Consent forms in research are a source of current and retrospective information for the subject, a 'prompt' for the person who is obtaining consent, and a documentation of the 'informed' consent process and its adequacy. Occasionally, these forms may be administered by inexperienced trainees or ancillary personnel, and thus stand virtually alone. Therefore, the forms must be inherently comprehensible to the subjects. To test whether this is the case, 65 new applications were randomly selected from 13 consecutive IRB agendas, and their consent documents were computer-analyzed (Flesch/Fry scoring) after correction for expected confounding features, such as lists, tables, and polysyllabic proper names and jargon. Mean U.S. school grade for 70% comprehension (Fry score) was 15.03 ± 0.19 (standard error of the mean), implying readability by 37.4 ± 1% of the U.S. adult population. In contrast, a consecutive sampling of 21 Ann Landers columns yielded a mean Fry score of 7.67 ± 0.5 (p < 0.01; readable by 75 ± 3%). Fifteen Reader's Digest articles yielded a mean Fry score of 9.95 ± 0.65 (p < 0.01; readable by 59.1 ± 3%), and 15 'Talk of the Town' columns from The New Yorker averaged a grade level of 13.3 ± 0.83; p < 0.01; readable by 42.7% ± 4.8%). No document was improved by more than one grade level by the IRB review process, and most were unchanged. The authors' conclusions are that: (1) most 'informed' consent documents are complex enough to be comprehensible to only a minority of patients or subjects; and (2) IRB review, although it probably improves consent form clarity for the educated reader, does little to make the forms more accessible to less well-educated subjects. Obtaining truly informed consent with current consent forms requires that they be presented in true dialogue by a person fluent in lay language and familiar with the details of the study being explained and with the elements of informed consent.
Bibliographical noteFunding Information:
This study was supported by a small grant from the Minnesota Medical Foundation. Portions of this work were presented at the May 1990 meeting of the American Federation for Clinical Research, Washington, D.C. Correspondence: Dale E. Hammerschmidt, MD, Chair, CUHSR Committee "01," Hematology Division, Box 480 UMHC, 500 S.E. Harvard. Street, Minneapolis, MN 55455.
- Human subjects
- Informed consent