Abstract
BACKGROUND: In the early 1990’s, the Cooperative Study of Sickle Cell Disease (CSSCD) estimated a median life expectancy of 42 years for males, and 48 years for females with sickle cell anemia. We used death certificate data from the late 1990’s and early 2000’s to examine age at death and contributing causes of death for persons with sickle cell disease (SCD).
METHODS: We used the National Center for Health Statistics Multiple Cause of Death (MCOD) files to examine age at death and contributing causes of death for persons in the U.S. with SCD during the years 1999 to 2004. The MCOD files contain data from all death certificates filed in the U.S. Each observation in the data has listed an underlying (primary) cause of death, as well as up to 20 conditions thought to contribute to the death. We used ICD-10 codes D570-D578 to identify all deaths attributed to SCD during the time period under study. Records with the ICD-10 code for sickle cell trait (D573) were excluded from further analyses. We used the Clinical Classification Software provided by the Healthcare Cost and Utilization Project to collapse all listed ICD-10 codes into smaller categories. Analyses of age at death were conducted using t-tests, median tests, ANOVA, and multiple linear regression as appropriate.
RESULTS: From 1999 to 2004, there were 4553 deaths in the U.S. attributed to SCD (mean = 759/yr, sd = 42.6). SCD was listed as the primary cause in 65% of the deaths. 95% of the deaths were attributed to HbSS disease, and approximately 1% of the deaths were attributed to double heterozygous sickle cell disorders (SC/SD/SE/Thal). 50.4% of the deaths were among males. 64% of the decedents had a high school education or less. 54% of the decedents lived in the South. 68% of the decedents died as inpatients in a hospital. The mean age at death for the time period was 38.2 years (sd = 15.6). There was no change in the mean age at death during the time period. Females were older than males at death (39.4 vs. 36.9, p < 0.0001). Those with HbSS were younger than those with a double heterozygous disorder (38 vs. 47, p < 0.02). Having SCD listed as the primary cause of death was associated with younger age at death (36.8 vs. 40.7, p < 0.0001). Decedents with at least some college education were older at death than those with high school educations or less (40.9 vs. 37.0 p < 0.0001). There were no regional differences in mean age at death. In a multivariate model of age at death with the predictors gender, region, education, and whether or not SCD was listed as the primary cause of death, being female and having some college education remained associated with older age at death, while having SCD listed as the primary cause of death remained associated with younger age at death. Septicemia, pulmonary heart disease, liver disease and renal failure were among the top contributing causes of death for adults, while septicemia, acute cerebrovascular disease and pneumonia were among the top contributing causes of death for kids.
CONCLUSIONS: Persons dying from SCD during 1999 to 2004 experienced ages at death that are not improved over those reported by the CSSCD, suggesting the continued need for societal efforts aimed at improving the quality of care for SCD, especially among adults with the condition. Educational attainment is associated with age at death among the SCD population, though it is not possible from the cross-sectional nature of this data to determine the causal directionality of this association.
Mortality Cause in Goldsmith Workers of Bangladesh: Findings From Verbal Autopsy