2021-10-07 464
A few days ago, Yao Chen, Vice President of Hainan Real World Data Research Institute, explored an integrated solution for hospital real world data collection, governance and management in the practice of many years of clinical research, and cooperated with Boao Lecheng Clinical Research Center and Hangzhou Laimai Medical Information Technology Co., Ltd. has cooperated to develop an innovative electronic source data record (ESR, eSource Record) tool. Most recently, the U.S. FDA released a draft guidance in September 2021, “Real-World Data: Evaluating Electronic Health Records and Medical Claims Data to Support Regulatory Decisions for Drugs and Biologics.” Many of the important topics mentioned in the guidelines are covered in the ESR, which strongly demonstrates the value of the Institute's work in this area.
Vice President of Hainan Real World Data Research Institute
Professor Yao Chen
Following the publication of "Using Real-World Evidence to Support Regulatory Decisions for Medical Devices" in 2017 and the "Real-World Evidence Program Framework" in 2018, the US FDA issued a draft guidance in September 2021, "Real-World Data: Evaluating Electronic Health Records and Medical Claims Data to Support Regulatory Decision-Making of Drugs and Biologics." FDA publishes this guidance as part of its Real World Evidence (RWE) program. The draft guidance aims to provide recommendations for sponsors, researchers and other stakeholders to use electronic health record (EHR) or medical claims data in clinical research to support regulatory decisions about efficacy or safety.
The guidance focuses on three issues related to the use of RWD collected from EHR and claims data, including the selection of data sources that appropriately address the research question, the development and validation of definitions of study design elements, and the source and quality of data throughout the study life cycle .
The FDA further explained that the guidance does not address study design or type of statistical analysis, nor does it endorse any particular type of data source or research methodology. "As with all study designs, it is important to ensure the reliability and relevance of the data used to help support regulatory decisions."
FDA notes that each data source should be evaluated to determine whether the available information is appropriate to address a particular research hypothesis. Since existing electronic healthcare data was not developed to support regulatory submissions to the FDA, it is important to understand the potential limitations of its use for this purpose. For studies that need to combine data from multiple data sources or study centers, given the potential heterogeneity of population characteristics, clinical practices, and coding across different data sources, FDA recommends demonstrating whether and how data from different sources were obtained in an acceptable manner. quality is integrated.
In the Study Design Elements section, FDA mentions that research questions of interest should be identified first and then the data sources and study design best suited to address those questions. Studies should not be designed to fit a particular data source, as the limitations of a particular data source may limit study design choices and limit the inferences that can be drawn.
The FDA noted that for prospective clinical studies where EHR use is recommended, the EHR system could be modified to collect additional patient data during routine care through an add-on module to the EHR system. However, EHR-based data collection may still be incomplete due to the limited ability to add modules to collect a large amount of additional information. The ESR is designed to be added to the EMR as an add-on module based on the data collected routinely in medical care, so that it can supplement the additional data required for research and effectively improve the data quality of clinical research.
FDA recommends addressing the comprehensiveness of data sources in obtaining care and outcomes relevant to the research question. The guide discusses points to consider when examining data quality during the data lifecycle, from the accumulation and management of data to its transformation and de-identification and its eventual storage in data warehouses and research-specific datasets. generate.
A few days ago, the author has explored an integrated solution for real-world data collection, governance and management in hospitals in the practice of many years of clinical research, and cooperated with Boao Lecheng Clinical Research Center and Hangzhou Laimai Medical Information Technology Co., Ltd. A type of Electronic Source Recording ( ESR, eSource Record) tool. ESR is committed to addressing data quality throughout the life cycle and integrating multi-source data, and follows the implementation standards for clinical research. The ESR solution includes five steps: research project preparation, pre-collection of medical history, writing of in-hospital medical records, out-of-hospital follow-up and eCRF traceability. Its functions cover the whole process of clinical research, mainly including clinical medical source data collection, data extraction and management, and automatic connection between hospital electronic medical records (EMR) and clinical research electronic data capture (EDC). ESR can also manage source data from different sources uniformly, so that data from other sources such as in-hospital, out-of-hospital, and process supervision can be backed up to form a certified copy database. Using artificial intelligence technology such as natural language processing, the certified copy database is managed to form a clinical research database, and the quality of the data is controlled through multiple links to ensure the quality of the data.
In order to facilitate clinicians to write electronic medical records ( EMR) completely and efficiently in the clinical medical process, ESR tools can combine the requirements of hospital medical record writing specifications, and customize data path prompts that conform to clinical habits according to the source data collection content of different clinical research programs. Through functions such as voice input and pre-filling of medical records for medical history consultation, in addition, out-of-hospital follow-up functions such as WeChat public account can be used to conveniently collect data outside the hospital. ESR can be automatically filled into eCRF from EMR's unstructured documents using Natural Language Processing (NLP) in real time, and it also supports source data traceability viewing. ESR takes into account the diversity of source data sources, the challenges of data interoperability and data standardization . By innovatively optimizing the source data collection process of clinical research, and following the eSource concept and GCP principle design, ESR can meet the ALCOA+CCEA standards for clinical research data quality, while improving the efficiency of clinicians in writing electronic medical records. By connecting EMR and EDC, ESR can flexibly respond to the current status of medical information level, implement simpler and easier to implement and promote, and has higher standardization and sustainability.
As can be seen, the design philosophy of the ESR is consistent with the themes the FDA has emphasized in this draft guidance. At the same time, it also provides high-quality hospital electronic medical record (EMR) source data for the currently under construction Hainan Provincial Clinical Real World Data Research Platform (Phase I), which solves the problem that restricts the development of my country's imported licensed pharmaceutical equipment projects in the Lecheng International Medical Tourism Pilot Zone. Bottlenecks in world research. This also further reflects that Hainan Lecheng Real World Data Research Institute is gradually moving towards the international frontier of real world research in a way of independent innovation and courage to open up.
Click the link below to read the original and translation of the FDA guidance:
