Beyond the hype: RWD is not just another acronym
Despite the promise, there are challenges in applying the theory of RWD analysis to today’s practices in providing care and conducting research.
The use of real-world data in clinical trials — and artificial intelligence to extract and curate this data — is getting a lot of attention as a promising new technique with significant potential to change the way clinical trials are run.
But there's debate about the extent to which it's truly a transformative revolution rather than a trendy topic and buzzword.
While the application of real-world data (RWD) is tied to potential benefits like advancing clinical research, improving patient safety and accelerating treatment development, concerns still remain about data quality, biases and regulatory hurdles.
What is the real value of RWD to clinical trials, and what issues must be addressed to realize its potential?
Some background basics
RWD is health information gathered from myriad sources outside the confines of a traditional clinical trial, including electronic health records, insurance claims and wearable devices, representing data generated by patients themselves through wearables and mobile apps. Combined, these often-disparate RWD datasets capture a patient’s health history. It’s a raw, unfiltered view of how patients experience diseases, respond to treatments and interact with the healthcare system in everyday life.
With the skillful application of advanced analytics, RWD can reveal insights and generate real-world evidence (RWE). While these terms are sometimes used interchangeably, RWD and RWE are not the same; RWE is a product of analyzed RWD.
AI-driven techniques including machine learning (ML) and natural language processing (NLP) can help identify new data relationships and patterns from RWD that then become RWE.
So, why all the hype?
As advanced cloud technologies have enabled the collection, storage and analysis of petabytes of information, life sciences companies began scouring RWD, seeking RWE for a wide variety of purposes, including retrospective and prospective studies, comparative effectiveness research, health economics and outcomes research and commercialization research.
The successful application of advanced analytics to RWD has a range of potential benefits.
Accelerated trial design. RWD can help identify patient populations, inform sample size calculations and refine study endpoints.
Increased diversity in trials. RWD can help identify and include diverse patient populations in trials.
Improved patient outcomes. RWD can be used to study the effectiveness and safety of a drug in real-world settings, potentially leading to more accurate and relevant RWE.
Cost efficiency. Leveraging existing RWD reduces the need for extensive data collection, saving time and resources.
Insights into long-term effects. RWD can provide information about long-term outcomes and safety of drugs after they are approved.
Putting theory into practice
As with any innovative field, there are often challenges in applying the promising theory of RWD analysis to the reality of today’s practices. There are significant challenges to collecting, aggregating and analyzing so much disparate data in a way that can ensure privacy, minimize bias and deliver meaningfully improved outcomes. RWD can be messy, incomplete and subject to various biases because of its observational nature and variations in data entry across healthcare systems.
Today’s healthcare system generates approximately a zettabyte (a trillion gigabytes) of data annually, and this amount is doubling every two years. This data sits in multiple silos, different languages, various formats, both structured and unstructured, (such as images and clinical notes) and is often coded differently. The heterogeneity of the data is its greatest richness and also its biggest challenge.
The richness and scale of RWD enables researchers to perform deep analysis that can identify patterns that would otherwise remain obscure. This can best be accomplished by training algorithms to perform such calculations. Yet there are several risks inherent in this process.
First is the challenge of identifying and integrating all relevant factors and variables, such as treatment patterns, drug availability, disease severity, care setting and comorbidities. Much of the RWD available through medical records or claims tends to be episodic, offering only a partial picture of the health landscape. Researchers need to be sensitive to these gaps and must actively seek out alternative data sources to compensate.
RWD is also subject to selection bias, as cohort selection and treatment decisions in clinical practice are not random. For example, when using claims data, researchers may overlook that a certain segment of the population is less likely to visit the doctor or has never made a claim.
It is incumbent on researchers to think carefully about the relationships between different variables and to systematically challenge their own biases to ensure the validity of any conclusions.
Signs of progress
In recognition of these challenges the industry has been evolving. Tokenization – the replacement of personally identifiable information with unique random codes – has become a valuable solution to accurately link real-world patient data from disparate healthcare systems without compromising patient confidentiality.
At the same time, regulatory authorities have issued guidance supporting the increased adoption of insights from unstructured RWD. This amounts to an endorsement for the application of high-quality, curated datasets for therapeutic areas that are disease-indication specific. By sourcing data from various health care settings and sources, these datasets remove knowledge gaps and help researchers to better understand diverse patient populations.
The embrace of AI-driven techniques to analyze RWD has become critical in the fight to reduce costs and complexities of clinical studies. However, it is not a panacea in its own right. Rather, these advances are a catalyst to generating insights. The application of RWD remains a tool — a very powerful tool — to empower business intelligence and essential, broad-based collaboration among researchers, medical experts, data scientists and other stakeholders.
With techniques such as tokenization expanding understanding of the patient journey, the effective use of RWD does suggest a paradigm shift, a significant advance in our ability to see beyond organizational silos, allocate resources efficiently and improve proactive planning in a way that is modernizing clinical trials, streamlining communications and enabling a data-driven approach to decision-making.
This approach not only streamlines the clinical development process but also paves the way for more informed decision-making in healthcare, ultimately contributing to the advancement of patient care and therapeutic interventions.
RWD has significant potential to transform clinical trials, but it's crucial to acknowledge both the benefits and challenges associated with its use. Overcoming the hurdles related to data quality, biases and regulatory compliance will be key to realizing the full potential of RWD in accelerating medical innovation.
Sujay Jadhav is the Chief Executive Officer at Verana Health.