Framing procedural risks to patients: is 99% safe the same as a risk of 1 in 100?

Acad Med. 2000 Aug;75(8):840-2. doi: 10.1097/00001888-200008000-00018.


Purpose: A patient's willingness to consent to a procedure may be influenced by various factors, including the patient's rapport with the physician, nonverbal cues he or she receives during the discussion of risks, and other elements of the discussion of risks. Previous reports address these influences, but the effect of the actual wording used to describe risks is unclear. The purpose of this study was to better understand how framing the risk involved in a procedure affects a patient's likelihood to consent to the procedure.

Method: In a 1997 study at the Cleveland Clinic Foundation, the authors randomly assigned 116 patients to view one of two short videos describing angioplasty and its associated risks. Sixty-three participants viewed the first video, which framed the procedure as 99% safe, and 53 viewed the second, which framed the likelihood of complication as 1 in 100. Participants were then asked to rate their consent to two hypothetical treatment scenarios on a four-point Likert-type scale (1 = definitely, 4 = definitely not).

Results: When asked to consent to a treatment scenario that would relieve chest pain but offer no survival benefit, respondents who viewed the first video were more likely to consent than were those who viewed the second (p<.001). There was no significant difference in the two groups' likelihoods to consent when the potential health benefit was to reduce the risk of future heart attack.

Conclusion: This study's finding provides evidence that how a physician describes a procedure's risks when obtaining a patient's informed consent significantly influences the likelihood of consent. This fact should be considered when teaching communication skills, including interviewing and patient education skills, so that patients will be more likely to make health care decisions that are consistent with their own values and beliefs.

Publication types

  • Clinical Trial
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Communication
  • Decision Making*
  • Humans
  • Informed Consent*
  • Physician-Patient Relations
  • Risk Assessment