Search

Medical systems in general, and patient treatment decisions and outcomes in particular, can be affected by bias based on gender and other demographic elements. As language models are increasingly applied to medicine, there is a growing interest in building algorithmic fairness into processes impacting patient care. Much of the work addressing this question has focused on biases encoded in language models—statistical estimates of the relationships between concepts derived from distant reading of corpora. Building on this work, we investigate how differences in gender-specific word frequency distributions and language models interact with regards to bias. We identify and remove gendered language from two clinical-note datasets and describe a new debiasing procedure using BERT-based gender classifiers. We show minimal degradation in health condition classification tasks for low- to medium-levels of dataset bias removal via data augmentation. Finally, we compare the bias semantically encoded in the language models with the bias empirically observed in health records. This work outlines an interpretable approach for using data augmentation to identify and reduce biases in natural language processing pipelines.

Veterans Health Administration (VHA) services are most frequently used by patients 65 years and older, an age group that is disproportionally affected by COVID-19. Here we describe a modular Clinical Trial Informatics Solution (CTIS) that was rapidly developed and deployed to support a multi-hospital embedded pragmatic clinical trial in COVID-19 patients within the VHA. Our CTIS includes tools for patient eligibility screening, informed consent tracking, treatment randomization, EHR data transformation for reporting and interfaces for patient outcome and adverse event tracking. We hope our CTIS component descriptions and practical lessons learned will serve as a useful building block for others creating their own clinical trial tools and have made application and database code publicly available.

Objective To best meet our point-of-care research (POC-R) needs, we developed ProjectFlow, a configurable, clinical research workflow management application. In this article, we describe ProjectFlow and how it is used to manage study processes for the Diuretic Comparison Project (DCP) and the Research Precision Oncology Program (RePOP). Materials and methods The Veterans Health Administration (VHA) is the largest integrated health care system in the United States. ProjectFlow is a flexible web-based workflow management tool specifically created to facilitate conduct of our clinical research initiatives within the VHA. The application was developed using the Grails web framework and allows researchers to create custom workflows using Business Process Model and Notation. Results As of January 2021, ProjectFlow has facilitated management of study recruitment, enrollment, randomization, and drug orders for over 10 000 patients for the DCP clinical trial. It has also helped us evaluate over 3800 patients for recruitment and enroll over 370 of them into RePOP for use in data sharing partnerships and predictive analytics aimed at optimizing cancer treatment in the VHA. Discussion The POC-R study design embeds research processes within day-to-day clinical care and leverages longitudinal electronic health record (EHR) data for study recruitment, monitoring, and outcome reporting. Software that allows flexibility in study workflow creation and integrates with enterprise EHR systems is critical to the success of POC-R. Conclusions We developed a flexible web-based informatics solution called ProjectFlow that supports custom research workflow configuration and has ability to integrate data from existing VHA EHR systems.

Objective Reducing risk of coronavirus disease 2019 (COVID-19) infection among healthcare personnel requires a robust occupational health response involving multiple disciplines. We describe a flexible informatics solution to enable such coordination, and we make it available as open-source software. Materials and Methods We developed a stand-alone application that integrates data from several sources, including electronic health record data and data captured outside the electronic health record. Results The application facilitates workflows from different hospital departments, including Occupational Health and Infection Control, and has been used extensively. As of June 2020, 4629 employees and 7768 patients and have been added for tracking by the application, and the application has been accessed over 46 000 times. Discussion Data captured by the application provides both a historical and real-time view into the operational impact of COVID-19 within the hospital, enabling aggregate and patient-level reporting to support identification of new cases, contact tracing, outbreak investigations, and employee workforce management. Conclusions We have developed an open-source application that facilitates communication and workflow across multiple disciplines to manage hospital employees impacted by the COVID-19 pandemic.

Objective Reducing risk of coronavirus disease 2019 (COVID-19) infection among healthcare personnel requires a robust occupational health response involving multiple disciplines. We describe a flexible informatics solution to enable such coordination, and we make it available as open-source software. Materials and Methods We developed a stand-alone application that integrates data from several sources, including electronic health record data and data captured outside the electronic health record. Results The application facilitates workflows from different hospital departments, including Occupational Health and Infection Control, and has been used extensively. As of June 2020, 4629 employees and 7768 patients and have been added for tracking by the application, and the application has been accessed over 46 000 times. Discussion Data captured by the application provides both a historical and real-time view into the operational impact of COVID-19 within the hospital, enabling aggregate and patient-level reporting to support identification of new cases, contact tracing, outbreak investigations, and employee workforce management. Conclusions We have developed an open-source application that facilitates communication and workflow across multiple disciplines to manage hospital employees impacted by the COVID-19 pandemic.

We completed a pilot study to guide the development of the VA Research Precision Oncology Data Commons infrastructure as a collaboration platform with the greater research community. Our results using a small subset of patients from the VA's Precision Oncology Program demonstrate the feasibility of our data sharing platform to build predictive models for lung cancer survival using machine learning, as well as highlight the potential of target genome sequencing data.

Background: There is wide interest in developing prognostic models in non–small-cell lung cancer (NSCLC) due to the heterogeneity of the disease. Models developed at other healthcare institutions may not be directly applicable for patients treated at the Department of Veterans Affairs (VA). External validation of a candidate prognostic model among VA patients would be crucial before it can be implemented to aid clinical decision-making. Methods: A prognostic model for mortality developed in the Military Health System (MHS) was applied to data from the VA Precision Oncology Data Repository (VA-PODR), which is available to researchers inside and outside the VA at the Veterans Precision Oncology Data Commons (VPODC). Measures of discrimination and calibration were calculated for the MHS model. The MHS model was also refitted in VA-PODR data using the same risk factors to compare the effect of specific factors and predictive performance when the model is developed using VA data. Results: Time-dependent AUC of the MHS prognostic model was 0.788, 0.806, 0.780, and 0.779 for predicting survival at 1, 2, 3, and 5 years following diagnosis, respectively. Significant discrepancies were found between predicted and observed rates of survival, particularly for later years. When the model is refit in VA-PODR data, it achieved cross-validated AUCs of 0.739, 0.773, 0.769, and 0.807 at the same time points, and discrepancies between predicted and observed survival were reduced. Conclusions: Validation of the MHS prognostic model in VA-PODR demonstrates that its discrimination remains strong when applied to VA patients. Nevertheless, further calibration to VA data may be needed to improve its risk estimation performance. This study highlights the utility of VA-PODR and the VPODC as a national resource for developing analytic tools that are well adapted to the Veteran population.

The Department of Veterans Affairs (VA) has a strong track record providing high-quality, evidence-based care to cancer patients. In order to accelerate discoveries that will further improve care for Veterans with cancer, the VA has partnered with the Center for Translational Data Science at the University of Chicago and the Open Commons Consortium to establish a data sharing platform, the Veterans Precision Oncology Data Commons (VPODC). The VPODC makes clinical, genomic, and imaging data from the VA available to the research community at large. In this paper, we detail our motivation for data sharing, describe the VPODC, and outline our collaboration model. By transforming VA data into a national resource for research in precision oncology, the VPODC seeks to foster innovation through collaboration and resource sharing that will ultimately lead to improved care for Veterans with cancer.

Summary The Veterans Affairs Precision Oncology Data Repository (VA-PODR) is a large, nationwide repository of de-identified data on patients diagnosed with cancer at the Department of Veterans Affairs (VA). Data include longitudinal clinical data from the VA's nationwide electronic health record system and the VA Central Cancer Registry, targeted tumor sequencing data, and medical imaging data including computed tomography (CT) scans and pathology slides. A subset of the repository is available at the Genomic Data Commons (GDC) and The Cancer Imaging Archive (TCIA), and the full repository is available through the Veterans Precision Oncology Data Commons (VPODC). By releasing this de-identified dataset, we aim to advance Veterans' health care through enabling translational research on the Veteran population by a wide variety of researchers., Highlights • Presents a large-scale repository on cancer patients with EHR, imaging, and genomic data • Describes a method for de-identification of different types of health data • Enables researchers to use data science tools to apply to the data, The Bigger Picture Accelerating the speed of innovation and discoveries in health care requires the liberation of real-world data from their silos. One of the greatest challenges in the application of artificial intelligence and machine learning to health care is the validation of new algorithms beyond where they were created. We present the Veterans Affairs Precision Oncology Data Repository (VA-PODR), a large-scale repository of de-identified data on patients diagnosed with cancer at the Department of Veterans Affairs (VA). VA-PODR includes longitudinal clinical, genomic, and imaging data originating from the VA's electronic health record system, the VA Central Cancer Registry, and other sources. VA-PODR enables researchers around the world to validate their algorithms and advance cancer research and health care in general. In addition, VA-PODR enhances Veterans' health care by facilitating development of algorithms that are well tuned to the Veteran population and ready for deployment inside the VA., The Veterans Affairs Precision Oncology Data Repository (VA-PODR) is a large, nationwide repository of real-world de-identified data on patients diagnosed with cancer at the Department of Veteran's Affairs. The repository contains longitudinal clinical data from the VA's nationwide electronic health record system and the VA Central Cancer Registry, targeted tumor sequencing data, and medical imaging data. By releasing this dataset, we aim to advance Veterans' health care through enabling translational research on the Veteran population by a wide variety of researchers.

Background & Aims Interactions between commensal microbes and the immune system are tightly regulated and maintain intestinal homeostasis, but little is known about these interactions in humans. We investigated responses of human CD4+T cellsto the intestinal microbiota. We measured the abundance of T cells in circulation and intestinal tissues that respond to intestinal microbes and determined their clonal diversity. We also assessed their functional phenotypes and effects on intestinal resident cell populations, and studied alterations in microbe-reactive T cells in patients with chronic intestinal inflammation. Methods We collected samples ofperipheral bloodmononuclear cells and intestinal tissues from healthy individuals (controls, n= 13−30) andpatients with inflammatory bowel diseases(n= 119; 59 withulcerative colitisand 60 with Crohn’s disease). We used 2 independent assays (CD154 detection andcarboxy-fluorescein succinimidyl esterdilution assays) and 9 intestinal bacterial species (Escherichia coli, Lactobacillus acidophilus, Bifidobacterium animalissubsplactis,Faecalibacterium prausnitzii, Bacteroides vulgatus, Roseburia intestinalis, Ruminococcus obeum,Salmonella typhimurium,andClostridiumdifficile) to quantify, expand, and characterize microbe-reactive CD4+Tcells. We sequencedT-cell receptorVβ genes in expanded microbe-reactive T-cell lines to determine their clonal diversity. We examined the effects of microbe-reactive CD4+Tcells on intestinal stromal andepithelial celllines.Cytokines,chemokines, and gene expression patterns were measured byflow cytometryand quantitativepolymerase chain reaction. Results Circulating and gut-resident CD4+T cells from controls responded to bacteria at frequencies of 40−4000 per million for each bacterial species tested. Microbiota-reactive CD4+T cells were mainly of a memory phenotype, present in peripheral blood mononuclear cells and intestinal tissue, and had a diverse T-cell receptor Vβrepertoire. These cells were functionally heterogeneous, produced barrier-protective cytokines, and stimulated intestinal stromal and epithelial cells viainterleukin17A,interferongamma, andtumor necrosis factor. In patients with inflammatory bowel diseases, microbiota-reactive CD4+T cells were reduced in the blood compared with intestine;T-cell responsesthat we detected had an increased frequency of interleukin 17A production compared with responses of T cells from blood or intestinal tissues of controls. Conclusions In an analysis of peripheral blood mononuclear cells and intestinal tissues from patients with inflammatory bowel diseases vs controls, we found that reactivity to intestinal bacteria is a normal property of the human CD4+T-cell repertoire, and does not necessarily indicate disrupted interactions between immune cells and the commensal microbiota. T-cell responses to commensals might support intestinal homeostasis, by producing barrier-protective cytokines and providing a large pool of T cells that react topathogens.

Interactions between the host and its microbiota are of mutual benefit and promote health. Complex molecular pathways underlie this dialog, but the identity of microbe-derived molecules that mediate the mutualistic state remains elusive. Helicobacter hepaticus is a member of the mouse intestinal microbiota that is tolerated by the host. In the absence of an intact IL-10 signaling, H. hepaticus induces an IL-23-driven inflammatory response in the intestine. Here we investigate the interactions between H. hepaticus and host immune cells that may promote mutualism, and the microbe-derived molecule(s) involved. Our results show that H. hepaticus triggers early IL-10 induction in intestinal macrophages and produces a large soluble polysaccharide that activates a specific MSK/CREB-dependent anti-inflammatory and repair gene signature via the receptor TLR2. These data identify a host-bacterial interaction that promotes mutualistic mechanisms at the intestinal interface. Further understanding of this pathway may provide novel prevention and treatment strategies for inflammatory bowel disease. We thank the High-Throughput Genomics Group at the Wellcome Trust Centre for Human Genetics for the generation of the Sequencing data. F.F. was supported by Cancer Research UK (OCRC-DPhil13-FF) and N.E.I. by the Kennedy Trust (KENN 15 16 03). M.M.-L. received a fellowship from the Spanish Ministry of Education, Culture, and Sport. This work was funded by the Wellcome Trust UK (095688/Z/11/Z), an ERC grant (Advanced Grant Ares(2013)3687660), and the Fondation Louis Jeantet Sí