Reporting Race and Ethnicity Data from Clinical Trials

Guidance for Industry: Collection of Race and Ethnicity Data in Clinical Trials

This guidance was released in September 2005 and is available at http://www.fda.gov/RegulatoryInformation/Guidances/ucm126340.htm.

This guidance follows a directive issued by the Office of Management and Budget (OMB) and was developed by an FDA working group that included the Office of the Commissioner (OC), the Center for Drug Evaluation and Research (CDER), the Center for Biologics Evaluation and Research (CBER), and the Center for Devices and Radiologic Health (CDRH).

Target audience

This guidance is relevant to medical–clinical staff who conduct, support, and report Investigational New Drug applications (INDs), New Drug Applications (NDAs), Biological License Applications (BLAs), Abbreviated New Drug Applications (ANDAs), and, for certain devices, Investigational Device Exemptions (IDEs) and Premarket Approval applications (PMAs) or Premarket Notifications (510K applications).

Laws and regulations referenced

• OMB Statistical Policy Directive 15, Race and Ethnic Standards for Federal Statistics and Administrative Reporting, 1997, reprinted in Appendix 2 of the guideline, available at http://www.whitehouse.gov/omb/fedreg_directive_15/

• OMB Provisional Guidance on the Implementation of the 1997 Standards for Federal Data on Race and Ethnicity, 2000, available at
http://www.whitehouse.gov/omb/bulletins/b00-02.html

• 63 FR 6854, codified at 21 CFR 312.33(a)(2) and 21 CFR 314.50(d)(5)(v) and (vi)(a)

• 21 CFR 312.120

• 21 CFR 314.106(b) and 601.2

• International Conference on Harmonization, M4 Common Technical Document for the Registration of Pharmaceuticals for Human Use

• Guidances for Industry and other documents (see Bibliography).


Summary

The background for this guidance is summarized in Appendix 1, History of Federal Efforts in Data Collection on Race, Ethnicity, and Other Subpopulations. In order to standardize data reporting across US government agencies, in 1997 OMB issued Statistical Policy Directive 15, Race and Ethnic Standards for Federal Statistics and Administrative Reporting. This guidance brings FDA into compliance and establishes 2 general goals:

• standardize the recommended OMB categories to ensure consistency in demographic subset analyses in applications submitted to FDA and

• make demographic subset analysis more useful by enabling analysts to evaluate potential variations in the safety and efficacy of pharmaceutical products among population subgroups.

In other words, standardized categories facilitate meaningful and potentially significant comparisons among populations and subpopulations of patients.

Rationale

This guidance attempts to standardize OMB categories for race and ethnicity and to facilitate comparisons of populations and subpopulations across studies. FDA notes that some population subgroups may respond differently to medications, but data regarding these responses cannot be meaningfully compared unless they are collected according to a standard. When data are captured by standardized instruments, FDA and others can more accurately analyze trends and enhance patient safety. As the guidance notes, the categories are not anthropologic or scientific but instead reflect sociocultural constructs (see Note below).

The guidance provides some flexibility by allowing investigators to drill down (eg, White can reflect origins in Europe, the Middle East, or North Africa, and Asian can reflect origins ranging from India to Japan).

Resulting recommendations

This guidance makes 5 recommendations:

1. Use a two-question format (1 question regarding race, 1 question regarding ethnicity; see Appendix 2). The ethnicity question should appear first.

2. Study participants should self-report race and ethnicity information whenever possible and should be allowed to report a multiracial identity. If the patient cannot answer these questions, ask a first-degree relative or other knowledgeable source.

3. To identify ethnicity, choose at minimum between the following:
• Hispanic or Latino
• Not Hispanic or Latino.

4. To identify race choose at minimum among the following:
• American Indian or Alaska Native
• Asian (nb broken out as a separate category and separated from Pacific Islander, which is a departure from earlier assessments)
• Black or African American
• Native Hawaiian or Other Pacific Islander.
• White

5. If the study requires more detailed race and ethnicity information, follow OMB Directive 15. Also available are Health Level Seven’s Reference Information Model Structural Vocabulary Tables. Health Level Seven (HL7) is accredited by the American National Standards Institute (more information is available at http:/hl7.org).

If sponsors use the electronic Common Technical Document from ICH E4, they can refer to section 2.7.4.1.3 and table 2.7.4.2, but FDA prefers its descriptors of race and ethnicity. FDA points out that this guidance does not address levels of enrollment necessary to show statistical significance. Finally, the agency notes that the guidance can be used in CDRH applications as appropriate.

Note

Race and ethnicity are fraught with misconceptions and historical baggage. The guidance notes that in June 2005 FDA approved a fixed-dose combination of isosorbide dinitrate and hydralazine hydrochloride (BiDil) to treat heart failure in African American patients. The agency also notes that subgroups respond differently to some medical products. Certain “intrinsic factors” (eg, genetics, metabolism, elimination) and “extrinsic factors” (eg, diet, environmental exposure, sociocultural issues) may be responsible and are subject to confounding. Others have reported that Asians and African Americans respond poorly to some antidepressants, antipsychotics, and beta blockers, possibly because they have low levels of an important metabolic enzyme (CYP 2D6). Advances in genomics may make racial and ethnic descriptors less and less important.

These and related issues are discussed in: Winker MA. Measuring race and ethnicity: why and how? JAMA. 2004;292(13):1612–1614.