Two common methods of age-adjustment or standardization are the direct and indirect methods. The direct method uses weighted averages (for instance, of age-specific rates) according to a predetermined formula based on the age distributions of the populations being compared. The rates actually observed in the populations are applied to an arbitrarily chosen "standard" population, for example, the population recorded at a census in Sweden in 1940, or a "theoretical" distribution constructed by imagining what the U.S. population might have been in 1960 if certain assumptions had been correct. If numbers in some age classes are too small for stable rates to be developed, or if the numbers are unknown, the indirect method is used. In this method, the ageadjusted rates of the standard population are projected onto the population being studied, and these rates are compared to what is actually observed in the population under study. The difference between what is observed and what would be expected if the study population had the same rates as standard population is expressed as a ratio, known as the standardized mortality ratio, or standardized incidence ratio.
John M. Last
(see also: Incidence and Prevalence; Rates; Rates: Adjusted; Rates: Age-Specific )