Why are there IRB’s at all?

Jim Fearon asks, “Why aren’t there IRB’s [institutional review boards] for the development industry?”

My question is: Why are there IRB’s at all? The IRB seems like some sort of parallel legal system. If people are doing something illegal, it should be stopped. If something is legal but should be illegal, the law should be changed. But I don’t know that we need a parallel legal system. Or maybe it is necessary for international development studies, given the weakness of international law (for example, the widespread worry that a person arrested for rape in the United States would not be extradited, if he were to somehow manage to escape back home to France).

P.S. More on the IRB process in the United States:

1. This article by Carl Elliott reminded me why institutional review boards (IRBs) are needed.

2. This site (via Seth Roberts) reminds me of why IRBs can be a bad thing.

For me, IRBs are typically a waste of time, nothing more, but for others they are a (potential) protection against health hazards and exploitation, and for others they are a barrier to research progress.

P.P.S. I do understand there is a connection between the IRB and the legal system, that one role of the IRB is to catch potential problems in research design before they break the law. So maybe my problem is more with the U.S. legal system than with IRB’s.

5 Responses to Why are there IRB’s at all?

  1. Ken May 27, 2011 at 12:25 am #

    My simple answer is that governments would probably do a bad job at regulating research designs. This is one of those areas (like in most other industries) where self-regulation through professional organizations generates greater efficiencies.

  2. Stephen May 27, 2011 at 1:15 am #

    Not everything that is lawful is ethical, too.

  3. Cyrus May 27, 2011 at 6:33 am #

    I agree with Ken that professional self regulation is the way to go here. The problem is that IRBs, in my experience, do not function that way either. The same IRB board typically covers an incredibly wide range of research and it’s therefore possible to have a dentist as head reviewer for a political science proposal (happened to me) or a political scientist heading review of a public health proposal, in which case the researcher has to waste time addressing what are often very misguided concerns. Having professional, discipline-specific standards would be a much better way go. In general I have found IRB processes to be mostly charades.

  4. Stephen Haptonstahl May 27, 2011 at 2:34 pm #

    I regularly run political science experiments, both “econ” style and, occasionally, “psych” style. Both IRBs I’ve worked with regularly have provided quick, efficient review of my applications. Both boards approve most of my apps without comment, but both boards occasionally catch something important that I missed, like being prepared for subjects to drop out during various parts of the experiment or having a clear debriefing script.

    IRBs can be a pain, but my experience is that they have been helpful.

    I appreciate the trust that the public can have knowing that someone is looking over my shoulder to make sure their rights are protected. Certainly most subjects don’t know about IRB or understand their purpose. However, most people assume that there is some kind of oversight, and IRB can do this job well.

    I have heard of IRBs that terrorize political science researchers because they use the same kind of scrutiny on econ experiments (that are often exempt from most IRB rules) and medical trials, which is, of course, silly. I hear from more experienced experimentalists that the cure is for (1) more social scientists to serve on IRBs and (2) social experimenters work with their IRB to educate them about what is and is not relevant ethically in social science experiments.

  5. Sebastian May 27, 2011 at 9:38 pm #

    I think both Stephens are right on.
    One issue were legality isn’t involved but ethics is concerns consent – Andrew will remember the study where academics were contacted for an on campus visit of a prospective students – I think the treatment was to randomize the apparent ethnicity and gender of said student. Andrew and a lot of other people were angry at the researchers for wasting their time and deceiving them – I think an IRB should exactly address these types of concerns. In this case the study had passed the IRB, but I think the fact alone that researchers have to justify this (and that the IRB can say no – and for all we know does often enough) is very helpful.