![[ Visit Client Website ]](images/banner.png)
BACKGROUND: The 122 Cities Mortality Reporting System (CMRS) provides timely pneumonia and influenza (P&I) mortality data, but deaths are counted for the week the death certificate was filed rather than the week of death. Also, the system only covers approximately 25% of the U.S. population and not all sites follow the standard case definition. The National Center for Health Statistics (NCHS) collects data from state vital statistics offices regarding all deaths. These data are considered the gold standard, but previously only final data were released resulting in a two year lag. We assessed real-time use of NCHS data for timeliness and reliability to determine its feasibility as an alternative to 122 CMRS.
METHODS: For NCHS data a P&I death is one with any mention of an ICD-10 multiple-cause code for pneumonia or influenza. We used NCHS data to retrospectively analyze all deaths that occurred between January 2011 and July 2013 with a completed record at the time of analysis. A record was considered complete on the date that fact of death, cause of death, and all coding was completed. Timeliness was assessed as the number of weeks from death occurrence to record completion. We assessed data reliability by determining how long after the week of death occurrence, the P&I death ratio was within 10% of the ratio once data was finalized.
RESULTS: During 2011-2013, 360,215 (77.6%) of 463,991 deaths had a known completion date. The average time from the occurrence of death to record completion was 12.4 weeks (range: 0-98 weeks). The lag improved from 16 weeks in 2011 to 6 weeks in 2013. A P&I ratio within 10% of the final ratio was obtained within 2 weeks of death occurrence for 83% of weeks and within 4 weeks for 91% of weeks.
CONCLUSIONS: NCHS data can produce a reliable P&I ratio approximately two weeks after the week of death, similar to the timeliness of the 122 CMRS. Given the increase in timeliness seen from 2011 to 2013, we think NCHS data could potentially be a replacement for the 122 CMRS. NCHS data would monitor deaths by date of death, provide 100% coverage, and use a standard case definition. Additionally, NCHS and its partners are actively working to improve timeliness by funding projects to increase the use of Electronic Death Registration systems. We will monitor both systems throughout the current influenza season to assess the feasibility of using NCHS mortality data for surveillance.