The independent and synergistic impacts of power outages and floods on hospital admissions for multiple diseases

Abstract

Highly destructive disasters such as floods and power outages (PO) are becoming more commonplace in the U.S. Few studies examine the effects of floods and PO on health, and no studies examine the synergistic effects of PO and floods, which are increasingly co-occurring events. We examined the independent and synergistic impacts of PO and floods on cardiovascular diseases, chronic respiratory diseases, respiratory infections, and food-/water-borne diseases (FWBD) in New York State (NYS) from 2002 to 2018. We obtained hospitalization data from the NYS discharge database, PO data from the NYS Department of Public Service, and floods events from NOAA. Distributed lag nonlinear models were used to evaluate the PO/floods-health association while controlling for time-varying confounders. We identified significant increased health risks associated with both the independent effects from PO and floods, and their synergistic effects. Generally, the Rate Ratios (RRs) for the co-occurrence of PO and floods were the highest, followed by PO alone, and then floods alone, especially when PO coverage is >75th percentile of its distribution (1.72% PO coverage). For PO and floods combined, immediate effects (lag 0) were observed for chronic respiratory diseases (RR:1.58, 95%CI: 1.24, 2.00) and FWBD (RR:3.02, 95%CI: 1.60, 5.69), but delayed effects were found for cardiovascular diseases (lag 3, RR:1.13, 95%CI: 1.03, 1.24) and respiratory infections (lag 6, RR:1.85, 95%CI: 1.35, 2.53). The risk association was slightly stronger among females, whites, older adults, and uninsured people but not statistically significant. Improving power system resiliency could be a very effective way to alleviate the burden on hospitals during co-occurring floods. We conclude that PO and floods have independently and jointly led to increased hospitalization for multiple diseases, and more research is needed to confirm our findings.

Keywords: Cardiovascular diseases; Chronic respiratory diseases; Floods; Food-/water-borne diseases; Power outage; Respiratory infections.

Fig. 1.

The annual variation of the number of division days for floods and PO in NYS, 2002–2018. Division days were defined as how many days of Floods or PO in a specific division.

Fig. 2.

Spatial distribution for floods and PO at the power-operating division level in NYS, 2002–2018. Panel A is for floods; panel B is for PO. Division days were defined as how many days of Floods or PO in a specific division.

Fig. 3.

Rate ratios of hospital admissions of CVD (A), chronic respiratory diseases (B), respiratory infection diseases (C), and FWBD (D) due to joint occurrence of floods and PO in different quantile coverages by lag days in NYS, 2002–2018.