow for identifying and translating a clinical need into a queryable format in our clinical data warehouse, creating data summaries and feeding this information back into clinical guideline creation. RESULTS:Clinical questions posed by the hospital medicine division were answered in a rapid time frame and informed creation of institutional guidelines for the care of patients with COVID-19. The cost of setting up a workflow, answering the questions, and producing data summaries required around 300 hours of effort and $300,000 USD. CONCLUSION:A key component of an LHS is the ability to learn from data generated during care delivery. There are rare examples in the literature and we demonstrate one such example along with proposed thoughts of ideal multidisciplinary team formation and deployment.

Our website uses cookies to enhance your experience. By continuing to use our site, or clicking "Continue," you are agreeing to our Cookie Policy | Continue

Background: Routinely collected real world data (RWD) have great utility in aiding the novel coronavirus disease (COVID-19) pandemic response [1,2]. Here we present the international Observational Health Data Sciences and Informatics (OHDSI) [3] Characterizing Health Associated Risks, and Your Baseline Disease In SARS-COV-2 (CHARYBDIS) framework for standardisation and analysis of COVID-19 RWD. Methods: We conducted a descriptive cohort study using a federated network of data partners in the United States, Europe (the Netherlands, Spain, the UK, Germany, France and Italy) and Asia (South Korea and China). The study protocol and analytical package were released on 11 th June 2020 and are iteratively updated via GitHub [4]. Findings: We identified three non-mutually exclusive cohorts of 4,537,153 individuals with a clinical COVID-19 diagnosis or positive test, 886,193 hospitalized with COVID-19 , and 113,627 hospitalized with COVID-19 requiring intensive services . All comorbidities, symptoms, medications, and outcomes are described by cohort in aggregate counts, and are available in an interactive website: https://data.ohdsi.org/Covid19CharacterizationCharybdis/. Interpretation: CHARYBDIS findings provide benchmarks that contribute to our understanding of COVID-19 progression, management and evolution over time. This can enable timely assessment of real-world outcomes of preventative and therapeutic options as they are introduced in clinical practice.

Training and evaluating language models increasingly requires the construction of meta-datasets --diverse collections of curated data with clear provenance. Natural language prompting has recently lead to improved zero-shot generalization by transforming existing, supervised datasets into a diversity of novel pretraining tasks, highlighting the benefits of meta-dataset curation. While successful in general-domain text, translating these data-centric approaches to biomedical language modeling remains challenging, as labeled biomedical datasets are significantly underrepresented in popular data hubs. To address this challenge, we introduce BigBIO a community library of 126+ biomedical NLP datasets, currently covering 12 task categories and 10+ languages. BigBIO facilitates reproducible meta-dataset curation via programmatic access to datasets and their metadata, and is compatible with current platforms for prompt engineering and end-to-end few/zero shot language model evaluation. We discuss our process for task schema harmonization, data auditing, contribution guidelines, and outline two illustrative use cases: zero-shot evaluation of biomedical prompts and large-scale, multi-task learning. BigBIO is an ongoing community effort and is available at https://github.com/bigscience-workshop/biomedical

Background: Given the costs of machine learning implementation, a systematic approach to prioritizing which models to implement into clinical practice may be valuable.Objective: The primary objective was to determine the health care attributes respondents at 2 pediatric institutions rate as important when prioritizing machine learning model implementation. The secondary objective was to describe their perspectives on implementation using a qualitative approach.Methods: In this mixed methods study, we distributed a survey to health system leaders, physicians, and data scientists at 2 pediatric institutions. We asked respondents to rank the following 5 attributes in terms of implementation usefulness: the clinical problem was common, the clinical problem caused substantial morbidity and mortality, risk stratification led to different actions that could reasonably improve patient outcomes, reducing physician workload, and saving money. Important attributes were those ranked as first or second most important. Individual qualitative interviews were conducted with a subsample of respondents.Results: Among 613 eligible respondents, 275 (44.9%) responded. Qualitative interviews were conducted with 17 respondents. The most common important attributes were risk stratification leading to different actions (205/275, 74.5%) and clinical problem causing substantial morbidity or mortality (177/275, 64.4%). The attributes considered least important were reducing physician workload and saving money. Qualitative interviews consistently prioritized implementations that improved patient outcomes.Conclusions: Respondents prioritized machine learning model implementation where risk stratification would lead to different actions and clinical problems that caused substantial morbidity and mortality. Implementations that improved patient outcomes were prioritized. These results can help provide a framework for machine learning model implementation.