Introduction
Integrating Real-World Data (RWD) and Real-World Evidence (RWE) into clinical research is gaining significant traction, due to the wide range of benefits it brings. However, integration of RWE into clinical research and trials can come with specific technical challenges such as handling unstructured data, standardizing data across multiple sources, and navigating regulatory and ethical concerns.
This blog discusses these challenges and offers solutions for overcoming these obstacles, providing practical strategies to ensure RWD/RWE can be leveraged effectively for more accurate insights and better outcomes in drug development and clinical trials.
What are the most frequent challenges in using RWE, and how can these be overcome?
Below, Andrew Elders, Senior Statistician at Phastar, discusses the issues that can be faced when working with RWE, as part of our recent webinar. Catch up on the full session here.
Effective handling of RWD often begins with robust data management and cleaning. Andrew also discussed the importance of thoughtful study design for primary data collection in real-world settings, including the development of case report forms (CRFs) tailored to answer specific research questions. Andrew elaborated on Phastar’s programming capabilities, including data engineering, algorithm development, and deriving unavailable parameters.
He outlined common challenges in working with RWD, such as unstructured formats, illogical dates, and treatment gaps. Missing data and various biases—such as selection or immortal time bias—were also highlighted as critical issues. Andrew explained that these challenges can often be mitigated through careful study design, alongside the application of statistical methods to address missing data and biases.
Below, we break down the key obstacles further and offer some practical solutions and examples for integrating RWE into clinical research and trials.
1. Ensuring Data Quality and Accuracy in RWE
Challenge: Variability and inconsistencies in data quality can arise when data is sourced from multiple places or systems, leading to discrepancies in accuracy and reliability. These differences can pose challenges in data integration and analysis, affecting the overall quality and integrity of the insights derived. [1] [2]
Solution: Employing advanced data cleaning and validation techniques to ensure data reliability. An example of this could be leveraging machine learning to identify and correct discrepancies in data, using Phastar’s data science capabilities.
2. Handling, Standardizing, and Integrating RWD
Challenge: Managing unstructured data from sources like physician notes and diagnostic images, while addressing heterogeneity in data formats, structures, and quality across diverse sources (e.g., EHRs, claims data, registries). These inconsistencies pose significant challenges for integration, standardization, and analysis.
Solution: Leveraging natural language processing (NLP) and AI tools to convert unstructured data into analyzable formats, alongside adopting standardized protocols and data models like the OMOP Common Data Model (CDM) for seamless integration and harmonization. [3]
3. Privacy, Security, and Regulatory Considerations Surrounding RWE
Challenge: Ensuring patient privacy is protected and securing sensitive health data are crucial steps in maintaining trust and compliance when utilizing RWE in clinical trials. [4] Additionally, navigating the evolving regulatory landscape and addressing uncertainties surrounding the use of RWE pose significant challenges for organizations aiming to leverage these data sources effectively.
Solution: Implementing robust measures when handling RWE such as data encryption, anonymization, and strict compliance with regulations like GDPR and HIPAA. Working with contract research organization (CRO) experts will help ensure implementation of robust data protection measures and maintain close collaboration with regulatory bodies to ensure compliance with evolving standards.
Conclusion
In conclusion, integrating RWD and RWE into clinical trials and research presents a unique set of challenges. From ensuring data quality and accuracy to managing unstructured and diverse data sources, each challenge requires tailored solutions to optimize the use of RWE in drug development.
Key strategies to overcome these obstacles include implementing advanced data cleaning and validation techniques, leveraging AI and natural language processing for handling unstructured data, and adopting standardized data models for integration. Moreover, addressing privacy, security, and regulatory concerns is crucial, necessitating robust data protection measures and close collaboration with regulatory bodies.
With the specialized expertise of CROs, RWE can be effectively utilized to complement traditional clinical trial data, paving the way for more comprehensive and real-world-applicable research findings.
References
[1] Bhatt, A. (2023). Data quality – The foundation of real-world studies. Perspectives in Clinical Research, 14(2), 92–94. https://doi.org/10.4103/picr.picr_12_23
[2] Rogers, J.R., Lee, J., Zhou, Z., Cheung, Y.K., Hripcsak, G., & Weng, C. (2020). Contemporary use of real-world data for clinical trial conduct in the United States: A scoping review. Journal of the American Medical Informatics Association, 28(1), 144–154. https://doi.org/10.1093/jamia/ocaa224
[3] Hallinan CM, Ward R, Hart GK, et al. Seamless EMR data access: Integrated governance, digital health and the OMOP-CDM. BMJ Health Care Inform. 2024;31(1): e100953. doi:10.1136/bmjhci-2023-100953
[4] Grimberg, F., Asprion, P.M., Schneider, B., Miho, E., Babrak, I., & Habbabeh, A. (2021). The Real-World Data Challenges Radar: A review on the challenges and risks regarding the use of real-world data. Digital Biomarkers, 5(2), 148–157. https://doi.org/10.1159/000516178
