Academia's Unconstitutional Restraining Order

Over at the Volokh Conspiracy the other day, Jim Lundgren explained a little-known (to me, anyway) routine practice of government-mandated censorship:

Imagine that I want to write a research article about government abuses of power. I plan to visit a library to look at the public papers of a living person (or a dead person whose papers might embarrass a living person). In most universities, I would be prohibited by federal law--as aggressively interpreted by the federal government--from going to that library without getting PRIOR APPROVAL of a committee set up under federal law, populated with some people outside my university, deciding whether I was allowed to visit the library and read the papers I want to read. If the government had only the desire to check into where I went and what I read after the fact, that would be a serious, though comparatively minor, restriction. No, I am required to get prior approval.

That is the system of Institutional Review Boards (IRBs, formerly called human subject committees) that operates at most major universities. If the government were checking up on me only after I went somewhere or after I read something in a library or after I talked to someone or asked people questions, that would be a big improvement over the current system. The federal government has interpreted its censorship power so broadly that, even for research that is supposed to be exempt from coverage under the federal statute, the federal government has insisted that a researcher get prior approval from an IRB that the work is indeed exempt. […]

How this massive system of goverment-sponsored censorship got going with little attention from Constitutional scholars (before Philip Hamburger) is a mystery to me. It is time for the courts to declare the IRB system what it is: unconstitutional.

Whole thing, worth reading in full, here.

UPDATE: Be sure to check the comments for testimonials from people who, unlike me, were familiar with the term "IRB" before this morning.

Comments

[thoreau]

I knew that IRB's had to OK studies with human subjects. But I thought that only referred to studies where you give them a pill or something. I had no clue that simply going to the library to examine somebody's records counted.

[Chuck]

Our university's IRB policy manual states the following under Definitions:

"Human Subjects Research is a systematic investigation designed to develop or contribute to general knowledge, which involves the collection of data from or about living human beings. Activities which meet this definition constitute "research," whether or not they are supported or funded under a program that is considered research for other purposes. For example, some "demonstration" and "service" programs may include research activities. It does not include research utilizing published or publicly available documents or research on elected or appointed public officials or candidates for public office." (boldface added)

Sorry for the long quote, but I wanted the context. Clearly, research involving public records does not fall under IRB purview.

I read the full post over on Volokh. The main case cited is that of Elizabeth Loftus, which is a particularly egregious example of IRB power being manipulated. But that doesn't necessarily constitute an indictment of IRBs in general. I find our own IRB to be overly bureaucratic, but I can also understand the university's perspective: without an IRB that passes federal muster, the institution is ineligible for most (all?) federal grants. IRBs are, as usual, a good idea made unwieldy through heavy government regulation and interference.

[Ben Rumson]

Education is enemy territory. A pretty good strategy, you'd think that someone planned it.

[thoreau]

Chuck-

Thanks for the info.

Ben-

Huh? Whose enemy? Whose strategy?

[Number 6]

Wait...what's to stop someone from simply checking out the material in question or readining at the library? Is it kept out of general circulation? Or is it a situation where the person would be censured after the fact?

[Number 6]

That should be "reading it at the library". Additional appy-polly-loggies for any other typos found in the above post, or any other, for that matter.

[Steven Horwitz]

Just want to second what Chuck said. My institution's IRB (a private school) has pretty much the same guidelines. We do not have an explicit exemption for public documents the way his does. However, knowing the work of the committee fairly well, I can tell you that, at least here, they have NO purview over such work. Our IRB is only interested in projects that collect data from living human beings - e.g., anything involving a survey, focus groups, a study on college student sleep habits, etc.. The kinds of things Lindgren is talking about, as well as the scholarly article he links to, certainly seem to be well beyond any reasonable understanding of what IRBs are supposed to do.

I can already see the development of a dangerous new meme in the conservative-libertarian blogosphere about IRBs being weapons of censorship or left-wing bias. Can we put a stop to it right now by making the same argument we've made about other analogous stories, namely that they are the exceptions rather than the rules? I sure hope so, because that's the truth as I see it.

[quasibill]

Slightly OT, but Chuck's comment could be a poster for those who oppose vouchers on the grounds that it will corrupt the private school market:

" I find our own IRB to be overly bureaucratic, but I can also understand the university's perspective: without an IRB that passes federal muster, the institution is ineligible for most (all?) federal grants."

Once the state starts handing out vouchers, and schools become dependent on them, all sorts of idiotic requirements will get grafted onto being eligible to receive them.

Similarly, IRBs are not per se a bad thing (there would probably be some equivalent in the absence of state interference), but get warped beyond belief once the government gets its tendrils in through bribery.

[bubba]

Interesting point on the vouchers, but private schools are already dependent on the State for accreditation. And, worst case you'd have a new third class of schools that are private, but with some State strings. More choice is better.

[Brian]

I'll add my voice to the tally of those well aware of IRBs (though I've never had to interact with them myself--thankfully) and having never heard of them covering anything like what's mentioned in the story.

[linguist]

My institution's IRB (a private school) has pretty much the same guidelines.

This has also been my experience.

I can already see the development of a dangerous new meme in the conservative-libertarian blogosphere about IRBs being weapons of censorship or left-wing bias.

Yes, I can see this beginning to happen too. Let's not get carried away. The whole IRB thing is to PROTECT THE SUBJECTS and their privacy. In most cases researchers can access public records.

What happens, in fact, is that when, say, a sociologist wants to review murder-suicide numbers throughout the 1980s for the entire, country, he states in writing his purpose in doing so, and what persons are likely to be included in that survey (murderers and victims). The IRB looks over the proposal, determines that in fact this research is not likely to infringe on anyone's rights, and approves it.

Everyone seems to HATE dealing with the IRB, feeling they have too much power.

Now with medical studies, of course, the process is MUCH more stringent. And I bet everyone here would be extremely upset if it weren't. So of COURSE researchers hate dealing with the IRB. They're a pain in the ass! It takes months to get things phrased just so and get approval, first from your department, then your university and then from the state IRB. This is still very preferable to having subjects' medical records misused!

[linguist]

Additionally, you all may not know that it's the IRB that approves the consent forms that subjects sign. I used to be the IRB enforcer for some human subject studies, and believe me, many subjects are NOT looking out for their rights so it's a damn good thing someone does.

[Nonymus]

I sit on an IRB at a large university and have dealt with dozens more. There are classes of research that involve human subjects tangentially- for instance reviewing peoples' medical records without contacting them- and directly- for instance interviewing public officials- that are exempt from IRB review. Paradoxically, my IRB (and I assume most) require certification that a study is exempt. There is no requirement to review studies that rely solely on published literature.

That said, IRBs come in many different temperments and a study that passes unscathed by one can be nixed by another. That doesn't speak to well about their protection function. Nevertheless, having seen what some researchers try to get approved, IRBs of some form or another are necessary. My take is that the top-down federal nature of the boards creates a lot of inefficiencies. However, even if IRBs were not a federal mandate they would continue to exist and to adhere largely to the same set of principles.

[Chuck]

Corsair09--

At my institution, there is an expedited review process so that frequently projects like surveys, that involve "minimal risk" to the subjects, can be reviewed and approved by the IRB chair. In such cases, the chair only sends it to the full committee if (s)he has concerns about the proposal, so your wife probably never sees 95% of those. The part of this that most faculty find annoying is that even anonymous surveys that do not produce any personally identifiable data have to go through the approval process. It's almost always a rubber stamp, but it is just one more bit of sand in the gears. On the other hand, I suppose it's a good thing that *somebody* is trying to err on the side of protecting private information.

If those doctors think the IRBs are a nuisance, they should be thankful they don't have to get proposals through the animal care and use committee.

One more general comment: as I hinted at above, an institution's IRB procedures can give a lot of individual power to the chair. Like other positions of power, this authority can be abused by the person wielding it. My own take on the Loftus situation, from not only on the Lindgren article but also from the earlier Skeptical Inquirer article it seems to be based on, is that this is a case of an IRB chair using his authority for some sort of hidden agenda. Loftus's work has been very controversial, and she has made some enemies in the psychology field.

[Curmudgeon]

Another thread (Hoax Watch) lauds the blogsosphere for its quick examination of items for plausibility/factuality. It looks like that's what's happening here.

At my alma mater it was called the ERB - Ethics Review Board - and it provided a variety of jumpable but nonetheless required hoops for research approvals and changes. Annoying, but to my eye appropriate.

[K. Toishi]

IRB... the bane of my existence during my dissertation.

[K. Toishi]

My institution also exempts the study of exisiting documents. Only research involving human subjects is under their jurisdiction. Alas, this does not mean I didn't curse IRB more than once during my diss.

Exemptions:

4 6.101(b)(4) Research involving the collection or study of existing data, documents, records, pathological specimens, or diagnostic specimens, if these sources are publicly available or if the information is recorded by the investigator in such a manner that participants cannot be identified, directly or through identifiers linked to the participants.

[joshua corning]

This looks like where privacy rights and free access to public information begin to show how they conflict. Anyway good luck with all that.

[Mike Swaim]

I'm involved in medical research, so I've had indirect exposure to my institution's IRB. From what I understand, if you're doing medical research in the US on human subjects, you have to have an IRB, and the IRB often has to sign off on any changes to your protocol. This can be a real hassle for smaller institutions that don't have their own IRBs (and either share one, or pay a commercial firm to act as their IRB).
If I remember correctly, IRBs were mandated as a reaction to some really nasty research done in the '60s and '70s. I took a class on human experimentation, and some of the researchers back then were doing some pretty frightening stuff.