TEF Health Education: Main title of the Knowledge Capsule

TEF Health Education: Main title of the Knowledge Capsule.

CHU GRENOBLE. [image] LTÉ. LABORATOIRE NATIONAL DE MÉTROLOGIE ET D' ESSAIS.

[Audio] Welcome to this exploration of Clinical Data Warehouses, or C-D-W-s-. In today's healthcare landscape, C-D-Ws are increasingly recognized as essential resources for driving research, innovation, and improved patient care. This presentation will provide you with a foundational understanding of C-D-W-s-, their capabilities, and the unique opportunities they present for industry partners like yourselves..

[Audio] Our journey today will begin by defining C-D-Ws and exploring the valuable insights that can be derived from the real-world data they contain. We will then delve into the various research and innovation applications of C-D-W-s-, highlighting the opportunities for collaboration between hospitals and industry partners. Finally, we will emphasize the importance of responsible data governance and the ethical considerations that are paramount when working with sensitive patient information..

[Audio] A Clinical Data Warehouse, or C-D-W--, serves a critical purpose in modern healthcare: enabling the secondary use of real-world patient data. This means that the information collected during routine clinical care can be repurposed for valuable research, quality improvement, and innovation initiatives. As you can see in this diagram, data from various hospital systems is extracted, transformed into a consistent and standardized format, and then loaded into the C-D-W--. This central repository of organized patient data, along with its comprehensive data catalogue, enables researchers to efficiently access and analyze information to answer important clinical questions, develop new therapies, and improve healthcare delivery..

[Audio] Before we delve into the specifics of C-D-W-s-, let's understand how they fit into a larger vision for healthcare—the Learning Health Information System, or L-H-I-S-. This system represents a continuous cycle where data generated from patient care is used to learn, improve, and refine treatments and healthcare delivery. The C-D-W is a crucial component of the L-H-I-S-, acting as the engine that drives this cycle. It allows us to analyze real-world data, identify trends, and generate knowledge that can be translated back into better care for patients..

[Audio] As we discuss C-D-W-s-, it's important to distinguish between two common data management approaches: data warehouses and data lakes. This table highlights their key differences. Data warehouses, with their structured nature and optimized query capabilities, are excellent for analysis and reporting. Data lakes, on the other hand, are more flexible and scalable, accommodating a wider range of data types, making them well-suited for data exploration and machine learning applications. Understanding these distinctions is crucial in the evolving landscape of healthcare data management..

[Audio] Modern data pipelines often utilize the Extract, Load, Transform (E-L-T--) approach to efficiently move data into C-D-W-s-. This process begins by extracting data from various hospital systems, such as electronic health records, laboratory systems, and administrative databases. The data is then loaded into the C-D-W in its raw, untransformed format. This allows for greater flexibility and scalability, as the data can be transformed and structured for specific analysis purposes within the C-D-W environment. It's important to distinguish E-L-T from the more traditional E-T-L (Extract, Transform, Load) approach. While E-T-L transforms data before loading it into the target system, E-L-T loads the raw data first and then performs transformations. E-L-T is often favored for C-D-Ws and data lakes, as it allows for more agility and scalability when handling large volumes of data..

[Audio] Clinical Data Warehouses are emerging as powerful tools for advancing epidemic surveillance and preparedness. The research presented in these publications demonstrates how C-D-Ws can be leveraged to address key limitations of traditional surveillance systems. Firstly, C-D-Ws provide a continuous stream of real-time, patient-level data, offering a more dynamic and granular view of disease trends compared to traditional methods, which often rely on delayed reporting. This real-time data capture is essential for early detection of potential outbreaks. Secondly, machine learning algorithms, when trained on comprehensive C-D-W datasets, can effectively predict epidemic activity. By integrating C-D-W data with external sources, such as internet search trends or social media data, we can further enhance the accuracy and predictive power of these models. These data-driven approaches are crucial for bridging the temporal gaps inherent in traditional surveillance systems, enabling a more rapid and proactive response to emerging health threats. The insights derived from CDW-based forecasting models can inform public health interventions, helping to optimize the timing and targeting of measures like vaccination campaigns or the allocation of healthcare resources. In conclusion, C-D-Ws hold significant potential to revolutionize epidemic surveillance and preparedness, paving the way for a more data-driven and proactive approach to safeguarding public health..

[image] Comparison of Unplanned 30-Day Readmission Prediction Models, Based on Hospital Warehouse and Demographic Data DHAL L UIN , BANN A V LEMORDANT.

[Audio] Voiceover: Clinical Data Warehouses are driving progress in personalized medicine by providing the data and analytical tools needed to tailor treatments to individual patients. C-D-Ws allow us to identify distinct patient subgroups based on factors like disease progression and genetic variations, revealing how these subgroups may respond differently to therapies. Machine learning algorithms, trained on C-D-W data, can then predict how individual patients will respond to specific treatments, guiding clinicians in selecting the most effective options. C-D-Ws also enable personalized drug dosage adjustments based on individual patient characteristics, optimizing efficacy and minimizing the risk of adverse effects. These advancements are paving the way for a future where treatments are increasingly tailored to each patient's unique needs and characteristics..

[Audio] Clinical Data Warehouses are not only valuable for evaluating care delivery but also for advancing our understanding of treatment effectiveness and safety. The research highlighted in these publications demonstrates how C-D-Ws can contribute to several key areas: Pharmacovigilance: By analyzing large-scale, real-world data on drug use and patient outcomes, C-D-Ws provide a powerful tool for monitoring drug safety and detecting potential adverse events that may not have been apparent in clinical trials. This information is crucial for informing safer prescribing practices and protecting patients from harm. Dosage Optimization: Determining the optimal drug dosage for individual patients is a complex challenge. C-D-Ws can help address this by analyzing real-world data on drug concentrations, patient characteristics, and treatment responses. This allows us to refine dosage guidelines, personalize treatment regimens, and improve the balance between efficacy and safety. Treatment Guideline Evaluation: When new treatment guidelines are introduced, it's essential to assess their impact on real-world clinical practice and patient outcomes. C-D-Ws provide a platform for conducting these evaluations, allowing us to determine whether guidelines are being followed, identify areas for improvement, and optimize treatment strategies..

[Audio] Beyond their applications in research and treatment evaluation, Clinical Data Warehouses are increasingly used to support a range of other critical functions within healthcare: Feasibility Studies and Patient Prescreening: C-D-Ws can streamline clinical trial recruitment by enabling researchers to rapidly identify potential participants who meet specific eligibility criteria. This prescreening process can significantly reduce the time and costs associated with finding suitable patients for clinical trials. eCRF and Cohort Database Population: C-D-Ws can efficiently populate electronic case report forms (eCRFs) and cohort databases for observational studies and registries. By extracting relevant patient data directly from the C-D-W--, we can reduce manual data entry, improve accuracy, and accelerate the research process. Optimizing Patient Coding: Accurate coding of patient diagnoses and procedures is essential for appropriate reimbursement and resource allocation within hospitals. C-D-Ws can be used to develop algorithms that automatically assign or validate I-C-D codes, improving coding accuracy, optimizing hospital financing, and ensuring resources are directed where they are most needed. These examples highlight the versatility and expanding role of C-D-Ws in healthcare. They are becoming essential tools for improving efficiency, supporting research, and optimizing resource allocation across a wide range of healthcare operations..

[Audio] What sets C-D-Ws apart is their ability to capture real-world data a rich source of information that reflects the complexities and nuances of actual clinical practice. Unlike traditional clinical trials, which are often limited in scope and duration, C-D-Ws provide a longitudinal perspective on patient care, allowing us to track patient journeys, treatments, and outcomes over extended periods. This data also encompasses a wider range of patients, capturing the diversity of the real-world population and providing insights that are more generalizable to everyday healthcare settings. As you can see from this table, both clinical trial data and real-world C-D-W data have their strengths. While clinical trials provide high internal validity due to their controlled environments, C-D-Ws offer high external validity, giving us a more accurate picture of how treatments perform in the real world..

[Audio] As we've seen, C-D-Ws are powerful tools for research and innovation, but they also contain sensitive patient information. Therefore, robust data governance frameworks are essential to ensure the responsible and ethical use of this data. Hospitals typically establish dedicated teams, often called Clinical Data Centers (CDCs), to oversee C-D-W management, data security, and access control. Steering Committees provide oversight and guidance on data use policies, while Scientific and Ethics Committees rigorously review research proposals to ensure they are ethically sound, scientifically valid, and protect patient privacy..

[Audio] To facilitate research while safeguarding patient privacy, hospitals often utilize datamarts secure subsets of data extracted from the main C-D-W and tailored to specific research areas. These datamarts contain only the data elements necessary for approved research projects. They are housed in secure environments with strict access controls, ensuring that researchers only access information relevant to their approved studies. This approach balances the need for data accessibility with the paramount importance of protecting patient confidentiality..

[Audio] Clinical Data Warehouses offer a valuable platform for collaboration between hospitals and industry partners. By working together, we can leverage the power of real-world data to advance healthcare in meaningful ways. This collaboration can take various forms, including joint research projects, product development initiatives, and the generation of real-world evidence to support regulatory submissions and demonstrate the value of new therapies..

[Audio] The future of C-D-Ws is full of exciting possibilities, but it's also important to acknowledge the challenges that lie ahead. While connecting C-D-Ws with diverse data sources offers a more holistic view of patient health, achieving seamless interoperability across different systems is a significant hurdle. Data standardization initiatives, like the Observational Medical Outcomes Partnership (O-M-O-P-) common data model, are emerging to address this challenge. Similarly, while the application of advanced analytics like (A-I ) and machine learning holds great promise, the reliability of these methods depends heavily on the quality of the underlying data. Robust data quality assessment and validation procedures are crucial to ensure the accuracy and trustworthiness of the results. And as we expand collaborative research networks, we must remain vigilant about protecting patient privacy, ensuring data security, and harmonizing data across different institutions and healthcare systems..

[Audio] The deployment of C-D-Ws is not happening in isolation. Both in France and across Europe, we are seeing a convergence towards a more interconnected, data-driven healthcare ecosystem. In France, a national strategy is underway to equip all university hospitals with C-D-W-s-, connect them through regional hubs, and support national-level initiatives like the Health Data Hub. At the European level, the European Health Data Space is being developed to create a secure and interoperable framework for sharing health data across borders. In this evolving landscape, CDWs play a critical role as the foundation for data collection, harmonization, and sharing. And it is the Clinical Data Centers (CDCs), with their expertise in data management, analysis, and ethical considerations, that bridge the gap between clinical practice, research, and data science, ensuring that C-D-W data is used responsibly and effectively to advance healthcare..

[Audio] As we've seen throughout this presentation, Clinical Data Warehouses hold immense potential to transform healthcare. They provide a unique window into the real world of patient care, enabling us to generate valuable insights, drive research and innovation, and ultimately, deliver better treatments and improved outcomes for patients. However, realizing the full potential of C-D-Ws requires a commitment to responsible data governance, ethical considerations, and strong collaborative partnerships. By working together, hospitals and industry partners can leverage the power of real-world data to address pressing healthcare challenges, accelerate scientific discovery, and create a brighter, healthier future for all..

© TEF-Health Consortium. Bibliography : 1. Bourg C, Le Tallec E, Curtis E, Lee C, Bouzille G, Oger E, et al. Heterogeneity of right ventricular echocardiographic parameters in systemic lupus erythematosus among four clinical subgroups, as stratified by clinical organ involvement in observational cohort. Open Heart. 2024 May 3;11(1):e002615. 2. Pardo I, Pierre-Jean M, Bouzillé G, Fauchon H, Corvol A, Prud’homm J, et al. Safety of subcutaneous versus intravenous ceftriaxone administration in older patients: A retrospective study. J Am Geriatr Soc. 2024 Apr;72(4):1060–9. 3. Pierre-Jean M, Marut B, Curtis E, Galli E, Cuggia M, Bouzillé G, et al. Phenotyping of heart failure with preserved ejection faction using electronic health records and echocardiography. Eur Heart J Open. 2024 Jan;4(1):oead133. 4. Mauguen C, Maruani A, Barbarot S, Abasq C, Martin L, Herbert J, et al. Factors associated with early relapse of infantile haemangioma in children treated for at least six months with oral propranolol: A case-control study using the 2014-2021 French Ouest DataHub. Ann Dermatol Venereol. 2023 Sep;150(3):189–94. 5. Pierre-Jean M, Bouzille G, Bories M, Le Corre P, Cuggia M. Leveraging Clinical Data Warehouses to Measure Impact of Update Prescription Guidelines of Polyvalent Immunoglobulins in 2018 in France. Stud Health Technol Inform. 2023 May 18;302:342–3. 6. Lalanne S, Bouzillé G, Tron C, Revest M, Polard E, Bellissant E, et al. Amoxicillin-Induced Neurotoxicity: Contribution of a Healthcare Data Warehouse to the Determination of a Toxic Concentration Threshold. Antibiotics (Basel). 2023 Mar 30;12(4):680. 7. Poirier C, Bouzillé G, Bertaud V, Cuggia M, Santillana M, Lavenu A. Gastroenteritis Forecasting Assessing the Use of Web and Electronic Health Record Data With a Linear and a Nonlinear Approach: Comparison Study. JMIR Public Health Surveill. 2023 Jan 31;9:e34982. 8. Guardiolle V, Bazoge A, Morin E, Daille B, Toublant D, Bouzillé G, et al. Linking Biomedical Data Warehouse Records With the National Mortality Database in France: Large-scale Matching Algorithm. JMIR Med Inform. 2022 Nov 1;10(11):e36711. 9. Fournier D, Jouneau S, Bouzillé G, Polard E, Osmont MN, Scailteux LM. Real-world safety profiles of pirfenidone and nintedanib in idiopathic pulmonary fibrosis patients. Pulm Pharmacol Ther. 2022 Oct;76:102149. 10. Bories M, Bouzillé G, Cuggia M, Le Corre P. Drug-Drug Interactions with Oral Anticoagulants as Potentially Inappropriate Medications: Prevalence and Outcomes in Elderly Patients in Primary Care and Hospital Settings. Pharmaceutics. 2022 Jul 5;14(7):1410. 11. Pierre-Jean M, Donal E, Cuggia M, Bouzillé G. Phenotyping of Heart Failure with Preserved Ejection Faction Using Health Electronic Records and Echocardiography. Stud Health Technol Inform. 2022 May 25;294:116–8. 12. Duclos C, Griffon N, Daniel C, Bouzillé G, Delamarre D, Darmoni S, et al. Reliability of Drug-Drug Interaction Measurement on Real-Word Data: The ReMIAMes Project. Stud Health Technol Inform. 2022 May 25;294:151–2. 13. Bannay A, Bories M, Le Corre P, Riou C, Lemordant P, Van Hille P, et al. Leveraging National Claims and Hospital Big Data: Cohort Study on a Statin-Drug Interaction Use Case. JMIR Med Inform. 2021 Dec 13;9(12):e29286. 14. Bérar A, Bouzillé G, Jego P, Allain JS. A descriptive, retrospective case series of patients with factitious disorder imposed on self. BMC Psychiatry. 2021 Nov 23;21(1):588. 15. Dhalluin T, Ansoborlo M, Rosset P, Thomazeau H, Cuggia M, Guillon L. Pilot Study of an e-Cohort to Monitor Adverse Event for Patient with Hip Prostheses from Clinical Data Warehouse. Stud Health Technol Inform. 2021 Nov 18;287:45–9. 16. Dhalluin T, Fakhiri S, Bouzillé G, Herbert J, Rosset P, Cuggia M, et al. Role of real-world digital data for orthopedic implant automated surveillance: a systematic review. Expert Rev Med Devices. 2021 Aug;18(8):799–810. 17. Duthe JC, Bouzille G, Sylvestre E, Chazard E, Arvieux C, Cuggia M. How to Identify Potential Candidates for HIV Pre-Exposure Prophylaxis: An AI Algorithm Reusing Real-World Hospital Data. Stud Health Technol Inform. 2021 May 27;281:714–8. 18. Ansoborlo M, Dhalluin T, Gaborit C, Cuggia M, Grammatico-Guilllon L. Prescreening in Oncology Using Data Sciences: The PreScIOUS Study. Stud Health Technol Inform. 2021 May 27;281:123–7. 19. Gangloff C, Rafi S, Bouzillé G, Soulat L, Cuggia M. Machine learning is the key to diagnose COVID-19: a proof-of-concept study. Sci Rep. 2021 Mar 30;11(1):7166. 20. Bories M, Bouzillé G, Cuggia M, Le Corre P. Drug-Drug Interactions in Elderly Patients with Potentially Inappropriate Medications in Primary Care, Nursing Home and Hospital Settings: A Systematic Review and a Preliminary Study. Pharmaceutics. 2021 Feb 16;13(2):266. 21. Poirier C, Hswen Y, Bouzillé G, Cuggia M, Lavenu A, Brownstein JS, et al. Influenza forecasting for French regions combining EHR, web and climatic data sources with a machine learning ensemble approach. PLoS One. 2021;16(5):e0250890. 22. Dhalluin T, Bannay A, Lemordant P, Sylvestre E, Chazard E, Cuggia M, et al. Comparison of Unplanned 30-Day Readmission Prediction Models, Based on Hospital Warehouse and Demographic Data. Stud Health Technol Inform. 2020 Jun 16;270:547–51. 23. Dalloux C, Claveau V, Cuggia M, Bouzillé G, Grabar N. Supervised Learning for the ICD-10 Coding of French Clinical Narratives. Stud Health Technol Inform. 2020 Jun 16;270:427–31. 24. Madec J, Bouzillé G, Riou C, Van Hille P, Merour C, Artigny ML, et al. eHOP Clinical Data Warehouse: From a Prototype to the Creation of an Inter-Regional Clinical Data Centers Network. Stud Health Technol Inform. 2019 Aug 21;264:1536–7. 25. Cuggia M, Combes S. The French Health Data Hub and the German Medical Informatics Initiatives: Two National Projects to Promote Data Sharing in Healthcare. Yearb Med Inform. 2019 Aug;28(1):195–202. 26. Poirier C, Lavenu A, Bertaud V, Campillo-Gimenez B, Chazard E, Cuggia M, et al. Real Time Influenza Monitoring Using Hospital Big Data in Combination with Machine Learning Methods: Comparison Study. JMIR Public Health Surveill. 2018 Dec 21;4(4):e11361. 27. Bouzillé G, Poirier C, Campillo-Gimenez B, Aubert ML, Chabot M, Chazard E, et al. Leveraging hospital big data to monitor flu epidemics. Comput Methods Programs Biomed. 2018 Feb;154:153–60. 28. Bouzille G, Jouhet V, Turlin B, Clement B, Desille M, Riou C, et al. Integrating Biobank Data into a Clinical Data Research Network: The IBCB Project. Stud Health Technol Inform. 2018;247:16–20. 29. Bouzillé G, Westerlynck R, Defossez G, Bouslimi D, Bayat S, Riou C, et al. Sharing Health Big Data for Research - A Design by Use Cases: The INSHARE Platform Approach. Stud Health Technol Inform. 2017;245:303–7..