![[ Visit Client Website ]](images/banner.png)
BACKGROUND: Maricopa County Department of Public Health (MCDPH) has conducted heat-associated death (HAD) surveillance since 2006. The main data sources for HAD are weekly line lists from the Office of Medical Examiner (OME) and death certificates (DC). MCDPH’s priority is to obtain complete information on activity prior to death, air-conditioning status, and decedent’s living situation. This information is often missing from the DC and the OME list. Therefore, the goal of this project was to review OME supporting documentation for additional information to determine which provides the most comprehensive data for surveillance.
METHODS: A retrospective chart review of all supporting OME documentation (preliminary report of death (PROD); autopsy report; toxicology report) was conducted for 106 HAD in 2011. The data collected from these sources were evaluated for completeness and accessibility, and compared to previous data collection methods (DC, OME list). If various documents provided information on the same variable, accessibility of the document determined which one would be used. Percent of completeness was calculated for every variable in each document. Accessibility was defined by data transmission (electronic stream, chart review) and formats (narrative, variable). A total of 223 variables were analyzed; several variables were excluded because they were not of interest for HAD surveillance or had a completion rate of less than 10%.
RESULTS: Based on the inclusion criteria, 97 variables were retained. The enhanced surveillance methodology included additional variables that were easy to access and complete (living situation, activity prior to death, medical history, alcohol or drug use at the time of death). Furthermore, some incomplete variables were enhanced as a result of the chart review; for example, air-conditioning status improved from 39% to 98% complete. For excluded variables, it was found that the autopsy and toxicology reports had many variables that were not useful for HAD surveillance such as lab, microbiology and histology values. Symptomatology (dizziness) and health status (home-bound) were examples of incomplete variables that were excluded.
CONCLUSIONS: Upon review, it was determined that the PROD in conjunction with the DC and OME list provided the most complete information for HAD surveillance. HAD surveillance was substantially improved by providing a comprehensive picture of the circumstances surrounding heat injury and death; this will improve the efficiency of the system by reducing time spent on data collection. However, the usefulness of additional data sources should be explored to provide a better understanding of the risk factors and circumstances of HAD.