Impacts of Climate Change and Extreme Weather on Injury: A Primer for Investigation Focusing on Hurricane-Related Impacts

Climate change and extreme weather can lead to injury-related morbidity and mortality through direct and indirect pathways, and it is imperative that we thoroughly investigate these impacts for public health planning, preparedness, and response activities. Injury can be defined as bodily harm from external forces or substances (including mechanical, thermal, electrical, chemical, or radiant exposures) or submersions. Climate-sensitive injuries can be further categorized by the type of impact. Specifically, climate and weather impacts can be acute or sustained events. Acute impacts are characterized by severe and sudden onset with typically shorter duration, and include heat or cold waves, wildfire, flooding, and storms. Sustained impacts are characterized by persistent or long-term effects that develop over longer periods, and include drought, more extreme temperature normal, and sea level rise.

A variety of data sources and epidemiologic methods are available to investigate these associations; though, each come with their own set of challenges. From the more obvious data sources such as vital statistics and hospitalization and emergency department visit data, to the lesser used sources, such as poison control center and syndromic surveillance data, these data have the potential to provide a comprehensive assessment of climate-sensitive injuries. Study designs commonly used in injury-related investigations include case-crossover, time series analyses, and other ecological designs. However, researchers must be aware of the challenges associated with each data source and design to ensure unbiased and externally valid associations. A data-related example includes the recent transition in International Classification of Diseases (ICD) coding structure from version 9 to 10, and how this affects the sensitivity and specificity of injury-related health outcomes. As for study designs, each has a specific set of assumptions or requirements (e.g., unique individuals must be identifiable in case-crossover analyses; mean, variance, and autocorrelation function must be constant over time in time series analyses), which may or may not be supported by the specific data sources available.

To further demonstrate these issues, we use hurricane-related injury as a case study. Morbidity and mortality associated with this acute climate impact may be both direct (injuries that occur immediately due to winds and storm surge) and indirect (injuries related to clean-up activities and damage to infrastructure). Additionally, these injuries may occur prior to the event (during evacuation or preparedness activities), during the event, or in the aftermath (during clean-up and response activities). Common types of injuries associated with hurricanes include drowning, unintentional trauma, electrocution, and carbon monoxide poisoning. Available data sources include vital statistics registries, hospitalization and emergency department visit data, syndromic surveillance, and poison center call data. Investigations have demonstrated significant increased risk of all-cause unintentional injury and carbon monoxide poisoning related to hurricane impacts.

In this presentation, specifically using hurricane-related injury, we will further clarify the causal pathway between acute and sustained climate impacts on injury-related morbidity and mortality; outline the numerous health data sources available for investigating these associations, including strengths, limitations, and specific issues in defining climate-sensitive injuries; identify specific epidemiologic and statistical methods appropriate for such investigations; and propose future research topics that are necessary for successful public health planning and mitigation activities.