How To... - Interpret Data
When are the health problems increasing or decreasing over time?
Surveillance data are analyzed by time and answer the question "when"
to determine if health problems are increasing or decreasing over time.
- The PedNSS and PNSS Summary of Trends in Health Indicators tables
can be used to compare prevalence of health indicators for the most
recent year of data to the previous year as well as examine the long
term trend in prevalence of the health indicator.
- Looking at only one year of data can be misleading because
prevalence often varies from year to year. One population may have a
higher prevalence one year, but a lower prevalence in either the
previous or subsequent years. Trend data can provide a more accurate
picture of whether or not the prevalence is changing over time.
- Change in the prevalence of a health indicator over time may be the
result of a change in the health indicator or simply a change in the
distribution of a specific demographic characteristic (age,
race/ethnicity, or education) of the population. The tables on Summary
of Trends in Racial and Ethnic Distributions or Age can be used to
determine if the population characteristics have remained stable or
changed over time.
- To determine if there is a true change in the prevalence of a
specific health indicator and that the change is not due to a change in
population characteristics such as age, race/ethnicity, or education you
may want to conduct an additional standardization analysis to control
for a specific person characteristic, thus, making the populations more
comparable from one time period to another and making the prevalence of
the health indicator of interest also comparable.
Learn how to calculate standardized rates using
an example of standardization of race/ethnicity for a population over
Change in Prevalence of a Health Indicator Over Time
An increase or decrease in the prevalence of a health indicator or its
trend over time can be portrayed either as absolute or relative change.
- Absolute change is one prevalence value minus another.
- Example: If the prevalence of obesity in 2001 was 12.9% and in 1992
it was 8.9%, the absolute change equals 12.9% minus 8.9% or an increase
- Relative change is the percentage change between the two years of
interest and is calculated by first obtaining the absolute change or
difference. This change or difference is divided by the prevalence of
the first or earlier year to obtain the relative change which is the percentage of
change over time.
- Example: Divide the absolute change (4.0%) by the first year prevalence
(8.9%) = .45 or 45% relative increase.
To determine if you should present the absolute or relative change
depends on the point you want to make. A general rule of thumb is that
when you are concerned about the magnitude of a public health problem,
such as how many people are affected, it may be appropriate to give
emphasis to absolute rather than relative changes.
On the other hand,
relative differences are of interest when you want to look at the
processes of causation, for example, to examine the effect of health care
or of a supposed risk factor or protective factor on the occurrence of a
health problem. It is not always easy to choose between the use of
absolute and relative change and sometimes both are important. However,
the most important question is whether the health problem is getting
better or worse.
back to top
Page last reviewed: October 29, 2009
Page last updated: October 29, 2009
Content Source: Division of Nutrition, Physical Activity and Obesity,
National Center for Chronic Disease
Prevention and Health Promotion