Minority children: over- or underrepresented in medical research?

Minority children: over- or underrepresented in medical research?

October 6, 2003

Despite efforts to include them in medical research, minority adults have historically been underrepresented. In a new study, however, published in the October issue of the journal Pediatrics, researchers from the University of Chicago show that African-American children are over-represented in research, including clinical trials. White and Hispanic children are underrepresented.

Although African-Americans have less access to health care in general, "our data suggest that black children do have fair access to the potential benefits of clinical trials," note the authors.

"Hispanic children may not," added co-author Lainie Ross, MD, PhD, associate professor of pediatrics and assistant director of the McLean Center for Clinical Medical Ethics at the University of Chicago, "but that may reflect language barriers and how race and ethnicity data are reported. We are currently trying to tease this out in a follow up study."

Twenty-five years ago, the primary concern of organizations that oversee human research was fairness in the distribution of risk. More recently, the pendulum has swung toward fairness in the distribution of benefits, with the chance to participate in clinical trials seen as a substantial benefit. Clinical trials, especially phase III trials, sometimes offer the best chance of therapeutic success.

For many years, children were "protected" from those benefits. A 1977 report recommended that research be done first on animals, then adults and only afterward on children. Since 1998, however, the National Institutes of Health have required researchers they fund to include children earlier in the process.

To measure minority participation, Ross and Catherine Walsh, a medical student at the University of Chicago, combed through all the full-length articles in the three main pediatric research journals (Pediatrics, Journal of Pediatrics, and Archives of Pediatrics and Adolescent Medicine) from July 1999 through June 2000. They found 128 articles that included data on race and ethnicity. The number of subjects ranged from eight (their lower limit) to 6,982, for a total of 58,413 and a per-study average of 575. They compared the percentages of racial and ethnic groups in medical research to census data.

Their first discovery was the enormous inconsistency in how race and ethnicity were assessed and reported, with many different labeling systems. Reporting of Hispanics was particularly varied and unreliable.

They were surprised to find that African-Americans were well represented, in fact over-represented, in clinical research. Although they make up 15 percent of the U.S. population, 26 percent of children involved in medical research were African-American, and 32 percent of those enrolled in clinical trials.

Although 69 percent of the U.S. population is white, only 54 percent of children in medical research were white and only 52 percent of those enrolled in clinical trials.

Both black and Hispanic children were over-represented in research on topics that were potentially stigmatizing, such as studies of child abuse, high-risk behaviors or HIV. Fifteen percent of the U.S. population is black but 30 percent of the children in potentially stigmatizing research studies were African-American. While 17 percent of the U.S. population is Hispanic, only 10 percent of the children in medical research were Hispanic, and 17 percent of those involved in potentially stigmatizing studies.

"There are possibly benign and not so benign explanations for these findings," Ross said. "Our data could not distinguish. Clearly, black and Hispanic children should be over-represented in AIDS research as they account for 82 percent of all reported pediatric AIDS cases. On the other hand, there are data to suggest that there is racial and ethnic bias in who is questioned about child abuse."

TableAll Studies(n=128)Clinical Trials(n=18)Potentially Stigmatizing Research(n=36)

Total # Children
r/e available




# White (%)

31,362 (54)

2,278 (52)

7,989 (45)

# Black (%)

15,259 (26)

1,405 (32)

5,323 (30)

# Hispanic (%)

5,894 (10)

287 (6.6)

3,116 (17)

# Other (%)

5,898 (10)

387 (8.9)

1,526 (8.5)

Why black children were more likely to be involved in clinical research is unclear. One possible explanation may simply be that academic medical centers are disproportionately located in urban centers where minorities live. The authors note that the ongoing shift of clinical research from major medical centers to smaller, suburban, private practices, could reverse this trend.

They conclude on a cautionary note. "Despite the participation of minority children in research, there are data that show that these benefits are not being translated into their clinical care. This should be a top priority."

The research was supported by the National Institutes of Health and the University of Chicago Office of Medical Education.