Re: Human Subjects Review/Offending Populations

[Mark Chaffin <mchaffin@xxxxxxxxxxxxxxxxx>]

I would disagree with the opinion that this type of research shouldn't
be done.  The result of a don't-ask-don't-tell policy in offender
research is that society probably is less safe.  Consider the benefits,
for example, of the landmark work of Gene Able and Judith Becker in
their NIH-funded project in the late 1980's.  This project, by the way,
would be an excellent example of how confidentiality issues and
reporting mandates can be managed in face-to-face interviews with
offending populations, and demonstrates that there is IRB and Federal
Government precendent for approving this type of research if it is
carefully undertaken.

There are a number of ways to manage the confidentiality issue.  First,
there is the Federal Certificate of Confidentiality which has already
been discussed.  Second, you can avoid discussing identifying victim
information in the interview (without which, there is no information to
report).  Third, especially sensitive information can be reported on a
highly confidential self-report questionnaire.  For a number of good
examples of how this can be handled, check the HIV literature, or look
at the procedure that is followed for blood donors.  Fourth, you can use
a "blind" interviewer--where the participant's identifying information
is not known or available to the interviewer.  This means having one
interviewer obtain identifying info, then having the blind interviewer
obtain sensitive info.  Fifth, you can eliminate the use of a "key,"
storing information solely on the basis of a participant ID with no link
to names, or, if you must have a key, you can use an encrypted one (as
many state health departments do in their HIV tracking).  

-- 
Mark Chaffin, Ph.D.
Center on Child Abuse and Neglect
University of Oklahoma Health Sciences Center
Dept. of Pediatrics, CHO 4B138
P.O. Box 26901
Oklahoma City, Okla.  73190
(405) 271-8858, FAX (405) 271-2931
mchaffin@etowah.uokhsc.edu