Download this Paper Open PDF in Browser

How Statistics Can Save Your Life or End It: A Course Module

10 Pages Posted: 6 Dec 2013  

Hershey H. Friedman

City University of New York (CUNY) - Department of Business Management

Martin Raphan

City University of New York (CUNY) - Department of Finance

Date Written: December 4, 2013

Abstract

Probability and statistics are powerful weapons that can help people make good decisions, but that can also be used to wreak havoc. This paper is the first in a series designed to provide examples to students of how statistics can be used and misused to mislead the public and result in poor decision-making. In this module, we discuss the use of statistics in evaluating healthcare decisions, and how they can mislead patients. This particular module, therefore, is also of use to patients who wish to understand how to use statistics to evaluate treatment options. A patient deciding whether to take a prescription drug or undergo a medical procedure should understand basic evaluation criteria like the number needed to treat (NNT) and number needed to harm (NNH), which are described in this paper. We also describe how information presented to consumers regarding the efficacy of drugs often focuses on the wrong statistical measures. Brochures and advertisements for medications often focus on the relative risk reduction (RRR), when NNT is more helpful in evaluating the efficacy of a treatment.

Keywords: Misuse of statistics, framing effects, healthcare statistics, treating the many to benefit the few, ARR, RRR, NNT, NNH

JEL Classification: C00, C10, C40, I20

Suggested Citation

Friedman, Hershey H. and Raphan, Martin, How Statistics Can Save Your Life or End It: A Course Module (December 4, 2013). Available at SSRN: https://ssrn.com/abstract=2363742 or http://dx.doi.org/10.2139/ssrn.2363742

Hershey H. Friedman (Contact Author)

City University of New York (CUNY) - Department of Business Management ( email )

Martin Raphan

City University of New York (CUNY) - Department of Finance ( email )

2900 Bedford Avenue
Brooklyn, NY 11210
United States

Paper statistics

Downloads
234
Rank
109,259
Abstract Views
805