Comparability Ratio About every 10-20 years the International Classification of Diseases (ICD) is revised to stay abreast of advances in medical science and changes in medical terminology. Each of these revisions produces breaks in the continuity of cause-of-death statistics. Discontinuities across revisions are due to changes in classification and rules for selecting underlying cause of death. Classification and rule changes impact cause-of-death trend data by shifting deaths away from some cause-of-death categories and into others. Comparability ratios measure the effect of changes in classification and coding rules. For causes shown in Table VI, comparability ratios range between 0.6974 and 1.1404. Influenza and pneumonia had the lowest comparability ratio (0.6974), indicating that Influenza and pneumonia is 30% less likely to be selected as the underlying cause-of-death in ICD-10 than in ICD-9. Pregnancy, childbirth, and the puerperium had the highest comparability ratio (1.1404), indicating that pregnancy is more than 14% more likely to be selected as the underlying cause using ICD-10 coding.
For selected causes of death, the ICD�codes used to calculate death rates for 1980 through 1998 differ from the ICD�codes most nearly comparable with the corresponding ICD� cause-of-death category, which also affects the ability to compare death rates across ICD revisions. Examples of these causes are ischemic heart disease; cerebrovascular diseases; trachea, bronchus and lung cancer; unintentional injuries; and homicide. To address this source of discontinuity, smortality trends for 1980�98 were recalculated, using ICD�codes that are more comparable with codes for corresponding ICD� categories. Table V shows the ICD�/p>codes used for these causes. Although this modification may lessen the discontinuity between the Ninth and Tenth Revisions, the effect on the discontinuity between the Eighth and Ninth Revisions is not measured.
Comparability ratios shown in Table VI are based on acomparability study in which the same deaths were coded by both the Ninth and Tenth Revisions. The comparability ratio was calculated by dividing the number of deaths classified by ICD� by the number of deaths classified by ICD� The resulting ratios represent the net effect of the Tenth Revision on cause-of-death statistics and can be used to adjust mortality statistics for causes of death classified by the Ninth Revision to be comparable with cause-specific mortality statistics classified by the Tenth Revision.
The application of comparability ratios to mortality statistics helps make the analysis of change between 1998 and 1999 more accurate and complete. The 1998 comparability-modified death rate is calculated by multiplying the comparability ratio by the 1998 death rate. Comparability-modified rates should be used to estimate mortality change between 1998 and 1999.
Caution should be taken when applying the comparability ratios presented in Table VI to age-, race-, and sex-specificmortality data. Demographic subgroups may sometimes differ with regard to their cause-of-death distribution, and this would result in demographic variation in cause-specific comparability ratios. SOURCE: Health, United States Related
Links
This page last reviewed
January 11, 2007
|