Tuesday, March 28, 2006

The Short Form

Recently, some colleagues and I had a discussion about the "short form" for documenting consent. As most of you know, the federal regulations allow the use of "A short form written consent document stating that the elements of informed consent required by [§46.116/>50.25] have been presented orally to the subject or the subject's legally authorized representative." [46.117(b)(2)/50.27(b)(2)] The short form may be used when the subject does not speak English or another language where there is a translated consent form. It may also be used when the subject is illiterate.

The short form is not used very much and many institutions don't use it all. This is probably because most people don't understand it and are afraid to use it. The problem is that there is not much guidance out there on the use of the short form. The only OHRP guidance on this is from 1995 (http://www.hhs.gov/ohrp/humansubjects/guidance/ic-non-e.htm) but I guess it still applies. It says that the short form should be in the subject's native language. It also says: "The IRB must receive all foreign language versions of the short form document as a condition of approval under the provisions of §46.117(b)(2). Expedited review of these versions is acceptable if the protocol, the full English language informed consent document, and the English version of the short form document have already been approved by the convened IRB."

I believe that IRBs should not be afraid to use the short form where it is appropriate since it is allowed in the regulations. The appropriate use is for subjects who speak an unanticipated language. If there is a reasonable expectation that there will be subjects recruited for the research who speak a certain language (most commonly, Spanish), then the consent form should be translated into that language. The short form should not be used in that case.

The problem is OHRP's guidance which requires IRB approval of the short form which should be in the subject's language. Although this can be done through expedited review, it can still be burdensome. This makes the use of the short form problematic for the situation where a subject shows up who speaks an unanticipated language. This, however, is guidance from OHRP and old guidance at that. I think an IRB could make a good case that the regulations do not require IRB approval of the short form or that the short form be in the subject's language. If a witness signs the short form indicating that the information in the English consent has been provided to the subject in his or her native language and that the subject consented to be in the research, that should be satisfactory documentation of informed.

I hope that OHRP comes up with new guidance on this topic and reconsiders the requirements of the old guidance. FDA guidance (http://www.fda.gov/oc/ohrt/irbs/informedconsent.html#nonenglish) does not require IRB approval or that the short form be in the subject's language.

Saturday, March 04, 2006

Mission Creep

The other day I was asked to comment on the "Mission Creep" white paper. The paper recommends "...focusing on those areas of research that pose the greatest risk, such as biomedical research, while removing or reducing scrutiny of many fields within the social sciences and humanities that pose minimal risk." This is based on the assumption that the primary purpose of IRB review is to protect subjects from risk. Beneficence, which addresses the relation between risk and benefit, is only one of the three Belmont principles. There are respect and justice issues even in minimal risk, innocuous research. I believe that all research activity, even innocuous research, requires some sort of independent, objective review. Reducing the oversight over minimal risk research jeopardizes

The problem that this paper is responding to is not IRB review per se, but IRB review that is unduly burdensome and that imposes unreasonable requirements. So, while claims that IRB review is not necessary for minimal risk research are not reasonable, IRBs are somewhat responsible for the attacks on IRB review. I have long argued that the regulations provide sufficient flexibility for the efficient and appropriate review of minimal risk research. IRB review of such research does not have to be burdensome or unreasonable if IRBs appropriately utilize the flexibility in the regulations. The horror stories that social and behavioral researchers quote in their attacks on IRB review are true. I have seen institutions where expedited review takes longer than full review and where IRBs have imposed written consent forms on telephone interview research. These are IRBs that a) don't understand social and behavioral research, b) don't know how to use or are afraid to use the flexibility in the regulations, and c) don't pay sufficient attention to the efficiency of their procedures.

The White Paper correctly recommends that IRBs receive better guidance on how to review non-biomedical research and that there needs to be increased sharing of best practices among IRBs. However, critics of IRB review should focus their attention on educating IRBs on the appropriate review of non-biomedical research rather than trying to reduce IRB oversight of such research.

Prisoner consent question

Recently I was giving a talk about prisoner research and had discussed OHRP's guidance on subjects who become incarcerated after enrolling in a research study. Someone from the audience asked me about what happens if a legally authorized representative who has consented on behalf of a subject becomes incarcerated. This was something I hadn't thought about and I doubt whether there is any guidance on it. My take on it was that, as long as consent had been appropriately obtained, there is no problem with the LAR being incarcerated. The requirements of Subpart C only apply to the subjects, not the LARs. However, if the subject needs to be re-consented there is a problem. I'm not sure if it's appropriate for an incarcerated LAR can give consent for a subject. On the other hand, the regulations don't address this, so I don't think there is a rule against it. I guess, before approving consent from an incarcerated LAR, I'd contact OHRP and get their opinion about it.