Accurate ascertainment of cause of death (COD) is needed to ensure validity of published data. Two systems in the US collect data on deaths related to vaccine-preventable diseases (VPD): the National Center for Health Statistics (NCHS) and the National Notifiable Disease Surveillance System. CSTE has endorsed validation of VPD-related rare deaths (01-ID-09, “Validation of Data Representing Deaths Due to Vaccine-Preventable Diseases”).
METHODS:
The National Center for Immunization and Respiratory Diseases (NCIRD) and NCHS established protocols in 2012 for validation of rare cause VPD death reports, changing to an “opt-out” design in 2014, with no states currently opting out. Under this protocol, once NCHS receives the death record from the state and codes the COD, reports of specified rare VPD deaths are provided to NCIRD for investigation with the appropriate state epidemiology and/or immunization offices. Rare VPD identified as underlying or contributing COD are included in the validation protocol, but only underlying COD are used in standard NCHS reports. Data for this report includes deaths occurring July 15, 2012 through July 15, 2015.
RESULTS:
Death reports were received for this validation process from 33 states. Of 115 death reports received (61 with the VPD listed as the underlying COD), 50 (36 underlying COD) were validated by NCIRD and state epidemiology/immunization programs, 56 (17 underlying COD) were identified by NCIRD as having a cause of death (COD) other than the VPD, and 9 (8 underlying) could not be investigated. Among 47 varicella deaths reported, 6 were validated as varicella, 33 were determined to have another COD, and 8 were not investigated. Among 21 pertussis reports, 17 were validated as pertussis and 4 were determined to have another COD. Of 9 mumps deaths reported, 1 was validated, 7 were determined to have another COD, and 1 was not investigated. The 5 deaths reported as measles were determined to have another COD. All 3 deaths reported as SSPE were validated. The 4 rubella deaths, the 2 polio deaths, and the 1 diphtheria death reported were found to have another COD. All 23 CRS death reports were validated.
CONCLUSIONS:
Rare-cause validation process requires cooperation of state vital records offices, COD certifiers, and NCHS staff who quickly identify deaths from rare causes. In addition, investigation of death reports by NCIRD and state epidemiology and/or immunization programs is required to make the national death surveillance system more accurate and usable for monitoring rare deaths due to VPDs.