[Skip to Navigation]
Sign In
March 2017

Transparent Reporting of Demographic Characteristics of Study Participants

Author Affiliations
  • 1Department of Dermatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois
  • 2Editor, JAMA Dermatology
  • 3Department of Dermatology, Wake Forest Baptist Medical Center, Winston-Salem, North Carolina
  • 4Department of Dermatology, Rush University Medical Center, Chicago, Illinois
JAMA Dermatol. 2017;153(3):263-264. doi:10.1001/jamadermatol.2016.5978

By 2060, the US Census Bureau projections indicate that the United States will no longer have a single majority population; rather, the nation will be composed of a “plurality” of races and ethnic groups.1 In the early 1990s, congressional legislation set forth guidelines to include women and encourage the inclusion of members of racial and ethnic minority groups in federally sponsored human participant research.2,3 While substantial gains have been made regarding the participation of women in clinical research, the number of minorities appears to be lagging. Compliance with congressional mandates as well as valid analysis of differences between sexes or races/ethnicities requires an accurate system of inclusion data tracking.

Investigators trying to comply with these requirements have found that collecting data on racial/ethnic characteristics was confounded by variations in language, definitions of race/ethnicity, and ever-changing reporting requirements. The result has been multiple variations in acceptable standards for collection of race/ethnicity data since implementation of the requirements, which initially divided the population into only 4 categories: American Indian or Alaskan Native, Asian or Pacific Islander, black, and white.4 The current revision by the US Department of Health and Human Services, Agency for Healthcare Research & Quality is based on a 2-question survey that begins by asking about Hispanic/Latino ethnicity and then about race, which is now divided into 6 categories (Table).5 Clearly, race/ethnicity standards are continuously evolving as deficiencies in our data collection methods are documented and new groups are recognized, for example those who identify as multiracial.

Table.  Two-Question Survey to Assess Self-identification of Race and Ethnicity
Two-Question Survey to Assess Self-identification of Race and Ethnicity

The recognition of the diversity of the US population and the unique challenges faced by each group has given rise to efforts to better understand their health behaviors, quantify and monitor outcomes, and decrease disparities. Consistency in the collection of race and ethnicity data are important to allow investigators to attenuate sample size limitations and collate data for a better estimate of outcomes in small but specific populations. Moreover, health issues that at the present time may not merit our attention could in the future become highly significant. The ability to collect and review health care outcomes by linking data from several small-sample-size studies could make a difference when attempting to understand and assess these trends.6

Another important issue is appropriate design of data collection instruments, especially when collecting information from non–English-speaking participants. Self-reported responses to survey questions soliciting self-identification of race and ethnicity may vary among cultures according to the influence of current sociopolitical issues and participant language preference.7 Other threats to validity include nonresponse to questions (which may result from participants’ lack of understanding) and equivalence. All participating racial/ethnic groups must be familiar with and understand both the questions and the response choices provided in the study. If familiarity is not established, then, in effect, the study ensures an unequal opportunity to participate. Participants’ response rate will also increase if the researchers provide an explanation for why the data are being collected. This explanation needs to be delivered by the team members who have been trained and use a prepared script to control for bias, especially if the concepts being studied have some cultural specificity.

Equivalence, “the extent to which items in surveys and interviews are understood to mean the same thing by all social groups,” must also be considered, since questions can mean different things to different racial/ethnic groups.8 For example, in regard to race/ethnicity questions, federal policy defines the term “Hispanic” as an ethnicity whose individuals can be of any race. However, it was found that two-thirds of Hispanics answering racial/ethnicity questions identified “Hispanic” as part of their racial background.9 Social scientists have long recognized this unique view among Hispanics, where their background was part of their racial identity and not separate.9 This perspective on Hispanic identity persists even after questions are reworded and terms defined in the body of government surveys. Government race/ethnicity categories often fall short for many in capturing self-reported identity, and data collection has proven challenging when trying to capture the unique voices of the US population on matters of race, ethnicity, health behaviors, and attitudes.

Nonetheless, researchers will need to overcome these barriers and recruit participants who reflect the diversity of the population to collect high-quality data. JAMA Dermatology will aspire to report the demographic characteristics of participants in every study that has a sufficient sample size to assure participant anonymity when race/ethnicity is reported. In the past, baseline characteristics of the population were reported as those relevant to the disease examined. Now, JAMA Dermatology’s Instructions for Authors are updated to include a new section, “Reporting Demographic Information for Study Participants,”10 which includes the following guidance: “Aggregate, de-identified demographic information (eg, age, sex, race/ethnicity, and socioeconomic indicators) should be reported for all research reports and systematic reviews along with all prespecified outcomes…. An explanation of who classified individuals as to race, ethnicity, or both, the classifications used, and whether the options were defined by the investigator or the participant should be included in the Methods section. The reasons that race/ethnicity was assessed in the study also should be described in the Methods section. Race/ethnicity of the study population should be reported in the Results section.”10

By making demographic characteristics of the research population truly meaningful, we hope to assist in the process of making informed policy decisions, tailoring medical research to the needs of the community, overcoming public mistrust, and better utilizing health resources. By changing the way in which demographic characteristics are reported, the journal will transparently present the findings needed to discharge our ethical obligation to better serve the health needs of the nation.

Back to top
Article Information

Corresponding Author: June K. Robinson, MD, Northwestern University Feinberg School of Medicine, 676 N St Clair St, #1267, Chicago, IL 60611 (june-robinson@northwestern.edu).

Published Online: February 1, 2017. doi:10.1001/jamadermatol.2016.5978

Conflict of Interest Disclosures: None reported.

Colby  SL, Ortman  JM. Projections of size and composition of the US population: 2014 to 2060. http://www.census.gov/content/dam/Census/library/publications/2015/demo/p25-1143.pdf. Accessed December 3, 2016.
103rd US Congress. S.1 (103rd): National Institutes of Health Revitalization Act of 1993. https://www.govtrack.us/congress/bills/103/s1. Accessed December 3, 2016.
National Institutes of Health. National Institutes of Health guidelines for the review of inclusion on the basis of sex/ethnicity and race in clinical research. https://grants.nih.gov/grants/peer/guidelines_general/Review_Human_Subjects_Inclusion.pdf. Accessed December 3, 2016.
Jackson  A. New OMB standards for federal data on race and ethnicity: HUD policy statement and implementing guidelines. https://archives.hud.gov/offices/cpd/systems/idis/library/raceethnicity.pdf. Accessed December 4, 2016.
Agency for Healthcare Research and Quality. Race, ethnicity, and language data: standardization for health care quality improvement: Institute of Medicine Report [content last reviewed October 2014]. https://www.ahrq.gov/research/findings/final-reports/iomracereport/index.html. Accessed December 3, 2016.
Moy  E, Arispe  IE, Holmes  JS, Andrews  RM.  Preparing the national healthcare disparities report: gaps in data for assessing racial, ethnic, and socioeconomic disparities in health care.  Med Care. 2005;43(3)(suppl):I9-I16.PubMedGoogle Scholar
Aday  LA, Chiu  GY, Andersen  R.  Methodological issues in health care surveys of the Spanish heritage population.  Am J Public Health. 1980;70(4):367-374.PubMedGoogle ScholarCrossref
Landrine  H, Corral  I.  Advancing research on racial-ethnic health disparities: improving measurement equivalence in studies with diverse samples.  Front Public Health. 2014;2:282. doi:10.3389/fpubh.2014.00282PubMedGoogle ScholarCrossref
Pew Research Center. Chapter 7: The many dimensions of Hispanic racial identity. http://www.pewsocialtrends.org/2015/06/11/chapter-7-the-many-dimensions-of-hispanic-racial-identity/. Accessed on December 7, 2016.
The JAMA Network. JAMA Dermatology Instructions for Authors [last updated December 21, 2016]. http://jamanetwork.com/journals/jamadermatology/pages/instructions-for-authors. Accessed January 11, 2017.