Research Shows How To Avoid Misleading With Statistics

Ann Arbor, MI, USA – Research conducted by the University of Michigan School of Public Health shows that when researchers compare two or more treatments, the choice of how to express results may affect their conclusion about which is best.

Using a mathematical model and a sophisticated software program, Edward Norton, PhD, and his team demonstrated that the choice between risk ratio, odds ratio and risk difference could result in different outcomes through what is called rank reversal. Previous research has suggested that rankings through indirect comparisons can be assumed to be consistent across the various methods.

Norton said, “The outcome of the team’s research impacts best practices for making indirect comparisons.” He continued on to say, “With the government funding more research that uses these different methods, researchers need to be aware that the method they choose could influence the conclusion.” The paper, “Rank Reversal in Indirect Comparisons,” is published in Value in Health.


Value in Health (ISSN 1098-3015) publishes papers, concepts, and ideas that advance the field of pharmacoeconomics and outcomes research as well as policy papers to help health care leaders make evidence-based decisions. The journal is published bi-monthly and has over 8,000 subscribers (clinicians, decision makers, and researchers worldwide).

International Society for Pharmacoeconomics and Outcomes Research (ISPOR) is a nonprofit, international, educational and scientific organization that strives to increase the efficiency, effectiveness, and fairness of health care resource use to improve health.

For more information: www.ispor.org

Posted in Methodological Articles, Value in Health Tagged with: , , ,