Skip to main content
JAMA Network logoLink to JAMA Network
. 2025 Aug 20;334(15):1390–1392. doi: 10.1001/jama.2025.13843

Threats of Weather Disasters for Drug Manufacturing Facilities in the US

Mahnum Shahzad 1,2,, Leticia M Nogueira 3, Anita Wagner 1,2
PMCID: PMC12368787  PMID: 40833740

Abstract

This study assesses the frequency with which climate-related disaster events affected counties with US drug production facilities.


In September 2024, Hurricane Helene triggered a nationwide shortage after hitting a Baxter facility in North Carolina that produces 60% of the country’s intravenous (IV) fluids.1 A similar IV shortage was caused when Hurricane Maria hit Puerto Rico in 2017. Climate change–driven extreme weather events impose new threats to established vulnerabilities in the US drug supply.2 Those threats must be examined to be appropriately mitigated, especially in light of the current administration’s recent executive order and policy proposals seeking to increase domestic pharmaceutical production.3 This study assessed the frequency with which climate-related disaster events affected counties with US drug production facilities.

Methods

We used archived versions of the US Food and Drug Administration (FDA) Drug Establishments Current Registration Site to identify all US-based drug production facilities that manufacture, prepare, propagate, compound, or process drugs distributed in the US from 2019-2024 (see eMethods in the Supplement). Counties with a Presidential Disaster Declaration from 2019 through 2024 were identified from the Federal Emergency Management Agency (FEMA) disaster declaration database. We included climate-related disaster events: fires, hurricanes, storms, tornadoes, and floods.4 We calculated the number of facilities that were in counties impacted by disasters, by type of production activity and by type of disaster over time. We used logistic regression to calculate the relative odds of disaster impact, by whether a county had drug production facilities, with year fixed effects and errors clustered at the county level. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline, and the Harvard Pilgrim Health Care Institute’s institutional review board determined the study to be nonhuman subjects research.

Results

There were 10 861 drug production facilities active from 2019 through 2024, ranging from 5063 in 2023 to 8790 in 2020, when 3860 facilities (43.9%) were in counties with at least one disaster declaration. Cumulatively, in the 6-year period, there were 6819 active facilities (62.8%) in counties when a disaster was declared, an average of 2146 active facilities (33.8%) annually over the study period (Figure 1). Facilities with all types of production activities were in counties with at least one disaster declaration between 2019 and 2024.

Figure 1. Number and Proportion of Drug Production Facilities in US Counties With at Least One Disaster Declaration.

Figure 1.

Manufacturing activities include facilities that list manufacture, active pharmaceutical ingredient manufacture, or positron emission tomography drug production in their list of activities. Analysis activities include facilities that analyze raw materials, active pharmaceutical ingredients, inactive ingredients, and the finished drug product. Packaging activities include facilities that pack, repack, or relabel. Other activities include facilities that list transfill, sterilize, particle size reduction, salvage, or distribution as their activities. “Exposed facilities” refers to facilities located in counties with at least one disaster declaration in a given year. Numbers next to the the bars represent percentages of facilities exposed to disasters

All types of climate-related disasters were declared in at least one county with an active drug production facility during the study period, although hurricanes were most common (Figure 2). Over the study period, there was no statistical difference in the likelihood that counties with and without drug production facilities experienced at least one disaster declaration (28.3% vs 28.4%; incidence rate ratio, 0.99; 95% CI, 0.91-1.08; P = .95).

Figure 2. Distribution of Disaster Declarations in Counties With and Without Drug Production Facilities in the US, 2019-2024.

Figure 2.

Each dot in all the panels represents a Federal Emergency Management Agency Presidential Disaster Declaration in counties with (red) and without (blue) drug production facilities. Storms include severe storms, tornadoes, and tropical storms. Noncontinental US states and territories, including Alaska, Hawaii, and Puerto Rico, are repositioned and rescaled beneath the continental US for clarity.

Discussion

In this nationwide assessment of the frequency with which climate-related disaster events affected counties with US drug production facilities, we determined that nearly two-thirds of all facilities were located in counties that experienced at least one disaster declaration. These disaster events posed risks of disruptions to facilities active in all aspects of the supply chain, from active pharmaceutical ingredient manufacture to packaging, disruptions that could be further magnified by the time needed for FDA reinspection of damaged facilities.

Limitations of this study include a lack of information on production volume, drug types, facility-level contributions to the US drug supply, or the proportion of drugs consumed in the US that are manufactured domestically.5 Furthermore, we used counties as the unit of analysis, which may mask variability in facilities’ vulnerability to disasters within counties. However, FEMA Presidential Disaster Declarations are only issued for major disasters,6 which typically disrupt transportation and resource availability and may indirectly affect production. Finally, data on how frequently disasters resulted in drug shortages are not readily available, hindering a nationwide assessment of weather-induced drug shortages and their impacts.

Drug shortages following disasters demonstrate how the pharmaceutical supply chain is not yet resilient to climate-related disruptions.2 A limited number of facilities manufacture significant shares of key therapeutics, increasing the risk of care disruption due to supply chain disruptions.1 These findings underscore the importance of recognizing climate-related vulnerabilities and the urgent need for supply chain transparency, for strategic allocation of production, and for disaster risk management strategies to prevent health care disruptions in the US.

Supplement 1.

eMethods

eReferences

jama-e2513843-s001.pdf (169.4KB, pdf)
Supplement 2.

Data Sharing Statement

jama-e2513843-s002.pdf (13.3KB, pdf)

References

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eMethods

eReferences

jama-e2513843-s001.pdf (169.4KB, pdf)
Supplement 2.

Data Sharing Statement

jama-e2513843-s002.pdf (13.3KB, pdf)

Articles from JAMA are provided here courtesy of American Medical Association

RESOURCES