Mortality among California’s sickle cell data collection cohort, 2005-2015 | oneSCDvoice
  • Join Today!

Become a member and connect with:

  • An Active Online Community
  • Articles and Advice on SCD
  • Help Understanding Clinical Trials
abstracts & posters

Mortality among California’s sickle cell data collection cohort, 2005-2015

key information

source: American Society of Hematology

year: 2017

authors: Susan T Paulukonis, Robert Ward Hagar


Information about age of death, cause of death, and related factors among those living with sickle cell disease (SCD) is typically limited to data collected from a clinical setting or from single source administrative data. These methods omit individuals who did not receive care in settings that collect and publish data on their patients with SCD, or patients without SCD listed as a cause of death, respectively. CDC has implemented the Sickle Cell Data Collection (SCDC) program in California and Georgia to establish true population-based surveillance systems at the state level in an effort to address such gaps in knowledge. These data allow investigation of healthcare utilization and outcomes for patients with SCD who receive care in all settings in these states. Hospital admissions, emergency department (ED), and Medicaid claims data for the years 2004 through 2015 were obtained by the SCDC programs in both states. Patients with a SCD International Classification of Diseases (ICD) 10 code having three or more healthcare encounters within any five-year period are included in the cohort presented here; this group was linked by social security number and date of birth to the state’s death records for the same period to identify deaths. The most common causes of death (COD) were determined using the underlying COD variable and categorized based on the ICD 10 broad coding scheme. SCD COD in the death record was determined by a search of underlying COD (ICD 10 codes). In death records after 2010, additional COD were included as text strings; these were examined for SCD specific COD 2011-2015. There were 5,248 persons meeting the case definition of SCD identified in the hospital, ED, and claims data during the 12-year time frame. Of these, matching death records were found for 543 of these patients. The mean age of death for the SCD patients was 44.2 years (median 45.7 years; SD 16.0 years). As shown in Table 1, 2% percent of deaths were among children under 10 years of age, and 5.5% were among those older than 10 and less than twenty years of age. Most deaths occurred in an inpatient setting (60%), followed by home (16%), and ED/outpatient setting (15%). Similar to previous reports, SCD ICD 10 codes or text strings for additional COD or other conditions were present in only 58% (n = 160 of 275 deaths) of death reports for years 2011-2015. The ICD 10 codes for SCD were present as the underlying COD in only 36% of death reports (all years). Circulatory-related COD, which can often include terminal events such as heart failure, made up 20% of death underlying causes. Death due to external injuries or events accounted for 9% of all deaths. SCD is not reliably coded in death certificates, making mortality and associated conditions challenging to monitor at a population level. Multi-source data collection allows for population based surveillance of health outcomes for those with SCD.

organisation: Public Health Institute, Alameda, CA; UCSF Benioff Children's Hospital Oakland, Oakland, CA

read more