A couple of weeks ago, the anti-vaccine movement took a swing for the fences and, as usual, made a mighty whiff that produced a breeze easily felt in the bleachers. In brief, a crew of anti-vaccine lawyers named headed by Mary Holland, co-author of Vaccine Epidemic: How Corporate Greed, Biased Science, and Coercive Government Threaten Our Human Rights, Our Health, and Our Children, published a highly touted (by Generation Rescue and other anti-vaccine groups, that is) “study” claiming to “prove” that the Vaccine Injury Compensation Program (VICP) had actually compensated children for autism. As is typical with such “studies” generated by the anti-vaccine movement, it was bad science, bad law, and just plain bad all around. The authors intentionally conflated “autistism-like” symptoms with autism, trying to claim that children with neurological injury with “autism-like” symptoms actually have autism. Never mind that there are specific diagnostic criteria for autism and that, if the children actually had autism many of the would have been given a diagnosis of autism. Never mind that what they were doing was akin to claiming that all patients with “Parkinson’s-like symptoms” have Parkinson’s disease. (Hint: They don’t.) Never mind that all they did was to demonstrate a prevalence of autism spectrum disorders among the VICP-compensated children that was clearly within the range of what would be anticipated if there were no relationship between vaccines and autism. Never mind all that. This was Holland’s big chance, but it went over like the proverbial led balloon. No one bit, other than FOX News.
The study rapidly faded into the obscurity it so richly deserves, in spite of mighty efforts by Generation Rescue, SafeMinds, and the likes of Ginger Taylor to keep it alive and use it as a rallying point to persuade legislators to pass anti-vaccine-friendly legislation. You could feel the frustration in its backers as Holland’s study, into which groups like Generation Rescue had apparently poured their hopes of being vindicated, crashed and burned.
However, there’s one aspect of this study that I didn’t discuss. In fact, I thought of it as I read it, but I wasn’t sure. Do you know what it is? I won’t try to keep you in suspense. (After that lengthy introduction, I’d better not.)
What I (and others) have noticed is that there was no statement in the article that approval had been obtained from the relevant institutional review boards (IRBs) to do human subjects research. For those not familiar with what an IRB is, an IRB is a committee that oversees all human subject research for an institution. It is the IRB’s responsibility to make sure that all studies are ethical in design and that they conform to all federal regulations. Basically, IRBs are charged with weighing the risks and benefits of proposed human subject research and making sure that (1) risks are minimized and that the risk:benefit ratio, at least as well as it can be estimated, is very favorable; (2) to minimize any pain, suffering or distress that might come about because of the experimental therapy; and (3) to make sure that researchers obtain truly informed consent. During the course of a study, regular reports must be made to the IRB, which can shut down any study in its institution if it has concerns about patient welfare.
Of course, I know this all because I happen to be involved in human subjects research. It’s part of what I do while doing research in breast cancer. Clinical trials are obviously human subjects research. After all, their very purpose is to test a new drug or treatment on human subjects in order to determine if it works. However, human subjects research encompasses a lot more than just clinical trials; in fact, it encompasses almost any research study that involves either human subjects or human subjects’ protected health information (i.e., medical charts). Basically, any human subject experimentation requires approval by a properly constituted IRB. There is no documentation in Holland et al that IRB approval had been obtained, either from the for NYU (given that Mary Holland is a research scholar at the NYU School of Law) and/or Pace University, given that Pace Law School is cited as having “provided significant research support for this study” in the footnotes of the paper:
Mary Holland, Research Scholar and Director of the Graduate Legal Skills Program, NYU School of Law; Louis Conte, independent investigator; and Robert Krakow and Lisa Colin, attorneys in private practice. Pace Law School provided significant research support for this study.
Autism News Beat and Sullivan are already all over this issue, but I thought I’d throw in a bit from my perspective, given that I actually have to work with IRBs and, before my career is over, will probably be roped into serving on an IRB. In any case, as is true of any researcher who is involved in human subjects research, I had to undergo specific training regarding the “rules of the road,” so to speak, as a requirement of being an investigator on any study involving human subjects research.
How is human subjects research defined? According to 32 CFR 219.102(d), research means “a systematic investigation, including research development, testing, and evaluation, designed to develop or contribute to generalizable knowledge,” and according to 32 CFR 219.102(f) a human subject is defined as ” living individual about whom an investigator (whether professional or student) conducting research obtains data through intervention or interaction with the individual, or identifiable private information.”
Under these criteria, I believe that the Pace study clearly qualifies as human subjects research. First of all, as poorly designed as the study was, it was clearly intended as a systematic investigation designed to contribute to generalizable knowledge and testing a hypothesis, namely that autism is associated with vaccine-induced brain injury. The authors claim to have found data in the VICP-compensated cases indicating that there is a higher incidence of autism in these children. Never mind that they found nothing of the sort, they set out to test that hypothesis and their methods were designed to support it. Moreover, as Autism News Beat pointed out, members of the Elizabeth Birt Center for Autism and Legal Advocacy (EBCALA) administered the Social Communication Questionnaire (SCQ) to the parents or caregivers of 22 children. This clearly makes the parents who answered the questionnaire (and their children) human subjects, as identifiable protected health information about the children is being solicited and used to test their hypothesis.
To be fair, I have to point out that there are types of studies that are exempt from full IRB approval. In fact, there are six categories of so-called “exempt” studies. I frequently use category four, which involves the use of already existing clinical information that has been de-identified. I also note that this particular exemption involves research that does not involve interacting with living humans about their health information; rather, these sorts of studies tend to involve chart reviews or the examination of human tissue specimens that have already been collected. Be that as it may, there is one potential exemption that EBCALA might claim. First, number two:
Research involving the use of educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, interview procedures or observation of public behavior, unless:
- information obtained is recorded in such a manner that human subjects can be identified, directly or through identifiers linked to the subjects; and
- any disclosure of the human subjects responses outside the research could reasonably place the subjects at risk of criminal or civil liability or be damaging to the subjects financial standing, employability, or reputation.
Unfortunately for Holland et al, the first “unless” probably kills this exemption. The information they obtained was obtained in a manner such that the human subjects can be identified. The second “unless” might or might not kill the exemption; I note that both criteria are used to decide whether an exemption is not permissible. However, even if this research could potentially be considered exempt – and I highly doubt that a reputable IRB operating within the law and the Common Rule would find it exempt, particularly since IRBs are charged with taking special care to protect human subjects considered to be vulnerable, such as children, and that the survey is in fact a screening/diagnostic tool, making the results protected health information – it should also be noted it is not the investigators who get to decide whether a protocol is exempt or not and therefore whether a protocol requires IRB evaluation and approval. Period. Investigators do not have the power to determine whether their research falls under one of these exempt categories; only the IRB can make that determination, usually in an expedited review that determines whether a potentially exempt protocol meets the standards of one of the categories that are exempt from full IRB approval. It’s right there in the rules:
Only the IRB may determine which activities qualify for an exempt review. Investigators do not have the authority to make an independent determination that research involving human subjects is exempt and must contact the IRB concerning the status of proposed research or changes in ongoing research.
Also note that the research cannot begin until after IRB approval is obtained.
Personally, unlike the case with the Geiers, my guess is that as lawyers Holland et al were simply ignorant that what they were doing constituted human subjects research and that’s why they didn’t bother with a little thing (to them) like obtaining IRB approval before doing their study. Indeed, when Autism News Beat inquired at Pace Law School about whether the EBCALA study had IRB approval, this is what he was told:
When asked if the Pace study had IRB approval, Pace Law spokesperson Lauren Rubenstein referred the question to the study’s co-author, Louis Conte. In an email, Rubenstein wrote “Louis Conte has told me that there was no human subjects research in this study.”
Which reveals Conte’s utter cluelessness about human subjects research. Investigators interacted with the parents of these children, administering a questionnaire to them for purposes of looking for data to support their hypothesis. It’s human subjects research until judged otherwise–by an IRB, not by Conte, Holland, or any of the other of the merry band of anti-vaccine lawyers who put this travesty of a study together. Not surprisingly, attempts to get Conte to respond to e-mails regarding IRB approval of the study were not returned.
Of course, ignorance of the law is no excuse; as lawyers the authors of Holland et al should know. I also find it rather hard to believe, whether or not they were ignorant of federal law and regulations regarding human subjects research, that at some point either Holland and her coauthors weren’t made aware somehow by someone that what they were doing constituted human subjects research under federal law. Surely someone must have realized this, given that questionnaires about protected health information were being administered to living human beings. If they did not, then, quite frankly, I find the level of incompetence and lack of concern about basic ethical standards that every investigator in the U.S. who engages in human subjects research knows as basic information because they are required to know them and abide by them as part of ethics and the law.
This lack of concern about such niceties of the ethical treatment of human subjects is nothing new in anti-vaccine studies. Andrew Wakefield demonstrated it with his callous treatment of children at a party, whom he basically bribed for blood samples and whose fear he made light of. The Geiers père et fils demonstrated it when they created an IRB to oversee their own research, sat on that IRB themselves with Dr. Geier himself chairing the committee (a massive conflict of interest), and packed it with their anti-vaccine cronies.
There’s one more “out” that Holland et al might try for, and that’s to say that their research was not federally funded and is therefore not covered by the Common Rule. However, most universities that accept federal funding for research agree to be bound by the Common Rule for all research carried out by their faculty, students, and trainees. Even if Holland et al tried to justify their failure to obtain IRB approval this way and argue that they are not legally bound to follow the Common Rule, their failure to obtain IRB approval for their study and submit to IRB oversight is a massive ethical failing. I characterized Holland et al as evidence that that when the anti-vaccine movement can’t win on science, it will appeal to the law. It also appears that when it can’t win on ethics, it will appeal to the law as well.
While Holland et al tout “unanswered” questions about autism and the VICP, the only question I have is: Will Pace University and NYU will investigate how such a study could be performed without its investigators obtaining the requisite IRB approval? Or will they sweep this massive ethical (and possibly legal) breach under the rug and hope it goes away?