The effects of power outages on cardiovascular outcomes in New York State

In 2020, U.S. electricity customers faced an average of 8 hours without power, the longest duration on record. Severe weather events can damage electricity grids and distribution pathways. Outages may be particularly problematic for cardiovascular health, as they disrupt access to temperature-controlling tools (e.g., air conditioners, fans). Higher temperatures can exacerbate cardiovascular disease (CVD). CVD affects 126.9 million American adults and is the leading cause of death and disability in the U.S., with different population groups affected. We seek to understand the impact of power outages on CVD and how this relationship varies across different population groups. A critical research gap remains about the health effects of power outages generally and the extent to which they affect acute CVD risk. Partly due to the lack of high spatiotemporal resolution measures for outage exposure, most prior studies have not measured neighborhood-level electricity availability. Additionally, previous studies did not consider the potential dual threat when outages co-occur with heat and cold or whether the effect of power outages on CVD outcomes varies across populations. In particular, we will consider and compare the CVD effects of power outages across different population groups. We will use 2017-2020 New York Statewide Planning and Research Co-operative System (SPARCS) hospitalization data and 30-minute resolution power outage data within 1,831 power operating units statewide to fill this gap. Using diagnostic codes, we will identify hourly hospitalizations for the acute cardiovascular outcomes, myocardial infarction and stroke. We plan to implement case-crossover study designs with conditional logistic models, which inherently control for non-time-varying individual-level confounding variables. Our overarching hypothesis is that more outage hours are associated with increased risk of CVD hospitalizations, particularly during hot and cold events, and that there will be differences in this exposure-response relationship. We aim to (1) investigate the association between power outages and CVD hospitalizations, (2) determine the impacts of exposure to power outages during hot and cold temperature events separately on CVD hospitalizations, and (3) identify whether the effects of outages on CVD hospitalizations vary across population groups. Collectively, we will provide crucial information that can inform future policies to reduce CVD-related morbidity.