Housing Displacement, Employment Disruption, and Mental Health After the 2023 Maui Wildfires

Ruben Juarez; Binh Le; Christopher Knightsbridge; Marsha Lowery; Alika K Maunakea

doi:10.1001/jamapsychiatry.2026.0044

. 2026 Mar 11;83(6):601–610. doi: 10.1001/jamapsychiatry.2026.0044

Housing Displacement, Employment Disruption, and Mental Health After the 2023 Maui Wildfires

Ruben Juarez ^1,^2,^✉, Binh Le ¹, Christopher Knightsbridge ³, Marsha Lowery ³, Alika K Maunakea ⁴

¹University of Hawai‘i Economic Research Organization, University of Hawai‘i at Manoa, Honolulu

²Department of Economics, University of Hawai‘i at Manoa, Honolulu

³Maui Wildfire Exposure Study, University of Hawai‘i, Honolulu

⁴Department of Anatomy, Biochemistry and Physiology, John A. Burns School of Medicine, University of Hawai‘i at Manoa, Honolulu

Accepted for Publication: December 18, 2025.

Published Online: March 11, 2026. doi:10.1001/jamapsychiatry.2026.0044

^✉

Corresponding Author: Ruben Juarez, PhD, University of Hawai‘i at Manoa, 2424 Maile Way, Saunders 542, Honolulu, HI 96822 (rubenj@hawaii.edu).

Author Contributions: Drs Juarez and Maunakea had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Juarez, Le, Maunakea.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Juarez, Le, Knightsbridge, Maunakea.

Critical review of the manuscript for important intellectual content: All authors.

Statistical analysis: Juarez, Le.

Obtained funding: Juarez, Maunakea.

Administrative, technical, or material support: Juarez, Knightsbridge, Lowery, Maunakea.

Supervision: Juarez, Maunakea.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was made possible through the generous financial support of the Hawai‘i Community Foundation (Maui Strong Fund), Kaiser Permanente, the State of Hawai‘i, and grant R61MD019793 from the US National Institutes of Health.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The findings and conclusions presented in this article are solely those of the authors and do not necessarily reflect the views of the funders.

Data Sharing Statement: See Supplement 2.

Additional Contributions: This large, community-based, multidisciplinary study would not have been possible without the unwavering support and commitment of numerous organizations, individuals, staff, and volunteers, as described in Juarez and colleagues.¹⁶

^✉

Corresponding author.

PMCID: PMC12980362 PMID: 41811306

Key Points

Question

What are the mental health consequences of wildfire exposure, housing displacement, and income disruption following the 2023 Maui wildfires?

Findings

In this cross-sectional study of 2453 adults, residents within wildfire burn zones had a 53% higher risk of depression and 67% higher risk of anxiety compared with unexposed individuals, with suicidal ideation not significantly higher within burn zones but elevated among residents outside burn zones. Housing displacement and income loss jointly accounted for more than half of the associations with depression and anxiety.

Meaning

In this study, wildfire exposure and its socioeconomic aftermath were linked to widespread graded psychological harms, underscoring the need to incorporate mental health care, stable housing, and economic recovery into disaster response frameworks.

Abstract

Importance

Climate-related disasters are escalating in frequency and severity, yet their population-level mental health impacts—especially in racially and ethnically diverse and geographically isolated settings—remain poorly characterized. The 2023 Maui wildfires, one of the deadliest US wildfires in more than a century, offer a critical opportunity to quantify these effects.

Objective

To examine associations between wildfire exposure and symptoms of depression, anxiety, and suicidal ideation and to assess indirect pathways of these associations through housing displacement and income disruption.

Design, Setting, and Participants

This multiarm, propensity-weighted cross-sectional study compared adults residing within burn zones and outside burn zones on Maui and unexposed residents from other Hawaiʻi counties. The analytic sample included wildfire-exposed and unexposed adults enrolled between January 2024 and February 2025 through the Maui Wildfire Exposure Study and the UHERO Rapid Health Survey. Residential addresses at the time of the fires were geocoded and linked to official burn zone perimeters to determine exposure status. Data were analyzed from May to September 2025.

Main Outcomes and Measures

The primary outcomes were self-reported depression (Center for Epidemiologic Studies Depression Scale), anxiety (7-item Generalized Anxiety Disorder scale), and suicidal ideation within the past 30 days, assessed using validated screening instruments.

Results

The analytic sample included 2453 adults (1535 wildfire exposed and 918 unexposed), among whom mean (SD) age was 50.8 (16.3) years and 1502 participants (61.2%) were women. Wildfire exposure was associated with higher risk of depression (risk ratio [RR], 1.53; 95% CI, 1.20-1.94) and anxiety (RR, 1.67; 95% CI, 1.14-2.45) compared with unexposed individuals. Although suicidal ideation was more frequent among burn zone residents (RR, 2.15; 95% CI, 0.72-6.44), this association was not statistically significant. Residents outside burn zones showed significantly higher risk of suicidal ideation (RR, 2.65; 95% CI, 1.21-5.77). Mediation analyses indicated that housing displacement and income loss jointly accounted for more than half of the associations with depression and anxiety.

Conclusions and Relevance

In this cross-sectional study, wildfire exposure and its socioeconomic consequences were associated with graded increases in psychological distress extending beyond the burn zone. These findings highlight the importance of integrating mental health care, housing stability, and economic recovery as central pillars of disaster response and climate resilience strategies.

This cross-sectional study conducted among Maui residents examines associations between wildfire exposure following the 2023 Maui wildfires and symptoms of depression, anxiety, and suicidal ideation.

Introduction

As climate-related disasters escalate in frequency and intensity, their mental health implications represent an urgent yet underquantified dimension of global health.^1,2 Extreme events, including wildfires,³ hurricanes,^4,5,6,7,8 and floods,^9,10 are consistently associated with elevated rates of depression, anxiety, posttraumatic stress disorder, and suicidality, particularly among individuals with direct exposure or preexisting social vulnerabilities.^11,12 However, most existing evidence relies on cross-sectional designs, convenience samples, or administrative records that may undercapture undiagnosed psychological distress—especially in rural and underserved populations with limited access to care. These methodological limitations restrict robust inference and obscure the true burden of disaster-related mental illness. Despite increasing global attention to this issue, population-based studies using rigorous analytic frameworks to evaluate disaster-related mental health in high-risk, structurally vulnerable US communities remain exceedingly rare.

The 2023 Maui wildfires—one of the deadliest in the US in more than a century—offer a critical case study. The fires devastated the town of Lahaina and surrounding communities, killing more than 100 people and destroying more than 2200 structures. Approximately 10 000 residents were displaced, with many losing their homes, employment, access to essential services, and social support networks.¹³ Recovery unfolded in a geographically isolated and racially and ethnically diverse setting already burdened by longstanding inequities in housing and mental health care, compounding the psychological strain. Although commentaries^14,15 have underscored likely long-term distress among displaced residents and Native Hawaiian communities, quantitative, population-level evidence has been lacking.

To address these gaps, this study used a cross-sectional, propensity score–weighted design to examine the associations between wildfire exposure, housing displacement, and employment disruption and mental health symptoms among a large, multilingual sample of Maui residents. Using data from the Maui Wildfire Exposure Study (MauiWES)¹⁶ and a demographically matched statewide comparison cohort from the UHERO Rapid Health Survey (UHERO-RHS),¹⁷ we evaluated how geographic proximity to burn zones and postdisaster socioeconomic instability relate to depression, anxiety, and suicidal ideation.

This analysis expands on prior work from MauiWES,¹⁶ which previously described midterm respiratory and psychological outcomes in a single-arm descriptive framework. The current study extends that work by expanding the cohort, introducing a population-based comparison group, and applying propensity score weighting to estimate relative risks, evaluate indirect pathways through displacement and income loss, and assess community-wide spillover effects. Collectively, these advances provide one of the first analytically weighted, population-based assessments of wildfire-related psychological distress in a high-risk US population, with implications for climate adaptation policy, trauma-informed disaster recovery, and equitable mental health response systems worldwide.

Methods

Study Design, Participants, and Ethical Considerations

This cross-sectional study assessed associations between wildfire exposure and mental health outcomes after the August 2023 Maui wildfires using 2 complementary sources: MauiWES (n = 1535) and the fifth wave of UHERO-RHS (n = 918). Both studies surveyed adults aged 18 years or older. MauiWES provided exposed participants; UHERO-RHS participants residing outside Maui served as unexposed controls. The final analytic sample included 2453 adults after excluding those with missing mental health or demographic data, as described in the Consolidated Standards of Reporting Trials (CONSORT)–style participant flow diagram (eFigure 1 in Supplement 1). The University of Hawaiʻi institutional review board approved the study, and written informed consent was obtained from all participants. Reporting followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for observational studies and the AGReMA (A Guideline for Reporting Mediation Analyses) short-form checklist.¹⁸

MauiWES Cohort (Exposed Group)

MauiWES is the largest disaster-related biosocial cohort in Hawai‘i, designed to evaluate intermediate and long-term health effects.¹⁶ Between January 2024 and February 2025 (6-18 months postfire, see eFigure 2 in Supplement 1), participants were enrolled through community-based, multilingual recruitment at shelters, hotel housing sites, and screening events. Materials and interpreter support were provided in English, Spanish, Tagalog, Ilocano, and Pacific Islander languages. Baseline assessments included validated mental health screening, clinical measures, and biospecimen collection (see the eAppendix in Supplement 1).

UHERO-RHS Cohort (Unexposed Control Group)

UHERO-RHS is a statewide longitudinal biennial survey that monitors health and social disparities across Hawaiʻi.¹⁷ For this study, participants were restricted to those living outside Maui County. Data were collected in fall 2024 using the same validated mental health instruments as MauiWES, thus ensuring comparability. Mental health measures in UHERO-RHS have remained stable across survey waves, providing a credible reference group.¹⁹

Exposure Classification

Residential addresses at the time of the fires were geocoded and linked to burn perimeter shapefiles for the Lahaina and Kula wildfires.^20,21 Individuals were classified into the following 3 mutually exclusive exposure groups: (1) burn zone, comprising residents whose primary address fell within an official burn perimeter; (2) outside burn zone, comprising Maui residents living elsewhere on the island; and (3) unexposed controls, comprising residents of other Hawaiian Islands from the UHERO-RHS. This classification enabled assessment of dose-response patterns in mental health by degree of exposure.

Measures

Outcomes and Covariates

Three primary mental health outcomes were assessed. Depression was measured with the 10-item Center for Epidemiologic Studies Depression Scale (CES-D; score ≥10 for clinically significant).²² Anxiety was measured with the 7-item Generalized Anxiety Disorder scale (GAD-7; score ≥10 for moderate to severe).²³ Suicidal ideation was assessed using a single self-report item on serious consideration of suicide in the past 30 days, consistent with US Centers for Disease Control and Prevention surveys.^24,25

Analyses adjusted for demographic and socioeconomic covariates linked to disaster-related mental health.¹⁷ Matching variables included age, sex, education, race and ethnicity, and Area Deprivation Index (ADI). Employment (employed, unemployed, retired) and housing status (original, temporary, new home) were added as postdisaster covariates. In mediation models, income change and housing displacement were evaluated as mediators. Full metrics and variable definitions are provided in the eAppendix in Supplement 1.

Statistical Analysis

Associations between wildfire exposure and mental health outcomes were estimated using a multiarm inverse probability of treatment weighting (IPTW) approach based on generalized boosted models.²⁶ This propensity score framework balanced covariates across the 3 exposure groups, enabling estimation of average treatment effects (ATE) with reduced confounding.²⁷

To improve comparability between exposed and unexposed participants, weighting was based on pre-exposure, time-invariant characteristics (age, sex, race and ethnicity, education), and neighborhood socioeconomic disadvantage, measured using the ADI for each participant’s census block group from 2020 (before the wildfire).^28,29,30 The ADI is a composite of 17 indicators of income, education, employment, and housing quality that provides a standardized, census-based measure of neighborhood socioeconomic context across US populations. Its inclusion helps reduce structural and regional confounding between Maui and other islands. The variation in ADI score is presented in eFigure 6 in Supplement 1. Participants outside the region of common support were excluded to ensure balance and overlap (caliper = 0.1).

Primary associations between wildfire exposure and mental health outcomes were estimated using weighted logistic regression models with a doubly robust specification, adjusting for both IPTW weights and residual covariate imbalances. This approach improves estimate precision and mitigates bias due to model misspecification.

Total effects of wildfire exposure were decomposed into natural direct and indirect components operating through housing displacement and income-related employment disruption. To explore potential mediating pathways, we conducted survey-weighted structural equation modeling to quantify indirect effects of displacement and employment disruption (proxied by income decrease) on psychological outcomes.³¹ Indirect, direct, and total effects were estimated simultaneously and reported with their 95% confidence intervals.

Causal interpretation of the mediation effects assumes (1) no unmeasured confounding of exposure-mediator, mediator-outcome, or exposure-outcome relations; (2) no mediator-outcome confounders affected by exposure; and (3) correct model specification. Potential violations of these assumptions are discussed in the Limitations section. Mediators and outcomes were measured concurrently 6 to 18 months postwildfire.

Robustness of findings was assessed through multiple sensitivity analyses, including estimating the ATE on the treated using IPTW restricted to exposed subgroups³² and conducting pairwise nearest-neighbor matching within a 0.1-caliper region of common support to validate comparability across exposure levels.³³ All analyses were conducted in R version 4.4.2 (R Foundation for Statistical Computing) using the twang, MatchIt, survey, and lavaan.survey packages. All P values were 2-sided, with statistical significance defined as P < .05.

Results

Baseline Demographic Characteristics and Covariate Balance

The Table summarizes baseline demographic and socioeconomic characteristics across exposure groups (unexposed controls, residents within wildfire burn zones, and residents outside burn zones). The analytic sample included 2453 adults (1535 wildfire exposed and 918 unexposed), among whom mean (SD) age was 50.8 (16.3) years and 1502 participants (61.2%) were women. Before weighting, groups differed meaningfully in age, sex, race and ethnicity, education, and employment status, with unexposed controls being older and more likely to identify as Asian. After applying ATE weighting based on pre-exposure characteristics, these baseline covariates were well balanced across groups (eFigure 3 in Supplement 1). Outcome comparisons presented below are based on the weighted analytic sample.

Table. Characteristics of Study Participants Before and After Weighting by Wildfire Exposure Group.

Characteristic	No. (%)^a
	Overall (N = 2453)	Unweighted sample			Weighted sample
	Overall (N = 2453)	Burn zone (n = 658)	Outside burn zone (n = 877)	Unexposed controls (n = 918)	Burn zone (n = 492.69)	Outside burn zone (n = 350.19)	Unexposed controls (n = 508.65)
Outcomes
Depression
No problem	1432 (58.38)	303 (46.05)	470 (53.59)	659 (71.79)	229 (46.41)	199 (56.82)	367 (72.17)
Depressive or highly depressive symptom	1021 (41.62)	355 (53.95)	407 (46.41)	259 (28.21)	264 (53.59)	151 (43.18)	142 (27.83)
Anxiety
Minimal or mild anxiety	1954 (79.66)	455 (69.15)	686 (78.22)	813 (88.56)	343 (69.64)	283 (80.77)	450 (88.45)
Moderate or severe anxiety	499 (20.34)	203 (30.85)	191 (21.78)	105 (11.44)	150 (30.36)	67 (19.23)	59 (11.55)
Suicidal ideation
No	2380 (97.02)	629 (95.59)	845 (96.35)	906 (98.69)	473 (95.95)	339 (96.76)	504 (99.03)
Yes	73 (2.98)	29 (4.41)	32 (3.65)	12 (1.31)	20 (4.05)	11 (3.24)	5 (0.97)
Covariates
Age, mean (SD), y	50.80 (16.29)	48.89 (15.64)	46.91 (15.51)	55.88 (16.16)	49.75 (16.63)	49.01 (14.95)	51.32 (16.02)
Gender
Female	1502 (61.23)	379 (57.60)	556 (63.40)	567 (61.76)	303 (61.44)	225 (64.28)	320 (62.94)
Male	951 (38.77)	279 (42.40)	321 (36.60)	351 (38.24)	190 (38.56)	125 (35.72)	189 (37.06)
Education
Some schooling, no high school diploma	207 (8.44)	104 (15.81)	94 (10.72)	9 (0.98)	47 (9.58)	30 (8.64)	23 (4.43)
High school diploma	505 (20.59)	179 (27.20)	236 (26.91)	90 (9.80)	112 (22.77)	72 (20.61)	89 (17.49)
Some college level or technical or vocational degree	713 (29.07)	196 (29.79)	269 (30.67)	248 (27.02)	149 (30.33)	104 (29.71)	148 (29.18)
Bachelor’s degree or higher	1028 (41.91)	179 (27.20)	278 (31.70)	571 (62.20)	184 (37.32)	144 (41.04)	249 (48.9)
Race and ethnicity^b
Asian
No	1670 (68.08)	539 (81.91)	724 (82.55)	407 (44.34)	357 (72.5)	252 (72.06)	326 (64.13)
Yes	783 (31.92)	119 (18.09)	153 (17.45)	511 (55.66)	136 (27.5)	98 (27.94)	182 (35.87)
Filipino
No	1920 (78.27)	508 (77.20)	638 (72.75)	774 (84.31)	378 (76.74)	271 (77.48)	398 (78.27)
Yes	533 (21.73)	150 (22.80)	239 (27.25)	144 (15.69)	115 (23.26)	79 (22.52)	111 (21.73)
Hispanic or Latino
No	2164 (88.22)	529 (80.40)	753 (85.86)	882 (96.08)	430 (87.14)	309.19 (88.24)	482.65 (94.88)
Yes	289 (11.78)	129 (19.60)	124 (14.14)	36 (3.92)	63 (12.86)	41 (11.76)	26 (5.12)
Native Hawaiian or Pacific Islander
No	1923 (78.39)	528 (80.24)	653 (74.46)	742 (80.83)	385 (78.21)	277 (79.23)	386 (75.93)
Yes	530 (21.61)	130 (19.76)	224 (25.54)	176 (19.17)	107 (21.79)	73 (20.77)	122 (24.07)
White
No	1439 (58.66)	392 (59.57)	510 (58.15)	537 (58.50)	283 (57.45)	194 (55.35)	279 (54.78)
Yes	1014 (41.34)	266 (40.43)	367 (41.85)	381 (41.50)	210 (42.55)	156 (44.65)	230 (45.22)
Other race^c
No	2292 (93.44)	601 (91.34)	805 (91.79)	886 (96.51)	449 (91.1)	322 (91.93)	492 (96.78)
Yes	161 (6.56)	57 (8.66)	72 (8.21)	32 (3.49)	44 (8.9)	28 (8.07)	16 (3.22)
House relocation
No	1652 (67.35)	89 (13.53)	647 (73.77)	916 (99.78)	81 (16.53)	268 (76.48)	508 (99.86)
Yes	801 (32.65)	569 (86.47)	230 (26.23)	2 (0.22)	411 (83.47)	82 (23.52)	1 (0.14)
Employment status
Employed	1488 (60.66)	354 (53.80)	597 (68.07)	537 (58.50)	260 (52.69)	247 (70.47)	338 (66.38)
Retired	454 (18.51)	77 (11.70)	99 (11.29)	278 (30.28)	76 (15.48)	42 (12.01)	112 (21.93)
Unemployed	511 (20.83)	227 (34.50)	181 (20.64)	103 (11.22)	157 (31.83)	61 (17.52)	59 (11.7)
Area Deprivation Index, mean (SD)	5.20 (2.67)	5.57 (2.43)	4.83 (2.35)	5.29 (3.05)	5.46 (2.35)	5.21 (2.51)	5.15 (2.76)

Open in a new tab

^{^a}

This Table presents continuous variables and categorical variables of key sociodemographic characteristics and mental health outcomes across treatment groups in both the unweighted and average treatment effect (ATE)–weighted analytic samples. Treatment groups include: (1) unexposed controls (non-Maui residents), (2) residents outside the wildfire burn zone, and (3) residents within the burn zone. The ATE-weighted sample included a total effective sample size of 1351.53 individuals: 508.65 in the unexposed group, 350.19 outside the burn zone, and 492.69 within the burn zone. Propensity score weighting was performed using generalized boosted models to achieve covariate balance across age, gender, education, race and ethnicity (Asian, Filipino, Hispanic or Latino, Native Hawaiian or Pacific Islander, or White), and Area Deprivation Index. After weighting, the treatment and unexposed controls were well balanced across baseline covariates.

^{^b}

Participants were asked to self-identify their racial or ethnic group through the question “Which 1 or more of the following would you say is your race?” For more details, see the eAppendix in Supplement 1.

^{^c}

Other races include self-reported Black or African American, American Indian or Alaska Native, and other race or ethnicity.

Mental Health Outcomes by Wildfire Exposure

Before weighting, the prevalence of adverse mental health outcomes varied substantially by exposure group. Clinically significant depression symptoms were reported by 259 of 918 unexposed controls (28.21%), 355 of 658 burn zone residents (53.95%), and 407 of 877 residents outside burn zones (46.41%). Clinically significant anxiety symptoms followed a similar gradient: 105 of 918 among unexposed controls (11.44%), 203 of 658 residents in the burn zone (30.85%), and 191 of 877 residents outside the burn zone (21.78%). Suicidal ideation was reported by 12 of 918 unexposed controls (1.31%), 29 of 658 burn zone residents (4.41%), and 32 of 877 among those outside burn zones (3.65%). After applying ATE weights, group differences persisted. Weighted prevalence of depression was 142 of 509 among unexposed controls (27.83%; 95% CI, 23.99%-31.66%), 264 of 493 among burn zone residents (53.59%; 95% CI, 49.16%-58.03%), and 151 of 350 among residents outside burn zones (43.18%; 95% CI, 38.24%-48.12%). Weighted anxiety prevalences were 59 of 509 (11.55%; 95% CI, 8.94%-14.17%), 150 of 493 (30.36%; 95% CI, 26.41%-34.30%), and 67 of 350 (19.23%; 95% CI, 16.06%-22.39%), respectively. Weighted suicidal ideation remained most frequent among burn zone residents (20 of 493; 4.05%; 95% CI, 2.51%-5.59%) and residents outside burn zones (11 of 350; 3.24%; 95% CI, 1.98%-4.49%) relative to unexposed controls (5 of 509; 0.97%; 95% CI, 0.35%-1.59%) (Figure 1).

Figure 1. — Bar graphs show the weighted prevalence (%) of depressive symptoms, anxiety symptoms, and suicidal ideation across 3 exposure groups: unexposed controls (dark blue), Maui residents outside the wildfire burn zones (orange), and residents within burn zones (light blue). Prevalence of all outcomes increased with proximity to the burn zones, consistent with a dose-response pattern. Error bars indicate 95% confidence intervals.

As illustrated in eFigures 4 and 5 in Supplement 1, both depression and anxiety scores demonstrated a dose-response gradient, with symptom severity increasing in proportion to wildfire exposure.

Weighted Associations Between Wildfire Exposure and Mental Health Outcomes

Wildfire exposure was significantly associated with increased risk of adverse mental health outcomes (Figure 2). In fully adjusted, propensity-weighted models, residents within wildfire burn zones had higher risk of depressive symptoms (risk ratio [RR], 1.53; 95% CI, 1.20-1.94) and moderate to severe anxiety (RR, 1.67; 95% CI, 1.14-2.45) compared with unexposed controls. Although suicidal ideation was more frequent among burn zone residents (RR, 2.15; 95% CI, 0.72-6.44), this association was not statistically significant.

Maui residents living outside burn zones also exhibited significantly higher risk of depression (RR, 1.43; 95% CI, 1.19-1.73), anxiety (RR, 1.41; 95% CI, 1.05-1.88), and suicidal ideation (RR, 2.65; 95% CI, 1.21-5.77) relative to unexposed individuals, indicating community-wide psychological effects extending beyond directly affected areas.

Structural and socioeconomic disruptions were associated with psychological distress. Participants residing in temporary housing had a significantly higher risk of anxiety (RR, 1.45; 95% CI, 1.07-1.96) and a higher, approaching statistical significance risk of depression (RR, 1.20; 95% CI, 0.98-1.46) compared with those in stable housing.

In contrast, employment status emerged as a strong protective factor against mental health outcomes: being employed was associated with substantially lower risk of depression (RR, 0.72; 95% CI, 0.64-0.82), anxiety (RR, 0.63; 95% CI, 0.52-0.78), and suicidal ideation (RR, 0.45; 95% CI, 0.26-0.78) (Figure 2; eTable 1 in Supplement 1).

Indirect Pathways Linking Wildfire Exposure to Mental Health Symptoms

To explore pathways through which wildfire exposure was associated with psychological outcomes, we conducted mediation analyses using survey-weighted structural equation modeling. Two mediators were examined—housing displacement and employment disruption (proxied by income decrease)—each hypothesized to transmit part of the association between exposure and mental health symptoms.

Among residents within the wildfire burn zone, exposure was associated with a significantly higher latent propensity for depressive symptoms relative to unexposed controls (total effect: β = 0.258; 95% CI, 0.198-0.317). Decomposition of the total effect indicated contributions from a direct pathway (β = 0.099; 95% CI, −0.003 to 0.201) and indirect pathways through housing displacement (β = 0.080; 95% CI, 0.005-0.156) and employment disruption (β = 0.079; 95% CI, 0.036-0.121), which jointly mediated the remainder. Combined, these mediators explained approximately 61.62% of the total wildfire effect on depressive symptoms and 77.66% on anxiety. Outside the burn zone, the corresponding proportions mediated were 54.54% for depression and 96.10% for anxiety. For suicidal ideation in the burn zone, the estimated direct effect was negative and the proportion mediated exceeded 100%, consistent with suppression or inconsistent mediation. Full mediation results for all mental health outcomes and exposure groups are presented in Figure 3 and eTable 2 in Supplement 1.

Figure 3. — Standardized coefficients from a survey-weighted structural equation model show total, direct, and indirect effects of wildfire exposure on mental health outcomes. Indirect effects are mediated through 2 pathways: housing instability and employment disruption (proxied by income decrease). Residents in the burn zone (A) experienced significant direct and indirect increases in depressive symptom burden, with housing displacement and employment disruption equally emerging as the mediators, compared to residents outside the burn zone (B). Full model results are reported in eTable 2 in Supplement 1.

Sensitivity and Supplementary Analyses

To assess the robustness of results, multiple sensitivity and supplementary analyses were performed using alternative weighting, matching, and population adjustments (eTables 3-9 in Supplement 1).

First, we re-estimated associations using ATE on the treated weights and pairwise nearest-neighbor matching restricted to participants within common support regions (eFigure 8, eTables 3 and 4 in Supplement 1). Both approaches produced results consistent with the primary weighted models, showing that direct and indirect wildfire exposure remained associated with higher prevalence of depression, anxiety, and suicidal ideation (eFigure 7, eTable 5 in Supplement 1). Pairwise matching reinforced that residents within burn zones experienced the highest mental health burden (eTable 6 in Supplement 1).

Second, to test generalizability to the broader state population, poststratification weights were applied based on the Hawaiʻi state population from the American Community Survey. ATE estimates remained consistent with primary results (eTable 7 in Supplement 1).

Third, to examine temporal stability, analyses were replicated using the third wave of UHERO-RHS, collected before the fires, as an unexposed reference group. Associations between wildfire exposure and both depression and suicidal ideation remained unchanged (eTable 8 in Supplement 1).

Finally, sensitivity analyses using E-values³⁴ indicated that the associations of wildfire exposure with depression and anxiety were moderately robust to unmeasured confounding. In contrast, the association with suicidal ideation was not statistically significant (eTable 9 in Supplement 1).

Across all sensitivity and supplementary analyses, results were consistent in magnitude and direction, supporting the robustness of associations for depression and anxiety, while effects on suicidal ideation were less stable.

Discussion

The 2023 Maui wildfires—among the deadliest in modern US history—underscore the severe mental health toll of climate-related disasters. In this racially and ethnically diverse, geographically isolated cohort, wildfire exposure—particularly when accompanied by housing or income disruption—was linked to higher prevalence of depression, anxiety, and suicidal ideation and was not limited to only those directly impacted.

These findings extend prior reports of cardiopulmonary and psychological effects among directly exposed residents,¹⁶ rising suicide deaths,³⁵ and crisis line calls.³⁶ Differences between our findings and prior reports of increased suicide mortality likely reflect differences in outcome definition, timing, and level of analysis. Prior work examined county-level suicide deaths during the acute wildfire period, whereas our study assessed individual-level, 30-day suicidal ideation 6 to 18 months postfire, a relatively infrequent outcome in population surveys, which limits precision due to small numbers of positive responses. Together, they depict a consistent pattern of heightened distress across clinical and community settings. Our analysis advances this evidence by quantifying population-level excess risk, identifying modifiable mediators (housing displacement and income loss), and showing that psychological distress extended beyond the burn zone.

Distress followed a graded, dose-response pattern, with the greatest burden among burn zone residents and attenuated yet significant effects among neighbors. Associations persisted after IPTW adjustment, suggesting they are unlikely to be due to confounding alone. Mediation analyses indicate that housing and employment disruption jointly explained much of the observed increase in depression and anxiety, highlighting that social and economic instability—rather than trauma exposure alone—drives population-level harm. Strengthening housing stability, employment continuity, and community cohesion may mitigate long-term impacts, consistent with a syndemic framework in which environmental shocks amplify social vulnerability.

Unlike administrative mortality or service utilization data,^35,36,37 our community-based screening captures clinically significant symptoms (including among displaced, uninsured, and multilingual residents) that are often undiagnosed and therefore not observable in administrative sources. The findings show that trauma-related symptoms extend beyond physical destruction, driven by community loss, housing instability, and economic disruption.

While disaster research often centers on posttraumatic stress disorder, our results suggest that depression, anxiety, and suicidality linked to socioeconomic disruption are more widespread, immediate, and modifiable and should be treated as core components of disaster recovery. Beyond social pathways, biological mechanisms may also contribute: wildfire smoke exposure is associated with systemic inflammation, oxidative stress, and neurobiological changes tied to depression and suicidality.^38,39,40 Future work integrating biomarker analysis from the MauiWES biorepository will clarify how environmental exposures interact with psychosocial stress to shape individual mental health trajectories.

The findings reinforce global calls to integrate mental health into climate resilience planning. Frameworks like the Sendai Framework for Disaster Risk Reduction,⁴¹ the Lancet Countdown on Health and Climate Change,⁴² and World Health Organization climate adaptation guidance⁴³ all emphasize psychological well-being as essential to preparedness and recovery. This study offers rare empirical evidence for that agenda, showing that mental health is core disaster infrastructure, not secondary to physical reconstruction.

In Hawaiʻi, where housing costs are among the nation’s highest and social supports are limited, these results underscore the need to pair long-term housing solutions with trauma-informed mental health care. Evidence-based, trauma-specific interventions—such as trauma-focused cognitive behavioral therapy and culturally grounded, community-based counseling—can be embedded within disaster response systems to address both acute and chronic distress. Transitional shelters alone are insufficient; recovery must treat the psychological toll of displacement as a central pillar of resilience.

Globally, as climate disasters intensify, few countries monitor or mitigate their mental health impacts. These findings demonstrate the feasibility and necessity of embedding mental health surveillance and trauma-informed interventions within disaster preparedness frameworks to identify vulnerable populations early and guide equitable recovery.

Strengths and Limitations

This study offers several methodological strengths. To our knowledge, it is among the first to apply multiarm propensity score weighting with a contemporaneous control group to estimate the mental health effects of wildfire exposure. Participants were matched not only on individual demographic characteristics, but also on neighborhood-level deprivation indices, reducing confounding and strengthening inference.

However, several limitations warrant consideration. Residual confounding from unmeasured factors—such as prior trauma, informal social support, or undiagnosed psychiatric conditions—cannot be ruled out. Regional differences in health care access or social cohesion may also contribute to bias despite adjustment using the ADI, a validated measure of neighborhood disadvantage. Because the ADI is derived from national socioeconomic indicators, it may underestimate contextual deprivation in Hawai‘i communities where high living costs, geographic isolation, and limited clinician availability are not fully captured. Notably, estimates were similar in magnitude and direction in sensitivity analyses conducted without adjustment for ADI. Selection bias is possible due to community-based recruitment, although it would likely attenuate rather than exaggerate group differences. Some variables, including housing and employment, functioned as both mediators and potential confounders, complicating interpretation. Mental health outcomes were self-reported, introducing possible recall or reporting bias, and data collected 6 to 18 months postfire cannot capture longer-term trajectories, although this period aligns with the peak prevalence of postdisaster distress observed in prior research.^1,44,45,46 Suicidal ideation was assessed using a single-item measure, which, while common in population-based studies, does not capture the multidimensional nature of suicidality; more comprehensive instruments should be considered in future work to strengthen measurement validity. Notably, associations involving suicidal ideation were less precise, likely reflecting limited power due to the low event frequency rather than the absence of an effect.

Interpretation of mediation findings depends on standard causal inference assumptions (no unmeasured confounding and correct model specification); concurrent measurement of mediators and outcomes limits temporal ordering. Exposure classification based on burn zone residence may not fully reflect individual pollutant exposure or fire intensity, but Maui’s small geographic scale likely minimizes misclassification.

Despite these limitations, the community-based design, representative sampling, and validated screening instruments strengthen internal validity and enhance generalizability to underserved and rural populations often missing from administrative datasets.

Conclusions

In conclusion, the 2023 Maui wildfires illustrate that climate-related disasters are associated with substantial and far-reaching mental health impacts. In this cross-sectional study, both direct and indirect exposure were associated with higher prevalence of depression, anxiety, and suicidal ideation, magnified by housing displacement and economic disruption. These findings show that psychological distress extends beyond physical burn zones, reflecting a community-wide toll.

Housing stability and employment continuity emerged as key modifiable determinants of postdisaster well-being. Effective recovery must therefore move beyond physical reconstruction to integrate mental health care, stable housing, and economic security as core components of disaster response. As climate extremes intensify, proactive surveillance, coordinated services, and culturally grounded care will be essential to rebuilding not only infrastructure, but also the psychological and social fabric of affected communities.

Supplement 1.

eAppendix. Study Description and Eligibility Criteria, Variable Description, and Sampling and Recruitment

eFigure 1. Data Cleaning and Flow Diagram

eFigure 2. Monthly Recruitment Distribution for the Maui Wildfire Exposure Study, January 2024-February 2025

eFigure 3. Covariate Balance Before and After Propensity Score Weighting

eFigure 4. Distribution of Mental Health Scores by Wildfire Exposure Group

eFigure 5. Geographic Variation in Mental Health Scores by ZIP Code

eFigure 6. Geographic Variation in Area Deprivation Index by Census Block Group in 2020

eTable 1. Association Between Wildfire Exposure and Mental Health Outcomes Using Propensity Score–Weighted Models

eTable 2. Mediation of Wildfire Exposure Effects on Mental Health Outcomes via Housing Relocation and Employment Disruption

eFigure 7. Sensitivity Analysis: Effects of Wildfire Exposure on Mental Health Using ATT Weighting

eFigure 8. Distribution of Wildfire Exposure Propensity Scores by Treatment Group

eTable 3. Covariate Balance by Treatment Group in Unweighted and ATT-Weighted Samples

eTable 4. Covariate Balance Across Treatment Groups Using Pairwise Propensity Score Matching

eTable 5. Association Between Wildfire Exposure and Mental Health Outcomes Using ATT Weighting

eTable 6. Effects of Wildfire Exposure on Mental Health Outcomes Using Pairwise Propensity Score Matching

eTable 7. Association Between Wildfire Exposure and Mental Health Outcomes in State Population-Weighted Sample Using Multi-Arm Propensity Score Matching

eTable 8. Association Between Wildfire Exposure and Mental Health Outcomes Using Pre-Wildfire Controls From UHERO-RHS Wave 3

eTable 9. Robustness Check With E-values for the Association Between Wildfires Exposure and Mental Health Outcomes

jamapsychiatry-e260044-s001.pdf^{(2.9MB, pdf)}

Supplement 2.

Data Sharing Statement

jamapsychiatry-e260044-s002.pdf^{(16.3KB, pdf)}

References

1.Goldmann E, Galea S. Mental health consequences of disasters. Annu Rev Public Health. 2014;35(1):169-183. doi: 10.1146/annurev-publhealth-032013-182435 [DOI] [PubMed] [Google Scholar]
2.Cianconi P, Betrò S, Janiri L. The impact of climate change on mental health: a systematic descriptive review. Front Psychiatry. 2020;11:74. doi: 10.3389/fpsyt.2020.00074 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Bryant RA, Gibbs L, Gallagher HC, et al. Longitudinal study of changing psychological outcomes following the Victorian Black Saturday bushfires. Aust N Z J Psychiatry. 2018;52(6):542-551. doi: 10.1177/0004867417714337 [DOI] [PubMed] [Google Scholar]
4.Abramson DM, Stehling-Ariza T, Park YS, Walsh L, Culp D. Measuring individual disaster recovery: a socioecological framework. Disaster Med Public Health Prep. 2010;4(S1)(suppl 1):S46-S54. doi: 10.1001/dmp.2010.14 [DOI] [PubMed] [Google Scholar]
5.Abramson DM, Garfield RM. On the edge: children and families displaced by Hurricanes Katrina and Rita face a looming medical and mental health crisis. Columbia University Mailman School of Public Health . April 17, 2006. Accessed October 6, 2025. doi: 10.7916/D88S4Z4B [DOI]
6.Merdjanoff A, Friedman S, Piltch-Loeb R, Abramson DM. The detrimental effect of high mobility on long-term mental health trajectories following Hurricane Katrina. Int J Mass Emerg Disasters. 2023;41(1):121-132. doi: 10.1177/02807270231171630 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Fussell E, Lowe SR. The impact of housing displacement on the mental health of low-income parents after Hurricane Katrina. Soc Sci Med. 2014;113:137-144. doi: 10.1016/j.socscimed.2014.05.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Hori M, Schafer M. Social costs of displacement in Louisiana after Hurricanes Katrina and Rita. Popul Environ. 2010;31:64-86. doi: 10.1007/s11111-009-0094-0 [DOI] [Google Scholar]
9.Golitaleb M, Mazaheri E, Bonyadi M, Sahebi A. Prevalence of post-traumatic stress disorder after flood: a systematic review and meta-analysis. Front Psychiatry. 2022;13:890671. doi: 10.3389/fpsyt.2022.890671 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Fernandez A, Black J, Jones M, et al. Flooding and mental health: a systematic mapping review. PLoS One. 2015;10(4):e0119929. doi: 10.1371/journal.pone.0119929 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Aung TW, O’Donnell K, De Luca S, Gunzler D. Depression and anxiety symptoms in adults displaced by natural disasters. JAMA Netw Open. 2025;8(8):e2528546. doi: 10.1001/jamanetworkopen.2025.28546 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Schwartz RM, Liu B, Lieberman-Cribbin W, Taioli E. Displacement and mental health after natural disasters. Lancet Planet Health. 2017;1(8):e314. doi: 10.1016/S2542-5196(17)30138-9 [DOI] [PubMed] [Google Scholar]
13.US Fire Administration . Preliminary after-action report: 2023 Maui wildfire. Federal Emergency Management Agency . Accessed July 25, 2025. https://www.usfa.fema.gov/blog/preliminary-after-action-report-2023-maui-wildfire/
14.Suresh A, Tu L, Stanford FC. The 2023 Maui wildfires and the mental health effects of climate-induced relocation. Am J Public Health. 2024;114(11):1184-1187. doi: 10.2105/AJPH.2024.307815 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Riley L, Greywolf C, Enos VP. The Lahaina fires lasted hours, but the trauma will take generations to heal. Am J Public Health. 2024;114(S3):S242-S245. doi: 10.2105/AJPH.2024.307628 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Juarez R, Phankitnirundorn K, Ozorio Dutra SV, Bond-Smith D, Lee AG, Maunakea AK. Health and social support in the aftermath of the Maui wildfires. JAMA Netw Open. 2025;8(8):e2525430. doi: 10.1001/jamanetworkopen.2025.25430 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Juarez R, Le B, Bond-Smith D, Bonham C, Sanchez-Johnsen L, Maunakea AK. Socioeconomic determinants of mental health outcomes among Hawaii adults. Front Public Health. 2025;13:1526687. doi: 10.3389/fpubh.2025.1526687 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Lee H, Cashin AG, Lamb SE, et al. ; AGReMA group . A guideline for reporting mediation analyses of randomized trials and observational studies: the AGReMA Statement. JAMA. 2021;326(11):1045-1056. doi: 10.1001/jama.2021.14075 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Juarez R, Bond-Smith D, Bonham C, et al. Hawai‘i’s health and wellbeing journey over time — monitoring equity and access. The Economic Research Organization at the University of Hawai‘i . Accessed October 8, 2025. https://uhero.hawaii.edu/wp-content/uploads/2025/05/HawaiisHealthAndWellbeingJourneyOverTime.pdf
20.Bay Area News Group. Map: Maui wildfires burning in Lahaina and upcountry. The Mercury News . August 9, 2023. Accessed April 25, 2025. https://www.mercurynews.com/2023/08/09/map-maui-wildfires-lahaina-upcountry/
21.Juliano TW, Szasdi-Bardales F, Lareau NP, et al. Brief communication: the Lahaina Fire disaster – how models can be used to understand and predict wildfires. Nat Hazards Earth Syst Sci. 2024;24(1):47-52. doi: 10.5194/nhess-24-47-2024 [DOI] [Google Scholar]
22.Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385-401. doi: 10.1177/014662167700100306 [DOI] [Google Scholar]
23.Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092-1097. doi: 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
24.Ivey-Stephenson AZ, Demissie Z, Crosby AE, et al. Suicidal ideation and behaviors among high school students — Youth Risk Behavior Survey, United States, 2019. MMWR Suppl. 2020;69(1):47-55. doi: 10.15585/mmwr.su6901a6 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Jones SE, Ethier KA, Hertz M, et al. Mental health, suicidality, and connectedness among high school students during the COVID-19 pandemic — Adolescent Behaviors and Experiences Survey, United States, January-June 2021. MMWR Suppl. 2022;71(3):16-21. doi: 10.15585/mmwr.su7103a3 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.McCaffrey DF, Griffin BA, Almirall D, Slaughter ME, Ramchand R, Burgette LF. A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med. 2013;32(19):3388-3414. doi: 10.1002/sim.5753 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Zhao QY, Luo JC, Su Y, Zhang YJ, Tu GW, Luo Z. Propensity score matching with R: conventional methods and new features. Ann Transl Med. 2021;9(9):812. doi: 10.21037/atm-20-3998 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Kind AJH, Buckingham WR. Making neighborhood-disadvantage metrics accessible - the neighborhood atlas. N Engl J Med. 2018;378(26):2456-2458. doi: 10.1056/NEJMp1802313 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Center for Health Disparities Research, University of Wisconsin School of Medicine and Public Health . About the Neighborhood Atlas® and Area Deprivation Index (ADI). Accessed October 6, 2025. https://www.neighborhoodatlas.medicine.wisc.edu/
30.Powell WR, Sheehy AM, Kind AJH. The Area Deprivation Index is the most scientifically validated social exposome tool available for policies advancing health equity. Health Affairs Forefront . July 20, 2023. Accessed October 6, 2025. https://www.healthaffairs.org/content/forefront/area-deprivation-index-most-scientifically-validated-social-exposome-tool-available
31.Oberski D. lavaan.survey: an R package for complex survey analysis of structural equation models. J Stat Softw. 2014;57(1):1-27. doi: 10.18637/jss.v057.i0125400517 [DOI] [Google Scholar]
32.Ridgeway G, McCaffrey D, Morral A, et al. Toolkit for weighting and analysis of nonequivalent groups: a guide to the twang package. RAND . Accessed October 6, 2025. https://www.rand.org/pubs/tools/TLA570-5.html
33.Caliendo M, Kopeinig S. Some practical guidance for the implementation of propensity score matching. J Econ Surv. 2008;22(1):31-72. doi: 10.1111/j.1467-6419.2007.00527.x [DOI] [Google Scholar]
34.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-value. Ann Intern Med. 2017;167(4):268-274. doi: 10.7326/M16-2607 [DOI] [PubMed] [Google Scholar]
35.Purtle J, Machavariani E, Rivera-González AC, et al. Increases in suicides and overdoses during the 2023 wildfires in Maui, Hawai’i. JAMA. 2025;334(11):1019-1021. doi: 10.1001/jama.2025.11146 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Rivera-González AC, Purtle J, Kaholokula JK, Stimpson JP, Ortega AN. Maui wildfire and 988 suicide and crisis lifeline call volume and capacity. JAMA Netw Open. 2024;7(11):e2446523. doi: 10.1001/jamanetworkopen.2024.46523 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Hoang T, Noy I, Van TL. Heterogenous mental health impacts of a forced relocation: the red zone in Christchurch (New Zealand). Health Econ. 2025;34(10):1782-1803. doi: 10.1002/hec.70004 [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Molitor D, Mullins JT, White C. Air pollution and suicide in rural and urban America: evidence from wildfire smoke. Proc Natl Acad Sci U S A. 2023;120(38):e2221621120. doi: 10.1073/pnas.2221621120 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Liu L, Urch B, Szyszkowicz M, et al. Influence of exposure to coarse, fine and ultrafine urban particulate matter and their biological constituents on neural biomarkers in a randomized controlled crossover study. Environ Int. 2017;101:89-95. doi: 10.1016/j.envint.2017.01.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Zundel CG, Ryan P, Brokamp C, et al. Air pollution, depressive and anxiety disorders, and brain effects: a systematic review. Neurotoxicology. 2022;93:272-300. doi: 10.1016/j.neuro.2022.10.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Aitsi-Selmi A, Egawa S, Sasaki H, Wannous C, Murray V. The Sendai Framework for Disaster Risk Reduction: renewing the global commitment to people’s resilience, health, and well-being. Int J Disaster Risk Sci. 2015;6(2):164-176. doi: 10.1007/s13753-015-0050-9 [DOI] [Google Scholar]
42.Romanello M, Napoli CD, Green C, et al. The 2023 report of the Lancet Countdown on health and climate change: the imperative for a health-centred response in a world facing irreversible harms. Lancet. 2023;402(10419):2346-2394. doi: 10.1016/S0140-6736(23)01859-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.World Health Organization . Mental health and climate change: policy brief. June 3, 2022. Accessed July 19, 2025. https://www.who.int/publications/i/item/9789240045125
44.Norris FH, Friedman MJ, Watson PJ. 60,000 Disaster victims speak: part ii. summary and implications of the disaster mental health research. Psychiatry. 2002;65(3):240-260. doi: 10.1521/psyc.65.3.240.20169 [DOI] [PubMed] [Google Scholar]
45.Booker JA, Fivush R, Graci ME, et al. Longitudinal changes in trauma narratives over the first year and associations with coping and mental health. J Affect Disord. 2020;272:116-124. doi: 10.1016/j.jad.2020.04.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Leviston Z, Stanley SK, Rodney RM, et al. Solastalgia mediates between bushfire impact and mental health outcomes: a study of Australia’s 2019–2020 bushfire season. J Environ Psychol. 2023;90:102071. doi: 10.1016/j.jenvp.2023.102071 [DOI] [Google Scholar]

Associated Data

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

Supplementary Materials