Evaluating risk: a primer for gastroenterologists
Eisen, G.M.; Sandler, R.S.; Wurzelmann, J.I.; Ransohoff, D.F.; Talal, A.
American Journal of Gastroenterology 90(5): 704-707
1995
ISSN/ISBN: 0002-9270 PMID: 7733072 Document Number: 439687
Objective: To review the various types of risk measures and to illustrate how interpretations may vary depending on how risk is presented. Methods: We define different types of risk estimates and provide examples of their use from the recent gastroenterotogical literature. Results: Calculations of absolute risk and the number of patients who must receive treatment to prevent one patient from experiencing a disease state (number needed to treat) provide useful information in assessing the utility of fecal occult blood testing for prevention of colorectal cancer and in estimating small bowel cancer rates for individuals with familial adenomatous polyposis. Conclusions: Measures such as relative risk and the odds ratio are best used in analyzing causality but may not be helpful in guiding clinical decisions. Absolute risk and attributable risk reveal the actual number of cases that can be explained by a given exposure and can thereby measure the impact of a clinical intervention for both the population and the patient. These measures should be the criteria for decision-making.