The crude death rate gives a general estimate of mortality in a population. However, it is unable to provide a completely accurate picture of the true nature of mortality in a population since the rate does not take into consideration differences of population composition. For example, two populations, with widely divergent crude mortality rates, may in fact have very similar patterns of mortality. A developing population with a low crude mortality rate may simply have a very young population, while an industrialized population with a higher crude mortality rate may be composed of more older individuals and have a larger number of adults dying.
The need for age adjustment becomes particularly important when cause-specific mortality is of interest. Unadjusted rates for chronic diseases (cardiovascular diseases, cancers, or chronic lower respiratory diseases) may appear to be higher for older populations when compared to a younger population. With age-adjustment those differences may be reduced or even reversed. A mechanism for adjusting the age structure differences is needed to determine if there really are mortality differences between two populations.
The method typically used to adjust for differences between populations is direct standardization. Direct standardization divides a population into smaller age groups, estimates mortality rates for each group, and applies these rates to a standard population. Typically, age groups no more than 10 years in length adequately provide accurate age-specific mortality rates. The age-specific rates observed in the population under study are multiplied by the number of people in the specified age group in the standard population. From the sum of those estimates, a single comparable measure of mortality is obtained.
By applying age-specific mortality rates to a standard population, direct standardization controls for differences in population composition. Mortality trends can be more accurately compared along geographic, temporal, or race/ethnicity lines, etc. In short, standardization lets us look at what the death rate would be in one population if that population had the same age structure as the standard population. Beginning with 1999 events, the United States year 2000 population is used as the standard for age-adjusting.
An example of how age-adjusted death rates are calculated is given below:
Example of Age-Adjusted Death Rate Calculations(based on 1999 data for Texas)
1999 TEXAS POPULATION ESTIMATE**
TEXAS RESIDENT DEATHS
TEXAS AGE-SPECIFIC DEATH RATES PER 100,000
US POP STANDARD MILLION
EXPECTED DEATHS STD. POP
|5 - 14
|15 - 24
|25 - 34
|35 - 44
|45 - 54
|55 - 64
|65 - 74
Age-Adjusted Death Rate = 9,062.1 per 1,000,000 population or 906.2 per 100,000 population
(Divide by 10 because the standard population has 1,000,000 people and we want to present our rate per 100,000).
* Total number of Texas Resident Deaths does not include those of unknown age.
** Source: Texas Vital Statistics 1999
2007 Annual Report List of Tables and References
Annual Reports for Other Years
Center for Health Statistics