OBJECTIVE: To apply component analysis, a structured approach to the ethical analysis of risks and potential benefits in research, to published emergency research using a waiver of/exception from informed consent. The hypothesis was that component analysis could be used with a high degree of interrater reliability, and that the vast majority of emergency research would comply with a minimal-risk threshold. METHODS: A Medline search and manual search were done to identify studies using a waiver of/exception from informed consent published between July 1996 and December 2000. A review panel of physicians and bioethicists independently classified nontherapeutic procedures in each study as minimal risk, probably minimal risk, or probably more than minimal risk. RESULTS: Seventy studies using a waiver of/exception from informed consent were identified. A majority of reviewers classified nontherapeutic procedures in 62 studies (88.6%) as minimal risk. Reviewers classified nontherapeutic procedures in six studies (8.6%) as minimal risk or probably minimal risk. In two studies (2.9%), nontherapeutic procedures were classified as probably more than minimal risk. The intraclass correlation coefficient was 0.89 (95% CI = 0.85 to 0.93), indicating very high interrater reliability. CONCLUSIONS: Component analysis can be used with high reliability to review emergency research and may improve the consistency of institutional review board review of emergency research. The vast majority of published emergency research performed using a waiver of/exception from consent complies with a properly-applied minimal-risk threshold. A minimal-risk threshold for nontherapeutic procedures protects subjects better than current U.S. regulations while permitting important emergency research to continue.
OBJECTIVE: To apply component analysis, a structured approach to the ethical analysis of risks and potential benefits in research, to published emergency research using a waiver of/exception from informed consent. The hypothesis was that component analysis could be used with a high degree of interrater reliability, and that the vast majority of emergency research would comply with a minimal-risk threshold. METHODS: A Medline search and manual search were done to identify studies using a waiver of/exception from informed consent published between July 1996 and December 2000. A review panel of physicians and bioethicists independently classified nontherapeutic procedures in each study as minimal risk, probably minimal risk, or probably more than minimal risk. RESULTS: Seventy studies using a waiver of/exception from informed consent were identified. A majority of reviewers classified nontherapeutic procedures in 62 studies (88.6%) as minimal risk. Reviewers classified nontherapeutic procedures in six studies (8.6%) as minimal risk or probably minimal risk. In two studies (2.9%), nontherapeutic procedures were classified as probably more than minimal risk. The intraclass correlation coefficient was 0.89 (95% CI = 0.85 to 0.93), indicating very high interrater reliability. CONCLUSIONS: Component analysis can be used with high reliability to review emergency research and may improve the consistency of institutional review board review of emergency research. The vast majority of published emergency research performed using a waiver of/exception from consent complies with a properly-applied minimal-risk threshold. A minimal-risk threshold for nontherapeutic procedures protects subjects better than current U.S. regulations while permitting important emergency research to continue.