Half-Truths and Consequences


Researchers did not tell parents the whole truth about medical experiments they were conducting on their children, documents obtained by the Voice show. They also indicate that the researchers, who were looking for predictors of aggression, originally planned to study only black and Hispanic children.

This new information adds more controversy to a research project already under fire. The furor erupted earlier this month when the New York Post reported that researchers at the New York State Psychiatric Institute and Columbia University gave 34 boys—all black or Hispanic—one dose of a drug called fenfluramine, best known for being half of the “fen-phen” diet combination pulled from the market last year because it damages heart valves. The NYSPI study was completed before the drug was banned, and the institute insists that a single dose of fenfluramine causes no significant side effects. But the study, which was partly funded with federal money, is under investigation by the national Office for Protection from Research Risks.

The fenfluramine procedure was just one element of a larger study, conducted by eminent researchers such as Gail Wasserman, Daniel Pine, and Michael Liebowitz. The study as a whole examined 126 younger brothers of convicted delinquents, ages six to 10, who were deemed “at risk” of following in their older siblings’ footsteps.

In applying for grants and publishing their findings, the researchers state unequivocally that they were seeking biological and environmental predictors of “antisocial behavior,” “aggression,” and “delinquency.” But when they went looking for research subjects, the description changed. “With so many children getting into trouble,” the researchers wrote in a recruitment letter to parents of potential subjects, “we are interested in learning about what keeps children out of trouble [emphasis in the original].”

In explaining the purpose of the study, the informed-consent form—which is supposed to fully and accurately describe the research—says only that the researchers are examining “the physical and emotional well-being of children whose families have been involved in the court system.” The fenfluramine procedure is described as follows: “This test will evaluate the responsiveness of a natural brain chemical called serotonin. Fenfluramine causes the release of serotonin, and its effects are measured in this test.” In fact, researchers wanted to see if levels of this brain chemical could predict aggressive behavior.

“Sugarcoated” and “misleading” is how Alvin Poussaint, a prominent psychiatrist from Harvard, describes the recruitment letter. The consent forms, he adds, “don’t really tell parents why their kids were being studied.” Boston University’s Leonard Glantz, an authority on experimentation on children, says, “If your child were in this study, and then you found out what they were really up to, would you feel you had been appropriately informed?”

In a written response to the Voice, the researchers call their language “neutral” and say they were trying not to stigmatize the children: “While it might have been most accurate to say, ‘There is evidence that your younger child is at risk for developing the same problems as his older brother,’ such a bald statement of probable risk could result in the type of detrimental labeling that we—and the critics of the study—are most concerned about.” Documents confirm that stigmatizing was a major concern.

But Poussaint counters, “If you want to give people full respect, honesty comes first. They didn’t protect against stigmatizing anyway. They basically said, ‘You’re a troubled family, and that’s why you’ve been selected.’ If you say that much, why not say more? Maybe there was the feeling that if we tell them more about what we’re doing, they may reject the study. So maybe it wasn’t just protecting the patients, but protecting themselves.”

Critics say that the partial information is particularly problematic, given other aspects of the study. The children were traced through court records, which ethicists say is a violation of privacy and may have created a coercive atmosphere in which parents felt pressured to enroll their children. In addition, the study paid parents—$125 for the fenfluramine test alone—and since many of the families were poor, critics have questioned whether this also constitutes coercion. These factors are “even more of a problem if the study was not described the way it really is,” says Poussaint.

The fenfluramine procedure required children to fast for 18 hours and have a catheter in their arms for six hours, and the drug itself involved possible side effects, including headache, nausea, and anxiety. The study’s ethical oversight panel—called an Institutional Review Board, or IRB—considered the drug too dangerous to give to a control group of children whose siblings were not delinquent. Ultimately, the IRB did deem the drug safe enough for control groups, but largely because they had tried it on boys in the study.

“Why was it okay to do it on those boys?” asks Glantz. The procedure offered no possible benefit to the children; it was given only to test a hypothesis. The researchers note that none of the boys suffered significant side effects. But to Glantz, that misses the point. “You can’t just say, ‘No one was harmed so it can’t be a bad thing.’ Harm is one issue, but a separate issue is use—is this a fair and legitimate use of children?”

Of the 126 boys in the study as a whole, 97 per cent were black or Hispanic. This skew, say the researchers, is an “unfortunate” accident of geography. Located in upper Manhattan, NYSPI worked with the Manhattan and Bronx family courts, where the vast majority of juveniles are minorities. NYSPI Deputy Director Jack Gorman quipped, “If our institute was located in Iowa, [the subjects] would have all been white.”

Yet in the original research plan, white children were explicitly excluded from the study. Only African American and Hispanic boys were to be allowed in. The IRB demanded that this provision be dropped, but even after race was stricken from the protocol, researchers continued to conceptualize the study racially. A subsequent grant application to the National Institute of Mental Health describes the study as “limited only to Black and Hispanic boys.” The research protocol states that it “is restricted to boys of two major minority populations in New York City because these restrictions maximize the sample’s risk for developing disruptive behaviors/conduct disorder.”

“If someone presented that to me, I would be very troubled,” says Poussaint. “There’s the implication that these problems are something special to black and Hispanic kids, and not to white kids who are delinquents. But there are a lot of white kids who are antisocial doing all kinds of things. Why are they so focused on blacks and Hispanics?”

The researchers deny “any ethnic-specific hypothesis,” adding that they recognized the racial skew and eventually sought to make the study more diverse. But that study was not funded.

Does race matter? The research found that witnessing violence increases the odds that a child will himself commit violence, and that inadequate parental supervision is more likely to lead to antisocial behavior. This is undoubtedly true for all races. Even the results of the fen-fluramine experiment—which suggest that harsh parenting can alter brain chemistry—would almost certainly be true across racial lines. Thus, the researchers maintain, it would be tragic if their findings are “contaminated by the unfounded allegations of bias.”

But that misses a critical point. Science begins with the selection of what and whom to study. “Whenever they look for predictors of violence,” says Poussaint, “they seem to target only low-income people. It’s possible to go into a suburban court and collect all kids who got into trouble with law, but they don’t tend to do that.”

Partly because researchers don’t do that, society continues to conceptualize criminals in terms of race—and that creates its own vicious circle. Why, after all, are black and brown kids so overrepresented in the juvenile courts to begin with? What makes police, social workers, and teachers see some troubled children as a menace and others as worth salvaging? And why did this study originally set out to look only at black and brown boys?

Research assistance: Sam Bruchey