To evaluate key attributes, strengths, and limitations of the American Red Cross (ARC) disaster-related mortality surveillance system implemented during Hurricane Ike in Texas 2008, and to provide recommendations for system improvement.

We evaluated key attributes of the ARC mortality surveillance system. Evaluation included interviews with stakeholders and linking ARC data with the Texas Department of State Health Services’ (DSHS) system for comparison.

During September 11 through October 6, 2008, the ARC identified 38 deaths, whereas DSHS identified 74 deaths related to Hurricane Ike (sensitivity = 47%; positive predictive value = 92%). The ARC had complete data on 61% to 92% of deaths, and an 83% to 97% concordance was observed between the 2 systems for key variables.

The ARC surveillance system is simple, flexible, and stable. We recommend establishing written guidelines to improve data quality and representativeness. As an important supporting agency in disaster situations and the sole source of data regarding disaster-related mortality in multiple states, improvement of the ARC system will benefit stakeholders and promote dissemination of useful information for preventing future deaths. (Disaster Med Public Health Preparedness. 2013;7:13-19)

The Texas Department of State Health Services (DSHS) implemented an active mortality surveillance system to enumerate and characterize hurricane-related deaths during Hurricane Ike in 2008. This surveillance system used established guidelines and case definitions to categorize deaths as directly, indirectly, and possibly related to Hurricane Ike.

The objective of this study was to evaluate Texas DSHS' active mortality surveillance system using US Centers for Disease Control and Prevention's (CDC) surveillance system evaluation guidelines.

Using CDC's Updated Guidelines for Surveillance System Evaluation, the active mortality surveillance system of the Texas DSHS was evaluated. Data from the active mortality surveillance system were compared with Texas vital statistics data for the same time period to estimate the completeness of reported disaster-related deaths.

From September 8 through October 13, 2008, medical examiners (MEs) and Justices of the Peace (JPs) in 44 affected counties reported deaths daily by using a one-page, standardized mortality form. The active mortality surveillance system identified 74 hurricane-related deaths, whereas a review of vital statistics data revealed only four deaths that were hurricane-related. The average time of reporting a death by active mortality surveillance and vital statistics was 14 days and 16 days, respectively.

Texas's active mortality surveillance system successfully identified hurricane-related deaths. Evaluation of the active mortality surveillance system suggested that it is necessary to collect detailed and representative mortality data during a hurricane because vital statistics do not capture sufficient information to identify whether deaths are hurricane-related. The results from this evaluation will help improve active mortality surveillance during hurricanes which, in turn, will enhance preparedness and response plans and identify public health interventions to reduce future hurricane-related mortality rates.

Choudhary E, Zane DF, Beasley C, Jones R, Rey A, Noe RS, Martin C, Wolkin AF, Bayleyegn TM. Evaluation of active mortality surveillance system data for monitoring hurricane-related deaths, Texas, 2008. Prehosp Disaster Med. 2012;27(4):1-6.