MORTALITY TECHNICAL NOTES

Cause of Death Codes and Classifications:

For 1999 to present, all causes of death are classified according to the ICD-10 classification system.  The ICD-9 coding system was used for deaths occurring between 1978 and 1998, so cause of death classifications before 1999 are not directly comparable to classifications for 1999 or later.

Cause-of-death coding: Mortality statistics for these tables were compiled in accordance with the World Health Organization (WHO) regulations, which specify that member nations classify causes of death by the current Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death.   The current version is the Tenth Revision of the International Classification of Diseases [ICD-10] (World Health Organization 1992).  Mortality statistics for 1999 and earlier were classified using the Ninth Revision [ICD-9].  (NCHS Data Definitions: Classification of Diseases)

Underlying causes of death: Tabulations of cause-of-death statistics are based solely on the underlying cause of death unless otherwise stated.  Every death is attributed to one underlying condition, based on information reported on the death certificate and using the international rules for selecting the "underlying cause of death" from the conditions stated on the death certificate.   The underlying cause is defined by the World Health Organization (WHO) as the disease or injury that initiated the train of events leading directly to death, or the circumstances of the accident or violence, which produced the fatal injury.  If more than one cause or condition of death is entered, the underlying cause is then determined by the sequence of conditions on the death certificate and selection rules of the ICD (Murphy 2000).  (NCHS Data Definitions: Cause of Death)

Related causes of death:   Examination of the combination of all listed causes can shed additional light on factors related to mortality.  Therefore, in Connecticut, we also present “related” causes of death categories for select diseases.  "Related" refers to the presence of a cause of death code in either the underlying or non-underlying (“contributing”) causes.   For example, "septicemia" is the count of death cases where septicemia appears as the underlying cause of death while "septicemia-related" is the count of death cases where septicemia appears in any of the causes of death listed on the death certificate (both underlying and contributing).

Mortality Statistics:

For 1999 to present, all rate calculations use the 2000 standard million population to age-adjust the mortality rates.  Reports before 1999 used the 1940 standard million population.  Age-adjusted rates before 1999 are not directly comparable to rates for 1999 or later, due to the change from the 1940 to the 2000 standard reference population as well as the change from ICD-9 to ICD-10 coding systems.

Number of deaths is the total number of deaths registered with the State of Connecticut where age at death was known.  The total death counts in these mortality tables may be slightly lower than the total counts in Connecticut's Registration Reports since deaths with unknown age are not included in these age-related calculations.

Crude death rate is a measure of the number of deaths in a population scaled to the size of that population per unit time.  The rate is calculated by dividing the number of deaths in a population in a year by the midyear resident population. Crude mortality rates in our tables are expressed in units of deaths per 100,000 individuals per year; thus, 95 deaths per year in a population of one million would correspond to a death rate of 9.5 per 100,000 persons.   When a rate is restricted to deaths in specific age, the rate is known as an age-specific rate. (NCHS Data Definitions: Crude Death Rate)

Age-adjusted mortality rates (AAMR; also known as age-adjusted death rates) are rates where the effect of differing age distributions between the groups has been removed.  They are used to compare the relative mortality risk across two or more population groups at the same point in time or to compare one population at two or more points in time.  Since the effect of age has been removed, these rates are called "age-adjusted" rates. This is a key difference between crude and age-adjusted rates.  More specifically, the adjusted rate estimates “what the crude rate would have been in the study population if that population had the same distribution as the standard population with respect to the variable(s) for which the adjustment or standardization was carried out” (Last, 1988).  Age-adjusted rates are computed by the direct method by applying age-specific rates in a population of interest to a standardized age distribution, in order to eliminate differences in observed rates that result from age differences in population composition.  Age-adjusted rates presented in the CT DPH Mortality tables are consistent with the methods used by the National Center for Health Statistics/Centers for Disease Control in their tabulation of U.S. rates. (NCHS Data Definitions: Age Adjustment)

AAMRs are widely used and provide a convenient summary measure of mortality for whole populations.   Nevertheless, they are estimates and are subject to certain limitations.  AAMRs are weighted statistical averages of the age-specific death rates, in which the weights represent the fixed population proportions by age (Murphy 2000). It is often assumed that groups whose AAMRs differ will have a similar pattern of differences across all age groups.   We recommend that users also check the age-specific rates underlying the reported AAMR differences between groups to determine whether the differences are relatively homogeneous across age groups or not.  Age-specific rates are available upon request.  Additional discussion about the relative strengths and weaknesses of age-standardization are available (see Janes 2000, p. 133).

Years of potential life lost (YPLL) is a measure of premature mortality.  It represents the number of years of potential life lost by each death before a predetermined end point (e.g., 65 or 75 years of age).  For example, the death of a person 15-24 years of age counts as 55.5 years of life lost.  The YPLL statistic is derived by summing age-specific years of life lost figures over all age groups up to 65 or 75 years.  YPLL is presented for persons less than 75 years of age because the average life expectancy in the United States is over 75 years.  For Connecticut, we also provide YPLL for persons less than 65 years of age.  (NCHS Data Definitions: YPLL)

Whereas the crude and adjusted death rates are heavily influenced by the large number of deaths among the elderly, the YPLL measure provides a picture of premature mortality by weighting deaths that occur at younger ages more heavily than those occurring at older ages. It thereby emphasizes different causes of death. For Connecticut, years of potential life lost are summarized both as the total years of life lost and as an age-adjusted YPLL rate.  Age-adjusted YPLLs are calculated using the methodology of Romeder and McWhinnie (1977).

References:

• Murphy, S.L. 2000. Deaths: Final data for 1998. National Vital Statistics Reports 48(11).
• Hyattsville, MD: National Center for Health Statistics. DHHS Publication No. (PHS) 2000-1120.
• Janes GR, et al. Descriptive epidemiology: analyzing and interpreting surveillance data.  In:  Teutsch SM, Churchill RE, editors. Principles and practice of public health surveillance. 2nd ed. New York: Oxford University Press; 2000.
• Last, J.M. (ed.) A dictionary of epidemiology.  Second edition. New York: Oxford University Press, 1988.
• Romeder, J.M. and McWhinnie, J.R. 1977. Potential Years of Life Lost between ages 1 and 70: An indicator of premature mortality for health planning. International Journal of Epidemiology 6:143-151.
• World Health Organization. 1992. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death, based on the recommendations of the Tenth Revision Conference, 1992. Geneva: World Health Organization.