Tropical cyclones, including hurricanes and tropical storms, are linked to increased rates of drug-related deaths up to three months after the storm passes—particularly in higher-income, White communities and among younger populations. The study of more than 30 years of data by researchers at Columbia University Mailman School of Public Health points to one overlooked health impact of climate change, which is leading to more active and severe storms.
The study is the first to quantify the association between tropical cyclone exposure and psychoactive drug–related deaths broadly and through a demographic and socioeconomic framing. Results are published in the journal JAMA Network Open .
Hurricanes and other tropical cyclones can trigger acute psychological distress that leads to substance use as a coping mechanism. They can also disrupt access to health care and substance use treatment, which may be life-threatening for individuals with severe substance-related conditions.
“When we think of the destructive impact of a hurricane or tropical storm, we often think of flooded neighborhoods and deaths by drowning, but there are other hidden harms linked to these storms, such as increases in drug-related deaths. Tropical cyclones are traumatic and disruptive, and for some people, the stress and instability that follow may push them to use drugs and alcohol in dangerous and deadly ways,” says Raenita Spriggs, MPH, a Columbia Mailman School doctoral candidate and the study’s first author.
“Tropical cyclones, which have increased in strength, intensity, duration, and activity over recent decades, may exacerbate the ongoing crisis of drug overdoses. It is critically important that policymakers and public health authorities integrate substance use and mental health services into climate disaster preparedness and response planning,” says Robbie M. Parks, PhD, assistant professor of environmental health sciences at Columbia Mailman and the study’s senior author.
The researchers found that the effect of tropical cyclones on drug-related death rates was more pronounced in wealthy, White communities. In contrast, in lower income communities of color, the associations were largely null, with some estimates suggesting reduced risk. To explain the difference, the researchers speculate that among wealthier, White communities with greater access to prescription drugs, misuse or overdose may rise when health care systems are disrupted and prescriptions become unmonitored. Additionally, when access to tranquilizers like Xanax, stimulants like cocaine, or alcohol are disrupted, individuals in these communities may turn to more dangerous or adulterated substances such as those containing fentanyl. In lower income communities of color, tropical cyclones may reduce individuals’ purchasing power and disrupt any existing connections to illicit drug markets. They also may also be more likely to die from other causes, exacerbated by tropical cyclone exposure.
Age also shaped vulnerability to tropical cyclone-related increases in drug-related deaths, with associations concentrated among younger populations. The largest relative increase was observed among individuals aged 15 to 29 years, among whom each additional day of hurricane exposure was associated with a 30 percent rise in death rates in the month of exposure. Elevated risks were also observed among adolescents and adults aged 15 to 44 years following exposure to all tropical cyclones. The authors note that younger adults may be particularly sensitive to cyclone-related disruptions, including employment instability, caregiving responsibilities, and interruptions in access to healthcare and substance use treatment.
The researchers examined death records from the National Center for Health Statistics (NCHS) between 1988 and 2019 restricted to counties that experienced at least one tropical cyclone during the study period in the mainland U.S. County-level annual population data were obtained from the US Census Bureau and the NCHS bridged-race dataset for 1988 to 2019. Data on tropical cyclones was modelled by co-authors based on wind field modelling and data obtained from the National Oceanic and Atmospheric Administration. Cyclone exposure was defined as the monthly number of days when a county experienced sustained winds of at least 34 knots. Psychoactive drug–related deaths were defined to include poisoning by and exposure to noxious substances (overdoses), mental and behavioral disorders due to psychoactive substance use, and alcohol-induced deaths.
They found that each additional tropical cyclone-exposed day per month was associated with a 3.84 percent increase in death rates in the month of exposure and a 3.76 percent increase in the month following exposure. Elevated risks persisted for months after exposure, with a 2.39 percent increase three months later. Although estimates for hurricanes alone were not statistically significant, the magnitude of the association was notable: each additional hurricane-exposed day was linked to a 7 percent increase in death rates in the month of exposure.
Across the 31-year study period and all exposed counties, the researchers estimated 1,235 excess psychoactive drug-related deaths associated with tropical cyclone exposure, an average of approximately 40 excess deaths per year. Excess deaths were calculated using nationally representative 2019 age-standardized death rates and county population sizes, with estimates conservatively limited to statistically significant storm impacts and accounting for how often counties were exposed to tropical cyclones.
“For every person who dies from drug-related causes in the days and weeks after a tropical cyclone, there are many others who may see other health-related impacts that accumulate over years. We haven’t begun to explore those impacts that occur further down the line, say a year or five years after a tropical cyclone. That’s something we’d like to investigate next,” Spriggs notes.
In earlier research, Robbie M. Parks published a study on how hurricanes are becoming more deadly , finding 83 percent of hurricane-related deaths occurring more recently and 94 percent in more socially vulnerable counties. He has also published research on geographic variation in hurricane evacuation patterns . Other studies examined the links between temperature and hospital visits from alcohol- and substance-related disorders , and interpersonal and self-inflicted violence. Another study identified a lack of attention to mental health in heat action plans .
Additional authors of the current study include: Victoria D. Lynch, Yuanyu Lu, Lincole Jiang, Katherine M. Keyes, Diana Hernández, Anne E. Nigra, Columbia Mailman School; Wil Lieberman-Cribbin, Northwell Health, Great Neck; Brooke Anderson, Colorado State University, Fort Collins; Marianthi-Anna Kioumourtzoglou, Brown University.
This study was supported by grants from the National Institute of Environmental Health Sciences (ES033742, ES009089) and the National Institute on Aging (AG093975).
Keyes reports receiving personal fees for consulting as an expert witness in national prescription opioid litigation during the conduct of the study. Parks reported receiving consultancy fees for work with the International Food Policy Research Institute.
JAMA Network Open
10.1001/jamanetworkopen.2025.60183
Observational study
People
Tropical Cyclone Exposure and Psychoactive Drug-Related Death Reates
20-Feb-2026