Los angeles county death certificate

Historically, efforts to assess disease and injury burden have been hampered by the lack of a single measure of burden that accounts for both morbidity and mortality. To address this limitation at the international level, the Disability-Adjusted Life Years (DALYs) measure was developed a decade ago to assess global patterns of disease and injury burden and to project future trends. (1,2) DALYs combine the impacts of premature mortality and disability associated with various health conditions. One DALY equates to one year of healthy life lost. DALYs have been used to measure disease burden in both developed and developing countries (3-5) and to assess the allocation of National Institutes of Health (NIH) funding in the United States. (6)

In this report, we describe the use of DALYs to assess the burden of disease and injury in a large urban population, compare the results to those of more traditional measures of disease burden (crude mortality and premature mortality), and discuss the potential benefits and limitations of using the DALYs methodology at the local health department level to assess and prioritize population health needs.

METHODS

DALYs measure population-level disease burden based on mortality and disability rates. They represent the sum of (a) the number of Years of Life Lost (YLLs) to premature mortality, summed across a given population, and (b) the number of Years Lived with Disability (YLDs), adjusted for level of disability and summed across the given population. Because one purpose of our study was to demonstrate that DALYs can be used domestically for assessing disease burden, we calculated DALYs for the Los Angeles County population in 1997 using the same methods as in the Global Burden of Disease (GBD) Study. (2)

Calculation of YLLs

The YLL component was calculated using morality rates derived from publicly available death records. All 1997 deaths records for individuals residing in Los Angeles County (N = 60,072) were eligible for inclusion in the analysis. Two hundred eighty-six (0.5%) were excluded because of insufficient information on the death certificate. The remaining 59,786 reports were grouped by underlying cause of death into 105 disease and injury categories as defined in the GBD Study. (2) The underlying cause of death is denoted using International Classification of Diseases, Ninth Revision (ICD-9) codes.

Standard life tables for males and females from the GBD study (indicating remaining life expectancy at each year of age) were then used to calculate YLLs. (2) These tables are based on different data sources for men and women. For women, the standard life tables use the maximum life expectancy observed internationally, i.e., 82.5 years for Japanese women. For men, the standard life tables are based on an ideal life expectancy (80.0 years) developed using models of the true biological male/female difference in life expectancy, controlling for other risk factors. (7)

YLLs were calculated for each disease and injury category as the difference between life expectancy and age at death in months, summed across the total county population, and by gender and race/ethnicity (i.e., African American, American Indian/Alaska Native, Asian/Pacific Islander, Latino, and white). We stratified our analyses according to race/ethnicity to examine potential disparities in disease burden. Data on race/ ethnicity were obtained directly from the death records, which are subject to misclassification. (8)

Calculation of YLDs

Calculation of YLDs requires estimates of disease and injury incidence, the frequency of associated disabilities, and the average duration and severity of the disabilities. (7) For the GBD Study, disease and disability incidence and duration for a range of conditions were estimated based on an exhaustive review of the published and unpublished literature and input from experts in epidemiology. (2) The severity of disabilities was quantified in the GBD study using disability weights, ranging from 0 (perfect health) to 1 (death), developed by an international panel of health care providers? For example, treated angina pectoris has a disability weight of 0.095 for all ages, whereas untreated unipolar major depression has a weight of 0.600 for all ages.

Because of the relative lack of disease incidence and disability data for the Los Angeles County population, we imputed disease and disability rates and assigned disability weights using the same methodology as in the GBD study, but with updated rates. To estimate YLDs in the county population. (2) we used updated age- and gender-specific YLD-to-YLL ratios and YLD rates for each disease and injury category for Established Market Economy (EME) countries obtained from the Harvard University School of Public Health Burden of Disease Unit (Personal communication, Catherine Michaud, MD, PhD, January 1999).

YLDs for each age/gender stratum in the county population were estimated in one of two ways: (1) if the EME YLD-to-YLL ratio was stable (defined as <10 in the GBD study), this ratio was multiplied by the county's YLLs to determine the county's YLDs; or (2) if the EME YLD-to-YLL ratio was unstable because of low mortality (defined as [greater than or equal to] 10 in the GBD study), the EME YLD rate was used to determine the county's YLDs by applying this rate to the county population estimated by the Census Bureau for 1997. (2) These YLD calculations were done for each age (<1, 1, 2-4, 5-9, 10-14, ... 80-84, [greater than or equal to] 85) and gender stratum and then summed across strata to obtain totals for each disease and injury category. We could not calculate separate YLDs by race/ ethnicity because we did not have comprehensive data on disease and disability rates by race/ethnicity.

Calculation of DALYs

Prior to summing the YLLs and YLDs within each disease and injury category, we applied the same age-weighting and discounting factors used in the GBD study to each age and gender stratum. (2) Age-weighting quantifies the perception that a year of healthy life has greater social value during early adulthood than during earlier or later life. Therefore, younger and older age categories are weighted somewhat less than 1.0, while age categories during early working adulthood are weighted greater than 1.0. The discounting factors account for the perception that a current year is worth more than a future year of healthy life and that future years need to be expressed in present value terms. YLLs and YLDs that extended beyond the current year into the future, therefore, were discounted, i.e., divided by [1.03.sup.n], where n is the number of years in the future, to produce a present value for YLLs and YLDs. This same calculation is used to calculate the present value of costs and benefits in cost-effectiveness studies. (9) We examined the sensitivity of our final results, as presented below, to removal of both age-weighting and discounting, and found no substantial differences in the findings.

After applying the age-weighting and discounting factors to each age and gender stratum, we summed the YLLs and YLDs within the 105 disease and injury categories to produce DALYs.

Rate calculations

Rates per 1,000 population were calculated for YLLs, YLDs, and DALYs by gender and race/ethnicity using 1997 county population projections based on the 1990 Census. (10) Rates were age-adjusted to the 1990 U.S. population.

RESULTS

Table 1 shows the 15 leading causes of disease and injury burden in the county population based on measures of crude mortality (number of deaths), premature mortality (YLLs), and DALYs. Although ischemic heart disease led the list for all three measures, there was considerable variation in the rank order of the other conditions. For example, alcohol dependence was 2nd on the DALYs list but only 29th based on YLLs and 39th based on crude mortality. Similarly, depression and osteoarthritis were among the leading causes of DALYs but accounted for very little burden based on crude mortality and YLLs. Conversely, pneumonia was 4th on the crude mortality list but dropped to 12th based on YLLs and 21st based on DALYs.

The 10 leading causes of DALYs by gender are shown in Table 2. Ischemic heart disease led the list in both males and females. However, homicide/other violence, drug overdose/other intoxication, motor vehicle crashes, and HI-V/AIDS accounted for higher percentages of DALYs in males than females. Diabetes, stroke, Alzheimer's/other dementias, and breast cancer accounted for higher percentages in females than males. Alcohol dependence, depression, and osteoarthritis were among the leading causes of DALYs in both groups.