Sensitivity of Disaster-Related Mortality Surveillance after Hurricane Sandy — New Jersey, 2012–2013

BACKGROUND: Hurricane Sandy made landfall in New Jersey on October 29, 2012. We assessed the sensitivity of the state’s mortality surveillance methodology for identifying storm-related deaths and evaluated whether New Jersey’s electronic death registration system (EDRS) could be used to improve mortality surveillance in the future.

METHODS: During the 3 weeks immediately following the storm, the Office of the State Medical Examiner (OSME) generated a line list of all storm-related deaths based upon reports from county and regional medical examiners (MEs). OSME subsequently updated its line list to include Sandy-related deaths identified through review of applications for funeral benefits submitted by families of victims to the Federal Emergency Management Agency (FEMA). We compared OSME’s original line list of Sandy-related deaths with its updated line list. In addition, we obtained death certificates from EDRS for all deaths on the updated line list and searched for specific keywords indicating disaster-relatedness (e.g., storm, disaster, Sandy, etc.).

RESULTS: OSME identified 36 Sandy-related deaths based upon reports from county and regional MEs; this number subsequently increased to 75 through review of FEMA funeral benefit applications. Therefore, the sensitivity of the surveillance system was 36/75 (48%). A keyword search of the 75 death certificates from EDRS revealed that 24/75 (32%) contained ≥1 keyword indicating Sandy-relatedness.

CONCLUSIONS: New Jersey’s mortality surveillance after Hurricane Sandy relied on reports about disaster-related deaths from county and regional MEs; this method was not sensitive for detecting Sandy-related deaths. New Jersey’s EDRS detected even fewer deaths on OSME’s line list. To improve disaster mortality surveillance using EDRS, we recommend training for health care providers and MEs on how to indicate disaster-relatedness on a death certificate.