most studies (66 percent), the reports did not indicate whether, or by whom, the research was approved (in terms of human subjects protections).21

A review of 10 years of correctional and scientific literature on HIV/ AIDS studies involving prisoners (Farley, unpublished, 2005) yielded similar findings. The studies reviewed lacked transparency. Fewer than one-third of the studies mentioned review by an IRB, and nearly one-half made no mention of informed consent.

Data Retrieval Needs Improving

“[Prisoners’] only single armor against being subjected to experimental abuse hangs on a single thread, on a single federal regulation in federally funded research only,” said Vera Hassner Sharav, founder and president of the Alliance for Human Research Protection. “Chimpanzees, by contrast, are protected by mandatory rules, oversight, and enforcement mechanisms since the Animal Welfare Act of 1966. The U.S. Department of Agriculture (USDA) must submit annual reports to Congress documenting the disposition of every chimp, dog, rabbit, and hamster. No one keeps track of how many human beings have died or been harmed in clinical research.” Sharav painted this stark comparison of protections for prisoners with protections for animals in research at the committee’s July 2005 meeting.

The dearth of information regarding the contemporary landscape of prisoner research led the committee to gather systematic information concerning the frequency and types of prisoner research currently being conducted and the research-related policies and procedures of state agencies that house large numbers of prisoners. It was conceded at the outset that the scope of the committee’s efforts in this regard would be limited. For example, the surveys of key personnel in prisons were limited to state DOCs and did not include the federal prison system or the myriad city, county, and municipal jails in which offenders may be at least temporarily


This does not necessarily mean, however, that human subjects reviews were not conducted or that appropriate approvals were not obtained. Journals and journal editors vary considerably in their requirements for reporting (or not) that the research had prior IRB or other human subjects review and approval.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement