Approaches to Measuring Differences in Health That Are Unaffected by the Prevalence of an Outcome

• Activity Number: 213
• Type: Roundtables
• Date/Time: Tuesday, July 31, 2007 : 7:00 AM to 8:15 AM
• Sponsor: Section on Health Policy Statistics
• Author(s): James P. Scanlan*+
• Companies: Attorney at Law
• Keywords: health disparities ; binary variables ; continuous variables ; mortality differences ; health inequality ; measurement

Abstract

Binary measures of group differences in health outcomes vary, depending on the prevalence of the outcome. For example, as mortality declines, relative differences in mortality tend to increase, while relative differences in survival tend to decline. Measures that are functions of binary outcomes (e.g., gini coefficients, longevity differences) also tend to change in some manner as the prevalence of an outcome changes. This renders problematic efforts to evaluate the size of health disparities at different times or in different settings with measures based on mortality or morbidity rates. It is possible that continuous variables may offer approaches to measuring the size of group differences in health that are independent of changes in prevalence of the factor being examined. But that possibility must be tested, and factors that are useful to examine need to be identified.