3026.0: Monday, October 22, 2001 - 1:30 PM

Abstract #28440

Ratios and differences are not good estimates of health disparities

Sunny Kim, PhD, Center for Biostatistics, The Ohio State University, M200 Starling-Loving, 320 West 10th Avenue., Columbus, OH 43210, (614)293-6897, Kim.747@osu.edu

Objectives: Health disparity among subgroup is a major health issue. One of the two primary goals for the Healthy People 2010 is to eliminate health disparities. In general, the most frequently used methods of calculating the disparities are the "differences" and the ratios" of two values. The disparity of infant mortality between races is frequently expressed as the ratio of the infant mortality of black and that of whites. Although ratios are the simple way to summarize and compare, it could be misleading in many cases. The purpose of this paper is to illustrate the instances that the ratios and differences do not have practical meaning and even be misleading. Methods: Using the infant mortality among blacks and whites, it is showed that the ratios and differences are not good estimates of health disparities. The infant mortality is used for illustration purpose, but a discussion of the ratio is equally applicable to other health measures including morbidity. Results: The ratio of infant mortality among blacks and whites is only appropriate if both groups are moving to desirable directions (i.e. infant mortality are decreasing). In all other cases, the ratio has no practical meaning and could be misleading. Conclusions: Although ratio is simple to use, we loose big part of information by making ratio. Accordingly, ratio could be misleading in many cases. The race specific mortality shows the more information on disparity among subgroups.

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to: 1. Recognize that ratios and differences are not At the conclusion of the session, the participant (learner) in this session will be able to: 1. The ratios and differences are not good estimates of health disparities in many cases 2. Recognize the strata specific rates are better to show the disparity among subgroups

Keywords: Biostatistics, Health Indices

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
Disclosure not received
Relationship: Not Received.

The 129th Annual Meeting of APHA