OHDSI

Polites: A Tool for the Automation of OHDSI Implementations #OHDSISocialShowcase #JoinTheJourney Lead: John Gresh Co-Author: Julia Skapik Background: Polites is a Java based tool that enables the automation of complete OHDSI implementations for either standalone (non-Docker) or Broadsea based OHDSI deployments. Broadsea provides a turn-key solution with unparalleled ease of use and a standardized stable production deployment that includes an instance of a Common Data Model replete with test data in PostgreSql. “Broadsea 3.0 provides a flexible approach to deploying OHDSI tools that are typically challenging to set up, and establishes a framework for supporting new OHDSI tools to come. Any site, regardless of size, can deploy a wide range of OHDSI tools on a laptop or a production server” using Broadsea. However, after initial deployment there are customizations that most implementations will want to make to create a production, test, or development instance of an OHDSI implementation. This includes the creation of an independent instance of the CDM that could be in the existing Broadsea instance of PostgreSql or in any of the other data management systems supported by OHDSI (including Oracle, Microsoft Sql Server, PostgreSql, Databricks, etc.). Creation of this environment entails numerous steps that need to integrated and executed flawlessly. This process can be time consuming and fraught with errors and includes the creation of multiple databases and/or schemas and other database objects, creation of multiple users with detailed specific privileges, the creation of meta data such as the CDM source record and webapi records, the importing of vocabulary data, the creation of sequences for primary keys for ETL processes, the creation of indexes and constraints, the importing of data, and the running of other processes such as Achilles. Polites provides a way to execute all of these processes and an interface that allows the processes to be selected individually or run in groups. https://lnkd.in/eQF2EuAe

We are THREE days away from the submission deadline for #OHDSIEurope2025! The 2025 Europe Symposium will be held July 5-7 in Hasselt, Belgium, and we want to see ?? there. FIRST we need to see your poster/demo abstract by Monday, March 31! Please use the submission link below and then join our OHDSI Belgium team at the "Old Prison" for our annual Europe Symposium! #JoinTheJourney Submission Form ?? https://lnkd.in/ecdGzxqT

A Oneshot and Lossless Federated Generalized Linear Mixed Effect Model #OHDSISocialShowcase #JoinTheJourney Lead/Presenter: Jiayi (Jessie) Tong Team: Jenna Reps, Juan Manuel Ramírez-Anguita, Milou Brand, PhD, Scott DuVall, Thomas Falconer, Alex Mayer Fuentes, Xing He, Miguel-Angel Mayer, Marc Suchard, Ross Williams, Jiang Bian, David Asch, Yong Chen Watch the presentation below ... you can learn more here: https://lnkd.in/e8tSfJCg. https://lnkd.in/e-CVqAcY

An Augmented Method for Empirical p-value Calibration in Observational Studies #OHDSISocialShowcase #JoinTheJourney Lead: Huiyuan Wang Team: Yuru Zhu, Martijn Schuemie, Yuqing Lei, Yong Chen Background: ??-values are ubiquitously utilized to make statistical assertions, particularly in evaluating the effects of medical products. However, their validity in observational studies is often compromised due to unmeasured confounding or biased selection. Existing methods for calibrating ??-values using multiple negative control outcomes (NCOs) fail to account for internal correlation structures, leading to potentially reduced power. https://lnkd.in/e7vV4suu

We are 5?? days away from the submission deadline for #OHDSIEurope2025! The 2025 Europe Symposium will showcase the amazing research happening within our community on July 7 in Hasselt, Belgium. We want to see your poster or software demo there, but FIRST we need to see your submission by Monday, March 31! Please send in your abstract before the deadline and then join our OHDSI Belgium team at the "Old Prison" for our annual Europe Symposium! #JoinTheJourney Submission Form ?? https://lnkd.in/ecdGzxqT

????????? ???????? ????????! ?? The abstract submission deadline for the ?????????? ???????????? ???????? ?????????????????? is approaching ?????????????? — the platform closes on ?????????? ????! We’re excited to welcome many poster exhibitors on July 7 in Hasselt, Belgium – don’t miss your chance to showcase your work and connect with the OHDSI community! Submit your abstract today and be part of a growing movement to turn health data into real-world impact! ?? ???????????????? ???????????????????? ????????: https://lnkd.in/ecdGzxqT ?? More info on our ?????????? ??????????????: https://lnkd.in/e_HpX86h #OHDSIEurope25 #DataSavesLives #JoinTheNetwork

A Systematic and Sustainable Solution for Assessing Network Data Quality #OHDSISocialShowcase #JoinTheJourney Lead: Kimberley Dickinson Team: Kaleigh Wieand, Charles Bailey, Hanieh Razzaghi Background: Performing research on multi-institutional Electronic Health Record (EHR) data can provide insight on a large, diverse patient population, but the quality of the research is highly dependent on the quality of the underlying data. Issues within an institution and discrepancies between institutions can be difficult to uncover and can lead to unexpected or biased results when an analysis is applied uniformly across a patient population. Many networks have developed approaches to assessing DQ, but they are often designed to address specific network concerns. In PEDSnet, a pediatric learning health system with a centralized database used for multidisciplinary healthcare research, we have developed a modular program that augments the capabilities of existing tools such as OHDSI’s Data Quality Dashboard (DQD) by not only visualizing findings but incorporating the DQ check process into a larger system of thresholding, communication, and resolution. The process has proven beneficial to our research, but is also designed for reproducibility and scalability. Our DQ program incorporates input from experts who transform local EHR data into an OMOP CDM and data scientists at the Data Coordinating Center (DCC) engaged in scientific research to develop and build checks and to detect and resolve issues. https://lnkd.in/eZ-nKUAp

Asieh Golozar, leader of the OHDSI #Oncology Workgroup, is spearheading one of our 14 guideline-driven evidence studies, focusing on bladder cancer. Join our March 25 community call (11 am ET) to explore her work with oncology data partners, emphasizing data characterization and fitness for use. Everybody is invited. Please use the link below to join this call. #JoinTheJourney #Data #RWD https://lnkd.in/esAm4vXf

A Collaborative Analytic Enclave for the Metabolic Dysregulation and Obesity Cancer Risk Program (MeDOC) Consortium: Extensions of the OMOP Common Data Model for Translational Research?#OHDSISocialShowcase #JoinTheJourney Lead: Madhan Subramanian Team: Nisha Grover, MPH, Maddie Wheeler, Marinella Temprosa Background: The MeDOC (Metabolic Dysregulation and Obesity Cancer Risk Program) Consortium, established in 2022, aims to advance our understanding of the underlying mechanisms that connect obesity, metabolic dysregulation, and cancer risk through individual and collaborative projects. The six-member consortium, sponsored by the National Cancer Institute (NCI), collects data for both human and pre-clinical translational research. This research focuses on the hallmarks of metabolic dysregulation encompassing hormones, microbiome, inflammation, immunity, glycemia, insulinemia, adipokines and lipids. In order to integrate data across all projects and provide an infrastructure to conduct collaborative analysis, we adopted the OMOP CDM and developed extensions to support the diverse data and biospecimens from both published and stored human and animal studies. Advances in biomedical research are galvanized by data-driven discoveries for which data mapping, harmonization, and documentation of study results, using FAIR principles, are essential components. Through these initiatives, MeDOC aims to enhance the standardization and interoperability of biomedical data, facilitating more efficient and effective research into the links between obesity, metabolic dysregulation, and cancer risk by creating tools for the opensource community. https://lnkd.in/edBnmEW4

Visual Acuity: A Case Study for a Complex Clinical Concept #OHDSISocialShowcase #JoinTheJourney Lead: Michelle Hribar Team: Robert Gale, Will Halfpenny, Brian Toy, Eric Brown, Sally Baxter, Kerry Goetz, OHDSI Eye Care & Vision Research WG Background: Visual acuity is one of the most important clinical measurements and outcomes in eye care and vision research, but it has several complexities that make it challenging to use in a standardized, observational health database. Visual acuity can be measured in multiple different ways, such as with corrective lenses and without, at distance or near, with the left eye, right eye, or both eyes, to name a few. With all these modifiers, creating fully pre-coordinated concepts would require thousands of concepts, which is not sustainable to maintain over time. Further, visual acuity can be measured using different methodologies and charts, resulting in different values and data types, such as “20/20”, “28 letters”, “hand motion”, “fix and follow”. Finally, visual acuity is entered into free text fields in electronic health records (EHRs), which results in a huge variety of data entries, many of them considered to be non-standard. https://lnkd.in/guQaNAdz