I run across a new study documenting discrimination against a minority group—usually African-Americans—almost every day. They are so commonplace that I seldom write about them, even though I know the cumulative effect of discrimination is devastating to its victims. However, since most of these studies are not controlled experiments, critics can usually offer alternative explanations that blame the victim. For example, if we find that black kids are expelled from schools at a much higher rate than white kids, a critic can always charge that they misbehave more often or that their misbehavior is more serious. While it’s sometimes possible to collect additional data that makes these explanations unlikely, they are hard to refute definitively.

I don’t think that reservation applies to a recent study by Dr. Monika Goyal and her colleagues in the Journal of the American Medical Association. It involves willingness to prescribe pain medication to black and white children suffering with appendicitis.

The data come from the National Hospital Ambulatory Medical Care Survey, a national probability survey of visits to hospital emergency departments between 2003 and 2010. The unwitting participants were about 940,000 children (mean age = 13.5) admitted with a diagnosis of appendicitis. The children were categorized as white, black or other. The main outcome measure was whether they received analgesic medication for their pain, and if so whether it was an opiate—generally acknowledged to be more effective—or a nonopiate, such as ibuprofen or acetaminophen. The effects of several control variables were statistically removed before analyzing the data: age, gender, ethnicity, triage acuity level, insurance status, geographic region, type of emergency department, year, and (most importantly) pain score on the 10-point Stanford Pain Scale.

Overall, 56.8% of the children received some type of pain medication and 41.3% received at least one opiate. These percentages are lower than is medically recommended. Not surprisingly, the higher the pain score, the greater the likelihood of receiving an analgesic.

The table shows the distribution of analgesia by race, holding pain level constant. The black-white difference in receiving any analgesia was not statistically significant; however, whites were more likely to receive a more effective opioid analgesic than blacks reporting the same pain level. (In case you were wondering, the analysis of ethnicity showed no significant discrimination against Hispanics.)

The data were further analyzed by looking at different levels of pain. Severe pain was defined as between “7” and “10” on the pain scale, while moderate pain was between “4” and “6.” Black and white children in severe pain were equally likely to get some pain medication, but whites were more likely to get opiates. Greater discrimination occurred among children with moderate pain. Black children were not only less likely to get opiates, they were also less likely to get anything at all. In other words, there are higher thresholds for both treating black children for pain, and for treating their pain with opiates.

The authors point out that previous ER studies have found that blacks of all ages and with various medical conditions were less likely to receive analgesics, but these studies can be explained away with victim-blaming rationalizations. For example, it was proposed that, since blacks were less likely to have health insurance, they used the emergency room for less serious conditions. However, all of these children had the same illness its severity was held constant. It has also been proposed that doctors are less willing to trust black patients with opiates due stereotypes about drug misuse. However, the current study did not involve prescriptions, and none of these children were sent home. Presumably, they all received appendectomies as soon as possible.

Since this study was published, it has been suggested that the findings reflect hospital policies rather than decisions by individual doctors. Maybe inner city hospitals that serve a higher percentage of black patients discourage their doctors from prescribing analgesics, especially opiates. It probably doesn’t matter to these kids whether they are denied pain relief by a person with a stethoscope or a person in a suit, although these two hypotheses do suggest different remedies.

In trying to understand this finding, I find myself drawn to some of the most depressing studies in all of social psychology—those involving dehumanization. Dehumanization refers to perceiving and treating another person as non-human—for example, as if he or she were an animal. Dehumanization is sometimes invoked as an explanation for extreme abuses, such as enslavement, torture and genocide. Ordinarily, when you see children in pain, you want to relieve their suffering if possible. Failure to do so suggests dehumanization of the victim. Studies show what appears to be dehumanization of black children (relative to white children) as early as age 10.

Social psychologist Jennifer Eberhardt and her colleagues have done studies suggesting that among white Americans, there is an unconscious association between black people and apes (called the “Negro-ape metaphor.”) To understand her studies, you must know about subliminal priming. A subliminal stimulus is an image presented very rapidly, below the threshhold of awareness. Studies show that subliminal primes improve the recognition of objects in the same or similar categories. Eberhardt has found that subliminally priming participants with images of black people improves their ability to recognize pictures of apes, and vice versa.

In one of her studies, participants were subliminally primed with images of either apes or large cats (lions, tigers, etc.) and shown a video of a policeman severely beating a suspect who they were informed was either black or white. Participants primed with ape images were more likely to see the beating of the black man as justified. This did not occur when they were primed with images of big cats, or when the suspect was said to be white.

Eberhardt did a content analysis of news articles showing that reporters were more likely to use ape metaphors when referring to convicted black murderers than convicted white murderers. Furthermore, those killers described as apelike were more likely to be executed by the state.

I suspect that dehumanization is one cause of the greater willingness of police to shoot and kill black suspects than white suspects in similar situations. Philip Atiba Goff and his colleagues were able to test police officers from a large urban department. The researchers had anonymous access to their personnel files, including their previous uses of force. The more strongly the officers associated black people with apes, the more frequently they had used force against black children, relative to children of other races, during their careers.

The destroyers are merely men enforcing the whims of our country, correctly interpreting its heritage and legacy. But all our phrasing—race relations, racial chasm, racial justice, racial profiling, white privilege, even white supremacy—serves to obscure that racism is a visceral experience, that it dislodges brains, blocks airways, rips muscle, extracts organs, cracks bones, breaks teeth. You must never look away from this. You must always remember that the sociology, the history, the economics, the graphs, the charts, the regressions all land, with great violence, on the body.

Ta-Nehisi Coates, Between the World and Me (p. 10)

It might also be a good idea to take a closer look at those political cartoons depicting President Obama as an ape.

We can only hope the publication of the Goyal study in such a prominent medical journal shames the profession into correcting this type of discrimination against black children. It is unacceptable.