Health Services

Biostatistics

Note on Changes in the Vital Statistics Data

Starting with the 1999 vital statistics data, two important changes were made in the way statistical data are classified and analyzed. These changes are very important to keep in mind when comparing rates calculated from 1999 or later years with rates from pre-1999 data.

The first change is that, starting with 1999, deaths are classified using a new system International Classification of Disease 10 (ICD 10) instead of the older ICD 9 system. For certain causes of death, like influenza and pneumonia, rates have been substantially reduced because deaths have been reclassified to other causes under ICD 10. For other causes of death, like cancer, total death rates have been little impacted but the internal distribution of deaths between different types of cancer have changed1. A dashed vertical line has been used in the mortality trend graphs in the SCDHS Annual Report to caution in making trend comparisons between rates compiled in the two ways. (Rigorous statistical comparisons between ICD 9 and ICD 10 compiled rates requires the use of a formula with a comparability ratio for that cause of death. Some of these ratios have been published by the National Center for Health Statistics, see footnote one below.)

A second major change was also made beginning with the 1999 data. The age adjustment standard for mortality data was changed from the 1940 US population to the 2000 US population. Age standards are used so that rates can be compared between populations that have different age distributions. Older populations generally have more disease, so one wants to control for differences in age when making population comparisons. Different data sets, mortality, cancer incidence, national health surveys, etc., had been using different standard populations (1940 US population, 1970 US population, etc.). Therefore it was decided in 1998 by the US Department of Health and Human Services to adopt a single age adjustment standard across all health data sets2. This change in the age adjustment standard only affects a single SCDHS Annual Report table: ‘Crude and Age-Adjusted Death Rates From Leading Causes by Township of Residence’. In this table the age-adjusted rates, given in parentheses, use the new standard. Caution should be exerted in comparing these age-adjusted rates with the corresponding age-adjusted rates in previous year editions of the SCDHS Annual Report. Rate differences could be due, entirely or in part, to differences in the age-adjustment standard or in changes in the disease classification system.


 1For a detailed discussion of the impact of this change see Anderson RN, Minino AM, Hoyert DL, and Rosenberg HM; ‘Comparability of Cause of Death Between ICD-9 and ICD-10: Preliminary Estimates’; National Vital Statistics Reports; Vol. 49, No. 2, May 18, 2001

2Klein, RJ, Schoenborn CA; ‘Age Adjustment Using the 2000 Projected US Population’; Healthy People 2010 Statistical Notes; No. 20, Jan. 2001