For a better experience, click the Compatibility Mode icon above to turn off Compatibility Mode, which is only for viewing older websites.

Heat and Infant Mortality in Philadelphia

Investigating whether high ambient temperatures are associated with increased risk of infant mortality

Woman holding infant


High ambient temperatures are associated with increased risk of mortality. Heat represents an important public health concern within cities, which are subject to the urban heat island effect. Lower socioeconomic and racial or ethnic minority groups have been shown to be particularly vulnerable to heat. Variations in heat-related vulnerability may be related to microclimates within cities, created by built environment characteristics, such as vegetation and tree canopy, which might also vary along sociodemographic lines. In addition to place-based variability, the adverse consequences of high heat are a particular concern for physiologically vulnerable subpopulations, including infants. Philadelphia, PA is a prime example of a city that is subject to the urban heat island effect. This pilot study investigates whether high ambient temperatures are associated with increased risk of infant mortality, and if the association is stronger in poorer, minority areas with less vegetation or tree canopy.

Research Methods

The research study uses information from a number of data sources to conduct statistical analyses to address study aims. The study population consists of any infant who died in Philadelphia between May 1 and September 30 during years 2000-2015. Death certificates will be used to determine the occurrence and dates of infant deaths, and to establish the infants' home addresses at the time of their death. Average daily temperature data is being linked to each infant's date of death. In addition, each home address is being assigned a geocode (i.e. a latitude and longitude) to link census tract level sociodemographic information, and to determine the amount of vegetation and tree canopy near each infant's home.

Statistical models will be used to compare the temperature on the days that infants died to the temperature on "control" days that were close in time to date of death. Analyses will be stratified by categories of census tract level sociodemographic characteristics, and by categories of vegetation (eg. low versus high). It is predicted that, on average, temperatures on the day an infant died will be higher than temperatures on comparison days. Researchers also hypothesized that these associations are stronger among infants living in census tracts with higher percentages of residents living below the poverty line, with less than a high school diploma, and who are black/African American. Lastly, these associations are expected to be stronger among infants who had less tree canopy and less vegetation near their homes.


Research Team

This research was supported by Urban Health Collaborative pilot funding in 2016.