FDA Real-World Evidence: What Does It Really Mean and How Does It Work?

Manufacturers want the least burdensome path to market, and Real-World Data (RWD)/Real-World Evidence (RWE) may be worth a strategic assessment. Real-World Evidence is “the clinical evidence about the usage and potential benefits or risks of a medical product derived from analysis of RWD”.  RWD is defined as “data relating to patient health status and/or the delivery of health care routinely collected from a variety of sources”.

FDA Real-World Data

Sources of FDA Real-World Data

Sources of real-world data are generally available outside of the structure of a formal clinical trial. The FDA considers RWD sources to include:

• Electronic Health Records (EHR)
• Medical claims and billing data
• Data from product and disease registries (e.g., the US Scientific Registry of Transplant Recipients (SRTR))
• Other sources (e.g., digital health technologies) that inform on health status
• Device generated data
• Public health surveillance
• Clinically annotated biobanks
• Medical device data repositories (e.g., imaging, EKG databases)
• Social Security Administration’s Death Master File

Uses for FDA Real-World Data

The FDA identifies multiple uses for RWD, including the following:

• Generating hypotheses to be tested in a clinical study
• Constructing performance goals that can be:
  • Used to inform the prior for a Bayesian analysis of a clinical trial
  • A part of a hierarchical model or a hybrid data synthesis
  • A concurrent control group as a mechanism for collecting data when a registry, electronic health record (EHR), claims data or other systemic data collection exists
• Generating primary clinical evidence to support a marketing application in its entirety
• Evidence to support interpretability of the primary clinical evidence (e.g., OUS population is representative of US population)
• Conducting post-market surveillance
• Gathering post-market data in lieu of pre-market data

FDA Real-World Evidence

As a subset of RWD, the FDA defines Real-World Evidence (RWE) as clinical evidence regarding the safe use and potential benefits or risks of a medical product derived from an analysis of RWD that can be used to inform or augment the FDA’s understanding of a device’s benefit/risk profile under real-world use conditions. RWD is considered RWE fit for purpose to support generation of clinical evidence for regulatory decision making if the data are relevant to and reliable for informing or supporting a regulatory decision.

FDA Post-Market RWD/RWE

The 2017 final guidance remains in effect until the December 2023 draft guidance is finalized. The FDA notes that it will take time to operationalize the recommendations of the guidance once finalized and expects a 60-day transition period.  Translating FDA guidance to actions is often an exercise in watching what the FDA does with the guidance in the “real world” of device regulatory submissions. A limited review of FDA medical device approvals indicates that the FDA has more often looked at RWD/RWE as part of post-market data collection rather than pre-market collection. Examples of language and recommendations from various devices in which the FDA mentions real-world data or evidence gathered from post-market device use include:

• “Applicant must conduct post-approval study: “Registry-based Real World Use Surveillance””
• “Requires post market approval Registry using product surveillance registry platform”
• “Requires post approval Validation Study”
• “Requires continued follow up/post approval study”
• “Requires traditional or registry-based post-approval study”

FDA Pre-Market RWD/RWE

Examples of device authorizations that leveraged pre-market RWD/RWE include literature reviews that indicated “no new or increased risks” in support of expanded labeling of an approved product, or evidence collection that was conducted within the context of FDA guidance.

Pre-market RWD data can be analyzed as a prospective study of retrospective data. The FDA expects the RWD to meet the standards of valid scientific evidence and requires:

• Controlled protocols presumably with standard data definitions and established data integrity controls
• Description of patient selection, “enrollment” criteria to minimize bias and represent the real-world population
• Compliance with human subject protection and privacy regulations as they apply to prospectively collecting existing retrospective data
• Complete and accurate records
• Data applicable to the US population

The prospective study requires a full protocol and statistical analysis plan (SAP) prior to accessing, retrieving, and analyzing the RWD to generate primary clinical evidence to support a marketing application. The protocol and SAP should address elements of a traditional clinical trial, and a Data Management Plan (DMP) should define data elements (data dictionary) and data integrity controls (e.g., the RWD was source data verified if it was collected for a different study/clinical investigation, or the RWD included in the RWE analysis will be source data verified as a study activity). A monitoring plan should describe systematic methods for data collection, verification, and any record retention necessary to ensure data quality assurance.

FDA Real-World Evidence Examples

Specific examples of FDA Real-World Evidence reviews for medical devices are described in this table: