News / TRIbune Newsletter

What is Real-World Data?

Real-world data (RWD) describes what is happening in routine clinical practice, outside the tightly-controlled clinical trial setting.¹ The FDA defines RWD as “data relating to patient health status and/or delivery of health care routinely collected from a variety of sources,”² which can include:

Database-specific sources: electronic health records (EHR), disease registries, medical claims databases and billing activities²
Patient-generated data: patient reported outcomes (PRO)³

A variety of digital health tools exist which can be utilized to collect RWD, and to turn qualitative data (such as PRO) into quantitative data. These include applications downloaded to personal electronic devices, social media platforms, wearable devices that sense activity and physiological data, as well as biological samples that can be drawn at home.

Current Applications of Real-World Data

RWD is used regularly in safety signal detection to monitor post-marketing safety data and serious adverse events. The Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and Research (CBER) use electronic health data stored in the Sentinel System as the primary source of pharmacoepidemiological data⁴ to inform their approval and post-approval decisions.

Congress passed the 21st Century Cares Act (2016), which placed additional focus on the use of RWD in the regulatory decision-making process. This prompted the FDA to develop the Real-World Evidence Framework⁴ (2018), which provides guidance on the use of RWD to generate valid, real-world evidence (RWE) used in the regulatory approval process.

One example of the successful use of RWD to generate RWE is the ADAPTABLE study.⁵ This ground-breaking pragmatic trial compared two commonly used aspirin doses (81mg vs 325mg) by randomizing 15,076 participants with a history of myocardial infarction or atherosclerotic cardiovascular disease to one of the two doses. The participants were identified from the National Patient-Centered Clinical Research Network (PCORnet) using electronic algorithms. The trial was integrated into routine clinical care with minimal inclusion/exclusion criteria and was performed entirely remotely with the use of multimodal, low-touch recruitment strategies, electronic informed consent, and a patient-specific access code that linked the patients to health-system data. EHR and claims data were used to capture study endpoints and patient-reported outcomes.The study showed that there were no significant differences in the efficacy or safety outcomes between the treatment groups, and that long-term adherence was better in the 81 mg dose group.

How Can We Apply Real-World Data to Clinical Trial Design?

RWD collected from various sources can help to document patient healthcare journeys and identify specific points in the timeline where innovative interventions could take place.¹ In this way, RWD has the potential to be woven into analysis infrastructure to support many types of study designs, and can also be applied throughout all phases of drug development.

Application of Real World Data in Clinical Trials

Targeted Product Profiles

During the planning stage of drug development, evidence collected from RWD can help to determine burden of both disease and unmet needs by refining estimates of prevalence and incidence.⁶ This can generate study hypotheses and provide further insights into the targeted indication, ultimately determining which drugs enter Phase I of development.

Sample Size, Recruitment, and Screening

The result of one meta-analysis⁶ estimates that decreasing sample size may decrease the time to complete future randomized clinical trials (RCTs) by 6 months, representing millions of dollars in the development process. Therefore, sample-size calculations and selection criteria are two important parts of the initial stages of clinical trial planning which could be better informed by RWD. Analyzing the size of the local population that satisfies the RCT eligibility and inclusion criteria using RWD has been shown to accurately predict whether a trial could attain its accrual goals.⁷

RWD can be used to digitalize aspects of the enrollment and intervention processes, which can help with the recruitment, screening, and retention of patients. One proposed example of a digital recruitment strategy is the use of social media outreach programs for new clinical trials to raise awareness of these studies.⁸ Other examples of digital tools include electronic consent forms using visual formats to aid in the informed consent process and remote monitoring tools to reduce in-person visits.

Trial Site Selection

RWD can optimize clinical trial site selection by identifying healthcare providers (HCPs) who have recently provided care for patients with the disease of interest,⁶ allowing for quicker enrollment of patients.

Baseline Demographics

A diverse study population is more likely to provide evidence of treatment across varying populations or to aid in the detection of rare adverse events (AEs) in only a subset of patients. RWD can be used to extract baseline medical data⁹ of the study population, including current medications and comorbidities, and can also increase diversity during recruitment by providing patient demographic data.

Biomarkers and Endpoints

During the later stages of drug development, RWD can identify characteristics that are associated with important clinical outcomes or can establish the validity of surrogate biomarkers^1,4 by providing insights into how clinical presentation changes at different stages of disease progression.⁶ An example of how RWD could be very useful in tracking clinical presentation and course of disease is relapsing-remitting multiple sclerosis (RRMS). RRMS presents differently in each affected person. RWD, such as a symptom-tracking diary or a mobile app, could help shed light on an individual’s course of disease and identify a relapse before permanent damage to the central nervous system occurs.

Standards of Care and Follow-Up

RWD can provide guidance on how to conduct a trial more efficiently by pinpointing clinical standard(s) of care. Data tokenization, the linking of patient-level RWD across multiple databases, provides a more comprehensive understanding of standards of care, including the visit frequency and the selection and timing of diagnostic lab tests. RWD can also guide follow-up assessments by providing insight on how patients change over time following initiation of an intervention⁹, including medication adherence (or discontinuation), or hospitalization frequency.

Limitations of Real-World Data

There are limited examples of using RWD to augment clinical trials due to several barriers, which include:

Lack of data collection standardization
Lack of third-party oversight
Communication and cooperation across many stakeholders
Retrospective data is prone to bias

Since RWD is collected during routine medical care and not specifically for research purposes, it often lacks standardization and quality controls compared to data that are collected during clinical trials, which are conducted according to Good Clinal Practice (GCP). In other words, the collection of RWD may not be standardized across institutions and data elements may be missing. This is especially true in diseases that lack clinical practice guidelines or with practice guidelines that frequently change (i.e., COVID-19). Also, the databases themselves may lack standardization. For example, there may be differences in the units of measure reported in EHRs or the frequency of visits. Some indications and interventions cannot be easily ascertained due to lack of qualifiers (i.e., indications that do not contain an ICD-9 or ICD-10 code) or lack of routine use in clinical practice (i.e., new drugs that have not yet secured FDA approval).¹⁰

Overcoming the Barriers: Moving Toward a New Future of Pragmatic Trial Design

The FDA has accepted the use of RWD to support drug product approvals, especially in the settings of oncology and rare disease, however it does not currently endorse any specific type or source of RWD.⁶ Historically, the agency has shown a preference for registry data and RWD that is collected prospectively because studies using these sources are more likely to be “clinical trial-like”.¹¹ The scientific community must develop means of ensuring that RWD used for regulatory purposes can be analyzed for quality, relevance, and reliability,¹¹similar to data collected from RCTs.

Recently (October 2022), the FDA announced the establishment of its new Advancing Real-World Evidence Program, which will allow Sponsors to meet with staff at CDER and CBER before protocol development or study initiation to discuss the use of RWE in medical product development. The aim is to improve the quality and acceptability of RWE-based approaches that can be used to meet the regulatory requirements of product labelling or post-approval obligations.

In general, a culture shift in research/academia and investments in innovative digital health technologies will be needed³ to fully embrace the possibilities of RWD.

About the Author

Erin DiMuzio, PharmD is a pharmacovigilance specialist with experience in processing clinical trial and post-marketing serious adverse event data in the areas of oncology, immunology, and hepatology. She earned her Doctor of Pharmacy degree from the University of Cincinnati and was a retail pharmacist for 8 years before earning a graduate certificate in medical writing and making the switch to PV.

Footnotes

Stroebel, H. (2022, March 29). How can I use real-world data in clinical trials? Clinical Leader. Retrieved August 26, 2022, from https://www.clinicalleader.com/doc/how-can-i-use-real-world-data-in-clinical-trials-0001#:~:text=Real-world data describes what,a variety of multiple sources.

U.S. Food and Drug Administration. (2022, May). Real-world evidence. Retrieved August 25, 2022,from https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence.

Stern, A.D., Bronneke, J., Debatin, J.F., Hagen, J., Matthies, H., Patel, S., et al. (2022, March). Advancing digital health applications: priorities for innovation in real-world evidence generation. Lancet Digital Health, 4, e200-206.

U.S. Food and Drug Administration. (2018, December). Framework for FDA’s Real-World Evidence Program. Silver Spring, MD: Author.

Jones, W.S., Mulder, H., Wruck, L.M., Pencina, M.J., et al. (2021, May). Comparative effectiveness of aspirin dosing in cardiovascular disease. NEJM, 384:1981-1990. doi:10.1056/NEJMoa2102137.

Dagenais, S., Russo, L., Madsen, A., Webster, J., & Becnel, L. (2022, January). Use of real-world evidence to drive drug development strategy and inform clinical trial design. Clinical Pharmacology & Therapeutics, 111(1), 77-89. doi:10.1002/cpt.2480.

Topaloglu, U., & Palchuk, M. (2018, February). Using federated network of real-world data to optimize clinical trials operations. Clinical Cancer Informatics, 1-10. doi:10.1200/CCI.17.00067.

Inan, O.T., Tenaerts, P., Prindiville, S.A. et al. (2020). Digitizing clinical trials. npj Digit. Med. 3(101). doi.org/10.1038/s41746-020-0302.

Rogers, J.R., Lee, J., Zhou, Z., Cheung, Y.K., Hripcsak, G., & Weng, C. (2021). 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. doi:10.1093/jamia/ocaa224.

Bartlett, V.L., Dhruva, S.S., Shah, N.D., Ryan, P., & Ross, J.S. (2019). Feasibility of using real-world data to replicate clinical trial evidence. JAMA Network Open, 2(10). doi:10.1001/jamanetworkopen.2019.12869.

Duggan, K.Z., & Gu, S. (2022, October 19). Real world evidence (RWE) in medical product submissions [Webinar]. FDAnews.