Mental Health After Hurricane Helene

Publication Date: 2026

Abstract

Click to View the Research Brief

Hurricane Helene devastated western North Carolina, contributing to unprecedented mental health challenges among community volunteers. This mixed-methods study examined mental health impacts six to eight months post-disaster among 373 volunteers using validated screeners (PTSD-4, GAD-4), exposure assessments, and narrative responses. Over half (>50%) reported at least one poor mental health indicator, 24.4% met all generalized anxiety disorder (GAD) and 5.4% met all post-traumatic stress disorder (PTSD) criteria. Generalized linear mixed models reveal hurricane exposures significantly predicted some mental health indicators. For instance, communication failures (OR: 10.03, 95% CI: 2.29-43.89) and fear for safety (OR: 2.74, 95% CI :1.52-4.95) associated with Hurricane Helene show the strongest associations with the PTSD indicator “poor sleep.” Prior mental health diagnosis consistently predicts adverse outcomes across all indicators (ORs: 1.74-4.74). Unexpectedly, most coping strategies were associated with poorer mental health. However, greenspace engagement (i.e., spending time outside) shows protective effects for both the mental health indicators avoidance (OR: 0.56, 95% CI: 0.31-1.00) and depression (OR: 0.24, 95% CI: 0.06-0.93). Content analysis of open-ended questions (n = 97) revealed persistent trauma, isolation, and infrastructure challenges, balanced by community cohesion themes. Findings inform the novel ReCoVER (Response, Coping, Volunteering, Exposure, Recovery) framework for disaster mental health response. Policy implications include prioritizing communication infrastructure resilience and establishing volunteer support protocols. Rural disaster response requires frameworks accounting for infrastructure vulnerability, spontaneous volunteerism, and limited mental health resources. Future longitudinal research should track recovery trajectories and test targeted interventions, particularly greenspace-based programs.

Introduction

Hurricane Helene, the most destructive tropical storm to hit the mainland United States since Hurricane Katrina in 2005, caused unprecedented flooding and damage in western North Carolina, a region rarely exposed to such extreme climate events (Cooper, 2024¹). The damages inflicted by Helene were compounded by isolated rural geographies and infrastructure failures, including washed-out roads and bridges and substantial damage to cellular towers and power lines. Over 100 people perished in western North Carolina and dozens of individuals were missing weeks after the storm. Community organizations rallied in the days, weeks, and months after the storm to organize relief efforts, deliver desperately needed supplies, help clear debris from flooded homes, assist in spreading reliable information, and support long-term recovery efforts (Cooper, 2024).

This on-the-ground research provides early evidence of the mental health impacts of Hurricane Helene. It addresses gaps in understanding mental health risk and resilience factors among rural volunteers impacted by disasters, particularly given limited communication and resource access, which intensifies mental health stressors. Using quantitatively derived survey data, this research helps develop a transdisciplinary framework that incorporates geographic, social, and resource-limited factors for an updated Substance Abuse and Mental Health Services Administration (SAMHSA) framework that accounts for volunteerism in rural settings. Qualitatively derived content analysis adds context to volunteer experiences following extreme weather events, notably storms (e.g., hurricanes) and inland flooding. Findings offer a community-informed mental health response model for rural volunteers, applicable to future disaster responses in remote and underserved areas, thus enhancing broader disaster resilience efforts.

Literature Review

Disasters and Mental Health

A large body of literature suggests that hurricane and inland flooding exposure are associated with adverse mental health outcomes, including anxiety, posttraumatic stress disorder (PTSD), and suicide-related outcomes (Orengo-Aguayo et al., 2019²; Runkle et al., 2021³; Wertis et al., 2023⁴). Both direct exposures, such as witnessing the destruction and experiencing personal losses, and indirect exposure, including media exposure and hearing secondhand accounts of trauma, can increase an individual's risk of developing poor mental health outcomes in the wake of a disaster (Berry et al., 2010⁵).

The SAMHSA post-disaster framework identifies the “honeymoon phase” as the period immediately following a disaster, during which communities come together to support one another. This model suggests that the honeymoon phase lasts up to a few weeks after the disaster and may be associated with lower levels of mental distress (DeWolfe, 2000⁶; Substance Abuse and Mental Health Services Administration [SAMHSA], n.d.⁷). Yet, recent research suggests immediate and prolonged mental health distress among exposed populations, suggesting that the honeymoon period may not apply across all geographic contexts and sub-populations and may vary in duration (e.g., days rather than weeks) (Runkle et al., 2021; Wertis et al., 2023). Climate change is a risk multiplier, such that individuals with a history of adverse mental health, those who hold systematically marginalized identities (i.e., race, ethnicity, gender identity, sexual orientation), and those who live in isolated and underfunded regions may be more likely to experience mental distress following exposure to a climate disaster (United Nations, n.d.⁸; Clayton and Crandon, 2025⁹; Torales et al., 2026¹⁰). Mental health response post-disaster is complex, and further investigation of individual and place-based risk and resilience factors is warranted.

Disasters and Volunteering

Local community volunteers are often first on-the-ground and continue disaster response weeks and months after state and federal resources have been exhausted (Federal Emergency Management Agency [FEMA], 2017¹¹). Volunteers who travel to disaster-affected areas have a tendency towards pro-social behaviors like helping others, which can buffer against some of the disaster's negative mental health consequences (Dass-Brailsford et al., 2011¹²). Many local volunteers engage in spontaneous volunteerism, meaning they volunteer with an organization intermittently, often immediately after the disaster, and do not have specialized training. This volunteering helps fill a critical need in disaster response (Dass-Brailsford et al., 2011), yet little is known about the mental health impacts among local community volunteers in disasters. An understanding of how to best support community volunteers, particularly in the context of mental health, is urgently needed as extreme weather events become more common and more communities are exposed to devastating disaster impacts.

Volunteering and Mental Health

Volunteering creates avenues for social cohesion, which can increase mental health resilience in the aftermath of a disaster (Pike et al., 2024¹³). However, volunteers are also exposed to extreme stressors, including direct and indirect exposure to grief, death, injury, and trauma, in addition to long days involving physical labor (Quevillon et al., 2016¹⁴). Spontaneous volunteerism is predictive of adverse mental health outcomes due to a lack of training and mental health support services among unpaid, informal volunteers (Fraser, 2024¹⁵; Thormar et al., 2010¹⁶).

Existing evidence on the mental health of volunteers suggests significant adverse mental health impacts, including PTSD, anxiety, and depression (Aldamman et al., 2019¹⁷; Naushad et al., 2019¹⁸; Quevillon et al., 2016; Thormar et al., 2010). However, not all volunteers experience adverse mental health outcomes. At the individual level, negative individual outlook, significant disaster exposure, and prior mental health concerns are potential predictors of adverse mental health post-disaster among volunteers (Aldamman et al., 2019; Fraser, 2024; Thormar et al., 2010). Countering this, higher levels of social support and engagement with effective coping strategies, like practicing self-compassion and allowing oneself to find joy amid loss and destruction, can reduce the risk of adverse mental health post-disaster and encourage positive mental health outcomes (Fraser, 2024; Gonzalez-Mendez & Díaz, 2021¹⁹; Quevillon et al., 2016). For example, after the 2010 Haiti Earthquake, local volunteers reported a sense of hope and purpose associated with their volunteer experience (Carlile et al., 2014²⁰). These findings suggest a complex association between local community volunteering and mental health post-disaster exposure.

Social Vulnerability Framework

Social vulnerability frameworks emphasize the underlying risk factors that can transform hazardous events into disasters (Clark-Ginsberg et al., 2024²¹; Griego et al., 2020²²). They broadly encompass individual and place-based factors, such as socioeconomic status, educational attainment, household composition, race and ethnicity, age, language, and the built environment (Centers for Disease Control and Prevention [CDC], 2024a²³; Cutter et al., 2003²⁴; Diez Roux & Mair, 2010²⁵). Indicators, such as social determinants of health (SDoH) (CDC, 2024a) and social vulnerability indices (e.g., the CDC/ATSDR SVI)(CDC, 2024b²⁶), outline potential risk factors to consider when investigating disaster exposure and impacts. In a disaster context, individual experience, including personal loss and damage, can also be integrated into a social vulnerability framework to understand how disaster exposure is related to underlying vulnerabilities and disaster impacts (Clark-Ginsberg et al., 2022; Griego et al., 2020). Consideration of pre-existing risk factors is essential when conducting research at the intersection of climate disasters and health outcomes, particularly among underserved and overburdened populations.

Rural mountain communities are overlooked and understudied in the context of disasters and health, resulting in a lack of knowledge about the mental health impacts of extreme weather events in these places (Sugg et al., 2023²⁷). This is particularly concerning, as rural communities often lack sufficient healthcare resources, resulting in substantial healthcare disparities (Iglehart, 2018²⁸). This trend is also specifically evident in the southern Appalachian Mountains (Hash et al., 2024²⁹; Ryan et al., 2022³⁰; Sugg et al., 2022³¹; Thornton & Deitz-Allyn, 2010³²).

Our study region, western North Carolina, is characterized by disproportionately higher mental health burdens compared to the rest of the state, including anxiety, depression, and suicide-related outcomes, driven in part by insufficient mental health resources, rural and isolated geographies, and socioeconomic disparities (Ryan et al., 2022; Sugg et al., 2022). As a risk multiplier, climate change is likely to exacerbate these disparities (United Nations, n.d., Torales et al., 2026), underscoring the need for participatory research that yields evidence-based recommendations for communities to better address and improve community mental health, particularly in the aftermath of a major hurricane.

This project aims to gain a deeper understanding of individual-level (e.g., age, race, prior mental health diagnosis) and disaster-related (e.g., exposure, volunteerism, coping strategies) mental health risk and resilience factors to better support local communities and volunteers in the aftermath of a disaster. By generating quantitative and qualitative data through volunteer surveys, this study considers numerous factors, such as volunteer experience and hurricane exposures, offering unique and critical insight into post-disaster community mental health. Results inform continued local response and provide evidence-based recommendations and future research directions to enhance community-driven mental health resilience efforts.

Research Questions

The following research questions guided this research:

1. What are the most common hurricane exposures, volunteer experiences, and coping strategies following Hurricane Helene among a cohort of community volunteers in western North Carolina?

2. What are the mental health impacts of Hurricane Helene among a cohort of community volunteers in western North Carolina?

3. How do hurricane exposure, volunteer activities, and coping strategies predict mental health in the aftermath of Hurricane Helene among a cohort of community volunteers in western North Carolina?

Research Design

Study Context

Hurricane Helene, a tropical storm system that occurred in the Appalachian region of western North Carolina caused catastrophic flooding and resulted in unprecedented destruction, surpassing the damage inflicted by previous storms in the state, such as Hurricane Matthew in 2016 and Hurricane Florence in 2018 (Governor's Recovery Office for Western North Carolina [GROW NC], 2025³³). The storm delivered upwards of 20 inches (500+ mm) of rain in less than 24 hours, as depicted in map “a” of Figure 1, and caused more than 1,000 landslides. The storm’s intensity was exacerbated by widespread infrastructure damages, as illustrated by the road closures of map “b” in Figure 1, and the region’s isolated geographies, depicted in map “c.” (GROW NC, 2025). Tens of thousands of homes were damaged or destroyed, and thousands of miles of roadways were impassable (Figure 1).

Figure 1. Maps Depicting Hurricane Helene’s Various Impacts and the Isolated Geographies of North Carolina

Severe damage to cellular towers, water lines, and power grid infrastructure led to widespread communication challenges and stressors related to obtaining basic needs for weeks after the storm (GROW NC, 2025). The hurricane acted as a risk multiplier, disproportionately impacting under-resourced, rural communities, exacerbating long-standing disparities in the Appalachian Mountains (Sugg et al., 2025³⁴).

Community Partnership

We leveraged an ongoing partnership with the Appalachian Medical Reserve Corps (APP MRC), a local branch of the national Medical Reserve Corps. The Medical Reserve Corps is a volunteer-based network of individuals housed under the Association for Strategic Preparedness and Response, who respond to disasters when needed, and Public Health AmeriCorps, a local branch of AmeriCorps focused on providing psychological first aid to rural and isolated communities in western North Carolina. Immediately after Hurricane Helene, APP MRC organized and operated the non-emergency call center, providing reliable information dissemination and wellness checks, coordinating resources, and directing volunteers to assist with debris cleanup, aid organization, and supply distribution. Based in western North Carolina, APP MRC is involved in coordinating with numerous organizations throughout the region to support long-term recovery efforts. Since ceasing operations of the storm-related call center, APP MRC has retained a number of the volunteers who became involved in the organization after Helene. The group has about 230 active volunteers who engage in regular training—including psychological first aid training—and volunteer activities. More than 800 people receive the organization’s monthly newsletter. APP MRC now supports disaster response, public health outreach, and resource mapping, and the organization includes a volunteer base that spans a wide range of individuals. At the time of writing this report, APP MRC will remain engaged in recovery efforts for the foreseeable future.

Survey

Survey Sampling

We employed purposive and snowball sampling to recruit a cohort of volunteers (n = 373) to assess short-term (i.e., six to eight months) mental health impacts, volunteer experience, motivation for volunteering, and hurricane exposure (e.g., flooding, damage to home, loss of job or work) related to Hurricane Helene in western North Carolina. Integrating transdisciplinary perspectives from geography, epidemiology, integrated climate science, and public health, we employed online survey sample collection using Qualtrics³⁵, whereby we collected de-identified demographic information (age, race, sex) and geographic indicators (i.e., zip code, county), in addition to information regarding storm impacts, volunteer experience, and mental health. A follow-up survey is planned for the one-year mark (October 2025).

To gain a deeper understanding of volunteers’ hurricane experiences, we analyzed written narratives of open-ended survey questions. These narratives further contextualize our findings (DeLyser et al., 2010³⁶). Open-ended questions also offer a unique opportunity for volunteers to share their experiences (Nurser et al., 2018³⁷). Leveraging this qualitative approach provides a more comprehensive understanding of volunteers' experience and outstanding needs.

Survey Measures

We drew from and adapted multiple validated survey instruments to assess the impacts of Helene on mental health, volunteer experience, and disaster impacts (See Appendix A for full survey instrument). Specifically, the first section of the survey asks participants about their hurricane experience using questions from the validated Hurricane Exposure Adversity and Recovery Tool (HEART) to discern differential hurricane exposure (Dodd et al., 2019³⁸). The second section asks participants reasons for volunteering using questions from the Volunteer Functions Inventory (VFI) survey (Clary et al., 1998³⁹). The third section applies validated survey measures from the Rapid Behavioral Health Assessment: Post-Disaster to assess posttraumatic stress disorder (PTSD-4), generalized anxiety disorder (GAD-4), and increased substance use (Goldmann et al., 2021⁴⁰).

Data Analysis

We conducted a descriptive statistics analysis for survey questions using the tableone package in R (R Core Team, 2022⁴¹; Yoshida, 2020⁴²), reporting the total (n) and percent (%) of the sample population for each survey response.

Generalized Linear Mixed Models

To understand how volunteerism, exposure, and coping strategies predict self-reported mental health indicators, we employed generalized linear mixed models (GLMMs), a type of multilevel modeling approach that addresses potential clustering in the data with random effects to account for repeated measures (i.e., billing zip code) (Brooks et al., 2024⁴³). We elected to run regression for the PTSD and GAD indicators with the largest sample sizes: for PTSD these were poor sleep (n = 138) and avoidance (n = 103); for GAD these were anxious/nervous (n = 206), and depressed/hopeless (n = 183). PTSD, anxiety, and depression are reported in prior work focused on volunteer settings (e.g., Quevillon et al., 2016).

We ran subset GLMM regressions using the MASS and glmmTMB packages in R (Brooks et al., 2024; R Core Team, 2022) to identify the best-fitting model based on Akaike Information Criterion (AIC) scores. Models adjusted for key variables, including individual-level demographics (e.g., prior mental health diagnosis, age, race, gender, income), hurricane exposure, volunteer activities, and coping strategies. We specified zip code as a random effect to account for geographic clustering of responses. Model diagnostics included checking for multicollinearity using variance inflation factors (VIF < 5 was considered acceptable) and testing for overdispersion. Results inform a transdisciplinary framework and offer evidence-based recommendations on how to best support community mental health in the aftermath of a disaster, particularly in the context of rural communities that rely on local volunteers during long-term disaster recovery efforts, as observed in western North Carolina following Hurricane Helene.

Content Analysis

We coded written narrative responses (n = 97; 26% of total sample) using content analysis (Willig & Rogers, 2017⁴⁴), labeling open-ended responses as fitting into one or more of four predetermined themes: mental health, exposure, volunteering, and positive reflection. We manually coded themes and labeled open-ended responses with binary (0/1) indicators to determine theme prevalence. The dplyr, rlang, and tidytext packages in R (Queiroz et al., 2024⁴⁵; R Core Team, 2022; Wickham et al., 2023⁴⁶) were used to determine which context keywords were mentioned most frequently in narrative statements.

Themes were used to contextualize hurricane exposures, volunteer experiences, positive reflections, and mental health concerns (Tennant al., 2021⁴⁷). The identified themes also inform subsequent tangible community mental health resilience recommendations, ongoing needs assessments, and knowledge exchanges.

Ethical Considerations and Researcher Positionality

The Appalachian State Institutional Review Board (IRB: HS-25-124) exempted this research (Research Determination Category 2). Participants’ protection was ensured by collecting only a few direct identifiers, reporting aggregated demographics, and instructing participants to complete the survey in a private setting using a private browser. We employ validated survey measures and include a “prefer not to answer” option in all questions. All participants were provided with a list of free crisis support resources with their consent forms, and we include local mental health resources in our report back (See Appendix B for the full report back email that we sent to participants).

All the researchers involved in this work were personally affected by Hurricane Helene. The researchers leverage a strong working partnership, coordinating with volunteer and recovery networks throughout the region.

Results

Quantitative Results

From April to June 2025, 373 volunteers completed the online survey. Table 1 shows the demographic characteristics of our survey participants, who are predominantly ages 18-24 (33.5%), women (66.8%), white (94.3%), and from lower-income backgrounds (e.g., 27.1% with an income of under $40,000). This racial/ethnic composition (94.3% white, 1.7% Black, 1.7% Hispanic/Latino) mirrors western North Carolina's demographics (approximately 89% white, 4.2% Black and 6.3% Hispanic/Latino) (U.S. Census Bureau, 2020⁴⁸), though our sample notably overrepresents younger adults relative to the general population.

Table 1. Select Demographic Characteristics of Survey Participants

As shown in Figure 2, responses clustered in Watauga and Buncombe counties, around the population centers of Boone and Asheville, though volunteers from many of the affected counties in the region are represented in our sample.

Figure 2. Map Depicting Count of Surveyed Volunteers by Zip Code of Residence

Hurricane Exposure

Table 2 highlights that most volunteers were exposed to Hurricane Helene (93% exposed). The most frequently reported exposures include power outages (90.9%), communication challenges (93.8%), fearing for the safety of self or a loved one (64.9%), and damages from flooding, landslides, and fallen trees (39.9%).

Table 2. Hurricane Exposure Among Volunteers

Mental Health and Coping Strategies

Table 3 shows the frequency of mental health and coping strategies reported by participants. The most commonly reported PTSD indicators were poor sleep (37.0%), as determined by response to the survey question “since Hurricane Helene, have you had nightmares or thought about Hurricane Helene when you did not want to,” followed by hurricane-related avoidance with 27.6% of respondents responding affirmatively to the question “since Hurricane Helene, have you tried hard not to think about or went out of your way to avoid situations that reminded you of Hurricane Helene?” A small proportion of respondents (5.4%) reported experiencing all four validated PTSD indicators (Poor sleep since Helene, avoidance since Helene, feeling guarded since Helene, and feeling detached since Helene). In response to the question “since Hurricane Helene how often have you been bothered by feeling down, depressed, or hopeless,” many respondents indicated feeling depressed or hopeless (49.1%) and even more (55.2%) reported feelings of anxiety in response to the question “since Hurricane Helene how often have you been bothered by feeling nervous, anxious, or on edge?”. Nearly a quarter (24.4%) of participants reported experiencing all four validated GAD indicators. In this sample, 42.6% of respondents reported a prior mental health diagnosis.

Table 3. Mental Health Outcomes and Coping Strategies Among Volunteers

Table 3 highlights that surveyed volunteers used a variety of coping mechanisms. Notably, volunteers reported turning to social outlets (55.2%), taking actions like volunteering (58.4%), reframing the storm in a more positive light (53.6%), and engaging in physical exercise (37.5%). Roughly one-third of respondents (33.0%) utilized greenspace by spending time outdoors and using parks, trails, and other access to nature to cope after the storm. Substance use and blaming oneself were reported among 17.7% of respondents; 9.9% reported increasing their substance use following Helene.

Volunteering

Table 4 summarizes the frequency of volunteer motivations and activities. The most common motivations for volunteering were wanting to help others (98.7%) followed by connections to community (60.9%) and engaging with volunteerism as a storm-related coping mechanism (53.1%). Most volunteers engaged in aid distribution (64.5%) and debris removal and clean-up (54.2%). A smaller proportion of participants engaged in volunteer coordination (27.9%), followed by mutual aid (13.1%) and repair and rebuilding efforts (13.1%). At the time of survey response, 31.9% of participants were still engaged in Helene-related volunteer work.

Table 4. Volunteer Activities and Motivations

Generalized Linear Mixed Models

Poor Sleep Indicator

Table 5 shows that poor sleep was the most frequently reported PTSD indicator (n = 138). The odds of reporting poor sleep were 2.02 times more likely for volunteers who reported a mental health diagnosis (OR: 3.02, CI: 1.71-5.33), compared to those who did not, and 1.21 times as likely among those who engaged in aid distribution (OR: 2.21, CI: 1.21-4.03), compared to those who did not. Measures of hurricane exposure, specifically communication issues (OR:10.03), experiencing injury or death (OR:1.85), and fearing for one's own or a loved one’s safety (OR:2.74) were significantly associated with higher odds of poor sleep, compared to those who did not experience that hurricane exposure. Those who engaged in substance use and self-blame as coping strategies were associated with 2.88 times the odds of poor sleep (OR: 3.88, CI: 1.86-8.07), while acceptance was associated with 1.22 times the odds of reporting poor sleep (OR: 2.22, CI: 1.22-4.01) (Table 5).

Table 5. Likelihood Volunteers Reported Poor Sleep Post-Helene

Avoidance Indicator

Avoidance after Helene, the second most reported PTSD indicator (n = 103), was associated with individual-level factors, notably prior mental health diagnosis (OR: 1.74, CI: 1.02-2.96) and identifying as a woman (OR: 1.79, CI :0.99-3.22), as reported in Table 6. Hurricane exposures, specifically being exposed to injury or death (OR: 2.33, CI: 1.15-4.71) and being trapped or having to evacuate (OR: 2.49, CI: 1.45-4.27), were also associated with higher odds. Like poor sleep, volunteers who engaged with substance use and self-blame coping strategies were 2.84 times more likely to report avoidance (OR: 3.84, CI: 1.99-7.42), whereas greenspace was associated with protective effects (OR: 0.56, CI: 0.31-1.00). Distracting oneself was associated with higher odds of reporting avoidance (OR: 1.95, CI: 1.06-3.60).

Table 6. Likelihood Volunteers Reported Avoidance Post-Helene

Feeling Anxious or Nervous

Feeling anxious or nervous was the most reported mental health concern after Helene (n = 206). Table 7 highlights that prior mental health remained a significant predictor, associated with 2.17 times the odds of reporting feeling anxious or nervous (OR: 3.17, CI: 1.48-6.80). Self-reporting lower income (i.e., <$40,000) was associated with 1.70 times the odds of reporting feeling anxious/nervous (OR: 2.70, CI: 0.99-7.42). Coping strategies also remained consistent predictors, with substance use and self-blame coping associated with 3.00 times the odds of reporting feeling anxious/nervous (OR: 4.00, CI: 1.79-8.94). Distraction (OR :2.46, CI: 21.15-5.30) and active engagement in recovery to cope (OR: 2.42, CI: 1.05-5.63) were also associated with higher odds.

Table 7. Likelihood Volunteers Reported Feeling Anxious or Nervous Post-Helene

Feeling Depressed or Hopeless

Feeling depressed or hopelessness was the second most identified GAD indicator (n = 183). The odds of feeling depressed or hopelessness were higher among volunteers with a prior history of mental health diagnosis, (OR: 4.74, CI: 1.67-13.46), as reported in Table 8. The most significant predictors of reporting feeling depressed or hopelessness are related to coping strategies: 5.55 times the odds for substance use and self-blame as coping (OR: 6.55, CI: 1.84-23.28) and 2.8 for creative outlets (OR: 3.85. CI :1.27-11.65). As for avoidance, utilizing greenspace was associated with protective associations (OR: 0.24, CI: 0.06-0.93).

Table 8. Likelihood Volunteers Reported Feeling Depressed or Hopeless Post-Helene

Qualitative Content Findings

In the online survey, 97 respondents (26%) provided narrative reflections on Hurricane Helene. Four predetermined themes were used to code the narratives: mental health, hurricane exposure, volunteering, and positive reflection. As shown in Figure 3, coding showed 34 respondents describing mental health, 47 describing hurricane exposure, 24 describing volunteering, and 16 describing a positive reflection.

Figure 3. Count of Narratives Mentioning Pre-Determined Themes

Next, we analyzed the content coded with each pre-determined them to assess the terms that respondents most frequently mentioned. Figure 4 shows the percentage of respondents who mentioned each term. We explain the results below.

Figure 4. Content Analysis Depicting Most Frequently Mentioned Terms

Mental Health

An estimated 35% of narratives mentioned mental health challenges. Figure 4 shows the most frequently mentioned terms, which included PTSD (10.3%), emotional (10.3%), isolation (6.9%), and those related to fear: scary (6.9%), terrifying (6.9%), worried (6.9%), and worry (6.9%).

Trauma and fear emerged as important contexts, with multiple volunteers identifying trauma related to collective loss. One person said, “it was terrifying… being… without power, internet and no drinking water for 2 months… the landscape, physical and emotional, is changed for the foreseeable future.” Another respondent described the “trauma of the damage to family and neighbors’ homes… destruction of our community, roads and bridges.”

Compounding stressors and triggers were also mentioned along with trauma, notably sleep disruption, weather events, and fear. . One respondent said of sleep disruption, “now months later there are sleepless nights when there is extreme wind and/or rain.” Of weather events, a respondent said, “during an ice storm in February [2025]…so many trees f[e]ll that had been damaged from Helene. It was terrifying.” Regarding fear, a respondent described how even months after the storm, “I was suffering from PTSD due to…local destruction, fear of…rain/wind…lack of ability to connect with family and community.”

Isolation also emerged as an important context after Hurricane Helene, with volunteers and community members facing significant communication barriers. As one participant described, “horrifying and terrible isolation…no communication until neighbors said an AM radio station was broadcasting where to get meal rations and water.” Another said, “isolation with no internet access, no cell service…was the worst.”

Hurricane Exposure

Nearly half (47%) of narratives mentioned hurricane exposure. Figure 4 illustrates that the most frequently cited exposures were water (30%), power (17.5%), damage (10%), road closure or damage (5%), flooded (5%), and service (5%).

Conversations most frequently mentioned basic services, like water and power, contextualizing widespread infrastructure failures and outages for months after the storm. For example, one participant said, “I had to evacuate…no water and power for days” and another mentioned that being “without power for 14 days…was difficult.” Others described transportation challenges due to infrastructure damage. As one participant mentioned, “I could not leave my town for several days because roads were damaged…there was no gas.”

Volunteering

Roughly 25% of narratives mentioned volunteering, with volunteered (27.8%), volunteering (16.7%), volunteer (16.7%), volunteer center or call center (11.1%), distribution (11.1%), and responder (11.1%), being the keywords mentioned most often, as shown in Figure 4. Volunteers often mentioned damages observed and the emotional toll of the devastation. “...still in shock of what so many have had to endure…really messed with my head, I am a helper and I couldn't help very much at all.” Another described volunteering as “eye opening” and being “humbled by the experience.”

Positive Reflection

Over 15 narratives (16.5%) shared positive reflections, and many mentioned feeling grateful (15.8%), helped (15.8%), and proud (10.5%), as shown in Figure 4. Narratives mentioned gratitude for community and neighbors, extending beyond the initial response period. As one participant mentioned, “I developed a bond with neighbors.”

Discussion

Drawing on a unique cohort of 373 community volunteers who responded to Hurricane Helene in western North Carolina, this research offers critical, novel insights into the short-term (six to eight months) mental health impacts of hurricane exposure. More than half of the participants report experiencing at least one poor mental health indicator, roughly one in five report all four generalized anxiety disorder (GAD-4) indicators, and one in twenty report all four post-traumatic stress disorder (PTSD-4) indicators. Contextual analysis suggests widespread and persistent distress related to Hurricane Helene, with participants identifying resurfacing, compounding, and unresolved trauma. Volunteers highlight community cohesion and connection in their positive reflections, noting ongoing support for one another in the face of widespread devastation and limited access to resources. Findings from this work not only contribute unique insights to the disaster mental health literature but also provide an urgently needed framework for assessing and addressing the mental health needs of volunteers and responders in rural communities, advancing hurricane and flood-related preparedness.

Specific hurricane exposures significantly predict indicators for PTSD (e.g., sleep, avoidance), most notably experiencing injury or death, being forced to evacuate or being trapped, and fearing for one's own or a loved one’s safety. These results confirm the importance of detailed hurricane exposure assessments and contribute novel insights into potential pathways from hurricane exposure to poor mental health, including disrupted sleep following storm-related traumas (Palmer et al., 2023⁴⁹). GAD indicators (e.g., anxiety, depression) are more consistently predicted by self-reported coping strategies, emphasizing that individual-level behavioral factors are important considerations in post-disaster mental health research (Fraser, 2024; Quevillon et al., 2016).

Volunteerism, specifically aid distribution, is associated with higher odds of reporting poor sleep. Contextual analysis suggests that volunteers who engaged in aid distribution may have witnessed widespread destruction, feeling that they were of little help in the face of such extensive damage and devastation. These findings suggest higher exposure among individuals involved in aid distribution with fewer volunteer-related benefits, compared to those engaged in volunteer coordination, mutual aid, rebuilding efforts, and debris removal. These findings are broadly consistent with the published literature, which suggests that spontaneous volunteering predicts adverse mental health (Fraser, 2024; Thormar et al., 2010).

Coping strategies exhibited nuanced and often unexpected patterns, such that many coping strategies, whether those with a negative connotation (e.g., substance use, self-blame) (Aldamman et al., 2019; Fraser, 2024; Fuchs et al., 2021⁵⁰; Thormar et al., 2010) or positive connotation (e.g., creative outlets, social support) (Fraser, 2024; Gonzalez-Mendez & Díaz, 2021; Quevillon et al., 2016), are associated with higher odds of reporting poor mental health. This counterintuitive finding requires careful interpretation, as cross-sectional data cannot establish whether coping strategies led to poor mental health or whether individuals experiencing distress tried multiple coping approaches. In addition, the timing of our assessment (six to eight months post-disaster) may capture ongoing coping efforts rather than their ultimate effectiveness. Other potential reasons include those employing multiple coping strategies may have experienced more severe hurricane exposure, representing unmeasured confounding despite our statistical controls.

A notable exception, spending time in greenspaces, reduced the odds of reporting avoidance and feeling depressed/hopelessness among our surveyed volunteers. Emerging evidence suggests volunteering in greenspaces (e.g., trails, parks) after hurricane exposure is associated with a higher likelihood of positive reflection (Miller, 2020⁵¹). An extensive body of work emphasizes the mental health benefits of greenspace exposure and engagement (Ryan, Sugg, et al., 2024⁵²; Ryan, Runkle, et al., 2024⁵³). Findings from this work contribute novel insights into self-reported greenspace usage following hurricane exposure, reporting consistent protective effects for a host of mental health indicators associated with PTSD and GAD.

Response, Coping, Volunteering, Exposure, Recovery (ReCoVER) Framework

Based on our empirical findings from volunteers assessed six to eight months post-Hurricane Helene, we propose the ReCoVER Framework (Response, Coping, Volunteering, Exposure, Recovery), a novel conceptual model that addresses critical gaps in understanding how disaster exposure and volunteer engagement shape mental health trajectories in rural communities. ReCoVER draws from Social Vulnerability Theory (Cutter et al., 2003), the SAMHSA Phases of Disaster model (SAMHSA, n.d.), and the Stress and Coping Theory (Lazarus & Folkman, 1984⁵⁴). While our data collection occurred during the recovery phase, participant narratives and symptom patterns provide retrospective insights into the full disaster timeline, from immediate response through extended recovery. The framework extends beyond traditional disaster mental health models like SAMHSA by centering the unique experiences of community volunteers who serve dual roles as both disaster survivors and responders (Thormar et al., 2010), integrating findings from our mixed-methods analysis of 373 volunteers to provide both theoretical advancement and practical implementation guidance.

The ReCoVER framework challenges existing assumptions about post-disaster mental health, particularly the SAMHSA "honeymoon phase" model (DeWolfe, 2000; SAMHSA, n.d.), which suggests communities experience initial cohesion and reduced distress immediately following disasters. Our findings, even at six to eight months post-event, reveal persistent mental health impacts that participants traced back to the immediate response period. This aligns with recent work suggesting immediate and prolonged mental health distress among exposed populations (Runkle et al., 2021; Wertis et al., 2023). Narratives describing "terrifying" experiences during the storm and ongoing trauma suggest that for rural communities facing unprecedented disasters, the honeymoon phase may never materialize. Instead, infrastructure failures, notably communication outages, emerged as primary and lasting drivers of psychological distress, positioning basic services as critical mental health infrastructure requiring protection and rapid restoration.

This novel ReCoVER framework, which is illustrated in Figure 5 integrates four core domains: (a) Disaster Exposure and Effects (e.g., injury, evacuation, fear, infrastructure), (b) Volunteer Engagement (e.g., aid distribution, rebuilding, coordination), (c) Coping (e.g., greenspace, creative outlets) and (d) Individual Context (e.g., demographics, prior mental health). Thus, this framework accounts for compounding impacts: direct trauma of disaster exposure and the complexity of community and individual-based responses, with re-traumatization proving persistent among our cohort of volunteers. As findings from this work emphasize, volunteering is not inherently protective, but rather is contextually complex, potentially dependent on role, coping, and exposure intensity. The ReCoVER framework integrates and extends existing models to enhance understanding and improve disaster mental health, public health preparedness, and recovery efforts.

Figure 5. The ReCoVER Conceptual Framework for Rural Disaster Recovery

The framework's implementation draws from our empirical findings and existing best practices (Pike et al., 2024; Quevillon et al., 2016). he recovery trajectory observed at six to eight months revealed the persistence of symptoms, with over 50% of volunteers reporting at least one poor mental health indicator, 24.4% meeting all GAD-4 criteria, and 5.4% meeting all PTSD-4 criteria. These rates exceed general population prevalence and align with literature on volunteer (e.g., community volunteer, first responder) mental health after a disaster (Naushad et al., 2019; Quevillon et al., 2016). Content analysis revealed ongoing triggers in 35% of narratives mentioning mental health, with volunteers describing "sleepless nights when there is extreme wind," suggesting evolution from acute trauma to chronic hypervigilance. Our planned one-year follow-up survey will allow for further examination into longitudinal mental health trajectories for community volunteers during long-term recovery efforts, implementing the ReCoVER framework at the one-year mark.

Infrastructure Failures and Compound Exposures

ReCoVER positions communication systems as critical mental health infrastructure. The exposure dimensions identified in our study reveal multiple, compounding pathways to mental health impacts, consistent with social vulnerability frameworks (Clark-Ginsberg et al., 2024; Cutter et al., 2003). Infrastructure failures dominate both statistical models and narrative accounts, with 93.8% experiencing communication outages and 90.9% experiencing power outages. Communication loss is the strongest predictor of poor mental health, transforming a disaster into a prolonged psychological crisis. This finding extends beyond traditional conceptualizations of disaster exposure (Berry et al., 2010) to position infrastructure as a mental health determinant.

Conclusions

Community volunteers report high levels of distress, with more than half indicating at least one poor mental health outcome. Hurricane exposures predict PTSD-related indicators, while individual coping strategies consistently predict anxiety and depression-related measures. Infrastructure failures, including power outages lasting for days, communication disruptions for months, and extended lack of potable water, are nearly universal hurricane exposures in our cohort. These exposures, particularly communication outages and being trapped due to road closures, significantly predict adverse mental health outcomes.

Implications for Practice or Policy

Our findings inform policy in three key ways. First, communication failure is the strongest predictor of the PTSD indicator for poor sleep, suggesting that infrastructure resilience directly impacts mental health. We suggest prioritizing the safeguarding of communication infrastructure (e.g., back-up power, satellite-based emergency communication networks), and strengthening local communication outlets (e.g., public radio stations), to ensure isolated populations can receive life-saving information in future disaster contexts. Second, given the high mental health burdens associated with certain volunteer activities (e.g. aid distribution), we recommend organizations establish support protocols for disaster volunteers and employ tiered volunteer roles, particularly for spontaneous volunteers, based on training and mental health preparedness. Lastly, our results stress the need for rural-specific frameworks that incorporate rural knowledge and lived experiences. Federal disaster response protocols need to account for the unique challenges mountainous, isolated communities face, where traditional "honeymoon phase" models do not apply.

Limitations

hile our sample's racial/ethnic composition aligns with western North Carolina demographics, the age distribution and potential self-selection biases may limit generalizability. The findings are most applicable to rural mountain communities with similar demographic profiles but may not extend to urban disaster contexts or communities with different age structures.

Future Research Directions

The unprecedented scale of Hurricane Helene's impacts in western North Carolina demands further attention. Our findings reveal that while communities have united in remarkable ways, the mental health toll on volunteers requires urgent, sustained, and evidence-based support systems. As climate disasters increase in frequency and intensity, protecting the mental health of community volunteers becomes essential for long-term disaster resilience.

Future research directions should prioritize longitudinal cohort studies that follow volunteers at six-month intervals to identify critical intervention windows. Our finding that fear persists months later ("sleepless nights when there is extreme wind") suggests that long-term monitoring is essential. Further research should integrate health data analysis whereby volunteer surveys are linked with secondary datasets (e.g., emergency department visits, prescription data) to capture community mental health service utilization patterns. Community-centered research should extend beyond surveys to include listening sessions, focus groups, and photovoice projects to understand organizational challenges, behavioral risk factors (e.g., coping strategies), and the interplays of trauma and resilience. Comparative studies can examine how the ReCoVER framework applies to other rural disasters (e.g., wildfires, tornadoes) and whether protective factors like greenspace remain consistent across disaster types. Finally, designing and testing evidence-based interventions, like greenspace-based volunteer programs and communication redundancy systems for isolated communities, can be a part of this next stage of research.

Author Acknowledgments. We thank the community volunteers for their time and insights and our community partners, Appalachian Medical Reserve Cops (APP MRC) and Public Health AmeriCorps. Maps created by Caleb Blackburn and Owen Watkins.  

References