Confusing Relative Risk with Absolute Risk Is Associated with More Enthusiastic Beliefs about the Value of Cancer Screening

Abstract

Background: Reviews of how data are presented in medical literature document that the benefit from an intervention is often exaggerated relative to the harm (e.g., relative risk for benefit and absolute risk for harm). Such mismatched presentations may create unwarranted enthusiasm, especially among those who misinterpret the statistics presented. The objective was to determine whether misinterpretation of risk data predicts enthusiasm for cancer screening.

Methods: The authors administered a survey with 14 items assessing beliefs about cancer screening and 6 items measuring data interpretation ability. Multiple linear regression was used to evaluate the association between data interpretation and enthusiasm for cancer screening, with adjustment for gender and year graduated from medical school.

Results: Eighty-eight of 139 physicians at a state-wide professional meeting returned completed surveys (63% response rate). Lower data interpretation scores were associated with higher enthusiasm for cancer screening scores (P = 0.004) in the adjusted primary analysis. Confusing relative risk with absolute risk appeared to drive the overall association.

Conclusions: Biased presentations of risk data could affect general beliefs about the value of cancer screening, especially among physicians who uncritically accept mismatched presentations of data.

Keywords: medical education; numeracy; physicians; risk communication; statistics.