1.2.3 Standardization of Rates
Crude rates, as measures of morbidity or mortality, can be used for population
description and may be suitable for investigations of their variations over time;
however, comparisons of crude rates are often invalid because the populations
may be di¤erent with respect to an important characteristic such as age, gen-
der, or race (these are potentialconfounders). To overcome this di‰culty, an
adjusted (or standardized) rate is used in the comparison; the adjustment
removes the di¤erence in composition with respect to a confounder.
Example 1.13 Table 1.7 provides mortality data for Alaska and Florida for
the year 1977.
Example 1.13 shows that the 1977 crude death rate per 100,000 population
for Alaska was 396.8 and for Florida was 1085.7, almost a threefold di¤erence.
However, a closer examination shows the following:
- Alaska had higher age-specific death rates for four of the five age groups,
the only exception being 45–64 years. - Alaska had a higher percentage of its population in the younger age
groups.
The findings make it essential to adjust the death rates of the two states in
order to make a valid comparison. A simple way to achieve this, called the
direct method, is to apply to a common standard population, age-specific rates
observed from the two populations under investigation. For this purpose, the
U.S. population as of the last decennial census is frequently used. The proce-
dure consists of the following steps:
- The standard population is listed by the same age groups.
- The expected number of deaths in the standard population is computed
TABLE 1.7
Alaska FloridaAge
Group
Number
of Deaths PersonsDeaths per
100,000Number
of Deaths PersonsDeaths per
100,000
0–4 162 40,000 405.0 2,049 546,000 375.3
5–19 107 128,000 83.6 1,195 1,982,000 60.3
20–44 449 172,000 261.0 5,097 2,676,000 190.5
45–64 451 58,000 777.6 19,904 1,807,000 1,101.5
65 þ 444 9,000 4,933.3 63,505 1,444,000 4,397.9
Total 1,615 407,000 396.8 91,760 8,455,000 1,085.3
16 DESCRIPTIVE METHODS FOR CATEGORICAL DATA