Research Conditions That Qualify for Emergency Exception from Informed Consent

Abstract

Medical research involving critically ill and injured subjects unable to provide informed consent can only be conducted under federal regulations that attempt to balance the need to develop lifesaving treatments with protection of research subjects' rights. Regulators, researchers, and medical ethicists have all struggled to define the conditions under which an emergency exception from informed consent is appropriate. Although research has been successfully conducted under the current regulations, confusion remains regarding the meaning of the regulations, the applicable conditions, and the best ways to balance the needs of future patients and the rights of research subjects. In May 2005, at the Academic Emergency Medicine Consensus Conference "Ethical Conduct of Resuscitation Research," a breakout session was held on the research conditions that qualify for the emergency exception from informed consent process. Several recommendations emerged: 1) The definition of "life-threatening condition" should be broadly interpreted to include serious disability as well as death. 2) Existing therapies should be considered "unsatisfactory," even if partially effective, when serious risk of morbidity or mortality remains even with the best available treatment or when the adverse effects of the best available treatment are serious. 3) Research with the emergency exception should be performed only if sufficient evidence exists that the proposed intervention has a reasonable chance of benefit. 4) More evaluation is needed to determine the degree to which the current rules impede research. 5) Application of the current regulatory framework for abbreviated or waived consent in emergency research should be encouraged. 6) Further study should also address variability among institutional review boards, the goals of community involvement, and how best to engage and educate the public in research efforts using emergency exception from informed consent.