Best Practices in the Real-World Data Lifecycle
The academic paper – Best Practices in the Real-World Data Lifecycle, published today in PLOS Digital Health – looks at how the use of Real World Data (RWD) can be expanded and successfully utilised by health services and organisations across the world over the next decade.
Real-World Data is health data that is routinely collected from a variety of sources and has great potential for health research and the development of new drugs and treatments to tackle diseases, including cancer. The paper outlines what will be needed to ensure the successful future of RWD and touches on a multitude of important topics, from data privacy to leveraging machine learning.
The work is the result of a unique collaboration that incorporates the diverse perspectives of many leaders key to the RWD space, including pharmaceutical representatives, academic institutions, and companies that specialise in health technology and real-world evidence.
Lead authors Joe Zhang, BM Bch, of Imperial College London, and Sanjay Budhdeo, MD, MSc, of University College London, said,
“Health data generation continues to increase, alongside the proliferation of digital health solutions. It is vital to increase sustainable data aggregation that can enhance direct patient care and enable equitable population health research and artificial intelligence development. This paper brings together viewpoints from stakeholders in diverse sectors to delineate key best practices for enhancing health data resources.”
Dr Charlie Davie, DATA-CAN hub director, added,
“The use of Real-World Data in cancer research is so important in continuing to improve patient outcomes and survival rates from cancer and other deadly diseases. Our research will help to harness the power of Real-World Data in improving patient care and treatment options.”
The paper sets out a framework of seven best practices that all stakeholders—including providers, payers, health systems, and academic organisations—should consider as they look to the future of RWD. They include:
- Adhering to international data standards
- Considering and customising quality assurance depending on use case
- Incentivising careful and complete data entry to maximise data’s value
- Using natural language processing to tap into unstructured data sources
- Implementing solutions that enable flexible analytics that can be used in nearly real time
- Ensuring data privacy and patient protection, while returning value to patients Actively working toward equity and representation within RWD datasets
To read the full paper, and see a list of contributors, visit PLOS Digital Health.