Compared with white American researchers, black American researchers are a third less likely to have an early-career National Institutes of Health (NIH) grant funded, according to an NIH-commissioned study published August 18 in Science. It’s a thorough study, experts say, but it leaves one major question unanswered: "Why?"

The difference persists even among black and white scientists who went to similar graduate schools, took part in the same NIH scientist training programs, have earned the same number of grants previously and have published the same number of scientific papers. "We have left no stone unturned in trying to find some explanatory variables," says Donna Ginther, a University of Kansas economist and lead author on the study. Ginther and her colleagues found the effect after examining grant applications from 40,069 scientists, submitted between 2000 and 2006.

This disparity in funding suggests U.S. biomedical science isn't benefiting from "the best and brightest minds" regardless of race, NIH director Francis Collins said during an August 17 press conference. "This situation is not acceptable." Not getting the particular type of grant studied, called an R01 grant, can hit a young researcher hard, adds Raynard Kington, a study co-author and former acting director for the NIH. "R01s are the coin of the realm," Kington explains. Whether a young scientist gets one often affects whether she or he gets a permanent or tenure-track job at a university.

The NIH will test its own review process to look for more answers, Collins says. Meanwhile, African-American researchers and students have a few ideas based on their own experiences.

If NIH reviewers can identify the applicants they screen, then one factor that may be at play is unconscious bias, Kington says. Unconscious bias is a well-documented effect in which people in a majority group, such as white Americans, subtly discriminate against African-Americans or Asian-Americans, for instance, even when they don't consciously hold racist or prejudicial views against that group or people of color in general. In the NIH's usual grant review process, applicants' names and ethnicities are removed, but in some cases it's still possible to guess who's applying. Applications from historically black universities are more likely to come from black researchers, for example. And in smaller fields of biology, reviewers might recognize applicants by the studies they propose because there are so few researchers in the field.

Many researchers say a lack of mentoring may hold back young black scientists. A commentary by Collins in the August 19 issue of Science mentions the issue. Evelynn Hammonds, dean of Harvard College, says she has seen many high-achieving African American undergraduates suffer from a lack of guidance; she has led diversity-increasing programs at Harvard and at MIT. Kasim Ortiz says the biggest challenge minority students face in graduate school is finding a mentor to walk them through the NIH grant-writing process. Ortiz is a doctoral student in health policy at the University of South Carolina and a historian for the National Black Graduate Student Association. "I think it's vitally important that institutions work to strategically develop pipelines to connect minority students and mentors early on," he says. Although all graduate students are on the look-out for a mentor, he finds it's especially difficult for his friends who are people of color.

Alondra Nelson, a Columbia University sociologist who studies race and science in the U.S., has a different view. Many black science students come from historically black liberal arts colleges, such as Spelman College and Morehouse College. While they get excellent educations there, they may not meet, network and collaborate with established biomedical scientists who manage large NIH grants, Nelson thinks. Without that early exposure to the "life of grant-writing," she says, they may be at a disadvantage compared with students from large research universities. On the other hand, students at historically black liberal arts colleges often work at research universities during the summer, Hammonds says, so she isn't sure how large an effect small colleges have. "That kind of question should be empirically examined."

Whatever the cause of the funding gap, action shouldn't wait for further study, Kington says. Of course, the NIH needs studies to identify where to target interventions. A planned NIH experiment that blinds reviewers to any identifying information should help answer whether reviewing scientists are succumbing to unconscious bias, for example. Meanwhile, "we don't have to wait to find the definitive, exact causal pathways" to try new ideas about how to close the funding gap, Kington says. There's enough evidence to get started now.