An Empirical Evaluation of Prompting Strategies for Large Language Models in Zero-Shot Clinical Natural Language Processing: Algorithm Development and Validation Study

BACKGROUND

Large language models (LLMs) have shown remarkable capabilities in natural language processing (NLP), especially in domains where labeled data are scarce or expensive, such as the clinical domain. However, to unlock the clinical knowledge hidden in these LLMs, we need to design effective prompts that can guide them to perform specific clinical NLP tasks without any task-specific training data. This is known as in-context learning, which is an art and science that requires understanding the strengths and weaknesses of different LLMs and prompt engineering approaches.

OBJECTIVE

The objective of this study is to assess the effectiveness of various prompt engineering techniques, including 2 newly introduced types-heuristic and ensemble prompts, for zero-shot and few-shot clinical information extraction using pretrained language models.

METHODS

This comprehensive experimental study evaluated different prompt types (simple prefix, simple cloze, chain of thought, anticipatory, heuristic, and ensemble) across 5 clinical NLP tasks: clinical sense disambiguation, biomedical evidence extraction, coreference resolution, medication status extraction, and medication attribute extraction. The performance of these prompts was assessed using 3 state-of-the-art language models: GPT-3.5 (OpenAI), Gemini (Google), and LLaMA-2 (Meta). The study contrasted zero-shot with few-shot prompting and explored the effectiveness of ensemble approaches.

RESULTS

The study revealed that task-specific prompt tailoring is vital for the high performance of LLMs for zero-shot clinical NLP. In clinical sense disambiguation, GPT-3.5 achieved an accuracy of 0.96 with heuristic prompts and 0.94 in biomedical evidence extraction. Heuristic prompts, alongside chain of thought prompts, were highly effective across tasks. Few-shot prompting improved performance in complex scenarios, and ensemble approaches capitalized on multiple prompt strengths. GPT-3.5 consistently outperformed Gemini and LLaMA-2 across tasks and prompt types.

CONCLUSIONS

This study provides a rigorous evaluation of prompt engineering methodologies and introduces innovative techniques for clinical information extraction, demonstrating the potential of in-context learning in the clinical domain. These findings offer clear guidelines for future prompt-based clinical NLP research, facilitating engagement by non-NLP experts in clinical NLP advancements. To the best of our knowledge, this is one of the first works on the empirical evaluation of different prompt engineering approaches for clinical NLP in this era of generative artificial intelligence, and we hope that it will inspire and inform future research in this area.

Mining Clinical Notes for Physical Rehabilitation Exercise Information: Natural Language Processing Algorithm Development and Validation Study

BACKGROUND

The rehabilitation of a patient who had a stroke requires precise, personalized treatment plans. Natural language processing (NLP) offers the potential to extract valuable exercise information from clinical notes, aiding in the development of more effective rehabilitation strategies.

OBJECTIVE

This study aims to develop and evaluate a variety of NLP algorithms to extract and categorize physical rehabilitation exercise information from the clinical notes of patients who had a stroke treated at the University of Pittsburgh Medical Center.

METHODS

A cohort of 13,605 patients diagnosed with stroke was identified, and their clinical notes containing rehabilitation therapy notes were retrieved. A comprehensive clinical ontology was created to represent various aspects of physical rehabilitation exercises. State-of-the-art NLP algorithms were then developed and compared, including rule-based, machine learning-based algorithms (support vector machine, logistic regression, gradient boosting, and AdaBoost) and large language model (LLM)-based algorithms (ChatGPT [OpenAI]). The study focused on key performance metrics, particularly F1-scores, to evaluate algorithm effectiveness.

RESULTS

The analysis was conducted on a data set comprising 23,724 notes with detailed demographic and clinical characteristics. The rule-based NLP algorithm demonstrated superior performance in most areas, particularly in detecting the "Right Side" location with an F1-score of 0.975, outperforming gradient boosting by 0.063. Gradient boosting excelled in "Lower Extremity" location detection (F1-score: 0.978), surpassing rule-based NLP by 0.023. It also showed notable performance in the "Passive Range of Motion" detection with an F1-score of 0.970, a 0.032 improvement over rule-based NLP. The rule-based algorithm efficiently handled "Duration," "Sets," and "Reps" with F1-scores up to 0.65. LLM-based NLP, particularly ChatGPT with few-shot prompts, achieved high recall but generally lower precision and F1-scores. However, it notably excelled in "Backward Plane" motion detection, achieving an F1-score of 0.846, surpassing the rule-based algorithm's 0.720.

CONCLUSIONS

The study successfully developed and evaluated multiple NLP algorithms, revealing the strengths and weaknesses of each in extracting physical rehabilitation exercise information from clinical notes. The detailed ontology and the robust performance of the rule-based and gradient boosting algorithms demonstrate significant potential for enhancing precision rehabilitation. These findings contribute to the ongoing efforts to integrate advanced NLP techniques into health care, moving toward predictive models that can recommend personalized rehabilitation treatments for optimal patient outcomes.

Neurological diagnoses in hospitalized COVID-19 patients associated with adverse outcomes: A multinational cohort study

Few studies examining the patient outcomes of concurrent neurological manifestations during acute COVID-19 leveraged multinational cohorts of adults and children or distinguished between central and peripheral nervous system (CNS vs. PNS) involvement. Using a federated multinational network in which local clinicians and informatics experts curated the electronic health records data, we evaluated the risk of prolonged hospitalization and mortality in hospitalized COVID-19 patients from 21 healthcare systems across 7 countries. For adults, we used a federated learning approach whereby we ran Cox proportional hazard models locally at each healthcare system and performed a meta-analysis on the aggregated results to estimate the overall risk of adverse outcomes across our geographically diverse populations. For children, we reported descriptive statistics separately due to their low frequency of neurological involvement and poor outcomes. Among the 106,229 hospitalized COVID-19 patients (104,031 patients ≥18 years; 2,198 patients <18 years, January 2020-October 2021), 15,101 (14%) had at least one CNS diagnosis, while 2,788 (3%) had at least one PNS diagnosis. After controlling for demographics and pre-existing conditions, adults with CNS involvement had longer hospital stay (11 versus 6 days) and greater risk of (Hazard Ratio = 1.78) and faster time to death (12 versus 24 days) than patients with no neurological condition (NNC) during acute COVID-19 hospitalization. Adults with PNS involvement also had longer hospital stay but lower risk of mortality than the NNC group. Although children had a low frequency of neurological involvement during COVID-19 hospitalization, a substantially higher proportion of children with CNS involvement died compared to those with NNC (6% vs 1%). Overall, patients with concurrent CNS manifestation during acute COVID-19 hospitalization faced greater risks for adverse clinical outcomes than patients without any neurological diagnosis. Our global informatics framework using a federated approach (versus a centralized data collection approach) has utility for clinical discovery beyond COVID-19.

Algorithmic Individual Fairness and Healthcare: A Scoping Review

Objective

Statistical and artificial intelligence algorithms are increasingly being developed for use in healthcare. These algorithms may reflect biases that magnify disparities in clinical care, and there is a growing need for understanding how algorithmic biases can be mitigated in pursuit of algorithmic fairness. Individual fairness in algorithms constrains algorithms to the notion that "similar individuals should be treated similarly." We conducted a scoping review on algorithmic individual fairness to understand the current state of research in the metrics and methods developed to achieve individual fairness and its applications in healthcare.

Methods

We searched three databases, PubMed, ACM Digital Library, and IEEE Xplore, for algorithmic individual fairness metrics, algorithmic bias mitigation, and healthcare applications. Our search was restricted to articles published between January 2013 and September 2023. We identified 1,886 articles through database searches and manually identified one article from which we included 30 articles in the review. Data from the selected articles were extracted, and the findings were synthesized.

Results

Based on the 30 articles in the review, we identified several themes, including philosophical underpinnings of fairness, individual fairness metrics, mitigation methods for achieving individual fairness, implications of achieving individual fairness on group fairness and vice versa, fairness metrics that combined individual fairness and group fairness, software for measuring and optimizing individual fairness, and applications of individual fairness in healthcare.

Conclusion

While there has been significant work on algorithmic individual fairness in recent years, the definition, use, and study of individual fairness remain in their infancy, especially in healthcare. Future research is needed to apply and evaluate individual fairness in healthcare comprehensively.

Meta-Analysis of the Performance of AI-Driven ECG Interpretation in the Diagnosis of Valvular Heart Diseases

Valvular heart diseases (VHDs) significantly impact morbidity and mortality rates worldwide. Early diagnosis improves patient outcomes. Artificial intelligence (AI) applied to electrocardiogram (ECG) interpretation presents a promising approach for early VHD detection. We conducted a meta-analysis on the efficacy of AI models in this context. We reviewed databases including PubMed, MEDLINE, Embase, Scopus, and Cochrane until August 20, 2023, focusing on AI for ECG-based VHD detection. The outcomes included pooled accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value. The pooled proportions were derived using a random-effects model with 95% confidence intervals (CIs). Study heterogeneity was evaluated with the I-squared statistic. Our analysis included 10 studies, involving ECG data from 713,537 patients. The AI algorithms mainly screened for aortic stenosis (n = 6), mitral regurgitation (n = 4), aortic regurgitation (n = 3), mitral stenosis (n = 1), mitral valve prolapse (n = 2), and tricuspid regurgitation (n = 1). A total of 9 studies used convolution neural network models, whereas 1 study combined the strengths of support vector machine logistic regression and multilayer perceptron for ECG interpretation. The collective AI models demonstrated a pooled accuracy of 81% (95% CI 73 to 89, I² = 92%), sensitivity was 83% (95% CI 77 to 88, I² = 86%), specificity was 72% (95% CI 68 to 75, I² = 52%), PPV was 13% (95% CI 7 to 19, I² = 90%), and negative predictive value was 99% (95% CI 97 to 99, I² = 50%). The subgroup analyses for aortic stenosis and mitral regurgitation detection yielded analogous outcomes. In conclusion, AI-driven ECG offers high accuracy in VHD screening. However, its low PPV indicates the need for a combined approach with clinical judgment, especially in primary care settings.

Role of Intraoperative Neuromonitoring to Predict Postoperative Delirium in Cardiovascular Surgery

OBJECTIVE

Postoperative delirium (POD) can occur in up to 50% of older patients undergoing cardiovascular surgery, resulting in hospitalization and significant morbidity and mortality. This study aimed to determine whether intraoperative neurophysiologic monitoring (IONM) modalities can be used to predict delirium in patients undergoing cardiovascular surgery.

DESIGN

Adult patients undergoing cardiovascular surgery with IONM between 2019 and 2021 were reviewed retrospectively. Delirium was assessed multiple times using the Intensive Care Delirium Screening Checklist (ICDSC). Patients with an ICDSC score ≥4 were considered to have POD. Significant IONM changes were evaluated based on a visual review of electroencephalography (EEG) and somatosensory evoked potentials data and documentation of significant changes during surgery.

SETTING

University of Pittsburgh Medical Center hospitals.

PARTICIPANTS

Patients 18 years old and older undergoing cardiovascular surgery with IONM monitoring.

MEASUREMENTS AND MAIN RESULTS

Of the 578 patients undergoing cardiovascular surgery with IONM, 126 had POD (21.8%). Significant IONM changes were noted in 134 patients, of whom 49 patients had delirium (36.6%). In contrast, 444 patients had no IONM changes during surgery, of whom 77 (17.3%) patients had POD. Upon multivariate analysis, IONM changes were associated with POD (odds ratio 2.12; 95% CI 1.31-3.44; p < 0.001). Additionally, baseline EEG abnormalities were associated with POD (p = 0.002).

CONCLUSION

Significant IONM changes are associated with an increased risk of POD in patients undergoing cardiovascular surgery. These findings offer a basis for future research and analysis of EEG and somatosensory evoked potential monitoring to predict, detect, and prevent POD.

Time-Series Aware Metrics for the Evaluation of Intraoperative Electroencephalography-Based Ischemia Detection

Continuous intraoperative monitoring with electroencephalo2 graphy (EEG) is commonly used to detect cerebral ischemia in high-risk surgical procedures such as carotid endarterectomy. Machine learning (ML) models that detect ischemia in real time can form the basis of automated intraoperative EEG monitoring. In this study, we describe and compare two time-series aware precision and recall metrics to the classical precision and recall metrics for evaluating the performance of ML models that detect ischemia. We trained six ML models to detect ischemia in intraoperative EEG and evaluated them with the area under the precision-recall curve (AUPRC) using time-series aware and classical approaches to compute precision and recall. The Support Vector Classification (SVC) model performed the best on the time-series aware metrics, while the Light Gradient Boosting Machine (LGBM) model performed the best on the classical metrics. Visual inspection of the probability outputs of the models alongside the actual ischemic periods revealed that the time-series aware AUPRC selected a model more likely to predict ischemia onset in a timely fashion than the model selected by classical AUPRC.

A broadly applicable approach to enrich electronic-health-record cohorts by identifying patients with complete data: a multisite evaluation

OBJECTIVE

Patients who receive most care within a single healthcare system (colloquially called a "loyalty cohort" since they typically return to the same providers) have mostly complete data within that organization's electronic health record (EHR). Loyalty cohorts have low data missingness, which can unintentionally bias research results. Using proxies of routine care and healthcare utilization metrics, we compute a per-patient score that identifies a loyalty cohort.

MATERIALS AND METHODS

We implemented a computable program for the widely adopted i2b2 platform that identifies loyalty cohorts in EHRs based on a machine-learning model, which was previously validated using linked claims data. We developed a novel validation approach, which tests, using only EHR data, whether patients returned to the same healthcare system after the training period. We evaluated these tools at 3 institutions using data from 2017 to 2019.

RESULTS

Loyalty cohort calculations to identify patients who returned during a 1-year follow-up yielded a mean area under the receiver operating characteristic curve of 0.77 using the original model and 0.80 after calibrating the model at individual sites. Factors such as multiple medications or visits contributed significantly at all sites. Screening tests' contributions (eg, colonoscopy) varied across sites, likely due to coding and population differences.

DISCUSSION

This open-source implementation of a "loyalty score" algorithm had good predictive power. Enriching research cohorts by utilizing these low-missingness patients is a way to obtain the data completeness necessary for accurate causal analysis.

CONCLUSION

i2b2 sites can use this approach to select cohorts with mostly complete EHR data.

Characterization of long COVID temporal sub-phenotypes by distributed representation learning from electronic health record data: a cohort study

Background

Characterizing Post-Acute Sequelae of COVID (SARS-CoV-2 Infection), or PASC has been challenging due to the multitude of sub-phenotypes, temporal attributes, and definitions. Scalable characterization of PASC sub-phenotypes can enhance screening capacities, disease management, and treatment planning.

Methods

We conducted a retrospective multi-centre observational cohort study, leveraging longitudinal electronic health record (EHR) data of 30,422 patients from three healthcare systems in the Consortium for the Clinical Characterization of COVID-19 by EHR (4CE). From the total cohort, we applied a deductive approach on 12,424 individuals with follow-up data and developed a distributed representation learning process for providing augmented definitions for PASC sub-phenotypes.

Findings

Our framework characterized seven PASC sub-phenotypes. We estimated that on average 15.7% of the hospitalized COVID-19 patients were likely to suffer from at least one PASC symptom and almost 5.98%, on average, had multiple symptoms. Joint pain and dyspnea had the highest prevalence, with an average prevalence of 5.45% and 4.53%, respectively.

Interpretation

We provided a scalable framework to every participating healthcare system for estimating PASC sub-phenotypes prevalence and temporal attributes, thus developing a unified model that characterizes augmented sub-phenotypes across the different systems.

Funding

Authors are supported by National Institute of Allergy and Infectious Diseases, National Institute on Aging, National Center for Advancing Translational Sciences, National Medical Research Council, National Institute of Neurological Disorders and Stroke, European Union, National Institutes of Health, National Center for Advancing Translational Sciences.

Clinical phenotypes and outcomes in children with multisystem inflammatory syndrome across SARS-CoV-2 variant eras: a multinational study from the 4CE consortium

Background

Multisystem inflammatory syndrome in children (MIS-C) is a severe complication of SARS-CoV-2 infection. It remains unclear how MIS-C phenotypes vary across SARS-CoV-2 variants. We aimed to investigate clinical characteristics and outcomes of MIS-C across SARS-CoV-2 eras.

Methods

We performed a multicentre observational retrospective study including seven paediatric hospitals in four countries (France, Spain, U.K., and U.S.). All consecutive confirmed patients with MIS-C hospitalised between February 1st, 2020, and May 31st, 2022, were included. Electronic Health Records (EHR) data were used to calculate pooled risk differences (RD) and effect sizes (ES) at site level, using Alpha as reference. Meta-analysis was used to pool data across sites.

Findings

Of 598 patients with MIS-C (61% male, 39% female; mean age 9.7 years [SD 4.5]), 383 (64%) were admitted in the Alpha era, 111 (19%) in the Delta era, and 104 (17%) in the Omicron era. Compared with patients admitted in the Alpha era, those admitted in the Delta era were younger (ES -1.18 years [95% CI -2.05, -0.32]), had fewer respiratory symptoms (RD -0.15 [95% CI -0.33, -0.04]), less frequent non-cardiogenic shock or systemic inflammatory response syndrome (SIRS) (RD -0.35 [95% CI -0.64, -0.07]), lower lymphocyte count (ES -0.16 × 109/uL [95% CI -0.30, -0.01]), lower C-reactive protein (ES -28.5 mg/L [95% CI -46.3, -10.7]), and lower troponin (ES -0.14 ng/mL [95% CI -0.26, -0.03]). Patients admitted in the Omicron versus Alpha eras were younger (ES -1.6 years [95% CI -2.5, -0.8]), had less frequent SIRS (RD -0.18 [95% CI -0.30, -0.05]), lower lymphocyte count (ES -0.39 × 109/uL [95% CI -0.52, -0.25]), lower troponin (ES -0.16 ng/mL [95% CI -0.30, -0.01]) and less frequently received anticoagulation therapy (RD -0.19 [95% CI -0.37, -0.04]). Length of hospitalization was shorter in the Delta versus Alpha eras (-1.3 days [95% CI -2.3, -0.4]).

Interpretation

Our study suggested that MIS-C clinical phenotypes varied across SARS-CoV-2 eras, with patients in Delta and Omicron eras being younger and less sick. EHR data can be effectively leveraged to identify rare complications of pandemic diseases and their variation over time.

Funding

None.

Sharing and Reusing Computable Phenotype Definitions

Background

A scalable approach for the sharing and reuse of human-readable and computer-executable phenotype definitions can facilitate the reuse of electronic health records for cohort identification and research studies.

Description

We developed a tool called Sharephe for the Informatics for Integrating Biology and the Bedside (i2b2) platform. Sharephe consists of a plugin for i2b2 and a cloud-based searchable repository of computable phenotypes, has the functionality to import to and export from the repository, and has the ability to link to supporting metadata.

Discussion

The i2b2 platform enables researchers to create, evaluate, and implement phenotypes without knowing complex query languages. In an initial evaluation, two sites on the Evolve to Next-Gen ACT (ENACT) network used Sharephe to successfully create, share, and reuse phenotypes.

Conclusion

The combination of a cloud-based computable repository and an i2b2 plugin for accessing the repository enables investigators to store and retrieve phenotypes from anywhere and at any time and to collaborate across sites in a research network.

Measuring and Reducing Racial Bias in a Pediatric Urinary Tract Infection Model

Clinical predictive models that include race as a predictor have the potential to exacerbate disparities in healthcare. Such models can be respecified to exclude race or optimized to reduce racial bias. We investigated the impact of such respecifications in a predictive model - UTICalc - which was designed to reduce catheterizations in young children with suspected urinary tract infections. To reduce racial bias, race was removed from the UTICalc logistic regression model and replaced with two new features. We compared the two versions of UTICalc using fairness and predictive performance metrics to understand the effects on racial bias. In addition, we derived three new models for UTICalc to specifically improve racial fairness. Our results show that, as predicted by previously described impossibility results, fairness cannot be simultaneously improved on all fairness metrics, and model respecification may improve racial fairness but decrease overall predictive performance.

ReDWINE: A clinical datamart with text analytical capabilities to facilitate rehabilitation research

Rehabilitation research focuses on determining the components of a treatment intervention, the mechanism of how these components lead to recovery and rehabilitation, and ultimately the optimal intervention strategies to maximize patients' physical, psychologic, and social functioning. Traditional randomized clinical trials that study and establish new interventions face challenges, such as high cost and time commitment. Observational studies that use existing clinical data to observe the effect of an intervention have shown several advantages over RCTs. Electronic Health Records (EHRs) have become an increasingly important resource for conducting observational studies. To support these studies, we developed a clinical research datamart, called ReDWINE (Rehabilitation Datamart With Informatics iNfrastructure for rEsearch), that transforms the rehabilitation-related EHR data collected from the UPMC health care system to the Observational Health Data Sciences and Informatics (OHDSI) Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) to facilitate rehabilitation research. The standardized EHR data stored in ReDWINE will further reduce the time and effort required by investigators to pool, harmonize, clean, and analyze data from multiple sources, leading to more robust and comprehensive research findings. ReDWINE also includes deployment of data visualization and data analytics tools to facilitate cohort definition and clinical data analysis. These include among others the Open Health Natural Language Processing (OHNLP) toolkit, a high-throughput NLP pipeline, to provide text analytical capabilities at scale in ReDWINE. Using this comprehensive representation of patient data in ReDWINE for rehabilitation research will facilitate real-world evidence for health interventions and outcomes.

Subtyping Social Determinants of Health in All of Us: Network Analysis and Visualization Approach

Background

Social determinants of health (SDoH), such as financial resources and housing stability, account for between 30-55% of people's health outcomes. While many studies have identified strong associations among specific SDoH and health outcomes, most people experience multiple SDoH that impact their daily lives. Analysis of this complexity requires the integration of personal, clinical, social, and environmental information from a large cohort of individuals that have been traditionally underrepresented in research, which is only recently being made available through the All of Us research program. However, little is known about the range and response of SDoH in All of Us, and how they co-occur to form subtypes, which are critical for designing targeted interventions.

Objective

To address two research questions: (1) What is the range and response to survey questions related to SDoH in the All of Us dataset? (2) How do SDoH co-occur to form subtypes, and what are their risk for adverse health outcomes?

Methods

For Question-1, an expert panel analyzed the range of SDoH questions across the surveys with respect to the 5 domains in Healthy People 2030 (HP-30), and analyzed their responses across the full All of Us data (n=372,397, V6). For Question-2, we used the following steps: (1) due to the missingness across the surveys, selected all participants with valid and complete SDoH data, and used inverse probability weighting to adjust their imbalance in demographics compared to the full data; (2) an expert panel grouped the SDoH questions into SDoH factors for enabling a more consistent granularity; (3) used bipartite modularity maximization to identify SDoH biclusters, their significance, and their replicability; (4) measured the association of each bicluster to three outcomes (depression, delayed medical care, emergency room visits in the last year) using multiple data types (surveys, electronic health records, and zip codes mapped to Medicaid expansion states); and (5) the expert panel inferred the subtype labels, potential mechanisms that precipitate adverse health outcomes, and interventions to prevent them.

Results

For Question-1, we identified 110 SDoH questions across 4 surveys, which covered all 5 domains in HP-30. However, the results also revealed a large degree of missingness in survey responses (1.76%-84.56%), with later surveys having significantly fewer responses compared to earlier ones, and significant differences in race, ethnicity, and age of participants of those that completed the surveys with SDoH questions, compared to those in the full All of Us dataset. Furthermore, as the SDoH questions varied in granularity, they were categorized by an expert panel into 18 SDoH factors. For Question-2, the subtype analysis (n=12,913, d=18) identified 4 biclusters with significant biclusteredness (Q=0.13, random-Q=0.11, z=7.5, P<0.001), and significant replication (Real-RI=0.88, Random-RI=0.62, P<.001). Furthermore, there were statistically significant associations between specific subtypes and the outcomes, and with Medicaid expansion, each with meaningful interpretations and potential targeted interventions. For example, the subtype Socioeconomic Barriers included the SDoH factors not employed, food insecurity, housing insecurity, low income, low literacy, and low educational attainment, and had a significantly higher odds ratio (OR=4.2, CI=3.5-5.1, P-corr<.001) for depression, when compared to the subtype Sociocultural Barriers. Individuals that match this subtype profile could be screened early for depression and referred to social services for addressing combinations of SDoH such as housing insecurity and low income. Finally, the identified subtypes spanned one or more HP-30 domains revealing the difference between the current knowledge-based SDoH domains, and the data-driven subtypes.

Conclusions

The results revealed that the SDoH subtypes not only had statistically significant clustering and replicability, but also had significant associations with critical adverse health outcomes, which had translational implications for designing targeted SDoH interventions, decision-support systems to alert clinicians of potential risks, and for public policies. Furthermore, these SDoH subtypes spanned multiple SDoH domains defined by HP-30 revealing the complexity of SDoH in the real-world, and aligning with influential SDoH conceptual models such as by Dahlgren-Whitehead. However, the high-degree of missingness warrants repeating the analysis as the data becomes more complete. Consequently we designed our machine learning code to be generalizable and scalable, and made it available on the All of Us workbench, which can be used to periodically rerun the analysis as the dataset grows for analyzing subtypes related to SDoH, and beyond.

Clinical Algorithms with Race: An Online Database

Some clinical algorithms incorporate a person's race, ethnicity, or both as an input variable or predictor in determining diagnoses, prognoses, treatment plans, or risk assessments. Inappropriate use of race and ethnicity in clinical algorithms at the point of care may exacerbate health disparities and promote harmful practices of race-based medicine. This article describes a comprehensive search of online resources, the scientific literature, and the FDA Drug Label Information that uncovered 39 race-based risk calculators, six laboratory test results with race-based reference ranges, one race-based therapy recommendation, and 15 medications with race-based recommendations. These clinical algorithms based on race are freely accessible through an online database. This resource aims to raise awareness about the use of race-based clinical algorithms and track the progress made toward eradicating the inappropriate use of race. The database will be actively updated to include clinical algorithms based on race that were previously omitted, along with additional characteristics of these algorithms.

A retrospective cohort analysis leveraging augmented intelligence to characterize long COVID in the electronic health record: A precision medicine framework

Physical and psychological symptoms lasting months following an acute COVID-19 infection are now recognized as post-acute sequelae of COVID-19 (PASC). Accurate tools for identifying such patients could enhance screening capabilities for the recruitment for clinical trials, improve the reliability of disease estimates, and allow for more accurate downstream cohort analysis. In this retrospective cohort study, we analyzed the EHR of hospitalized COVID-19 patients across three healthcare systems to develop a pipeline for better identifying patients with persistent PASC symptoms (dyspnea, fatigue, or joint pain) after their SARS-CoV-2 infection. We implemented distributed representation learning powered by the Machine Learning for modeling Health Outcomes (MLHO) to identify novel EHR features that could suggest PASC symptoms outside of typical diagnosis codes. MLHO applies an entropy-based feature selection and boosting algorithms for representation mining. These improved definitions were then used for estimating PASC among hospitalized patients. 30,422 hospitalized patients were diagnosed with COVID-19 across three healthcare systems between March 13, 2020 and February 28, 2021. The mean age of the population was 62.3 years (SD, 21.0 years) and 15,124 (49.7%) were female. We implemented the distributed representation learning technique to augment PASC definitions. These definitions were found to have positive predictive values of 0.73, 0.74, and 0.91 for dyspnea, fatigue, and joint pain, respectively. We estimated that 25 percent (CI 95%: 6-48), 11 percent (CI 95%: 6-15), and 13 percent (CI 95%: 8-17) of hospitalized COVID-19 patients will have dyspnea, fatigue, and joint pain, respectively, 3 months or longer after a COVID-19 diagnosis. We present a validated framework for screening and identifying patients with PASC in the EHR and then use the tool to estimate its prevalence among hospitalized COVID-19 patients.

Survey and Evaluation of Hypertension Machine Learning Research

Background Machine learning (ML) is pervasive in all fields of research, from automating tasks to complex decision-making. However, applications in different specialities are variable and generally limited. Like other conditions, the number of studies employing ML in hypertension research is growing rapidly. In this study, we aimed to survey hypertension research using ML, evaluate the reporting quality, and identify barriers to ML's potential to transform hypertension care. Methods and Results The Harmonious Understanding of Machine Learning Analytics Network survey questionnaire was applied to 63 hypertension-related ML research articles published between January 2019 and September 2021. The most common research topics were blood pressure prediction (38%), hypertension (22%), cardiovascular outcomes (6%), blood pressure variability (5%), treatment response (5%), and real-time blood pressure estimation (5%). The reporting quality of the articles was variable. Only 46% of articles described the study population or derivation cohort. Most articles (81%) reported at least 1 performance measure, but only 40% presented any measures of calibration. Compliance with ethics, patient privacy, and data security regulations were mentioned in 30 (48%) of the articles. Only 14% used geographically or temporally distinct validation data sets. Algorithmic bias was not addressed in any of the articles, with only 6 of them acknowledging risk of bias. Conclusions Recent ML research on hypertension is limited to exploratory research and has significant shortcomings in reporting quality, model validation, and algorithmic bias. Our analysis identifies areas for improvement that will help pave the way for the realization of the potential of ML in hypertension and facilitate its adoption.

A new method for estimating the probability of causal relationships from observational data: Application to the study of the short-term effects of air pollution on cardiovascular and respiratory disease

In this paper we investigate which airborne pollutants have a short-term causal effect on cardiovascular and respiratory disease using the Ancestral Probabilities (AP) procedure, a novel Bayesian approach for deriving the probabilities of causal relationships from observational data. The results are largely consistent with EPA assessments of causality, however, in a few cases AP suggests that some pollutants thought to cause cardiovascular or respiratory disease are associated due purely to confounding. The AP procedure utilizes maximal ancestral graph (MAG) models to represent and assign probabilities to causal relationships while accounting for latent confounding. The algorithm does so locally by marginalizing over models with and without causal features of interest. Before applying AP to real data, we evaluate it in a simulation study and investigate the benefits of providing background knowledge. Overall, the results suggest that AP is an effective tool for causal discovery.

Informative missingness: What can we learn from patterns in missing laboratory data in the electronic health record?

BACKGROUND

In electronic health records, patterns of missing laboratory test results could capture patients' course of disease as well as ​​reflect clinician's concerns or worries for possible conditions. These patterns are often understudied and overlooked. This study aims to identify informative patterns of missingness among laboratory data collected across 15 healthcare system sites in three countries for COVID-19 inpatients.

METHODS

We collected and analyzed demographic, diagnosis, and laboratory data for 69,939 patients with positive COVID-19 PCR tests across three countries from 1 January 2020 through 30 September 2021. We analyzed missing laboratory measurements across sites, missingness stratification by demographic variables, temporal trends of missingness, correlations between labs based on missingness indicators over time, and clustering of groups of labs based on their missingness/ordering pattern.

RESULTS

With these analyses, we identified mapping issues faced in seven out of 15 sites. We also identified nuances in data collection and variable definition for the various sites. Temporal trend analyses may support the use of laboratory test result missingness patterns in identifying severe COVID-19 patients. Lastly, using missingness patterns, we determined relationships between various labs that reflect clinical behaviors.

CONCLUSION

In this work, we use computational approaches to relate missingness patterns to hospital treatment capacity and highlight the heterogeneity of looking at COVID-19 over time and at multiple sites, where there might be different phases, policies, etc. Changes in missingness could suggest a change in a patient's condition, and patterns of missingness among laboratory measurements could potentially identify clinical outcomes. This allows sites to consider missing data as informative to analyses and help researchers identify which sites are better poised to study particular questions.

Acute respiratory distress syndrome after SARS-CoV-2 infection on young adult population: International observational federated study based on electronic health records through the 4CE consortium

PURPOSE

In young adults (18 to 49 years old), investigation of the acute respiratory distress syndrome (ARDS) after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been limited. We evaluated the risk factors and outcomes of ARDS following infection with SARS-CoV-2 in a young adult population.

METHODS

A retrospective cohort study was conducted between January 1st, 2020 and February 28th, 2021 using patient-level electronic health records (EHR), across 241 United States hospitals and 43 European hospitals participating in the Consortium for Clinical Characterization of COVID-19 by EHR (4CE). To identify the risk factors associated with ARDS, we compared young patients with and without ARDS through a federated analysis. We further compared the outcomes between young and old patients with ARDS.

RESULTS

Among the 75,377 hospitalized patients with positive SARS-CoV-2 PCR, 1001 young adults presented with ARDS (7.8% of young hospitalized adults). Their mortality rate at 90 days was 16.2% and they presented with a similar complication rate for infection than older adults with ARDS. Peptic ulcer disease, paralysis, obesity, congestive heart failure, valvular disease, diabetes, chronic pulmonary disease and liver disease were associated with a higher risk of ARDS. We described a high prevalence of obesity (53%), hypertension (38%- although not significantly associated with ARDS), and diabetes (32%).

CONCLUSION

Trough an innovative method, a large international cohort study of young adults developing ARDS after SARS-CoV-2 infection has been gather. It demonstrated the poor outcomes of this population and associated risk factor.

Long-term kidney function recovery and mortality after COVID-19-associated acute kidney injury: An international multi-centre observational cohort study

Background

While acute kidney injury (AKI) is a common complication in COVID-19, data on post-AKI kidney function recovery and the clinical factors associated with poor kidney function recovery is lacking.

Methods

A retrospective multi-centre observational cohort study comprising 12,891 hospitalized patients aged 18 years or older with a diagnosis of SARS-CoV-2 infection confirmed by polymerase chain reaction from 1 January 2020 to 10 September 2020, and with at least one serum creatinine value 1-365 days prior to admission. Mortality and serum creatinine values were obtained up to 10 September 2021.

Findings

Advanced age (HR 2.77, 95%CI 2.53-3.04, p < 0.0001), severe COVID-19 (HR 2.91, 95%CI 2.03-4.17, p < 0.0001), severe AKI (KDIGO stage 3: HR 4.22, 95%CI 3.55-5.00, p < 0.0001), and ischemic heart disease (HR 1.26, 95%CI 1.14-1.39, p < 0.0001) were associated with worse mortality outcomes. AKI severity (KDIGO stage 3: HR 0.41, 95%CI 0.37-0.46, p < 0.0001) was associated with worse kidney function recovery, whereas remdesivir use (HR 1.34, 95%CI 1.17-1.54, p < 0.0001) was associated with better kidney function recovery. In a subset of patients without chronic kidney disease, advanced age (HR 1.38, 95%CI 1.20-1.58, p < 0.0001), male sex (HR 1.67, 95%CI 1.45-1.93, p < 0.0001), severe AKI (KDIGO stage 3: HR 11.68, 95%CI 9.80-13.91, p < 0.0001), and hypertension (HR 1.22, 95%CI 1.10-1.36, p = 0.0002) were associated with post-AKI kidney function impairment. Furthermore, patients with COVID-19-associated AKI had significant and persistent elevations of baseline serum creatinine 125% or more at 180 days (RR 1.49, 95%CI 1.32-1.67) and 365 days (RR 1.54, 95%CI 1.21-1.96) compared to COVID-19 patients with no AKI.

Interpretation

COVID-19-associated AKI was associated with higher mortality, and severe COVID-19-associated AKI was associated with worse long-term post-AKI kidney function recovery.

Funding

Authors are supported by various funders, with full details stated in the acknowledgement section.

A Framework for Modeling and Interpreting Patient Subgroups Applied to Hospital Readmission: Visual Analytical Approach

BACKGROUND

A primary goal of precision medicine is to identify patient subgroups and infer their underlying disease processes with the aim of designing targeted interventions. Although several studies have identified patient subgroups, there is a considerable gap between the identification of patient subgroups and their modeling and interpretation for clinical applications.

OBJECTIVE

This study aimed to develop and evaluate a novel analytical framework for modeling and interpreting patient subgroups (MIPS) using a 3-step modeling approach: visual analytical modeling to automatically identify patient subgroups and their co-occurring comorbidities and determine their statistical significance and clinical interpretability; classification modeling to classify patients into subgroups and measure its accuracy; and prediction modeling to predict a patient's risk of an adverse outcome and compare its accuracy with and without patient subgroup information.

METHODS

The MIPS framework was developed using bipartite networks to identify patient subgroups based on frequently co-occurring high-risk comorbidities, multinomial logistic regression to classify patients into subgroups, and hierarchical logistic regression to predict the risk of an adverse outcome using subgroup membership compared with standard logistic regression without subgroup membership. The MIPS framework was evaluated for 3 hospital readmission conditions: chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), and total hip arthroplasty/total knee arthroplasty (THA/TKA) (COPD: n=29,016; CHF: n=51,550; THA/TKA: n=16,498). For each condition, we extracted cases defined as patients readmitted within 30 days of hospital discharge. Controls were defined as patients not readmitted within 90 days of discharge, matched by age, sex, race, and Medicaid eligibility.

RESULTS

In each condition, the visual analytical model identified patient subgroups that were statistically significant (Q=0.17, 0.17, 0.31; P<.001, <.001, <.05), significantly replicated (Rand Index=0.92, 0.94, 0.89; P<.001, <.001, <.01), and clinically meaningful to clinicians. In each condition, the classification model had high accuracy in classifying patients into subgroups (mean accuracy=99.6%, 99.34%, 99.86%). In 2 conditions (COPD and THA/TKA), the hierarchical prediction model had a small but statistically significant improvement in discriminating between readmitted and not readmitted patients as measured by net reclassification improvement (0.059, 0.11) but not as measured by the C-statistic or integrated discrimination improvement.

CONCLUSIONS

Although the visual analytical models identified statistically and clinically significant patient subgroups, the results pinpoint the need to analyze subgroups at different levels of granularity for improving the interpretability of intra- and intercluster associations. The high accuracy of the classification models reflects the strong separation of patient subgroups, despite the size and density of the data sets. Finally, the small improvement in predictive accuracy suggests that comorbidities alone were not strong predictors of hospital readmission, and the need for more sophisticated subgroup modeling methods. Such advances could improve the interpretability and predictive accuracy of patient subgroup models for reducing the risk of hospital readmission, and beyond.

Hospitalizations Associated With Mental Health Conditions Among Adolescents in the US and France During the COVID-19 Pandemic

Importance

The COVID-19 pandemic has been associated with an increase in mental health diagnoses among adolescents, though the extent of the increase, particularly for severe cases requiring hospitalization, has not been well characterized. Large-scale federated informatics approaches provide the ability to efficiently and securely query health care data sets to assess and monitor hospitalization patterns for mental health conditions among adolescents.

Objective

To estimate changes in the proportion of hospitalizations associated with mental health conditions among adolescents following onset of the COVID-19 pandemic.

Design, Setting, and Participants

This retrospective, multisite cohort study of adolescents 11 to 17 years of age who were hospitalized with at least 1 mental health condition diagnosis between February 1, 2019, and April 30, 2021, used patient-level data from electronic health records of 8 children's hospitals in the US and France.

Main Outcomes and Measures

Change in the monthly proportion of mental health condition-associated hospitalizations between the prepandemic (February 1, 2019, to March 31, 2020) and pandemic (April 1, 2020, to April 30, 2021) periods using interrupted time series analysis.

Results

There were 9696 adolescents hospitalized with a mental health condition during the prepandemic period (5966 [61.5%] female) and 11 101 during the pandemic period (7603 [68.5%] female). The mean (SD) age in the prepandemic cohort was 14.6 (1.9) years and in the pandemic cohort, 14.7 (1.8) years. The most prevalent diagnoses during the pandemic were anxiety (6066 [57.4%]), depression (5065 [48.0%]), and suicidality or self-injury (4673 [44.2%]). There was an increase in the proportions of monthly hospitalizations during the pandemic for anxiety (0.55%; 95% CI, 0.26%-0.84%), depression (0.50%; 95% CI, 0.19%-0.79%), and suicidality or self-injury (0.38%; 95% CI, 0.08%-0.68%). There was an estimated 0.60% increase (95% CI, 0.31%-0.89%) overall in the monthly proportion of mental health-associated hospitalizations following onset of the pandemic compared with the prepandemic period.

Conclusions and Relevance

In this cohort study, onset of the COVID-19 pandemic was associated with increased hospitalizations with mental health diagnoses among adolescents. These findings support the need for greater resources within children's hospitals to care for adolescents with mental health conditions during the pandemic and beyond.

SurvMaximin: Robust federated approach to transporting survival risk prediction models

OBJECTIVE

For multi-center heterogeneous Real-World Data (RWD) with time-to-event outcomes and high-dimensional features, we propose the SurvMaximin algorithm to estimate Cox model feature coefficients for a target population by borrowing summary information from a set of health care centers without sharing patient-level information.

MATERIALS AND METHODS

For each of the centers from which we want to borrow information to improve the prediction performance for the target population, a penalized Cox model is fitted to estimate feature coefficients for the center. Using estimated feature coefficients and the covariance matrix of the target population, we then obtain a SurvMaximin estimated set of feature coefficients for the target population. The target population can be an entire cohort comprised of all centers, corresponding to federated learning, or a single center, corresponding to transfer learning.

RESULTS

Simulation studies and a real-world international electronic health records application study, with 15 participating health care centers across three countries (France, Germany, and the U.S.), show that the proposed SurvMaximin algorithm achieves comparable or higher accuracy compared with the estimator using only the information of the target site and other existing methods. The SurvMaximin estimator is robust to variations in sample sizes and estimated feature coefficients between centers, which amounts to significantly improved estimates for target sites with fewer observations.

CONCLUSIONS

The SurvMaximin method is well suited for both federated and transfer learning in the high-dimensional survival analysis setting. SurvMaximin only requires a one-time summary information exchange from participating centers. Estimated regression vectors can be very heterogeneous. SurvMaximin provides robust Cox feature coefficient estimates without outcome information in the target population and is privacy-preserving.

The All of Us Research Program: Data quality, utility, and diversity

The All of Us Research Program seeks to engage at least one million diverse participants to advance precision medicine and improve human health. We describe here the cloud-based Researcher Workbench that uses a data passport model to democratize access to analytical tools and participant information including survey, physical measurement, and electronic health record (EHR) data. We also present validation study findings for several common complex diseases to demonstrate use of this novel platform in 315,000 participants, 78% of whom are from groups historically underrepresented in biomedical research, including 49% self-reporting non-White races. Replication findings include medication usage pattern differences by race in depression and type 2 diabetes, validation of known cancer associations with smoking, and calculation of cardiovascular risk scores by reported race effects. The cloud-based Researcher Workbench represents an important advance in enabling secure access for a broad range of researchers to this large resource and analytical tools.

•

The All of Us Research Program has released data for over 315,000 participants

•

Demonstration projects support the utility and validity of the All of Us dataset

•

The cloud-based Researcher Workbench provides secure, low-cost compute power

The engagement of participants in the research process and broad availability of data to diverse researchers are essential elements in building precision medicine equitably available for all. The NIH has established the ambitious All of Us Research Program to build one of the most diverse health databases in history with tools to support research to improve human health. Here, we present the initial launch of the Researcher Workbench with data types including surveys, physical measurements, and electronic health record data with validation studies to support researcher use of this novel platform. Broad access for researchers to data like these is a critical step in returning value to participants seeking to support the advancement of precision medicine and improved health for all.

International electronic health record-derived post-acute sequelae profiles of COVID-19 patients

The risk profiles of post-acute sequelae of COVID-19 (PASC) have not been well characterized in multi-national settings with appropriate controls. We leveraged electronic health record (EHR) data from 277 international hospitals representing 414,602 patients with COVID-19, 2.3 million control patients without COVID-19 in the inpatient and outpatient settings, and over 221 million diagnosis codes to systematically identify new-onset conditions enriched among patients with COVID-19 during the post-acute period. Compared to inpatient controls, inpatient COVID-19 cases were at significant risk for angina pectoris (RR 1.30, 95% CI 1.09–1.55), heart failure (RR 1.22, 95% CI 1.10–1.35), cognitive dysfunctions (RR 1.18, 95% CI 1.07–1.31), and fatigue (RR 1.18, 95% CI 1.07–1.30). Relative to outpatient controls, outpatient COVID-19 cases were at risk for pulmonary embolism (RR 2.10, 95% CI 1.58–2.76), venous embolism (RR 1.34, 95% CI 1.17–1.54), atrial fibrillation (RR 1.30, 95% CI 1.13–1.50), type 2 diabetes (RR 1.26, 95% CI 1.16–1.36) and vitamin D deficiency (RR 1.19, 95% CI 1.09–1.30). Outpatient COVID-19 cases were also at risk for loss of smell and taste (RR 2.42, 95% CI 1.90–3.06), inflammatory neuropathy (RR 1.66, 95% CI 1.21–2.27), and cognitive dysfunction (RR 1.18, 95% CI 1.04–1.33). The incidence of post-acute cardiovascular and pulmonary conditions decreased across time among inpatient cases while the incidence of cardiovascular, digestive, and metabolic conditions increased among outpatient cases. Our study, based on a federated international network, systematically identified robust conditions associated with PASC compared to control groups, underscoring the multifaceted cardiovascular and neurological phenotype profiles of PASC.

Changes in laboratory value improvement and mortality rates over the course of the pandemic: an international retrospective cohort study of hospitalised patients infected with SARS-CoV-2

OBJECTIVE

To assess changes in international mortality rates and laboratory recovery rates during hospitalisation for patients hospitalised with SARS-CoV-2 between the first wave (1 March to 30 June 2020) and the second wave (1 July 2020 to 31 January 2021) of the COVID-19 pandemic.

DESIGN, SETTING AND PARTICIPANTS

This is a retrospective cohort study of 83 178 hospitalised patients admitted between 7 days before or 14 days after PCR-confirmed SARS-CoV-2 infection within the Consortium for Clinical Characterization of COVID-19 by Electronic Health Record, an international multihealthcare system collaborative of 288 hospitals in the USA and Europe. The laboratory recovery rates and mortality rates over time were compared between the two waves of the pandemic.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome was all-cause mortality rate within 28 days after hospitalisation stratified by predicted low, medium and high mortality risk at baseline. The secondary outcome was the average rate of change in laboratory values during the first week of hospitalisation.

RESULTS

Baseline Charlson Comorbidity Index and laboratory values at admission were not significantly different between the first and second waves. The improvement in laboratory values over time was faster in the second wave compared with the first. The average C reactive protein rate of change was -4.72 mg/dL vs -4.14 mg/dL per day (p=0.05). The mortality rates within each risk category significantly decreased over time, with the most substantial decrease in the high-risk group (42.3% in March-April 2020 vs 30.8% in November 2020 to January 2021, p<0.001) and a moderate decrease in the intermediate-risk group (21.5% in March-April 2020 vs 14.3% in November 2020 to January 2021, p<0.001).

CONCLUSIONS

Admission profiles of patients hospitalised with SARS-CoV-2 infection did not differ greatly between the first and second waves of the pandemic, but there were notable differences in laboratory improvement rates during hospitalisation. Mortality risks among patients with similar risk profiles decreased over the course of the pandemic. The improvement in laboratory values and mortality risk was consistent across multiple countries.

International comparisons of laboratory values from the 4CE collaborative to predict COVID-19 mortality

Given the growing number of prediction algorithms developed to predict COVID-19 mortality, we evaluated the transportability of a mortality prediction algorithm using a multi-national network of healthcare systems. We predicted COVID-19 mortality using baseline commonly measured laboratory values and standard demographic and clinical covariates across healthcare systems, countries, and continents. Specifically, we trained a Cox regression model with nine measured laboratory test values, standard demographics at admission, and comorbidity burden pre-admission. These models were compared at site, country, and continent level. Of the 39,969 hospitalized patients with COVID-19 (68.6% male), 5717 (14.3%) died. In the Cox model, age, albumin, AST, creatine, CRP, and white blood cell count are most predictive of mortality. The baseline covariates are more predictive of mortality during the early days of COVID-19 hospitalization. Models trained at healthcare systems with larger cohort size largely retain good transportability performance when porting to different sites. The combination of routine laboratory test values at admission along with basic demographic features can predict mortality in patients hospitalized with COVID-19. Importantly, this potentially deployable model differs from prior work by demonstrating not only consistent performance but also reliable transportability across healthcare systems in the US and Europe, highlighting the generalizability of this model and the overall approach.

A Novel Personalized Random Forest Algorithm for Clinical Outcome Prediction

Machine learning algorithms that derive predictive models are useful in predicting patient outcomes under uncertainty. These are often "population" algorithms which optimize a static model to predict well on average for individuals in the population; however, population models may predict poorly for individuals that differ from the average. Personalized machine learning algorithms seek to optimize predictive performance for every patient by tailoring a patient-specific model to each individual. Ensembles of decision trees often outperform single decision tree models, but ensembles of personalized models like decision paths have received little investigation. We present a novel personalized ensemble, called Lazy Random Forest (LazyRF), which consists of bagged randomized decision paths optimized for the individual for whom a prediction will be made. LazyRF outperformed single and bagged decision paths and demonstrated comparable predictive performance to a population random forest method in terms of discrimination on clinical and genomic data while also producing simpler models than the population random forest.

Distinguishing Admissions Specifically for COVID-19 From Incidental SARS-CoV-2 Admissions: National Retrospective Electronic Health Record Study

BACKGROUND

Admissions are generally classified as COVID-19 hospitalizations if the patient has a positive SARS-CoV-2 polymerase chain reaction (PCR) test. However, because 35% of SARS-CoV-2 infections are asymptomatic, patients admitted for unrelated indications with an incidentally positive test could be misclassified as a COVID-19 hospitalization. Electronic health record (EHR)-based studies have been unable to distinguish between a hospitalization specifically for COVID-19 versus an incidental SARS-CoV-2 hospitalization. Although the need to improve classification of COVID-19 versus incidental SARS-CoV-2 is well understood, the magnitude of the problems has only been characterized in small, single-center studies. Furthermore, there have been no peer-reviewed studies evaluating methods for improving classification.

OBJECTIVE

The aims of this study are to, first, quantify the frequency of incidental hospitalizations over the first 15 months of the pandemic in multiple hospital systems in the United States and, second, to apply electronic phenotyping techniques to automatically improve COVID-19 hospitalization classification.

METHODS

From a retrospective EHR-based cohort in 4 US health care systems in Massachusetts, Pennsylvania, and Illinois, a random sample of 1123 SARS-CoV-2 PCR-positive patients hospitalized from March 2020 to August 2021 was manually chart-reviewed and classified as "admitted with COVID-19" (incidental) versus specifically admitted for COVID-19 ("for COVID-19"). EHR-based phenotyping was used to find feature sets to filter out incidental admissions.

RESULTS

EHR-based phenotyped feature sets filtered out incidental admissions, which occurred in an average of 26% of hospitalizations (although this varied widely over time, from 0% to 75%). The top site-specific feature sets had 79%-99% specificity with 62%-75% sensitivity, while the best-performing across-site feature sets had 71%-94% specificity with 69%-81% sensitivity.

CONCLUSIONS

A large proportion of SARS-CoV-2 PCR-positive admissions were incidental. Straightforward EHR-based phenotypes differentiated admissions, which is important to assure accurate public health reporting and research.

Synergies between centralized and federated approaches to data quality: a report from the national COVID cohort collaborative

OBJECTIVE

In response to COVID-19, the informatics community united to aggregate as much clinical data as possible to characterize this new disease and reduce its impact through collaborative analytics. The National COVID Cohort Collaborative (N3C) is now the largest publicly available HIPAA limited dataset in US history with over 6.4 million patients and is a testament to a partnership of over 100 organizations.

MATERIALS AND METHODS

We developed a pipeline for ingesting, harmonizing, and centralizing data from 56 contributing data partners using 4 federated Common Data Models. N3C data quality (DQ) review involves both automated and manual procedures. In the process, several DQ heuristics were discovered in our centralized context, both within the pipeline and during downstream project-based analysis. Feedback to the sites led to many local and centralized DQ improvements.

RESULTS

Beyond well-recognized DQ findings, we discovered 15 heuristics relating to source Common Data Model conformance, demographics, COVID tests, conditions, encounters, measurements, observations, coding completeness, and fitness for use. Of 56 sites, 37 sites (66%) demonstrated issues through these heuristics. These 37 sites demonstrated improvement after receiving feedback.

DISCUSSION

We encountered site-to-site differences in DQ which would have been challenging to discover using federated checks alone. We have demonstrated that centralized DQ benchmarking reveals unique opportunities for DQ improvement that will support improved research analytics locally and in aggregate.

CONCLUSION

By combining rapid, continual assessment of DQ with a large volume of multisite data, it is possible to support more nuanced scientific questions with the scale and rigor that they require.

Distinguishing Admissions Specifically for COVID-19 from Incidental SARS-CoV-2 Admissions: A National EHR Research Consortium Study

Admissions are generally classified as COVID-19 hospitalizations if the patient has a positive SARS-CoV-2 polymerase chain reaction (PCR) test. However, because 35% of SARS-CoV-2 infections are asymptomatic, patients admitted for unrelated indications with an incidentally positive test could be misclassified as a COVID-19 hospitalization. EHR-based studies have been unable to distinguish between a hospitalization specifically for COVID-19 versus an incidental SARS-CoV-2 hospitalization. From a retrospective EHR-based cohort in four US healthcare systems, a random sample of 1,123 SARS-CoV-2 PCR-positive patients hospitalized between 3/2020â€"8/2021 was manually chart-reviewed and classified as admitted-with-COVID-19 (incidental) vs. specifically admitted for COVID-19 (for-COVID-19). EHR-based phenotyped feature sets filtered out incidental admissions, which occurred in 26%. The top site-specific feature sets had 79-99% specificity with 62-75% sensitivity, while the best performing across-site feature set had 71-94% specificity with 69-81% sensitivity. A large proportion of SARS-CoV-2 PCR-positive admissions were incidental. Straightforward EHR-based phenotypes differentiated admissions, which is important to assure accurate public health reporting and research.

Artificial intelligence in clinical and translational science: Successes, challenges and opportunities

Artificial intelligence (AI) is transforming many domains, including finance, agriculture, defense, and biomedicine. In this paper, we focus on the role of AI in clinical and translational research (CTR), including preclinical research (T1), clinical research (T2), clinical implementation (T3), and public (or population) health (T4). Given the rapid evolution of AI in CTR, we present three complementary perspectives: (1) scoping literature review, (2) survey, and (3) analysis of federally funded projects. For each CTR phase, we addressed challenges, successes, failures, and opportunities for AI. We surveyed Clinical and Translational Science Award (CTSA) hubs regarding AI projects at their institutions. Nineteen of 63 CTSA hubs (30%) responded to the survey. The most common funding source (48.5%) was the federal government. The most common translational phase was T2 (clinical research, 40.2%). Clinicians were the intended users in 44.6% of projects and researchers in 32.3% of projects. The most common computational approaches were supervised machine learning (38.6%) and deep learning (34.2%). The number of projects steadily increased from 2012 to 2020. Finally, we analyzed 2604 AI projects at CTSA hubs using the National Institutes of Health Research Portfolio Online Reporting Tools (RePORTER) database for 2011-2019. We mapped available abstracts to medical subject headings and found that nervous system (16.3%) and mental disorders (16.2) were the most common topics addressed. From a computational perspective, big data (32.3%) and deep learning (30.0%) were most common. This work represents a snapshot in time of the role of AI in the CTSA program.

Categorizing metadata to help mobilize computable biomedical knowledge

INTRODUCTION

Computable biomedical knowledge artifacts (CBKs) are digital objects conveying biomedical knowledge in machine-interpretable structures. As more CBKs are produced and their complexity increases, the value obtained from sharing CBKs grows. Mobilizing CBKs and sharing them widely can only be achieved if the CBKs are findable, accessible, interoperable, reusable, and trustable (FAIR+T). To help mobilize CBKs, we describe our efforts to outline metadata categories to make CBKs FAIR+T.

METHODS

We examined the literature regarding metadata with the potential to make digital artifacts FAIR+T. We also examined metadata available online today for actual CBKs of 12 different types. With iterative refinement, we came to a consensus on key categories of metadata that, when taken together, can make CBKs FAIR+T. We use subject-predicate-object triples to more clearly differentiate metadata categories.

RESULTS

We defined 13 categories of CBK metadata most relevant to making CBKs FAIR+T. Eleven of these categories (type, domain, purpose, identification, location, CBK-to-CBK relationships, technical, authorization and rights management, provenance, evidential basis, and evidence from use metadata) are evident today where CBKs are stored online. Two additional categories (preservation and integrity metadata) were not evident in our examples. We provide a research agenda to guide further study and development of these and other metadata categories.

CONCLUSION

A wide variety of metadata elements in various categories is needed to make CBKs FAIR+T. More work is needed to develop a common framework for CBK metadata that can make CBKs FAIR+T for all stakeholders.

Multinational characterization of neurological phenotypes in patients hospitalized with COVID-19

Neurological complications worsen outcomes in COVID-19. To define the prevalence of neurological conditions among hospitalized patients with a positive SARS-CoV-2 reverse transcription polymerase chain reaction test in geographically diverse multinational populations during early pandemic, we used electronic health records (EHR) from 338 participating hospitals across 6 countries and 3 continents (January-September 2020) for a cross-sectional analysis. We assessed the frequency of International Classification of Disease code of neurological conditions by countries, healthcare systems, time before and after admission for COVID-19 and COVID-19 severity. Among 35,177 hospitalized patients with SARS-CoV-2 infection, there was an increase in the proportion with disorders of consciousness (5.8%, 95% confidence interval [CI] 3.7-7.8%, pFDR < 0.001) and unspecified disorders of the brain (8.1%, 5.7-10.5%, pFDR < 0.001) when compared to the pre-admission proportion. During hospitalization, the relative risk of disorders of consciousness (22%, 19-25%), cerebrovascular diseases (24%, 13-35%), nontraumatic intracranial hemorrhage (34%, 20-50%), encephalitis and/or myelitis (37%, 17-60%) and myopathy (72%, 67-77%) were higher for patients with severe COVID-19 when compared to those who never experienced severe COVID-19. Leveraging a multinational network to capture standardized EHR data, we highlighted the increased prevalence of central and peripheral neurological phenotypes in patients hospitalized with COVID-19, particularly among those with severe disease.

International Changes in COVID-19 Clinical Trajectories Across 315 Hospitals and 6 Countries: Retrospective Cohort Study

Background

Many countries have experienced 2 predominant waves of COVID-19–related hospitalizations. Comparing the clinical trajectories of patients hospitalized in separate waves of the pandemic enables further understanding of the evolving epidemiology, pathophysiology, and health care dynamics of the COVID-19 pandemic.

Objective

In this retrospective cohort study, we analyzed electronic health record (EHR) data from patients with SARS-CoV-2 infections hospitalized in participating health care systems representing 315 hospitals across 6 countries. We compared hospitalization rates, severe COVID-19 risk, and mean laboratory values between patients hospitalized during the first and second waves of the pandemic.

Methods

Using a federated approach, each participating health care system extracted patient-level clinical data on their first and second wave cohorts and submitted aggregated data to the central site. Data quality control steps were adopted at the central site to correct for implausible values and harmonize units. Statistical analyses were performed by computing individual health care system effect sizes and synthesizing these using random effect meta-analyses to account for heterogeneity. We focused the laboratory analysis on C-reactive protein (CRP), ferritin, fibrinogen, procalcitonin, D-dimer, and creatinine based on their reported associations with severe COVID-19.

Results

Data were available for 79,613 patients, of which 32,467 were hospitalized in the first wave and 47,146 in the second wave. The prevalence of male patients and patients aged 50 to 69 years decreased significantly between the first and second waves. Patients hospitalized in the second wave had a 9.9% reduction in the risk of severe COVID-19 compared to patients hospitalized in the first wave (95% CI 8.5%-11.3%). Demographic subgroup analyses indicated that patients aged 26 to 49 years and 50 to 69 years; male and female patients; and black patients had significantly lower risk for severe disease in the second wave than in the first wave. At admission, the mean values of CRP were significantly lower in the second wave than in the first wave. On the seventh hospital day, the mean values of CRP, ferritin, fibrinogen, and procalcitonin were significantly lower in the second wave than in the first wave. In general, countries exhibited variable changes in laboratory testing rates from the first to the second wave. At admission, there was a significantly higher testing rate for D-dimer in France, Germany, and Spain.

Conclusions

Patients hospitalized in the second wave were at significantly lower risk for severe COVID-19. This corresponded to mean laboratory values in the second wave that were more likely to be in typical physiological ranges on the seventh hospital day compared to the first wave. Our federated approach demonstrated the feasibility and power of harmonizing heterogeneous EHR data from multiple international health care systems to rapidly conduct large-scale studies to characterize how COVID-19 clinical trajectories evolve.

An atomic approach to the design and implementation of a research data warehouse

Objective

As a long-standing Clinical and Translational Science Awards (CTSA) Program hub, the University of Pittsburgh and the University of Pittsburgh Medical Center (UPMC) developed and implemented a modern research data warehouse (RDW) to efficiently provision electronic patient data for clinical and translational research.

Materials and Methods

We designed and implemented an RDW named Neptune to serve the specific needs of our CTSA. Neptune uses an atomic design where data are stored at a high level of granularity as represented in source systems. Neptune contains robust patient identity management tailored for research; integrates patient data from multiple sources, including electronic health records (EHRs), health plans, and research studies; and includes knowledge for mapping to standard terminologies.

Results

Neptune contains data for more than 5 million patients longitudinally organized as Health Insurance Portability and Accountability Act (HIPAA) Limited Data with dates and includes structured EHR data, clinical documents, health insurance claims, and research data. Neptune is used as a source for patient data for hundreds of institutional review board-approved research projects by local investigators and for national projects.

Discussion

The design of Neptune was heavily influenced by the large size of UPMC, the varied data sources, and the rich partnership between the University and the healthcare system. It includes several unique aspects, including the physical warehouse straddling the University and UPMC networks and management under an HIPAA Business Associates Agreement.

Conclusion

We describe the design and implementation of an RDW at a large academic healthcare system that uses a distinctive atomic design where data are stored at a high level of granularity.

Classification of Twitter Vaping Discourse Using BERTweet: Comparative Deep Learning Study (Preprint)

Background

Objective

Methods

Results

Conclusions

Predicting outcomes in central venous catheter salvage in pediatric central line-associated bloodstream infection

OBJECTIVE

Central line-associated bloodstream infections (CLABSIs) are a common, costly, and hazardous healthcare-associated infection in children. In children in whom continued access is critical, salvage of infected central venous catheters (CVCs) with antimicrobial lock therapy is an alternative to removal and replacement of the CVC. However, the success of CVC salvage is uncertain, and when it fails the catheter has to be removed and replaced. We describe a machine learning approach to predict individual outcomes in CVC salvage that can aid the clinician in the decision to attempt salvage.

MATERIALS AND METHODS

Over a 14-year period, 969 pediatric CLABSIs were identified in electronic health records. We used 164 potential predictors to derive 4 types of machine learning models to predict 2 failed salvage outcomes, infection recurrence and CVC removal, at 10 time points between 7 days and 1 year from infection onset.

RESULTS

The area under the receiver-operating characteristic curve varied from 0.56 to 0.83, and key predictors varied over time. The infection recurrence model performed better than the CVC removal model did.

CONCLUSIONS

Machine learning-based outcome prediction can inform clinical decision making for children. We developed and evaluated several models to predict clinically relevant outcomes in the context of CVC salvage in pediatric CLABSI and illustrate the variability of predictors over time.

Evaluation of eye tracking for a decision support application

Eye tracking is used widely to investigate attention and cognitive processes while performing tasks in electronic medical record (EMR) systems. We explored a novel application of eye tracking to collect training data for a machine learning-based clinical decision support tool that predicts which patient data are likely to be relevant for a clinical task. Specifically, we investigated in a laboratory setting the accuracy of eye tracking compared to manual annotation for inferring which patient data in the EMR are judged to be relevant by physicians. We evaluated several methods for processing gaze points that were recorded using a low-cost eye-tracking device. Our results show that eye tracking achieves accuracy and precision of 69% and 53%, respectively compared to manual annotation and are promising for machine learning. The methods for processing gaze points and scripts that we developed offer a first step in developing novel uses for eye tracking for clinical decision support.

A simple electronic medical record system designed for research

With the extensive deployment of electronic medical record (EMR) systems, EMR usability remains a significant source of frustration to clinicians. There is a significant research need for software that emulates EMR systems and enables investigators to conduct laboratory-based human–computer interaction studies. We developed an open-source software package that implements the display functions of an EMR system. The user interface emphasizes the temporal display of vital signs, medication administrations, and laboratory test results. It is well suited to support research about clinician information-seeking behaviors and adaptive user interfaces in terms of measures that include task accuracy, time to completion, and cognitive load. The Simple EMR System is freely available to the research community and is on GitHub.

Validation of an internationally derived patient severity phenotype to support COVID-19 analytics from electronic health record data

OBJECTIVE

The Consortium for Clinical Characterization of COVID-19 by EHR (4CE) is an international collaboration addressing coronavirus disease 2019 (COVID-19) with federated analyses of electronic health record (EHR) data. We sought to develop and validate a computable phenotype for COVID-19 severity.

MATERIALS AND METHODS

Twelve 4CE sites participated. First, we developed an EHR-based severity phenotype consisting of 6 code classes, and we validated it on patient hospitalization data from the 12 4CE clinical sites against the outcomes of intensive care unit (ICU) admission and/or death. We also piloted an alternative machine learning approach and compared selected predictors of severity with the 4CE phenotype at 1 site.

RESULTS

The full 4CE severity phenotype had pooled sensitivity of 0.73 and specificity 0.83 for the combined outcome of ICU admission and/or death. The sensitivity of individual code categories for acuity had high variability-up to 0.65 across sites. At one pilot site, the expert-derived phenotype had mean area under the curve of 0.903 (95% confidence interval, 0.886-0.921), compared with an area under the curve of 0.956 (95% confidence interval, 0.952-0.959) for the machine learning approach. Billing codes were poor proxies of ICU admission, with as low as 49% precision and recall compared with chart review.

DISCUSSION

We developed a severity phenotype using 6 code classes that proved resilient to coding variability across international institutions. In contrast, machine learning approaches may overfit hospital-specific orders. Manual chart review revealed discrepancies even in the gold-standard outcomes, possibly owing to heterogeneous pandemic conditions.

CONCLUSIONS

We developed an EHR-based severity phenotype for COVID-19 in hospitalized patients and validated it at 12 international sites.

Use, Impact, Weaknesses, and Advanced Features of Search Functions for Clinical Use in Electronic Health Records: A Scoping Review

Objective  Although vast amounts of patient information are captured in electronic health records (EHRs), effective clinical use of this information is challenging due to inadequate and inefficient access to it at the point of care. The purpose of this study was to conduct a scoping review of the literature on the use of EHR search functions within a single patient's record in clinical settings to characterize the current state of research on the topic and identify areas for future study.

Methods  We conducted a literature search of four databases to identify articles on within-EHR search functions or the use of EHR search function in the context of clinical tasks. After reviewing titles and abstracts and performing a full-text review of selected articles, we included 17 articles in the analysis. We qualitatively identified themes in those articles and synthesized the literature for each theme.

Results  Based on the 17 articles analyzed, we delineated four themes: (1) how clinicians use search functions, (2) impact of search functions on clinical workflow, (3) weaknesses of current search functions, and (4) advanced search features. Our review found that search functions generally facilitate patient information retrieval by clinicians and are positively received by users. However, existing search functions have weaknesses, such as yielding false negatives and false positives, which can decrease trust in the results, and requiring a high cognitive load to perform an inclusive search of a patient's record.

Conclusion  Despite the widespread adoption of EHRs, only a limited number of articles describe the use of EHR search functions in a clinical setting, despite evidence that they benefit clinician workflow and productivity. Some of the weaknesses of current search functions may be addressed by enhancing EHR search functions with collaborative filtering.

International Analysis of Electronic Health Records of Children and Youth Hospitalized With COVID-19 Infection in 6 Countries

IMPORTANCE

Additional sources of pediatric epidemiological and clinical data are needed to efficiently study COVID-19 in children and youth and inform infection prevention and clinical treatment of pediatric patients.

OBJECTIVE

To describe international hospitalization trends and key epidemiological and clinical features of children and youth with COVID-19.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cohort study included pediatric patients hospitalized between February 2 and October 10, 2020. Patient-level electronic health record (EHR) data were collected across 27 hospitals in France, Germany, Spain, Singapore, the UK, and the US. Patients younger than 21 years who tested positive for COVID-19 and were hospitalized at an institution participating in the Consortium for Clinical Characterization of COVID-19 by EHR were included in the study.

MAIN OUTCOMES AND MEASURES

Patient characteristics, clinical features, and medication use.

RESULTS

There were 347 males (52%; 95% CI, 48.5-55.3) and 324 females (48%; 95% CI, 44.4-51.3) in this study's cohort. There was a bimodal age distribution, with the greatest proportion of patients in the 0- to 2-year (199 patients [30%]) and 12- to 17-year (170 patients [25%]) age range. Trends in hospitalizations for 671 children and youth found discrete surges with variable timing across 6 countries. Data from this cohort mirrored national-level pediatric hospitalization trends for most countries with available data, with peaks in hospitalizations during the initial spring surge occurring within 23 days in the national-level and 4CE data. A total of 27 364 laboratory values for 16 laboratory tests were analyzed, with mean values indicating elevations in markers of inflammation (C-reactive protein, 83 mg/L; 95% CI, 53-112 mg/L; ferritin, 417 ng/mL; 95% CI, 228-607 ng/mL; and procalcitonin, 1.45 ng/mL; 95% CI, 0.13-2.77 ng/mL). Abnormalities in coagulation were also evident (D-dimer, 0.78 ug/mL; 95% CI, 0.35-1.21 ug/mL; and fibrinogen, 477 mg/dL; 95% CI, 385-569 mg/dL). Cardiac troponin, when checked (n = 59), was elevated (0.032 ng/mL; 95% CI, 0.000-0.080 ng/mL). Common complications included cardiac arrhythmias (15.0%; 95% CI, 8.1%-21.7%), viral pneumonia (13.3%; 95% CI, 6.5%-20.1%), and respiratory failure (10.5%; 95% CI, 5.8%-15.3%). Few children were treated with COVID-19-directed medications.

CONCLUSIONS AND RELEVANCE

This study of EHRs of children and youth hospitalized for COVID-19 in 6 countries demonstrated variability in hospitalization trends across countries and identified common complications and laboratory abnormalities in children and youth with COVID-19 infection. Large-scale informatics-based approaches to integrate and analyze data across health care systems complement methods of disease surveillance and advance understanding of epidemiological and clinical features associated with COVID-19 in children and youth.

Bayesian network models with decision tree analysis for management of childhood malaria in Malawi

BACKGROUND

Malaria is a major cause of death in children under five years old in low- and middle-income countries such as Malawi. Accurate diagnosis and management of malaria can help reduce the global burden of childhood morbidity and mortality. Trained healthcare workers in rural health centers manage malaria with limited supplies of malarial diagnostic tests and drugs for treatment. A clinical decision support system that integrates predictive models to provide an accurate prediction of malaria based on clinical features could aid healthcare workers in the judicious use of testing and treatment. We developed Bayesian network (BN) models to predict the probability of malaria from clinical features and an illustrative decision tree to model the decision to use or not use a malaria rapid diagnostic test (mRDT).

METHODS

We developed two BN models to predict malaria from a dataset of outpatient encounters of children in Malawi. The first BN model was created manually with expert knowledge, and the second model was derived using an automated method. The performance of the BN models was compared to other statistical models on a range of performance metrics at multiple thresholds. We developed a decision tree that integrates predictions with the costs of mRDT and a course of recommended treatment.

RESULTS

The manually created BN model achieved an area under the ROC curve (AUC) equal to 0.60 which was statistically significantly higher than the other models. At the optimal threshold for classification, the manual BN model had sensitivity and specificity of 0.74 and 0.42 respectively, and the automated BN model had sensitivity and specificity of 0.45 and 0.68 respectively. The balanced accuracy values were similar across all the models. Sensitivity analysis of the decision tree showed that for values of probability of malaria below 0.04 and above 0.40, the preferred decision that minimizes expected costs is not to perform mRDT.

CONCLUSION

In resource-constrained settings, judicious use of mRDT is important. Predictive models in combination with decision analysis can provide personalized guidance on when to use mRDT in the management of childhood malaria. BN models can be efficiently derived from data to support clinical decision making.