Biased science

Science is embarrassingly white. That’s a huge problem.

Google Image Search

Type the word scientist into Google image search, and you’ll overwhelmingly get a bunch of white faces in return. Even the cartoon scientists are white.

Sadly, this is a microcosm of the state of science today. Researchers are largely monochromatic. It’s no matter that the country is becoming more multicultural and interracial. What that means is that science—the very thing that’s supposed to be unbiased—ends up reinforcing our prejudices and injustices. The consequences are far-reaching in their economic, intellectual and social impact.

For example, as many as three quarters of Pacific Islanders don’t metabolize a commonly prescribed anti-blood clotting drug properly due to a genetic condition prevalent in that population. That means the drug doesn’t do much for them in terms of preventing heart attacks, strokes or even death. The study that paved the way for that medication’s widespread use was 95% white. African-Americans have some of the highest rates of cancer, but their inclusion in cancer-related clinical trials is lower than for whites, according to a 2014 study. The most widely used asthma medications don’t work as well on African-Americans and Latinos.

“If you don’t diversify something like science, society doesn’t change and the people who have power and are in charge of knowledge generation, they do not change,” said Carl Hart, a neuroscientist at Columbia University. “If things do not change, the people who have been subjugated previously will continue to be subjugated.”

In other words, we should view access to science and science education as a civil rights issue.

Science isn’t abstract. It’s not something that exists in the white ivory tower of academia. It affects our lives every day—and will continue to do so. The phone you’re holding, the car you drive, the booze you drink were all crafted by science. The processed food you snack on while you Netflix & chill is the meticulous brain child of scientists. And so are the medications and treatments doctors use to keep you and your family healthy.

That’s where some of the more troubling consequences of an all-white establishment really come into play. In a paper published Tuesday in the journal PLoS Medicine, an international group of scientists call attention to the problem, describing the status and repercussions of the unbearable whiteness of biomedical research.

“This country is plagued by racial and ethnic disparities in some of the deadliest and most debilitating diseases, including cancer, cardiovascular disease, diabetes and asthma,” said Dr. Esteban González Burchard, a pulmonologist at the University of California, San Francisco and one of the paper’s lead authors, in a statement. “The only way we will make progress in decreasing that burden of disease is by understanding why it occurs. And we can’t understand that without including diverse communities in our research.”

That means recruiting more minorities to participate in research and to design the research itself.

Since 1993—the year the government mandated that federally funded clinical research include more minorities and women—less than 2% of more than 10,000 cancer clinical trials funded by the National Cancer Institute included enough minority participants to meet the agency’s own criteria and goals for inclusion. And less than 5% of respiratory research included racial and ethnic minorities, the authors wrote. Even some of the more notable efforts, like UCSF’s new genetic dabatase, in which 1 in 5 people come from traditionally underrepresented groups, still don’t stack up to reality. The U.S. population is 40% non-white.

Think about that: Almost half of the U.S. population is missing out on some of the benefits of the research they’re bankrolling with their tax dollars. That number is only expected to grow if the representation of minorities in research doesn’t improve. As medicine moves toward being a data-driven practice this will become more problematic. In January, President Barack Obama announced the Precision Medicine Initiative to weave together the genetic, sensor, lifestyle, environmental, microbiome and medical-record data of 1 million volunteers into one database. The idea is that from all this data, trends about what causes diseases will emerge, enabling new, more efficient types of treatments. Machines will be making some of these calls, and they’re only as unbiased as the data from which they learn. Bias in, bias out.

As González Burchard notes, diseases don’t affect everyone equally. The way a virus or cancer affects you is the sum total of your age, the environment you live in and your genetics. How your body responds to medications also depends on these factors. That means if researchers don’t study a potential treatment in people like you, you might be SOL when you get sick.

The higher cost of recruiting minorities, who may live farther from academic research centers or not speak the language of study organizers, is often flagged as one of the underlying causes for the exclusion of minorities from studies. That argument, however, doesn’t hold water. It’s lazy.

“If you thought it was important that your science was done correctly, you’d figure out what to do. You’d partner and do all kinds of things. That just doesn’t play as an explanation,” said Shirley Malcom, the director of education and human resources programs at the American Association for the Advancement of Science, who wasn’t involved in writing the PLoS Medicine paper. “I don’t see how anybody can say that the science is good if you’re not looking at all of these different aspects—like race and ethnicity.”

She points to the increased inclusion of women, both in research studies and recruitment into science education programs as a sign that change can happen.

“We have made headway in that regard, though we still have a lot way to go,” she said. But I don’t feel that we have made as much headway with regard to racial minorities.”

This bad science costs all tax payers money, though—regardless of their background. Uncle Sam doesn’t care what color you are, just that he collects. The PLoS Medicine authors note that eliminating racial and ethnic health disparities would have reduced healthcare spending by more than $1.2 trillion between 2003 and 2006. That’s because some of the very diseases where the research skews the most white are the ones that affect minorities the most, leaving them with less viable options for care. (To put that in context, in 2015, the federal deficit was $412 billion.) Add that to the fact that minorities are less likely to be insured and to have access to high-quality healthcare, and you close the pernicious loop of what some have called medical apartheid.

So what gives? It may be as simple as a case of blissful ignorance, A.K.A. white privilege.

“The people who are actually required to do the implementation [of recruiting diverse sets of people], their group is doing fine. There’s no reason to do anything in many of their minds. They have no understanding of the larger issues that are at play,” said Hart, the Columbia neuroscientist, in reference to the way minorities have been historically shut out and continue to be shut out of science and other social benefits. “It doesn’t bother people in the dominant culture because they continue to win.”

Here’s where the other part of the equation, the lack of minority researchers, comes into play. The NIH does have programs to up the number of minorities in research, like grant programs specifically designed to fund minority researchers. (I benefited from one while I was a PhD student in neurobiology at Columbia University.) Even so, less than 2% of NIH grants list a black scientist as the principal investigator. Similarly low proportions list Latinos and Native Americans. Fewer non-white applicants get RO1s—the most coveted type of grants in biomedical research—than whites, according to a Freedom of Information Act request. Without an RO1, setting up your own lab is close to impossible. Blacks or African Americans and Hispanics, respectively, represented only 4.3% and 7.2% of PhDs in biomedical sciences in 2013, although they represented 13.9% and 17.2% of the US population during the same period, the researchers wrote.

This lack of diversity among researchers—says Malcom, the education program chief at the American Association for the Advancement of Science—is harder to fix. It’s reflective of a bigger issue: the institutional oppression of minority students. Students in lower-income school districts are more likely to come from minority groups. These groups have less access to good science, math and engineering teachers. That holds them back for the rest of their lives, closing off opportunities their white counterparts have, like lucrative professions.

Professional, scientific, and technical services were some of the top contributors to the increase in economic growth in the U.S. in 2015, according to the Bureau of Economic Analysis. But the representation of minorities, like Latinos, in these sectors was dismally low:

science hispanic

Right now, we have affirmative action in place to try to close the opportunity gap and give minorities access to these fields. But that policy may not be around much longer, if the Supreme Court rules in favor of Abigail Fisher, a white female student who claims she was unjustly denied admission to the University of Texas at Austin in favor for less qualified minority students. That could have profound consequences for how state medical schools and graduate programs recruit a diverse group of students to their campuses, says Malcom.

In the end, that will end up affecting our science and healthcare. There’s research that shows that a diversity of opinions leads to innovation and better workplaces. “There’s a whole question of trying to develop cultural competency among physicians,” Malcom told me. “If I’m going to treat all different kinds of people, how do I gain insights into social factors that may inform the way I diagnose and treat different people?”

When we go to the doctor, we want them to understand the whole of us, that means having a scientific and cultural knowledge base from which to craft a diagnosis tailored to us as individuals. We’re failing miserably at that. If we want to right the ills of discrimination, we can’t let that happen. Diversity in health and science has to be a priority. Plus the authors note, recruiting a diverse group of individuals to clinical studies benefits everyone, not just that group:

table1PLoS Medicine

In an address to the National Academy of Science in 2013, President Barack Obama said “our university system is the crown jewel of our economy as well as our civilization. And that’s what’s allowing us to continually replenish our stock of people who are willing to dream big dreams and reach higher than anybody else.”

People of all colors should have access to that dream. Right now, that’s just not the case. Just look at what Justice Gregory Scalia said on NPR about blacks and science, in reference to the Fisher case:

“They come from lesser schools where they do not feel that they’re — that they’re being pushed ahead in — in classes that are too — too fast for them,” Scalia said. “Most of the black scientists in this country don’t come from schools like the University of Texas.”

That’s the kind of false thinking that will continue to segregate science.