According to the 21st Century Cures Act, the United States (US) Congress defined RWE as data regarding the usage, or the potential benefits or risks, of a drug derived from sources other than traditional clinical trials. The Food & Drug Administration (FDA) has elaborated further that “real-world data (RWD) are the data relating to patient health status and/or the delivery of health care routinely collected from a variety of sources.” Through analytics, “real-world evidence (RWE) is the clinical evidence regarding the usage and potential benefits or risks of a medical product derived from analysis of RWD.”

Nowadays, artificial intelligence (AI), through machine learning (ML), deep learning (DL),digital innovation such as digital therapeutics (DTx), and decentralized clinical trials (DCTs) would be the enabler for the healthcare industry to provide optimal solutions to patients that are much faster, more precise and more efficient.

AI Priorities in Healthcare

Although RWE can help understand patients, health conditions, and healthcare resource usage beyond randomized controlled trials (RCTs), its use in a regulatory capacity is in its infancy. Those who generate evidence and those who interpret and use it in a practical sense must keep in mind the limitations of the source data and analytical approaches used. Similar to data obtained from RCTs, transparency of methodology, use of best practices, conduct of handling protected health information, are essential when using RWD for the purpose of AI.

Therefore, it is quite important to consider prioritizing through well-constructed frameworks, data quality and accessibility, and international collaborations. For regulatory purposes, in particular, early engagement with regulators will support subsequent efforts to obtain and analyze observational data. Finally, in an era of digital innovation, AImay enable extensive collection, aggregation, analyses, and interpretations to generate evidence and insights.

Furthermore, regarding the definition and applications of AI, the purpose of AI for drugs versus device regulations, the availability of quality data fed into explainable algorithms, as well as ethical conduct must be carefully considered.

Barriers When Applying AI

Potential bottlenecks exist for RWE generation for value-added purposes, e.g., supporting regulatory submissions, label expansions, value-based contracts, comparative effectiveness research, etc. For example, claims data are typically generated for insurance billing purposes, not adjudicated in terms of data quality, and medical errors can exist. Refinement, or training, of analytic algorithms is essential to improving the accuracy, quality and speed of RWD analytics.

Researchers must first ensure that the data obtained are complete and relevant to the condition, patient population, and treatment analyzed. Unstructured data such as texts may contain relevant information for only certain sub-populations or information may be entered for some patients but not others.

Even structured data pose challenges in the application of RWD analytics and AI since data may have inconsistent terms, different formats between sources, and have incomplete or messy information. These situations might lead to inaccuracy in the analyses and convergence of the algorithms.

In parts of the world, data aggregation may pose an equally significant challenge. Legal barriers around data privacy, practical barriers related to data storage across multiple organizations, and economic barriers involving lack of incentives for organizations to collaborate and share data, all affect the availability of data to which analytic tools and algorithms can be applied. Additional challenges are likely to be encountered in this fast-moving field.

It is critical to identify multi-disciplinary partnerships and talents who can be skillful in harnessing big data

It is also worth emphasizing a well-known saying, “garbage, in garbage out.” Data standards not only cover the quality aspect but also common data model applications. It is also critical to identify multi-disciplinary partnerships and talents who can be skillful in harnessing big data.

Future of AI-Driven RWE

AI can be useful in a variety of ways, e.g., process automation, medicines regulation, monitoring medication adherence, digital innovation via eHealth, mHealth and telehealth, and sophisticated algorithms. Besides usefulness in harnessing RWE, AI can play a critical role in optimizing RCTs and generating evidence through pragmatic clinical trials (PCTs).

If challenges and barriers can be overcome successfully, with large-volume data shown to provide sufficiently accurate and comprehensive evidence-generation, AI has the potential to shorten the timeline for clinical trial design and regulatory approval, and to uncover patterns in large sets of data that would otherwise not be observed.

Finally, while the use of AI to capture, amalgamates, standardize, and analyze RWD is still evolving; it has a potential to support the increased availability of data to improve global health and healthcare now and well into the future.

Disclaimer

Dr. Zou is an employee of Viatris. The views expressed are her own and do not necessarily represent those of her employer. Editorial support was not provided.