As part of the Real-World Evidence Program,¹ the U.S. Food and Drug Administration (FDA) has released the final guidance, “Real-World Data: Assessing Electronic Health Records and Medical Claims Data To Support Regulatory Decision-Making for Drug and Biological Products.”² Unlike well-controlled clinical studies, real-world data can be associated with issues related to completeness, accuracy, traceability, and relevance. Recognizing these shortcomings, this final guidance provides a multitude of considerations for using electronic health records (EHRs) and medical claims data in clinical studies to support a regulatory decision concerning the safety or effectiveness of a drug product, including selection of data sources to address different study questions, development and validation of definitions for study design elements, and data traceability and quality.³

The guidance specifically addresses the proposed use of EHRs and medical claims data in clinical studies involving drugs (i.e., human drugs and biological products).⁴ Medical claims data, also known as administrative healthcare claims data, refers to information that is submitted to a patient’s insurer for payment for treatments and other services.⁵ Both EHRs and medical claims data are sources of real-world data (RWD), from which real-world evidence (RWE) can be derived based on analysis of RWD to support a regulatory decision regarding safety or effectiveness of a drug product.⁶

The guidance outlines key issues for determining the reliability and relevance of the data and that should be addressed in the study protocol, including:

1. the appropriateness and potential limitations of the data source for the study question(s) and to support key study elements;
2. the time periods for ascertaining study design elements;
3. the conceptual definitions and operational definitions for study design elements (e.g., inclusion and exclusion criteria for the study population, exposures, outcomes, and covariates) and the results of validation studies; and
4. the quality assurance and quality control (QA/QC) procedures for data accrual, curation, and incorporation into the final study-specific dataset.

We encourage sponsors who are interested in using EHRs or medical claims data to support FDA regulatory decisions to review this guidance before initiating such studies to help ensure the FDA’s acceptance of the data and to maximize the completeness and accuracy of such RWD in providing reliable clinical evidence of safety or effectiveness of a drug product.

Data Sources

Given that EHRs and medical claims data are collected during routine patient care and without a prespecified research protocol, the FDA recognizes their limitations when they are used for research purposes or to support regulatory submissions.⁷ Sponsors should be prepared to demonstrate that the proposed data sources for their study are relevant to answering the specific study question(s), such as description of the detail and completeness needed to capture the study populations, exposures, key covariates, outcomes of interest, and the relevant time periods, among other important parameters.⁸ Sponsors should consider, among other things:

• the rationale for selecting the data sources to address a specific study question, background information on the healthcare system that generates the electronic healthcare data, and the generalizability of the study findings from the data sources;
• the comprehensiveness of the data source in capturing care and outcomes relevant to the study question(s), such as study populations, exposures, outcomes of interest, and key covariates;
• data variability over time and across different data systems and structures;
• missing data and the risk of duplicate records;
• the relevant time periods of the RWD;
• the sample size, quality of the data, data heterogeneity, and different clinical and coding practices across different data sources or systems;
• the data recording practices of the healthcare system; and
• validation of study variables.⁹

Study Design

The FDA also recommends that sponsors ascertain and validate their study design elements. Some considerations for study design and validation are to:

• clearly define the pertinent time periods (e.g., for identifying the study population; defining inclusion and exclusion criteria; assessing exposure, outcome, and covariates; following up with patients; and the “washout” time period (if applicable));
• determine the methods for selecting, and describe the operational definitions relevant to, the inclusion and exclusion criteria;
• define, ascertain, and validate the “exposure” (e.g., the duration, dose, formulation, and strength of a drug or treatment being evaluated in the study);
• define, ascertain, and validate the clinical outcome; and
• define, ascertain, and validate the key covariates, including confounders and effect modifiers.¹⁰

Data Quality and Traceability Throughout the Data Life Cycle

The quality of data may vary or fluctuate throughout the data life cycle.¹¹ The data life cycle spans multiple phases: i) data accrual from the original source data; ii) curation of data to the clinical data repository; iii) transformation and de-identification of data (where necessary); iv) creation of a data warehouse; and v) production of a study-specific dataset for analysis.¹² The FDA recommends that sponsors use consensus-based standards to review the completeness, accuracy, and plausibility of the data, which includes verifying the data against the original source and on an ongoing basis at each phase of the life cycle.¹³

The FDA also recommends that sponsors ensure “traceability” of EHRs and medical claims data. Data traceability refers to the method (e.g., audit trail) that enables the understanding of the data’s origin and how it entered the EHR or medical claim.¹⁴ Sponsors should consider developing procedures for: i) characterizing the data for completeness, conformance, and plausibility; ii) documenting the QA/QC plan; and iii) documenting the data management process.¹⁵

The FDA encourages sponsors proposing to use EHRs and medical claims data in clinical studies to support a regulatory decision to submit their protocol and statistical analysis plan before beginning their study, as well as consider seeking the agency’s feedback.¹⁶

Contact Us

For questions regarding FDA regulatory strategy and recent developments in FDA regulation, including the FDA’s Real World Evidence activities, please contact Eva Yin, Jonathan Trinh, or any member of Wilson Sonsini’s FDA regulatory, healthcare, and consumer products practice.

---

[1] See U.S. Food and Drug Amin., Framework for FDA’s Real-World Evidence Program (Dec. 2018), https://www.fda.gov/media/120060/download.

[2] 89 Fed. Reg. 60430, 60430 (July 25, 2024); U.S. Food and Drug Admin., Real-World Data: Assessing Electronic Health Records and Medical Claims Data to Support Regulatory Decision-Making for Drug and Biological Products: Guidance for Industry (July 2024), https://www.fda.gov/media/152503/download (hereinafter “EHR and MCD Guidance”).

[3] 89 Fed. Reg. at 60430; EHR and MCD Guidance, at 2.

[4] The final guidance does not apply to medical devices. EHR and MCD Guidance, at 1 n.3.

[5] Id. at 2 n.6.

[6] Id. at 2. RWD is defined as “data relating to patient health status or the delivery of health care routinely collected from a variety of sources,” including data derived from electronic health records, medical claims data, data from product and disease registries, patient-generated data including from in-home use settings, and data gathered from other sources that can inform on health status, such as digital health technologies. Id. By comparison, RWE is defined as “the clinical evidence regarding the usage and potential benefits or risks of a medical product derived from analysis of RWD.” Id. (emphasis added).

[7] Id. at 4 and 6.

[8] Id. at 6.

[9] See id. at 5–14.

[10] See id. at 15–28.

[11] Id. at 28.

[12] Id.

[13] Id. at 28–29.

[14] Id. at 3 n.11.

[15] Id. at 30–33.

[16] Id. at 3.