Built environment features: Predictors or effect modifiers of post-disaster mental health outcomes? A systematic review.

Built environment features: Predictors or effect modifiers of post-disaster mental health outcomes? A systematic review - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Soc Sci Med . Author manuscript; available in PMC: 2026 Jun 4. Published in final edited form as: Soc Sci Med. 2026 Feb 26;397:119130. doi: 10.1016/j.socscimed.2026.119130 Built environment features: Predictors or effect modifiers of post-disaster mental health outcomes? A systematic review Yue Zhang a, * , Dongying Li a , Galen Newman a , Yining Liu a , Emily Besse b Author information Article notes Copyright and License information a Department of Landscape Architecture and Urban Planning, Texas A&M University, College Station, TX, the United States b Department of Psychological & Brain Sciences, Texas A&M University, College Station, TX, the United States * Corresponding author. Department of Landscape Architecture and Urban Planning, Texas A&M University, 789 Ross Street, 3137 TAMU, College Station, TX, 77840, the United States. yz94@tamu.edu (Y. Zhang). Issue date 2026 May. PMC Copyright notice PMCID: PMC13233002  NIHMSID: NIHMS2169307  PMID: 41795408 The publisher's version of this article is available at Soc Sci Med Abstract Emerging research has examined the mental health impact of built environment features in the context of climate disasters; however, evidence remains fragmented and heterogeneous. While existing reviews have synthesized particular built environment characteristics such as greenspace and housing under extreme weather events, a comprehensive synthesis examining diverse built environment features in relation to disaster-related mental health outcomes is lacking. This systematic review synthesized empirical evidence to assess these associations and identify research gaps and priorities for future study. Drawing upon seven electronic databases: Web of Science, Medline, ProQuest, EMBASE, PsycINFO, CINAHL, and Environmental Complete and following PRISMA guidelines, this review identified a final set of 22 articles. Almost half of the included studies were published between 2020 and 2024; most were cross-sectional studies conducted in the United States. Overall, the association of built environment features with mental health has been examined in diverse disasters, such as hurricanes, drought, heatwaves, and floods. Mental health assessments have focused on a range of outcomes, notably PTSD and psychological distress. We identified 12 studies that examined the role of built environment features in predicting climate disaster-related mental health outcomes, considering urbanicity, land cover, housing type, and characteristics of neighborhood vulnerability. Ten studies examined the role of four features in modifying the direction and strength of the relationship between climate disasters and mental health, namely urbanicity, land cover, neighborhood deprivation, and sky view factor. The synthesized evidence provides new insights into the psychological effect of built environments and policy implications for disaster preparedness and recovery. Keywords: Built environment, Mental health, Climate disasters, Systematic review 1. Introduction With ongoing global warming, climate hazards such as hurricanes, floods, heat, and wildfires are projected to be more frequent and intense. The effects of these extreme events are also anticipated to be exacerbated by population growth and urban development ( Qian et al., 2022 ). Experiencing natural disasters has direct impacts on mental health, for example, promoting depression, anxiety, and suicide ( Cianconi et al., 2020 ); overall, the prevalence of mental diseases in disaster-affected populations is two to three times higher than in the general population ( Math et al., 2015 ). This significant link of disaster exposure with poor mental health outcomes has been consistently demonstrated in systematic reviews and meta-analyses. For example, a meta-analysis of 27 articles published between 1980 and 2017 found that psychological and psychiatric disorders increased after natural disasters ( Beaglehole et al., 2018 ). Similarly, a significant link exists between natural disaster exposure and poor mental health, with flooding identified as the most frequent calamity associated with adverse mental health outcomes ( Keya et al., 2023 ). Meanwhile, meta-analyses focusing on heat events have indicated extremely high temperature to be significantly associated with a higher risk of mental disorders and suicide ( Li et al., 2023 ). It has also been reported that marginalized groups experience elevated mental and physical health issues due to weather events ( Benevolenza and DeRigne, 2019 ). Built environment features are likewise a key social determinant of health, influencing mental health through their impacts on access to amenities, exposure to stressors, and community social connections. For example, urbanicity was found to be positively associated with depression and mental disorders ( Sui et al., 2022 ). A systematic review of 12 studies further found housing disadvantages to be positively associated with mental health challenges ( Singh et al., 2019 ). Conversely, in a review of 263 studies, Wendelboe-Nelson et al. (2019) reported that about 70% demonstrated a positive association between green space characteristics and mental health and well-being. However, despite such findings highlighting the potential of the built environment to influence emotional and social resilience in the face of challenges, only a handful of studies to date have examined the role of built environments in mental health amid climate-related disasters. Existing theories and evidence indicate two distinct mechanisms through which built environment features are associated with mental health outcomes before and after disasters. First, built environment features can directly impact mental health by promoting or hindering nature-based restoration, physical activity, and social interactions ( R. Zhang et al., 2021 ), which effects may occur both in routine circumstances and during disasters. For example, stress reduction theory highlights that direct nature exposure can promote physiological and psychological relaxation and facilitate the reduction of stress ( Ulrich, 1984 ), while attention restoration theory suggests that natural environments facilitate mental recovery from cognitive fatigue by engaging the mind in a way that can promote restoration ( Kaplan and Kaplan, 1989 ; Kaplan, 1995 ). These theories provide a foundational framework for conceptualizing the direct predictor role of specific built environment features in mental health improvement. Moreover, emerging evidence has suggested that green space ( Baylan et al., 2025 ; Webb et al., 2024 ), as well as other features such as housing and access to functional amenities ( Owen and Crane, 2022 ; Schroeder et al., 2021 ; Stroebe et al., 2021 ), were associated with mental health recovery following traumatic events, including disaster experiences. Second, enabling or disabling environmental factors can impact the coping mechanisms of individuals and communities when facing disasters, helping buffer or exacerbate disaster-associated mental effects. In particular, the conservation of resources theory identifies that disaster survivors may experience psychological distress and trauma when primary resources such as stable housing, financial stability, and social networks are destroyed or lost ( Hobfoll, 1989 ). For example, prior research has examined and supported the hypothesis that disaster survivors who were able to conserve place-based resources, such as homes and access to healthcare, were at lower risk of adverse mental health outcomes ( Fussell and Lowe, 2014 ). Accordingly, implementation of built environment interventions such as infrastructure reconstruction, housing solutions, and resource distributions would modify the mental health impacts of disasters by buffering resource loss, supporting coping processes, and reducing vulnerability during recovery, thereby attenuating disaster-related mental health risks. Consistent with this mechanism, existing research on the role of housing in protecting health outcomes has also emphasized that lack of access to secure, affordable, and suitable housing might amplify health risks associated with climate change and extreme weather events ( Cartwright et al., 2025 ; Li et al., 2025 ). However, important questions remain: which of the two hypothesized mechanisms operates during climate-related disasters, and which one predominates? Whether built environment features function as direct determinants of mental health outcomes or as modifiers of disaster-related impacts has important implications for linking these pathways to appropriate health-promotion intervention strategies and climate adaptation policy responses. For example, the direct-effect mechanism would suggest disaster preparedness strategies aimed at improving baseline environmental conditions, such as green space availability and housing quality, to reduce mental health risks and support mental wellbeing when facing climate disasters ( Li et al., 2021 ; Moise, 2016 ). While effect-modification pathways would highlight the importance of adaptation policies such as resilient infrastructure development and targeted recovery planning to mitigate the adverse mental health consequences of climate hazards ( Xu et al., 2024 ). Additionally, understanding which mechanism predominates can guide the prioritization of interventions and support more effective allocation of disaster preparedness and recovery resources by targeting the pathways most relevant to mental health vulnerability in disaster contexts. To seek answers from existing literature, we first conducted a comprehensive review. Three extant reviews were found that provide important relevant information; however, few prior reviews have addressed our question on the two roles of the built environment. Venkataramanan et al. (2019) reviewed studies on the effects of green infrastructure for stormwater and flood management on health outcomes. However, the authors found no studies that connected green infrastructure for risk management to mental and physical health. Poter et al. (2023) focused on research assessing the impact of nature-led recovery on the mental health of people and their communities post-natural disasters, and emphasized the importance of connecting people with the natural environment for better recovery processes. Finally, Yang et al. (2023) conducted a narrative review on the potential of residential greenness to alleviate depression by reducing heat stress, which aimed to clarify the relationships between heat events, residential greenness, and depression. Given the lack of extant reviews, it is crucial to establish the existence and magnitude of the two hypothesized roles of built environment features in mental health within the context of climate-related disasters. Existing reviews have mostly focused on the mental health benefits of green infrastructure and natural environments; the associations between other built environment features, such as neighborhood characteristics, and mental health remain unexplored. Moreover, considering that the built environment can not only enhance but also hinder mental health ( Sullivan and Chang, 2011 ), it is warranted to conduct a review identifying potential adverse effects of built environments on mental health in relation to climate disasters. An integrated synthesis examining how diverse built environment features relate to mental health outcomes in the context of climate disasters is therefore necessary to identify which place-based characteristics could exacerbate vulnerability or promote resilience, under what climate disaster conditions these effects would be most pronounce, and how such evidence could inform urban planning, disaster risk reduction, and climate adaptation strategies that account for mental health impacts. Thus, the current systematic review aims to address these research gaps and identify modifiable built environment factors that may impact mental health outcomes associated with climate-related disasters. The specific research objectives are to: 1) examine the presence of direct associations and effect modifications of the built environment on disaster-related mental health outcomes, 2) identify the types of built environment factors that are associated with the mental health consequences of climate hazards, and 3) summarize research and knowledge gaps in existing literature and suggest opportunities for future research. 2. Method We conducted a systematic review following the Cochrane Handbook ( Higgins, 2008 ) and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines ( Page et al., 2021 ), emphasizing a four-step process: identification, screening, eligibility, and inclusion. The protocol was registered at the International Prospective Register of Systematic Reviews (PROSPERO CRD42024570239). This systematic review mandated critical appraisal of study quality and risk of bias, enabling a rigorous and high-quality synthesis of evidence to address the study’s specific research questions. The rigorous quality assessment incorporated in this review would strengthen the credibility and reliability of the synthesis by ensuring that findings were interpreted in relation to the methodological robustness of the included studies. In addition, by providing a transparent and critically appraised synthesis of the evidence, this systematic review would generate findings that were suitable for informing policy responses. As such, evidence-based conclusions would be grounded not only in the quantity of available research but also in its methodological quality. 2.1. Study eligibility criteria Eligibility criteria were structured following the population, exposure, comparator, and outcome (PECO) framework to characterize the inclusion and exclusion criteria of the study. Population. The review included studies involving the general human population. Animal studies were excluded. We did not restrict demographic and socioeconomic characteristics or geographical areas. Exposure. Studies examining the role of built environments in the context of climate-related disasters were included. Climate-related disasters are defined as extreme weather events caused by climate change, including avalanches, coastal hazards, drought, fire, flood, fog, heat, high wind events, and winter weather. Natural disasters caused by geological movements, such as earthquakes and tsunamis, were excluded. Built environment features refer to the physical characteristics of places, including rural-urban contexts, spatial configurations, spatial organizations, land use, urban design, transportation, mobility, buildings, facilities, green and blue space, neighborhood disadvantages and disorders, and neighborhood change. Eligible exposure measurements considered both built environment-related factors and the occurrence of climate extremes. Comparators. The review included studies considering a control population or those reporting different levels of built environment exposure within the same population. Outcome. The mental-health-related outcomes considered morbidity, mortality, and self-reported psychological states, including mental and behavioral disorders (ICD-10: F00-F99), suicide-related behaviors (ICD-10: T14.91, R45.851, and X60-X84), general mental health conditions, and psychological affect. Furthermore, we included only observational studies that quantitatively examined mental health outcomes in relation to both built environment features and climate-related disasters and were published in peer-reviewed journals written in English. Studies were excluded if they were not written in English, not published in a peer-reviewed journal, or if they reported experimental or quasi-experimental studies, reviews, case studies, commentaries, or editorials. 2.2. Search strategy The systematic search was conducted across seven electronic databases, including Web of Science, Medline, ProQuest, EMBASE, PsycINFO, CINAHL, and Environmental Complete. The search strategy was developed using the PECO framework, with search terms focused on two primary domains: exposure (i.e., built environments and climate-related disasters) and outcome (i.e., mental health). To define the context of climate-related disasters, we included 62 terms, such as “ climate disaster,” “storm surge,” “drought,” “wildfire,” “heatwave,” “thunderstorm,” “ haze,” and “ cold wave.” To identify built environment exposure, we included 117 terms, for example “ built environment,” “transportation,” “ infrastructure,” “ green space,” and “ residence characteristics.” Search terms for mental health included 257 terms considering psychological affect and clinical diagnosis, including “ happiness,” “sadness,” “ dementia,” “ substance abuse,” “ schizophrenia,” “ mood disorders,” and “ depression.” Wildcards were used to account for varied forms of the keywords. Within the same theme, we connected search terms with the “ OR” operator to capture all particular terms; when combining separate themes, we utilized the “ AND” operator to ensure a more targeted literature search. The exact syntax of queries was tailored to the specific search rules of each database. Additional studies were identified through a backward citation search, which involved reviewing all articles from the reference lists of the final set of search result studies. We removed duplicates already present in the search results and integrated the two sets. To evaluate the sensitivity of the proposed search strategy, we tested it against nine benchmark papers selected for high relevance to our research topic or for recognition and frequent citation within the academic community. Our search strategy effectively identified all benchmark papers. The initial search of all databases was conducted by two researchers (YZ, DL) on August 1st, 2024, with no restriction on publication year. A second search was performed on April 19th, 2025, to capture additional literature published between August 2024 and April 2025. Details of the search strategies are provided in the Supplementary Material , including an example search syntax, a list of benchmark papers, and Table S1 listing the full keywords for the two primary domains. 2.3. Study selection and data extraction After importing the retrieved articles into Endnote 21 and removing duplicates, we performed screening and eligibility determination by examining the title, abstract, and full text. The included articles were selected based on eligibility criteria. Three researchers (YZ, YL, and EB) engaged in title and abstract screening and subsequent full-text screening. At each phase, each record was reviewed by two researchers who made independent selection decisions. Any disagreements were addressed through discussion or consultation with a third researcher (YZ, YL, EB, or DL). We developed a descriptive information spreadsheet in Microsoft Excel to extract and tabulate information from each included study. The following study characteristics were extracted: authors, publication year, citation details, study location (country, state/province, county, city), climate zone, population, sample size, type of study, unit of analysis, spatial unit, temporal unit, exposure measurement, outcome measurement, confounder/covariates, study objectives, and results. Regarding the exposure measurement, we identified the type of climate-related disaster, details on disaster measures, specific built environment factors, the method of measuring built environment exposure, the role of the built environment variable, and details on built environment measures. Outcome measurements included mental health outcomes, the method of measuring mental health outcomes, time from disaster to mental health outcome assessment, and details on mental health measures. We extracted results as reported in each study, including all findings compatible with each outcome. This included data from multiple time points or subgroup analyses for the same outcome. Two researchers (YZ and YL) independently conducted the information extraction and cross-checking steps. Any discrepancies in data collection were resolved by a third researcher (DL). 2.4. Risk of bias assessment To assess bias in the included studies, we developed a risk of bias (RoB) assessment rubric by referring primarily to relevant items from the Office of Health Assessment and Translation (OHAT) Risk of Bias Rating Tool for Human and Animal Studies ( OHAT, 2019 ), with inclusion of items from the risk of bias assessment instrument for systematic reviews informing World Health Organization global air quality guidelines ( World Health Organization, 2020 ) and from the National Institutes of Health National Heart, Lung, and Blood Institutes Study Quality Assessment Tools ( National Institutes of Health, 2014 ). This rubric encompassed six domains: selection bias, confounding bias, exposure assessment, outcome measurement, attrition/exclusion bias, and selective reporting bias. Each domain includes one or two subdomains, which were rated on a three-point scale: low, moderate, and high risk. For example, within the confounding domain, we assessed whether the study considered potential covariates/confounding variables with valid measurement methods and whether these variables were adjusted statistically for their impact on the exposure-outcome relationship. Studies that considered self-identified characteristics (i.e., age, gender, race/ethnicity, and socioeconomic status) as confounders or covariates and used well-established surveys to collect data were rated as having a low RoB. However, studies that considered confounding variables related to post-disaster housing instability and mobility challenges were rated as having a moderate RoB when self-report data were adopted. Within the exposure assessment domain, we assessed whether the study examined different levels of exposure in relation to the outcome and whether the characterization of exposure variables was valid and reliable. For example, studies exploring the relationship between built environment features and mental health outcomes in the context of disasters were rated as having a high RoB if there were no variations in built environment exposure level. Likewise, if characteristics of built environments were measured by means of self-reported surveys rather than official data, the study was considered to pose a high RoB. Within the outcome assessment domain, we considered whether assessors were blinded to participant exposure status and whether the outcome measurement was valid. For example, studies evaluating subjective well-being with valid scales were categorized as low RoB, while studies measuring mental disorders with invalid scales were considered to have a moderate RoB. Within selective reporting domain, we evaluated whether all outcomes outlined in the methods were reported. Studies that only reported statistically significant results were rated as having a moderate RoB. While studies that failed to report pre-specified outcomes were categorized as high RoB. The details of the assessment rubric are stated in Supplementary Material Table S2 . The overall judgment of each domain was determined by combining the ratings from its subdomains. For domains with a single subdomain, we carried forward the subdomain rating as the overall rating. For domains with two subdomains, we applied the following criteria: if all subdomains received a low RoB rating, the domain was classified as low. If at least one subdomain was assessed as moderate RoB and none rated high, the domain was considered as moderate RoB. If one subdomain was rated as high RoB while the other was rated as low RoB, the domain was also categorized as moderate RoB. However, if at least one subdomain was rated as high RoB without any subdomain considered as low RoB, the domain was assigned a high RoB. We presented the RoB assessment results at both subdomain and domain levels to better reflect the methodological strengths and weaknesses of each study. We did not exclude any articles based on RoB results. Each article was independently assessed by two researchers (YZ and YL), with conflicts resolved by discussion with a third researcher (DL). 2.5. Evidence synthesis Due to the anticipated significant methodological and statistical heterogeneity among the included studies, it was not feasible to statistically pool study results through meta-analysis. Instead, we conducted a narrative synthesis to summarize and interpret the findings across studies. We tabulated and grouped studies by the following themes: study characteristics (publication year, population characteristics, study type, and geographical and climate zone distributions, types of climate-related disasters, types of built environment features, and outcome measurements), and specific roles of built environment features (prediction and effect modification). We compared the study characteristics qualitatively to explore potential causes of heterogeneity. In addition, differences in measurement methods for climate disaster exposure, built environment features, and mental health outcomes, as well as timing of outcome assessment, were considered to help explain heterogeneity in findings across studies. Finally, we narratively appraised the consistency of results across studies and the methodological quality of the included evidence to inform the certainty and confidence in the findings. In the narrative synthesis, we summarized effect measures as presented in the original studies. For each outcome, we reported the direction and magnitude of associations using mean difference, standardized or unstandardized regression coefficients, adjusted or unadjusted odds ratios, relative risks, incidence rate ratio, or percentage change, along with measures of precision (95% confidence intervals) or statistical significance (p-values). When multiple effect estimates were reported, adjusted measures were prioritized. No transformation or conversion of effect measures was performed. Excel PivotTables were used to tabulate and summarize data, and Adobe Illustrator to visualize the results. 3. Results 3.1. Study selection A total of 22,643 records were identified through the database search. An additional 573 records were identified through backward citation searching and other sources. After removing duplicates, 10081 unique records remained for title and abstract screening, of which 10,032 were excluded. Subsequently, full-text screening was conducted for 49 records, resulting in 22 studies eligible for inclusion in this review. The majority of studies were excluded after the full-text screening because they did not clearly specify built environment characteristics as an exposure variable ( Dargin and Mostafavi, 2020 ; M. X. Zhang et al., 2021 ) or did not define the context of climate disasters ( Arifwidodo and Chandrasiri, 2020 ; Tupinier Martin et al., 2024 ). The PRISMA flow diagram outlining the literature selection process and results is presented in Fig. 1 . Fig. 1. Open in a new tab Study selection process following the PRISMA guideline. 3.2. Study characteristics Table 1 summarizes individual study characteristics for 22 included studies in this review. The earliest article included in the review was published in 2007. From 2020 to 2024, there has been a steady increase in the number of relevant publications, with almost half of the included studies (45.6%) being published during this period. The vast majority of included research focused on all ages or adults (n = 19, 86.3%), while three studies (13.7%) examined both adolescents and adults. In terms of study type, the greatest proportion were cross-sectional studies (n = 9, 40.9%), followed by cohort studies (n = 4, 18.2%), time-series studies (n = 5, 22.7%), case-crossover studies (n = 2, 9.1%), and finally ecological studies (n = 2, 9.1%). Among studies considering temporal resolutions, most (55.6%) examined daily trends, with a few considering monthly (33.3%) or yearly patterns (11.1%). Table 1. Summary of individual study characteristics (N = 22). Citation Geographic region Population Sample size Study design Study objectives (exposure-outcome relationship) Disaster type BE factors Role of BE Exposure measurement method Mental health outcomes Outcome measurement method Time after disaster Austin et al. (2018) NSW, Australia Farmers: a cohort of farmers who lived or worked on a farm. (aged 18+) 664 persons Cohort Drought-psychological distress Drought Remoteness Predictor Disaster: Official data BE: Official data General psychological distress Self-report, K10 1 year Burrows et al. (2023) LA, USA Low-income mothers: a cohort of low-income mothers who survived Hurricane Katrina in New Orleans. (aged 19–34) 229 persons Cohort Residential greenness-psychological distress Hurricane Katrina Residential greenness Predictor Disaster: Research-defined BE: Official data General psychological distress Self-report, K6 NR Desalvo et al. (2007) LA, USA Employees: all faculty, staff, and administrators who were employed at Tulane University and survived Hurricane Katrina. (aged 18+) 1,542 persons Cross-sectional Hurricane-PTSD Hurricane Katrina Post-disaster housing type Predictor Disaster: Research-defined BE: Self-report PTSD Self-report, PCL-17 6 months Fussell and Lowe, (2014) LA, USA Low-income parents: a parents who survived Hurricane Katrina and completed pre- and post-disaster assessments. (aged 19–34) 392 persons Cohort Post-disaster housing displacement-mental health Hurricane Katrina Pre- and post- disaster housing types Predictor Disaster: Research-defined BE: Self-report General psychological distress, PTSD, and stress Self-report, K6, IES-R, PSS 8–18 months Gros et al. (2012) TX, USA General population: adults residing in Galveston and Chambers counties in Eastern Texas who were victims of Hurricane Ike. (aged 18+) 1,249 persons Cross-sectional Post-disaster loss of services-psychiatric symptoms Hurricane Ike Urban vs. non-urban Effect modifier Disaster: Research-defined BE: Official data Depression, PTSD, and worry Self-report, CES-D, PSWQ, PCL-17 12–13 months Hanigan et al. (2012) NSW, Australia General population: deaths from suicide for NSW from 1970 to 2007. (all ages) NR Time-series Drought-suicide Drought Urban vs. rural Effect modifier Disaster: Official data BE: Official data Suicide Clinical diagnosis 1 month Hanigan et al. (2022) NSW, Australia General population: deaths from self-inflicted injuries for NSW from 1970 to 2007. (all ages) 18,942 persons Time-series Drought-suicide Drought Urban vs. rural Effect modifier Disaster: Official data BE: Official data Suicide Clinical diagnosis 1 month Ho et al. (2018) Hongkong, China General population: deaths from 2007 to 2014 in Hongkong. (all ages) 284,477 persons Time-series with case-only Haze-mental-health-related mortality Haze Sky view factor and vegetation coverage Effect modifier Disaster: Official data BE: Official data Mental and behavioral disorders (F00–F99) Clinical diagnosis 0–3 days Lavigne et al. (2023) Alberta and Ontario, Canada General population: mental-disorders-related emergency room visits in Alberta and Ontario from 2004 to 2020. (all ages) 9,958,759 records Case-crossover Extreme temperature-mental and behavioral disorders Extreme temperature Neighborhood deprivation, residential green space and urbanization Effect modifier Disaster: Official data BE: Official data Mental and behavioral disorders (F00–F99) Clinical diagnosis 5 days Li et al. (2021) TX, USA General population: residents living in 30 block groups in Houston which were heavily impacted during Hurricane Harvey. (aged 18+) 272 Cross-sectional Neighborhood greenness-PTSD & Severity of hurricane-PTSD Hurricane Harvey Neighborhood greenness Predictor Disaster: Self-report BE: Self-report & official data PTSD Self-report, IES-R, PCL-5 2 years Lu and Singelmann (2009) LA and Houston, USA General population: hurricane evacuees residing in Houston apartment complexes and in various locations in southern Louisiana. (aged 16-87) 682 persons Cross-sectional Residential environment-stress level Hurricane Katrina & Hurricane Rita Post-disaster housing type Predictor Disaster: Research-defined BE: Research-defined Anxiety Self-report, BSI NR Moise (2016) New Orleans, USA General population: from private and public hospitals in the state, whose diagnosis categories were substance abuse disorders. (all ages) 1,401 persons Ecological Hurricane-substance abuse disorders Hurricane Katrina Blighted properties and neighborhood recovery Predictor Disaster: Research-defined BE: Official data Substance abuse disorders Clinical diagnosis 3 years Newman et al. (2022) TX, USA General population: residents living in 30 block groups in Houston which were heavily impacted during Hurricane Harvey. (aged 18+) 257 persons Cross-sectional Neighborhood vacancy-PTSD & Neighborhood vacancy-health-related quality of life (HRQOL) Hurricane Harvey Neighborhood vacancy, tree cover Predictor Disaster: Self-report BE: Self-report & official data PTSD & HRQOL Self-report, PCL-5, HRQOL-14 2 years Obrien et al. (2014) Nation-wide, General population: a 5,012 Cohort Drought-mental Drought Urban vs. rural Effect Disaster: General Self-report, NR Pan et al. (2024) Anhui, China General population: individuals for whom the primary diagnosis was definite schizophrenia in Anhui Province from 2017 to 2020. (all ages) 160,736 persons Case-crossover Combined effects of PM2.5 and heatwaves-schizophrenia Heatwave Land cover Effect modifier Disaster: Official data BE: Official data Schizophrenia Clinical diagnosis 2–4 days Phan Minh et al. (2016) Hanoi, Vietnam General population: all patients with mental health disorders in the north of Vietnam, including Hanoi. (all ages) 21,443 records Time-series Heatwaves-mental health Heatwave Urban vs. rural Effect modifier Disaster: Official data BE: Official data Mental and behavioral disorders (F00–F99, except F60–F69) Clinical diagnosis 3–7 days Rahman et al. (2022) Nation-wide, Indonesia General population: individuals from the Indonesian Life Survey Family 5 who had experienced climate disasters. (aged 18+) 7,110 persons Cross-sectional Climate disasters-subjective well-being Combined climate disasters Urban vs. rural Effect modifier Disaster: Self-report BE: Disaster: Research-defined Happiness & life satisfaction Self-report, Non-valid scale NR Schwartz et al. (2018) NY, USA General population: adults residing in Queens, Staten Island, and Long Island. (aged 18+) 1,615 persons Cross-sectional Post-disaster housing displacement-mental health Hurricane Sandy Post-disaster housing type Predictor Disaster: Research-defined BE: Self-report Depression, anxiety, PTSD and substance use Self-report, PHQ-4, PCL-17, TQS, NIDA-Modified ASSIST V2.0 NR Shabani et al. (2024) Lorestan, Khuzestan, and Golestan, Iran General population: individuals affected by the flood who were randomly selected from three heavily affected provinces of Iran. (aged 16+) 2,305 persons Crosssectional Flood-PTSD Flood Urban vs. rural, post-disaster housing type Predictor Disaster: Self-report BE: Self-report PTSD Self-report, IES-R 5 months Xiong et al. (2016) Kelantan and Kuala Lumpur, Malaysia General population: adults who have faced at least one flood incident within the past year, and whose physical and mental health are sufficiently good to participate in the research. (aged 20–59) 200 persons Crosssectional Locality-PTSD Flood Urban vs. rural Effect modifier Disaster: Self-report BE: Research-defined PTSD Self-report, IES-R 1 year Xu et al. (2024) Murray–Darling Basin, Australia General population: people who died from suicide from 2006 to 2016. (all ages) NR Ecological Drought and hotter temperaturesuicide Drought Vegetation coverage Effect modifier Disaster: Official data BE: Official data Suicide Clinical diagnosis 1 month Zhang et al. (2020) NY, USA General population: emergency department visits for several health conditions fromYork 2005 State. to 2014 in New (all ages) 1411657 records Time-series combined with machine learning Community factormental health Hurricane Sandy Areas prone to flooding, developed areas, & green coverage Predictor Disaster: Research-defined BE:Official data Mental and behavioral disorders Clinical diagnosis 1–3 days Open in a new tab Note. NR: not reported; BE: built environment. 3.2.1. Geographical and climate zone distribution Fig. 2 presents the geographical and climate zone distributions of the included studies. A large portion of studies were conducted in the United States (n = 10, 45.5%), followed by Australia (n = 5, 22.7%), China (n = 2, 9.1%), and finally one paper each from Vietnam, Indonesia, Malaysia, and Iran. With respect to climate, most US-based studies (80%) focused on the Gulf Coast, including cities in Louisiana and Texas, which are characterized by a humid subtropical climate. Only two were sited in New York, where the climate ranges from humid subtropical to humid continental. Most Asian studies (83.3%) primarily focused on the humid subtropical and tropical climates. Studies conducted in Iran, Australia, and Canada encompassed regions with diverse and varied climatic conditions. Fig. 2. Open in a new tab Geographical and climate zone distributions of included studies. 3.2.2. Type of climate-related disasters assessed Among disaster types examined, hurricanes were the most highly represented (n = 10, 45.5%), with five studies (22.7%) examining exposure to drought, three (13.6%) the impact of extreme heat events, and two (9.1%) flood-affected regions. One study (4.5%) specifically evaluated exposure to haze, and one (4.5%) combined climate disasters. Studies focusing on hurricane-impacted areas were all concentrated in the US, involving hurricanes such as Katrina, Rita, Ike, Sandy, and Harvey. Regarding methods for assessing exposure to disasters, nine hurricane-related studies (90%) recruited participants from regions affected by hurricanes ( Burrows et al., 2023 ; Gros et al., 2012 ; Lu and Singelmann, 2009 ), but did not measure disaster exposure. The same method was adopted in the two studies on disastrous floods ( Shabani et al., 2024 ; Xiong et al., 2016 ). In contrast, severeness of extreme weather events (e.g., drought, heatwave, and haze) was objectively measured by official data. Five studies conducted in Australia adopted agricultural drought indices to define exposure, including the Hutchinson Drought Severity Index ( Obrien et al., 2014 ) and the Palmer Drought Severity Index ( Xu et al., 2024 ). Based on weather data such as rainfall and air temperature, these indices can identify different levels of dryness and capture the duration of drought conditions. Three studies on heat likewise used threshold or percentile-based temperature metrics ( Lavigne et al., 2023 ; Pan et al., 2024 ; Phan Minh et al., 2016 ). The study on haze exposure used the definition provided by the local observatory, which was depended on meteorological parameters such as relative humidity, precipitation, and fog ( Ho et al., 2018 ). 3.2.3. Type of built environment assessed All included studies measured at least one built environment factor, with five (22.7%) assessing multiple features. More than one-third of studies (36.4%) focused on the effect of urbanicity or remoteness of the neighborhood. For example, research in the US used the census definition of metropolitan statistical areas to classify participants into urbanized and non-urbanized areas ( Gros et al., 2012 ), while studies in Australia, Vietnam, and Malaysia adopted the local administrative divisions ( Phan Minh et al., 2016 ; Xiong et al., 2016 ). Urbanization in the Canadian study was defined by calculating the proportion of area characterized as urban within a specific buffer ( Lavigne et al., 2023 ). A number of studies (27.3%) focused on the role of green spaces in residential surroundings. Six studies measured green space objectively, among which one included an additional perceived measure of the quality of neighborhood greenness using the Neighborhood Open Space (NOS) scale ( Li et al., 2021 ). The Normalized Difference Vegetation Index (NDVI) emerged as the most popular metric for green space assessment, being used in two-thirds of relevant studies. One study specifically compared the NDVI with the Enhanced Vegetation Index (EVI), revealing the differences in their respective impacts ( Burrows et al., 2023 ). Another commonly used metric was the percentage of vegetation coverage, defined as the proportion of vegetated area, such as tree canopy, within a specific buffered area ( Newman et al., 2022 ). Two studies (9.1%) measured not only the percentage of green space cover but also blue and developed space cover ( Pan et al., 2024 ; Zhang et al., 2020 ). Neighborhood deprivation and vulnerabilities were included in four studies (18.2%). Neighborhood deprivation was assessed based on a deprivation index derived from national census data ( Lavigne et al., 2023 ). Neighborhood vacancy was determined from official land-use data ( Newman et al., 2022 ). Neighborhood blight and recovery were defined by address-level data ( Moise, 2016 ). Finally, the measure of flood-prone areas was objectively derived from geospatial data ( Zhang et al., 2020 ). Several studies (18.2%) measured participants’ housing type before or after climate disasters. Most studies used self-reported surveys to assess the pre-disaster housing situation (e.g., homeowner, renter, public housing resident, or living with family/friends) or the post-disaster residency type (e.g., original home, new home, with friends/relatives, or temporary housing). However, one study operationalized the post-disaster exposure by selecting participants based on the type of house they lived in ( Lu and Singelmann, 2009 ). In one study, the sky view factor describing urban morphology was derived from both airborne LiDAR data and a city building map, providing an accurate representation of the built environment ( Ho et al., 2018 ). 3.2.4. Outcome measurements Various mental health outcomes were assessed in the included studies, with four (18.2%) examining more than one health outcome. The most frequently assessed outcome was PTSD (n = 8, 36.4%), followed by general psychological distress (n = 4, 18.2%), mental and behavioral disorders (MBDs) (n = 4, 18.2%), suicide (n = 3, 13.6%), stress and anxiety (n = 2, 9.1%), depression (n = 2, 9.1%), and substance use (n = 2, 9.1%). Only one study (4.5%) evaluated schizophrenia. Two studies (9.1%) examined subjective well-being, including happiness, life satisfaction, and health-related quality of life (HRQOL). Studies focusing on MBDs assessed the health outcomes within seven days, identifying the acute effects of environmental exposure. Research on suicide also examined outcomes over relatively short timeframes, such as daily and monthly intervals. However, assessments of PTSD and general psychological distress utilized longer timeframes of five months to two years. The longest timeframe was used in assessing substance abuse disorder, with studies extending up to three years. More than half of the included studies (54.5%) measured mental health outcomes by means of validated scales. For example, instruments for PTSD assessment included the PTSD checklist (e.g., PCL-17 and PCL-5) and the Impact of Event Scale-Revised (IES-R). General psychological distress was primarily evaluated with the Kessler Psychological Distress Scale (e.g., K-10 and K-5). Instruments used to measure stress, anxiety, depression, worry, and health-related quality of life comprised the Patient Health Questionnaire (PHQ), Perceived Stress Scale (PSS), Brief Symptom Inventory (BSI), Center for Epidemiologic Studies Depression (CES-D), Penn State Worry Questionnaire (PSWQ), and CDC HRQOL-14. Instruments for assessing substance use included Tobacco Questions for Surveys (TQS) and modified version of the WHO’s Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST). Only one study employed non-validated scales to assess happiness and life satisfaction ( Rahman et al., 2022 ). A substantial minority of studies (40.9%) relied on morbidity and mortality data extracted from medical records to measure mental health outcomes. For example, studies on suicide risk assessed outcomes using data on deaths from self-harm injuries ( Hanigan et al., 2022 ; Xu et al., 2024 ), while studies on MBDs frequently utilized emergency department (ED) visits or hospital admissions ( Lavigne et al., 2023 ; Moise, 2016 ). 3.3. The role of built environment features Fig. 3 illustrates all built environment factors assessed and their roles in the reviewed literature. Twelve included studies (54.5%) examined whether a specific factor was a predictor of mental health outcomes associated with exposure to climate-related disasters. Ten included studies (45.5%) evaluated whether a specific factor modified the strength or direction of the relationship between climate-related disasters and mental health outcomes. Fig. 3. Open in a new tab The roles of built environment features. 3.3.1. Predictive roles in mental health consequences of climate disasters The included studies assessed the following characteristics of built environments: remoteness, geographical locality or urbanicity, land cover (i.e., green space and developed areas), pre- and post-disaster housing type, neighborhood vacancy, neighborhood blight, flood-prone areas, and neighborhood recovery ( Table 2 ). No significant evidence was found regarding the mental health impacts of remoteness and pre-disaster housing type. However, land cover, post-disaster housing, and neighborhood characteristics pursuant to vulnerability were significant predictors of mental health conditions in relation to exposure to climate-related disasters. Although multiple built environment features have been found to be associated with mental health in the context of climate-related disasters, the number of studies examining some features is limited, with evidence extracted from only one or two studies. Table 2. Summary of results revealing predictive roles of built environment factors in the mental health consequences of climate-related disasters (N = 12). Citation Disaster type Built environment factor Mental health outcomes Covariates/confounders Results Austin et al. (2018) Drought Remoteness General psychological distress Sex, age, employment, financial security, marital status Remoteness: General psychological distress was not influenced by remoteness. Shabani et al. (2024) Flooding Geographical locality (urban or rural) & postdisaster housing type PTSD Age, gender, marital status, employment, education, assets, preexisting mental disorders, height of incoming floods, property damage, compensation for flood damage, access to health care services, temporary stay in a camp Locality: There is no significant association between locality (urban or rural residency) and PTSD. Post-disaster housing type: There is no significant association between living in a camp after the flood and PTSD. Xiong et al. (2016) Flood Urbanicity (urban vs. rural) PTSD Age, sex, ethnicity, religion, marital status, education, income, family members Urbanicity: 1) The prevalence of PTSD differed significantly between rural and urban areas ( X 2 (1, n = 200) = 16.783, p < 0.001). 2) The mean score of intrusion subscale between rural and urban respondents was significantly different. Mean score for intrusion was higher for rural (mean = 2.521) than urban (mean = 1.714) respondents ( p < 0.001; mean difference = 0.807, 95% CI: 0.461, 1.153). For the other two subscales of PTSD which are hyperarousal and avoidance, independent t -test showed no significant difference in mean. Burrows et al. (2023) Hurricane Katrina Residential greenness General psychological distress Age, race, social support, benefits receipt, mobility Residential greenness: 1) Cross-sectional residential greenness (living in areas of high vs. low residential greenness as measured by NDVI and EVI) was not significantly associated with psychological distress; 2) Change in level of greenness as measured by EVI (comparing pre- and post-Katrina neighborhoods) was associated with increased odds of distress at the first poststorm survey, compared to moving between or staying within low-greenness neighborhoods (low-high: OR = 3.48, 95% CI: 1.40, 8.62; high-low: OR = 2.60, 95% CI: 1.05, 6.42); 3) Change in residential greenness was not significantly associated with psychological distress when using NDVI. Li et al. (2021) Hurricane Harvey Neighborhood greenness PTSD Education, employment, poverty, assistance receipt, income, household structure Neighborhood greenness: 1) NDVI was significantly negatively related to hurricane-related distress ( B = −0.155, 95% CI: −0.276, −0.034); 2) NDVI was not associated with probable PTSD (OR = 0.818, 95% CI: −0.247, 2.707). Perceived quality of neighborhood green space: 1) Perceived quality of neighborhood green space was correlated with lower odds of probable PTSD (OR = 0.610, 95% CI: 0.422, 0.833). 2) Perceived quality was significantly associated with hyperarousal ( B = −0.137, 95% CI: −0.264, −0.010), but not with intrusion or avoidance 3) Perceived quality of green space had a statistically significant total effect ( β = −0.181, 95% CI: −0.325, −0.038) and direct effect ( β = −0.168, 95% CI: −0.319, −0.016) on probable PTSD, and a marginally significant partial mediation through emotional resilience ( β = −0.039, 95% CI: −0.079, 0.002). 4) The total effect of perceived quality of green space on hyperarousal was significant ( β = −0.121, 95% CI: −0.242, −0.001) and transmitted through emotional resilience ( β = −0.041, 95% CI: −0.080, −0.003). 5) Perceived quality of greenness had a significant effect on cohesion ( β = 0.293, 95% CI: 0.175, 0.410), but the path from cohesion to hyperarousal was insignificant, leading to an insignificant indirect effect ( β = 0.022, 95% CI: −0.014, 0.059). Zhang et al. (2020) Hurricane Sandy Areas prone to flooding, developed areas, & green coverage Mental and behavioral disorders Temperature, dew point, PM 2.5 , holidays, day of the week Landscape indicators: 1) Green coverage showed a low and negative contribution to vulnerability for mental disorders (−5.74%). 2) Developed areas made a low and negative contribution to vulnerability for mental disorders (−3.78%). 3) Areas prone to flooding had a great and positive contribution to vulnerability for mental disorders (27.43%). Newman et al. (2022) Hurricane Harvey Neighborhood vacancy & tree cover PTSD & healthrelated quality of life Age, gender, race/ethnicity, income, neighborhood deprivation Tree cover: No significant relationship was found with PTSD, nor with HRQOL. Neighborhood vacancy: 1) Percentage of vacant land area (objectively-measured) was a significant predictor of PTSD (OR = 1.477, 95% CI: 1.045, 2.089). 2) Perceived neighborhood vacancy was significantly associated with higher odds of PTSD (OR = 1.385, 95% CI: 1.038, 1.847). 3) The relationship between perceived vacancy and HRQOL approached statistical significance (OR = 1.289, 95% CI: 0.960, 1.731). 4) Objectively measured vacancy was not a significant predictor. Moise (2016) Hurricane Katrina Blighted properties & neighborhood recovery (nondisplaced residents & building demolition) Substance abuse disorders Age, sex, race, poverty, housing units, water depth Blighted properties: Post-Katrina, blighted block groups were significantly associated with higher rates of hospitalizations for substance abuse disorders ( b = 0.092, p < 0.001). Neighborhood recovery: 1) Nondisplaced residents: post-Katrina, areas with greater percentage of nondisplaced residents were significantly associated with higher rates of hospitalization for substance abuse disorders ( b = 0.03, p = 0.01); 2) Building demolition: areas with a greater percentage of completed demolitions had lower risk of substance abuse disorder hospitalizations ( b = −0.052, p = 0.02). Fussell and Lowe, (2014) Hurricane Katrina Pre-disaster housing type General psychological distress, PTSD, & stress Age, race, marital status, household structure, vehicle, public assistance receipt, housing damage, hurricanerelated traumas, pre-hurricane health Pre-disaster housing type: Housing before Hurricane Karina was not a significant predictor of psychological distress, PTSD, or stress. Desalvo et al. (2007) Hurricane Katrina Post-disaster housing type PTSD Age, sex, race, education, income, evacuation experiences, current commute, traumatic experiences Post-disaster housing type: Living in a temporary trailer after Hurricane Katrina was associated with higher likelihood of PTSD, compared to living in the same dwelling as before the disaster (OR = 2.01, 95% CI: 1.15, 3.53). Lu and Singelmann (2009) Hurricane Katrina & Hurricane Rita Post-disaster housing type Anxiety Sex, race, marital status, employment, education, income, poster-disaster residential stability Post-disaster housing type: Living in a Louisiana trailer park, as opposed to an apartment in Houston, increased anxiety levels ( b = 0.471, p < 0.01). Schwartz et al. (2018) Hurricane Sandy Post-disaster housing type Depression, anxiety, PTSD and substance use Age, gender, education, race/ethnicity, pre-existing mental health diagnosis, and the number of days displaced, time between hurricane exposure and questionnaire completion Post-disaster housing type: Staying with friends and family versus being at a shelter was significantly associated with a 48 percent decreased odds of having PTSD symptoms (AOR: 0.52, 95% CI: 0.31–0.88) and a 48 percent decreased odds of being a current tobacco user (AOR: 0.52, 95% CI: 0.30–0.92). There were no statistically significant associations between type of displacement and all other outcomes. Open in a new tab Notes: OR: odds ratio; AOR: adjusted odds ratio; β : standardized regression coefficient; B , b : unstandardized regression coefficients; CI: confidence interval, p : p-value. Urbanicity/geographical locality. Research on the relationship between locality and mental health in the context of climate-related disasters is very limited. Xiong et al. (2016) demonstrated a significant urban-rural difference in PTSD symptoms after flooding, with rural residents experiencing more severe intrusion symptoms compared to their urban counterparts. Shabani et al. (2024) also found that rural residence was associated with increased odds of PTSD compared to urban residence; however, this association was not statistically significant (AOR = 1.30, 95% CI: 0.77, 2.18). Land cover/green spaces. The majority of studies identified an association between increased green space and reduced PTSD. Li et al. (2021) found that the perceived quality of green space was associated with lower odds of probable PTSD (OR = 0.610, 95% CI: 0.422, 0.833) and lower hurricane-related distress ( B = −0.137, 95% CI: −0.264, −0.010), while quantity of green space (as measured by NDVI) was only associated with hurricane-related distress ( B = −0.155, 95% CI: −0.276, −0.034). Zhang et al. (2020) revealed that both green coverage and developed areas make negative contributions to vulnerability to mental and behavioral disorders. However, Burrows et al. (2023) demonstrated that change in greenness level (comparing pre- and post-hurricane neighborhoods), either low to high or high to low, was significantly associated with increased psychological distress (low-high: OR = 3.48, 95% CI: 1.40, 8.62; high-low: OR = 2.60, 95% CI: 1.05, 6.42), provided EVI was the measure considered; significant results were not found for NDVI. However, a fourth study demonstrated inconsistent results, namely that tree canopy cover was not associated with PTSD or health-related quality of life in disaster-affected areas ( Newman et al., 2022 ). Housing type. A few studies suggested significant differences in the mental disorders experienced by hurricane evacuees in relation to housing type. For example, Lu and Singelmann (2009) found that post-hurricane living in a trailer park, compared to an apartment, was associated with an increase in anxiety level ( b = 0.471, p < 0.01). Similarly, Desalvo et al. (2007) indicated that people living in a temporary trailer after a hurricane were at higher risk of PTSD compared to those who remained in their original residence (OR = 2.01, 95% CI: 1.15, 3.53). Characteristics of the neighborhood. Objectively measured neighborhood vacancy was significantly associated with higher odds of PTSD (OR = 1.477, 95% CI: 1.045, 2.089), but not a significant predictor of HRQOL ( Newman et al., 2022 ). In hurricane-affected communities, neighborhood blight was significantly associated with higher rates of hospitalization for substance abuse ( b = 0.092, p < 0.001) ( Moise, 2016 ).However, neighborhoods having a higher percentage of flood-prone area exhibited greater vulnerability to combined MBDs alongside lower vulnerability to substance abuse disorders ( Zhang et al., 2020 ). Regarding the impact of neighborhood recovery from disasters, results varied according to the indicator. For example, areas with a greater proportion of nondisplaced residents significantly associated with higher substance abuse risk ( b = 0.03, p = 0.01), while those having a greater percentage of completed demolitions demonstrated a lower risk of substance abuse disorders ( b = −0.052, p = 0.02) ( Moise, 2016 ). 3.3.2. Effect modifiers of the climate disasters–mental health relationship Collectively, the included studies examined effect modification by four built environment features: urban-rural differences/urbanization, neighborhood deprivation, land cover (i.e., green space, blue space, and grey space), and sky view factor ( Table 3 ). No evidence supported that sky view factor could modify the association of mental health with disaster exposure; conversely, evidence confirmed that urbanicity, green space, blue space, and neighborhood deprivation all play significant roles in modifying that relationship. Although existing studies have examined the effect modification by these built environment features, the research attention has been primarily paid to urbanicity and green space, with limited evidence on other features. Table 3. Summary of results revealing effect modification by built environment factors on the association between climate-related disasters and mental health (N = 10). Citation Disaster type Built environment factor Mental health outcomes Covariates/confounders Results Gros et al. (2012) Hurricane Ike Urbanicity (urban vs. non-urban) Depression, PTSD, and worry Age, race/ethnicity, sex, education, income Urban settings: 1) Loss of services was significantly associated with PTSD symptoms was significant ( B = 0.03, p < 0.05); 2) Loss of services was not significantly associated with depression. Non-urban settings: 1) Loss of services had a significant positive relation with PTSD symptoms ( B = 0.08, p < 0.01); 2) Loss of services was significantly positively related with and depression for those in non-urban settings ( B = 0.03, p < 0.01). The effect of loss of services on worry did not vary across urban and non-urban settings. Hanigan et al. (2012) Drought Urbanicity (urban vs. rural) Suicide Age, sex, season, time trend Urban settings: The analysis for the urban population found no association between drought and suicide. Rural settings: 1) An increased relative risk of suicide of 15% (95% CI: 8%–22%; p < 0.0001) was observed among rural males aged 30–49 years when the drought index rose from the first quartile to the third quartile; 2) The risk of suicide for rural females aged >30 y declined with increased values of the drought index (95% CI: −1.32 to −0.01; p < 0.05); 3) An increase in drought index values was significantly associated with an increased risk of suicide among rural males aged 10–29 years ( p < 0.01). Hanigan et al. (2022) Drought Urbanicity (urban vs. rural) Suicide Age, sex, season, time trend Urban settings: Drought was not associated with urban suicide rates. Rural settings: 1) Drought and suicide had a linear relationship among rural males aged 30–49 yrs for interquartilerange (IQR) increases in the drought index (RR = 1.14, 95% CI: 1.07, 1.22; p < 0.0001); 2) Among rural females of 30–49 years, the risk of suicide decreased with IQR increases in drought duration (RR = 0.86, 95% CI: 0.80, 0.91; p < 0.05); 3) Drought-associated risk was also identified among rural males aged 10–29 yrs ( p < 0.01). Obrien et al. (2014) Drought Urbanicity (urban vs. rural) General psychological distress Age, sex, relationship status, income Urban settings: Distress was not significantly associated with any dryness pattern (very dry drought, long drought, constant drought, and constant and recent long drought). Rural settings: Exposure to long and constant drought was significantly associated with increased distress ( B = 2.49, 95% CI: 0.59, 4.39). Phan Minh et al. (2016) Heatwaves Urbanicity (urban vs. rural) Mental and behavioral disorders Season, day of the week, time trend Urban settings: 1) All MBDs (F01-F99) did not have significant association with heatwaves. 2) High relative risk of mental retardation (F70-F79) was identified when the daily maximum temperature at lags reached or exceeded 35 ◦ C for 3 consecutive days among residents in urban areas (RR = 1.83, 95% CI: 1.11, 3.04). Rural settings: 1) The highest relative risk was observed when the daily maximum temperature at lags reached or exceeded 35 ◦ C for 7 consecutive days (F00-F99, except F60–69: RR = 1.69, 95% CI: 1.08, 2.64; F0–9: RR = 7.82, 95% CI: 2.66, 22.96); 2) The number of mental disorder cases increased among residents in rural areas with at least 3 days of heat in the 95th percentile (RR = 1.36, 95% CI: 1.00, 1.86). Rahman et al. (2022) Combined climate disasters Urbanicity (urban vs. rural) Happiness & life satisfaction Age, gender, marital status, family members, education, health condition, income, assets Urban settings: Climate disasters were not significantly associated with urban resident happiness and life satisfaction. Rural settings: Climate disasters were associated with decreases in happiness ( B = −0.0080, p < 0.1) and life satisfaction ( B = −0.0079, p < 0.05). Lavigne et al. (2023) Extreme temperature Neighborhood deprivation, residential green space, & urbanization Mental and behavioral disorders Relative humidity, mean temperature, temperature range Urbanization: Evidence was not clear. Neighborhood deprivation: Impacts of heat on MBDs were lowest among those residing in materially and socially privileged neighborhoods (OR = 1.042, 95% C I: 0.971, 1.119) and highest among those living in materially and socially deprived neighborhoods (OR = 1.181, 95% CI: 1.129, 1.231) ( p -value for trend = 0.02). Residential green space: Greenness exposure in the residential environment, specifically increasing tree canopy coverage, appeared to attenuate the impacts of heat on MBD ED visits ( p -value for trend = 0.01). Pan et al. (2024) Heatwaves Land cover Schizophrenia PM 2.5 Land cover: In the context of heatwave exposure, combined heatwave and PM 2.5 exposure was associated with increased schizophrenia risk, but individuals residing in areas rich in blue spaces had the lowest risk. 1) Green space: (percent change: 1.464%, 95% CI: 0.932, 1.997) 2) Blue space: (percent change: 0.122%, 95% CI: 0.01, 0.235) 3) Grey space: (percent change: 4.169%, 95% CI: 3.965, 4.373) Xu et al. (2024) Drought Vegetation coverage Suicide Season, year Vegetation coverage: Geen space coverage moderates the relationship between extreme drought and suicide rate (the marginal effects of extreme drought on suicide rates became smaller in areas with higher NDVI value). Ho et al. (2018) Haze Sky view factor (SVF) & vegetation coverage Mental and behavioral disorders (MHDs) Air temperature, relative humidity, NO2, O3, SO2, anthropogenic heat, day of week, season Vegetation coverage: Compared with all areas across Hongkong, areas with lower vegetation (percentage of vegetation <50th percentile) were associated with higher likelihood of mortality related to MHDs on a 1-day lag (regular hazy day: OR = 1.387, 95% CI: 1.155, 1.666; hazy day with higher ozone: OR = 1.922, 95% CI: 1.298, 2.846; cold hazy day: OR = 1.873, 95% CI: 1.122, 3.124). Sky view factor: A high-density environment (SVF <50th percentile) had no contribution to mental issues during all haze events. Open in a new tab Notes: OR: odds ratio; RR: relative risk; B : unstandardized regression coefficient; CI: confidence interval; p : p-value. Urbanicity. The majorities of the included studies suggested the adverse impacts of disasters on mental health to be less pronounced in urban regions compared to non-urban regions ( Gros et al., 2012 ; Phan Minh et al., 2016 ). For example, Obrien et al. (2014) identified a significant relationship of drought with increased risks of psychological distress among rural residents ( B = 2.49, 95% CI: 0.59, 4.39), while observing decreased risks of distress in urban regions, although the latter result was not statistically significant ( B = −0.81, 95% CI: −1.70, 0.09). Rahman et al. (2022) also found that climate disasters did not significantly affect urban residents’ happiness and life satisfaction but were associated with decreases in rural residents’ subjective well-being (happiness: B = −0.0080, p < 0.1; life satisfaction: B = −0.0079, p < 0.05). Land cover/green spaces. Significant effects of green and blue spaces on mitigating the mental health impacts of climate hazards were reported in three studies. One found that increasing tree canopy appeared to mitigate the negative impacts of heat on mental health ( Lavigne et al., 2023 ), while the second found a lower level of greenness exposure to be associated with higher likelihood of mental health-related mortality during hazy days (OR = 1.387, 95% CI: 1.155, 1.666) ( Ho et al., 2018 ). Finally, Pan et al. (2024) identified that under combined exposure to heatwaves and PM 2 . 5 , individuals residing in areas rich in blue spaces had the lowest risk of schizophrenia (percentage change: 0.122%, 95% CI: 0.01, 0.235), while those in environments dominated by built-up areas experienced the highest vulnerability (percentage change: 4.169%, 95% CI: 3.965, 4.373). Characteristics of the neighborhood. Only one study evaluated effect modification by neighborhood deprivation. The evidence suggested that the impact of heat on MBDs was minimal among people residing in privileged neighborhoods (OR = 1.042; 95% CI: 0.971–1.119) and most severe among those in deprived neighborhoods (OR = 1.181; 95% CI: 1.129–1.231) ( Lavigne et al., 2023 ). 3.4. Risk of bias Fig. 4 presents the results of the RoB assessment. Of the 22 reviewed studies, only one (4.5%) was rated as having moderate RoB in the selection domain; that study conducted two rounds of participant recruitment with an interval of seven to nine months between the phases. However, the evidence provided in the study indicated that the lag between the two surveys did not corrupt the results. In the confounding bias domain, four studies (18.2%) were assessed as having moderate RoB as they adopted self-report surveys to measure variables related to evacuation processes, displacement characteristics, housing damage, or characteristics of particular disasters; hence, recall bias and subjectivity might have limited their reliability and accuracy. In the exposure assessment domain, five studies (22.7%) were classified as having moderate RoB due to not directly measuring built environment exposure or disaster severity, which might introduce perception bias. In the outcome assessment domain, 13 studies (59.1%) were assessed as having moderate RoB. Of these, 12 studies assessing mental diseases relied on self-reported surveys rather than clinical diagnosis, while one study evaluating subjective well-being depended on a non-valid instrument. In the attrition/exclusion domain, only one study (4.5%) categorized as high RoB; this was due to large attrition (about 75%) without well-documented methods for addressing that problem. Three studies (13.6%) were assessed as having moderate RoB because of sample exclusion from analyses; while the proportion was acceptable (about 10%), insufficient information was provided concerning the exclusion and its rationale. In the selective reporting domain, all studies were assessed as having low RoB. Detailed evaluations of each study are provided in Supplementary Material Figure S1 and Figure S2 . Fig. 4. Open in a new tab Results of the risk of bias assessment. 4. Discussion 4.1. Key findings This systematic review synthesized evidence regarding the association of built environments with mental health outcomes in diverse disaster scenarios, including hurricanes, drought, heatwaves, floods, and haze. The mental health impacts examined included PTSD, general psychological distress, suicide, anxiety, depression, substance abuse, schizophrenia, ED visits/hospital admissions related to MBDs, and subjective well-being. In addition, most of the included cross-sectional studies adjusted for demographic factors (e.g., age, gender/sex, race/ethnicity, marital status, and family structure), socioeconomic factors (e.g., education, employment, income, assets, poverty level, commuting, assistance receipt, access to health care services), health-related factors (e.g., pre-existing diseases), or disaster exposure-related factors (i.e., traumatic experiences, property damage, floodwater depth, evacuation, and displacement) as covariates or confounding variables. Some of the time-stratified studies considered time-related variables such as season, long-term trend, day of the week, weather variables, and air pollutant concentrations. In this review, we identified extant observational studies to document two roles of built environment factors: 1) predicting mental health outcomes in relation to disaster exposure, involving such features as remoteness, urbanicity (urban vs rural), land cover (i.e., green space and developed area), housing type, neighborhood vulnerability (i.e., vacancy, blight, and flood-prone area), and neighborhood recovery, and 2) modifying the strength and direction of the relationship between mental health outcomes and climate-related disaster exposure, involving urbanicity, land cover (i.e., green space, blue space, and grey space), neighborhood deprivation, and sky view factor. In the predictor role, while diverse built environment features have been observed to be associated with mental health in relation to climate-related disasters, the evidence on each specific feature remains limited. Most studies have focused on green space and post-disaster housing type, with findings that are inconsistent across studies. In the role of effect modifier, some research has assessed the effect modification by built environment factors; however, such examinations have primarily focused on features such as urbanicity and land cover, highlighting a critical gap and the need for broader investigation of other built environment features. In particular, we observed several contradictions between the direct effect and modifying effects of built environment features on mental health outcomes in the context of climate-related disasters. For example, green space demonstrated mixed results of direct associations with climate disaster-related mental health ( Burrows et al., 2023 ; Li et al., 2021 ), but more consistently attenuated adverse mental health impacts when considered in relation to specific climate hazards ( Ho et al., 2018 ; Lavigne et al., 2023 ). Similarly, built-up intensity appeared protective in a direct-predictor analysis ( Zhang et al., 2020 ), while exacerbating vulnerability under hazard-specific conditions such as heatwaves ( Pan et al., 2024 ). These divergent patterns underscore the inherent complexity of relationships between built environment features and mental health outcomes in the climate disaster context. Discrepancies between these two roles do not necessarily indicate inconsistency in the evidence, but rather point to distinct underlying mechanisms operating at different stages of exposure and response. When examined as direct predictors, built environment characteristics estimate overall associations with mental health outcomes, whereas as effect modifiers they capture differential vulnerability by indicating how environmental context alters the mental health impacts of disaster exposure. Therefore, failure to distinguish these dual roles may lead to incomplete or misleading interpretation of the impact of built environment features on climate disaster-related mental health. Overall, positive impacts were reported for natural environments. For example, some studies examining the effect of nature exposure suggested that greater access to green spaces would lower the likelihood of mental disorders such as PTSD in disaster-affected areas ( Li et al., 2021 ; Zhang et al., 2020 ). Similarly, others suggested that increases in natural resources such as green and blue spaces would mitigate the adverse effects of climatic hazards on mental health, for example mitigating MBD-related mortality risk during hazy days, or attenuating MBD-related ED visits during heatwaves ( Ho et al., 2018 ; Pan et al., 2024 ). The synthesized evidence aligns with the findings in a previous systematic review on the impact of nature-led recovery for post-disaster health, emphasizing that natural environments could benefit community connectedness, resilience, and spiritual and emotional nourishment ( Porter et al., 2023 ). Existing research indicates that housing instability is a key determinant of mental health( Kang, 2022 ; Suglia et al., 2011 ), and the manner by which a post-disaster housing strategy is planned and designed may influence the recovery process of disaster survivors ( Biswas, 2019 ). Consistent with these reports, the reviewed studies on post-disaster temporary housing such as trailers demonstrated that instable residential environments would increase the risk of mental diseases like PTSD and anxiety among individuals affected by climate-related disasters ( Desalvo et al., 2007 ; Lu and Singelmann, 2009 ). Finally, most included studies examining the urban-rural differences in mental health outcomes associated with climatic hazards revealed rural residents to be at higher risk of mental illnesses compared to urban residents ( Hanigan et al., 2022 ; Obrien et al., 2014 ). These consistent findings may be explained by rural areas facing different opportunities and challenges in disaster management and often having a lower capacity for disaster resilience due to limited financial resources, low population density, and insufficient social networks for mitigation practice, rebuilding efforts, and resilience support ( Kapucu et al., 2013 ; Parsons et al., 2021 ). However, an existing systematic review and meta-analysis on urban-rural differences in psychiatric disorders found the prevalence of psychiatric disorders such as mood disorders and anxiety to be significantly higher in urban areas, suggesting an association of urbanization with increased mental illnesses in normal circumstances ( Peen et al., 2010 ). These differential findings highlight the unique challenges that climate-related disasters pose to the mental health of rural populations. Further research remains needed to better understand the urban-rural differences in specific mental disorders within the context of climate hazards. 4.2. Limitations and future directions This is the first systematic review to consider both a wide range of climatic disasters and built environment features, with a particular emphasis on the impact of built environments on mental health outcomes in the specific context of climate-related disasters. While this review utilized a comprehensive search strategy to synthesize evidence on this topic, strictly following the Cochrane and PRISMA guidelines, it also has several limitations that should be considered. First, the literature search was restricted to quantitative studies written in English and published in peer-reviewed journals; the exclusion of non-English publications, non-peer-reviewed articles (e.g., dissertations, conference publications, and reports), and published studies using experimental or quasi-experimental study designs may cause the possibility of publication bias. Also, given the predefined scope and search strategy, some relevant studies may not have been captured. Second, while a meta-analysis could minimize the influence of subjective synthesis using statistical methods and produce robust evidence, the heterogeneity in exposure and outcome measures and variety of methods adopted across the included studies precluded conducting a meta-analysis. Our literature search did not find any studies conducted in the European, South American, or African countries. The lack of research from these diverse geographical and cultural contexts presents a challenge for understanding regional differences in the impacts of built environments on mental health outcomes in relation to climate-related disasters. Expanding the body of literature into such underrepresented regions is essential for addressing knowledge gaps and providing a more comprehensive understanding of geographic variations. Additionally, among the various built environment characteristics examined, evidence on the effects of features such as neighborhood vacancy, neighborhood blight, and neighborhood deprivation was derived from only one or two studies. Relying on a single study might include methodological bias and raises concerns about reliability. Thus, replication is warranted that can assess the consistency of results and strengthen the evidence base so as to advance evidence-informed policymaking. Moreover, most reviewed studies focused on the mental health impact of built environments among populations affected by hurricanes, floods, heat waves, and drought. Growing evidence has indicated that climatic disasters such as wildfires, typhoons, and landslides also lead to adverse mental disorders ( Burrows et al., 2021 ; Tito et al., 2024 ; To et al., 2021 ); accordingly, more studies are needed to clarify the influences of built environment factors on mental health outcomes across a wide range of disaster types. In addition, existing literature displayed inconsistent consideration and measurement of disaster severity, which might constrain interpretation of the relationships between built environment features and mental health outcomes. Disaster exposure was often characterized by binary indicators (affected vs. unaffected), with limited attention to variations in intensity, duration, or cumulative impacts. This lack of standardization may contribute to heterogeneous and conflicting findings. For example, prior research on the direct association between green space and hurricane-related mental health has reported mixed results ( Burrows et al., 2023 ; Li et al., 2021 ), which might be partially attributed to inadequate consideration of the discrepancy in different hurricanes’ characteristics. Without accounting for variations in disaster severity, it remains unclear whether these inconsistencies reflected true effects of built environment features or underlying differences in disaster exposure. Future research should therefore also adopt more standardized measures of disaster characteristics to enable stronger comparability across studies. Finally, subjective well-being is an important psychological and sociological parameter that encompasses both mental health and life quality in the form of life satisfaction and emotional responses to ongoing life ( Diener et al., 2018 ). In the context of disaster impact research, assessment of both mental health outcomes and subjective well-being is valuable for capturing the full spectrum of disaster recovery and informing health-promoting interventions. Although we included subjective well-being in this review, only a limited number of studies employed such constructs. Additional research focusing on subjective well-being is critical, with a particular emphasis on using validated evaluation tools to ensure the accuracy of findings and support policy decisions. Collectively, differences in methodologies such as contextual settings and disaster characteristics, as well as limited availability of research on this topic, may contribute to inconsistent findings across the extant literature. To better understand the complexity of the relationships between built environment features and mental health outcomes in disaster contexts, future research is therefore required not only to more clearly distinguish dual roles of built environment features but also to apply more consistent and transparent measurement strategies and examine a broader range of disaster types, built environment features, and mental health outcomes. Addressing these issues would enhance interpretability, strengthen comparability across studies, and improve the applicability of findings for disaster risk reduction and adaptation planning. In addition to methodological issues, this review has identified research and knowledge gaps, unanswered questions that can advance and direct future research efforts. 4.2.1. Long-term patterns of built environment features’ effects on mental health and well-being following climate-related disaster exposure Natural disasters may increase the risk for psychiatric disorders, but the prevalence of such disorders tends to decline over time, suggesting that most people are resilient to the mental health effects of disaster exposure ( Pietrzak et al., 2012 ). However, longitudinal research on mental health trajectories has indicated that the impact of natural disaster exposure on mental health might persist over time, particularly among severely affected populations ( Bryant et al., 2014 ; Schwartz et al., 2017 ). As such, the psychological effects of environmental factors would also persistently play important roles in survivors’ lives. However, in this review, the identified cross-section studies always focused on the effect of built environment features on mental health at a specific point in time. There is yet a lack of evidence on how these features impact the mental health recovery process of disaster-affected populations over time. In addition, trajectories of psychological recovery after disasters vary by age, with children and adolescents reporting higher levels of depression and anxiety symptoms than adults at all timepoints ( Newnham et al., 2022 ). Prior research also demonstrated that experiencing one or more natural disasters during childhood would increase the risk of developing mental health disorders in adulthood ( Adebäck et al., 2018 ; Maclean et al., 2016 ). However, the literature included in this review primarily examined the mental health consequences of adults, with susceptible groups such as children and adolescents being underrepresented. Also, evidence is lacking on the psychological effects of built environment exposure over the course of young disaster survivors’ later lives. How does the impact of built environment features on mental health in relation to climatic disasters change during the recovery process? Does the impact of built environments on post-disaster mental health recovery differ between children and adults? Does exposure to a built environment feature in relation to early childhood disaster exposure impact later mental health outcomes? 4.2.2. The role of built environment features in coping with mental health burdens associated with cumulative or multiple disaster exposure Climate extremes are becoming more frequent due to climate change. Most regions have experienced significant impacts of recurring climate risks; for example, since 2000, the state of Florida has been struck by thirteen hurricanes, Louisiana by twelve, and Texas by nine ( Hurricane Research Division, 2023 ). In addition, there is increasing evidence indicating the potential for natural disasters to overlap, cascade, and compound ( AghaKouchak et al., 2020 ; Pescaroli and Alexander, 2018 ), which demonstrates that people are encountering multiple disasters. Finally, recent research has revealed the impacts of multiple-disaster scenarios on mental health. For example, compounding events caused by co-occurring heatwaves and drought were linked to a threefold increase in ED visits due to suicide and mood disorders ( Sewell et al., 2024 ). Therefore, exposure to multiple disasters is likely to pose a greater risk for adverse mental health outcomes than exposure to a single disaster. However, little is known about the effects of built environment features on the mental health outcomes associated with exposure to multiple disaster scenarios. How do built environment features (e.g., housing, green space, or neighborhood characteristics) impact the mental health effects associated with multiple disaster exposure (e.g., repeated, concurrent, or cascading disasters)? Does the impact of built environment features on mental health differ between exposure to a single disaster and exposure to multiple disasters? In which type of disaster scenario does the built environment have the most significant impact on mental health? 4.3. Policy implications Despite the limitations of this review and of the included studies, the evidence synthesized in this review suggests that disaster preparedness and recovery policies can be strengthened by adopting a more differentiated approach to built environment interventions, informed by the varied pathways through which environmental characteristics influence mental health outcomes in the context of climate-related disasters. Features such as urbanicity and green and grey infrastructures demonstrated both direct associations with disaster-related mental health and the capacity to modify the psychological impacts of climate hazards, underscoring their relevance across multiple phases of disaster management. These findings complement existing disaster risk reduction and climate adaptation policies by highlighting the value of incorporating land use patterns, environmental quality, and settlement structure into preparedness planning and recovery frameworks as modifiable components of mental health resilience, rather than treating them solely as contextual background conditions. Moreover, housing and neighborhood characteristics such as housing type, neighborhood blight, and vacancy, appeared to exert more consistent direct impacts on mental health in disaster-affected settings, while also conditioning vulnerability under hazard exposure. This evidence reinforces recovery-oriented policy priorities that emphasize housing stability, neighborhood revitalization, and mitigation of environmental deterioration, and suggests that these domains represent important leverage points for reducing disaster-related mental health risks and supporting sustained recovery. The evidence collectively extends existing policy frameworks by demonstrating how long-term environmental improvement and hazard-specific mental health protection can be more effectively aligned through the integration of built environment considerations into disaster preparedness and recovery planning. 5. Conclusion Overall, this systematic review synthesizes evidence on the roles of built environment features in mental health outcomes following climate-related disasters. Despite heterogeneity and methodological challenges, these included studies provide new insights into the psychological effects of built environments and policy implications for disaster preparedness and recovery. Future research is required to improve methodologies, address knowledge gaps, and ultimately develop a better understanding of how various built environment features impact mental health and well-being amid increasingly severe climatic hazards driven by climate change. Supplementary Material ZHANGetal2026_Supplementary NIHMS2169307-supplement-ZHANGetal2026_Supplementary.doc (804.5KB, doc) Acknowledgments This research was funded by the National Institute on Minority Health and Health Disparities (1R01MD16587), National Institute of Environmental Health Sciences (P30ES029067, P42ES027704), and National Academy of Sciences, Engineering, and Medicine (2000013443). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.socscimed.2026.119130 . Footnotes Ethnics approval Ethics approval is not required, as this study is a systematic review based solely on published results and does not involve data collection from human subject. CRediT authorship contribution statement Yue Zhang: Conceptualization, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing. Dongying Li: Conceptualization, Funding acquisition, Methodology, Visualization, Writing – original draft, Writing – review & editing. Galen Newman: Conceptualization, Methodology, Visualization, Writing – original draft, Writing – review & editing. Yining Liu: Conceptualization, Methodology, Visualization, Writing – original draft, Writing – review & editing. Emily Besse: Conceptualization, Methodology, Visualization, Writing – original draft, Writing – review & editing. Data availability No data was used for the research described in the article. References Adebäck P, Schulman A, Nilsson D, 2018. Children exposed to a natural disaster: psychological consequences eight years after 2004 tsunami. Nord. J. Psychiatr. 72 (1), 75–81. [ DOI ] [ PubMed ] [ Google Scholar ] AghaKouchak A, Chiang F, Huning LS, Love CA, Mallakpour I, Mazdiyasni O, Moftakhari H, Papalexiou SM, Ragno E, Sadegh M, 2020. Climate extremes and compound hazards in a warming world. Annu. Rev. Earth Planet Sci. 48 (1), 519–548. [ Google Scholar ] Arifwidodo SD, Chandrasiri O, 2020. Urban heat stress and human health in Bangkok, Thailand. Environ. Res. 185, 109398. 10.1016/j.envres.2020.109398. [ DOI ] [ PubMed ] [ Google Scholar ] Austin EK, Handley T, Kiem AS, Rich JL, Lewin TJ, Askland HH, Askarimarnani SS, Perkins DA, Kelly BJ, 2018. Drought-related stress among farmers: findings from the Australian rural mental health study. Med. J. Aust. 209 (4), 159–165. 10.5694/mja17.01200. [ DOI ] [ PubMed ] [ Google Scholar ] Baylan E, Yalçin EA, Ozcan Z, 2025. The role of green spaces in post-disaster recovery: an exploratory study among migrant survivors of the 6 february earthquakes in edi˙rne, Turki˙ye. J. Lands. Res. Practices (Jolarp) 7 (1), 27–35. [ Google Scholar ] Beaglehole B, Mulder RT, Frampton CM, Boden JM, Newton-Howes G, Bell CJ, 2018. Psychological distress and psychiatric disorder after natural disasters: systematic review and meta-analysis. Br. J. Psychiatr. 213 (6), 716–722. [ DOI ] [ PubMed ] [ Google Scholar ] Benevolenza MA, DeRigne L, 2019. The impact of climate change and natural disasters on vulnerable populations: a systematic review of literature. J. Hum. Behav. Soc. Environ. 29 (2), 266–281. [ Google Scholar ] Biswas A, 2019. Exploring Indian post-disaster temporary housing strategy through a comparative review. Int. J. Disaster Resil. Built Environ. 10 (1), 14–35. [ Google Scholar ] Bryant RA, Waters E, Gibbs L, Gallagher HC, Pattison P, Lusher D, MacDougall C, Harms L, Block K, Snowdon E, 2014. Psychological outcomes following the victorian black Saturday bushfires. Aust. N. Z. J. Psychiatr. 48 (7), 634–643. [ DOI ] [ PubMed ] [ Google Scholar ] Burrows K, Desai MU, Pelupessy DC, Bell ML, 2021. Mental wellbeing following landslides and residential displacement in Indonesia. SSM-Mental Health 1, 100016. [ Google Scholar ] Burrows K, Fong KC, Lowe SR, Fussell E, Bell ML, 2023. The impact of residential greenness on psychological distress among hurricane katrina survivors. PLoS One 18 (5). 10.1371/journal.pone.0285510. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cartwright A, Khalatbari-Soltani S, Zhang Y, 2025. Housing conditions and the health and wellbeing impacts of climate change: a scoping review. Environ. Res, 120846 [ DOI ] [ PubMed ] [ Google Scholar ] Cianconi P, Betrò S, Janiri L, 2020. The impact of climate change on mental health: a systematic descriptive review. Front. Psychiatr. 11, 490206. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Dargin JS, Mostafavi A, 2020. Human-centric infrastructure resilience: uncovering well-being risk disparity due to infrastructure disruptions in disasters. PLoS One 15 (6), e0234381. 10.1371/journal.pone.0234381. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Desalvo KB, Hyre AD, Ompad DC, Menke A, Tynes LL, Muntner P, 2007. Symptoms of posttraumatic stress disorder in a New Orleans workforce following hurricane katrina. J. Urban Health 84 (2), 142. 10.1007/s11524-006-9147-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Diener E, Oishi S, Tay L, 2018. Advances in subjective well-being research. Nat. Hum. Behav. 2 (4), 253–260. [ DOI ] [ PubMed ] [ Google Scholar ] Fussell E, Lowe SR, 2014. The impact of housing displacement on the mental health of low-income parents after Hurricane Katrina. Soc. Sci. Med. 113, 137–144. 10.1016/j.socscimed.2014.05.025. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Gros DF, Price M, Gros KS, Paul LA, McCauley JL, Ruggiero KJ, 2012. Relations between loss of services and psychiatric symptoms in urban and non-urban settings following a natural disaster. J. Psychopathol. Behav. Assess. 34 (3), 343. 10.1007/s10862-012-9290-9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Hanigan IC, Butler CD, Kokic PN, Hutchinson MF, 2012. Suicide and drought in New South Wales, Australia, 1970–2007. Proc. Natl. Acad. Sci. USA 109 (35), 13950. http://proxy.library.tamu.edu/login?url=https://www.proquest.com/docview/1037345573?accountid=7082&bdid=105002&_bd=%2Ffb811GooSsJEIrPyPZ4iwZ4Hng%3D . [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Hanigan IC, Hanigan IC, Chaston TB, 2022. Climate change, drought and rural suicide in New South Wales, Australia: future impact scenario projections to 2099. Int. J. Environ. Res. Publ. Health 19 (13), 7855. 10.3390/ijerph19137855. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Higgins JP, 2008. Cochrane Handbook for Systematic Reviews of Interventions Version 5.0. 1. The Cochrane Collaboration. www.cochrane-handbook.org . [ Google Scholar ] Ho HC, Wong MS, Yang L, Shi W, Yang J, Bilal M, Chan T-C, 2018. Spatiotemporal influence of temperature, air quality, and urban environment on cause-specific mortality during hazy days. Environ. Int. 112, 10–22. [ DOI ] [ PubMed ] [ Google Scholar ] Hobfoll SE, 1989. Conservation of resources: a new attempt at conceptualizing stress. Am. Psychol. 44 (3), 513. [ DOI ] [ PubMed ] [ Google Scholar ] Hurricane Research Division, 2023. Detailed list of Continental United States hurricane impacts/landfalls 1851–1965, 1983–2023. https://www.aoml.noaa.gov/hrd/hurdat/UShurrs_detailed.html . [ Google Scholar ] Kang S, 2022. The cumulative relationship between housing instability and mental health: findings from the panel study of income dynamics. J. Soc. Distress Homeless 31 (2), 191–203. [ Google Scholar ] Kaplan S, 1995. The restorative benefits of nature: toward an integrative framework. J. Environ. Psychol. 15 (3), 169–182. [ Google Scholar ] Kaplan R, Kaplan S, 1989. The Experience of Nature: a Psychological Perspective. Cambridge university press. [ Google Scholar ] Kapucu N, Hawkins CV, Rivera FI, 2013. Disaster preparedness and resilience for rural communities. Risk Hazards Crisis Publ. Pol. 4 (4), 215–233. [ Google Scholar ] Keya TA, Leela A, Habib N, Rashid M, Bakthavatchalam P, 2023. Mental health disorders due to disaster exposure: a systematic review and meta-analysis. Cureus 15 (4). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Lavigne E, Maltby A, Côté J-N, Weinberger KR, Hebbern C, Vicedo-Cabrera AM, Wilk P, 2023. The effect modification of extreme temperatures on mental and behavior disorders by environmental factors and individual-level characteristics in Canada. Environ. Res. 219, 114999. [ DOI ] [ PubMed ] [ Google Scholar ] Li A, Toll M, Chapman R, Howden-Chapman P, Hernández D, Samuelson H, Woodward A, Bentley R, 2025. Housing at the intersection of health and climate change. Lancet Public Health 10 (10), e865–e873. [ DOI ] [ PubMed ] [ Google Scholar ] Li D, Newman G, Zhang T, Zhu R, Horney J, 2021. Coping with post-hurricane mental distress: the role of neighborhood green space. Soc. Sci. Med. 281. 10.1016/j.socscimed.2021.114084. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Li D, Zhang Y, Li X, Zhang K, Lu Y, Brown RD, 2023. Climatic and meteorological exposure and mental and behavioral health: a systematic review and meta-analysis. Sci. Total Environ. 892, 164435. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Lu A, Singelmann J, 2009. Location matters? A comparison of placement conditions for hurricane evacuees in Houston and Louisiana. Int. J. Mass Emergencies Disasters 27 (3), 191–217. http://proxy.library.tamu.edu/login?url=https://www.proquest.com/docview/743058755?accountid=7082&bdid=75230&_bd=ZqgBZADMXITyM%2F3a93pcqdXrka0%3D . [ Google Scholar ] Maclean JC, Popovici I, French MT, 2016. Are natural disasters in early childhood associated with mental health and substance use disorders as an adult? Soc. Sci. Med. 151, 78–91. [ DOI ] [ PubMed ] [ Google Scholar ] Math SB, Nirmala MC, Moirangthem S, Kumar NC, 2015. Disaster management: mental health perspective. Indian J. Psychol. Med. 37 (3), 261–271. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Moise IK, 2016. Hospitalizations for substance abuse disorders before and after Hurricane Katrina: spatial clustering and area-level predictors. Prev. Chronic Dis. 13. New Orleans, 2004 and 2008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] National Institutes of Health, 2014. Quality assessment tool for observational cohort and cross-sectional studies. National Heart, Lung, and Blood Institute. https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools . [ Google Scholar ] Newman G, Li DY, Park Y, 2022. The relationships between neighbourhood vacancy, probable PTSD, and health-related quality of life in flood-disaster-impacted communities. Urban Stud. 59 (15), 3077–3097. 10.1177/00420980221083101. [ DOI ] [ Google Scholar ] Newnham EA, Mergelsberg EL, Chen Y, Kim Y, Gibbs L, Dzidic PL, DaSilva MI, Chan EY, Shimomura K, Narita Z, 2022. Long term mental health trajectories after disasters and pandemics: a multilingual systematic review of prevalence, risk and protective factors. Clin. Psychol. Rev. 97, 102203. [ DOI ] [ PubMed ] [ Google Scholar ] Obrien LV, Berry HL, Coleman C, Hanigan IC, 2014. Drought as a mental health exposure. Environ. Res. 131, 181–187. 10.1016/j.envres.2014.03.014, 181–187. [ DOI ] [ PubMed ] [ Google Scholar ] Ohat N, 2019. Handbook for Conducting a Literature-based Health Assessment Using OHAT Approach for Systematic Review and Evidence Integration. US Department of Health and Human Services. Available online: https://ntp. niehs. nih. gov/whatwestudy/assessments/noncancer/handbook/index. html . [ Google Scholar ] Owen C, Crane J, 2022. Trauma-informed design of supported housing: a scoping review through the lens of neuroscience. Int. J. Environ. Res. Publ. Health 19 (21), 14279. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. bmj 372. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Pan R, Song J, Yi W, Liu J, Song R, Li X, Liu L, Yuan J, Wei N, Cheng J, Huang Y, Zhang X, Su H, 2024. Heatwave characteristics complicate the association between PM(2.5) components and schizophrenia hospitalizations in a changing climate: leveraging of the individual residential environment. Ecotoxicol. Environ. Saf. 271, 115973. 10.1016/j.ecoenv.2024.115973. [ DOI ] [ PubMed ] [ Google Scholar ] Parsons M, Reeve I, McGregor J, Hastings P, Marshall GR, McNeill J, Stayner R, Glavac S, 2021. Disaster resilience in Australia: a geographic assessment using an index of coping and adaptive capacity. Int. J. Disaster Risk Reduct. 62, 102422. [ Google Scholar ] Peen J, Schoevers RA, Beekman AT, Dekker J, 2010. The current status of urban-rural differences in psychiatric disorders. Acta Psychiatr. Scand. 121 (2), 84–93. [ DOI ] [ PubMed ] [ Google Scholar ] Pescaroli G, Alexander D, 2018. Understanding compound, interconnected, interacting, and cascading risks: a holistic framework. Risk Anal. 38 (11), 2245–2257. [ DOI ] [ PubMed ] [ Google Scholar ] Phan Minh T, Rocklöv J, Kim Bao G, Kullgren G, Nilsson M, 2016. Heatwaves and hospital admissions for mental disorders in Northern Vietnam. PLoS One 11 (5), e0155609. 10.1371/journal.pone.0155609. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Pietrzak RH, Tracy M, Galea S, Kilpatrick DG, Ruggiero KJ, Hamblen JL, Southwick SM, Norris FH, 2012. Resilience in the face of disaster: prevalence and longitudinal course of mental disorders following hurricane Ike. PLoS One 7 (6), e38964. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Porter JE, Dabkowski E, Ghasemirdekani M, Barbagallo MS, James MH, Prokopiv V, Wright W, 2023. The impact of nature-led recovery initiatives for individual and community health post disaster: a systematic literature review. Rev. Environ. Health 38 (4), 637–646. [ DOI ] [ PubMed ] [ Google Scholar ] Qian Y, Chakraborty T, Li J, Li D, He C, Sarangi C, Chen F, Yang X, Leung LR, 2022. Urbanization impact on regional climate and extreme weather: current understanding, uncertainties, and future research directions. Adv. Atmos. Sci. 39 (6), 819–860. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Rahman MS, Rahman MS, Novil Dedy A, Novil Dedy A, Moh S, Subagio H, Malik A, Triastono J, Renie O, Kilmanun JC, da Silva H, Pesireron M, Rein Estefanus S, Yusuf Y, 2022. Climate disasters and subjective well-being among urban and rural residents in Indonesia. Sustainability 14 (6), 3383. 10.3390/su14063383. [ DOI ] [ Google Scholar ] Schroeder K, Noll JG, Henry KA, Suglia SF, Sarwer DB, 2021. Trauma-informed neighborhoods: making the built environment trauma-informed. Prev. Med. Rep. 23, 101501. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schwartz RM, Gillezeau CN, Liu B, Lieberman-Cribbin W, Taioli E, 2017. Longitudinal impact of hurricane sandy exposure on mental health symptoms. Int. J. Environ. Res. Publ. Health 14 (9), 957. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Schwartz RM, Rasul R, Kerath SM, Watson AR, Lieberman-Cribbin W, Liu B, Taioli E, 2018. Displacement during hurricane sandy: the impact on mental health. J. Emerg. Manag. 16 (1), 17–27. 10.5055/jem.2018.0350. [ DOI ] [ PubMed ] [ Google Scholar ] Sewell K, Paul S, De Polt K, Sugg MM, Leeper RD, Rao D, Runkle JD, 2024. Impacts of compounding drought and heatwave events on child mental health: insights from a spatial clustering analysis. Discover mental health 4 (1), 1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Shabani A, Rasoulian M, Naserbakht M, Mitra Hakim S, Hajebi A, Tiyuri A, Seyed Abbas M, 2024. Prevalence and determinants of post-traumatic stress disorder five months after the 2019 huge flooding in Iran. BMC Public Health 24, 1. 10.1186/s12889-024-17861-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Singh A, Daniel L, Baker E, Bentley R, 2019. Housing disadvantage and poor mental health: a systematic review. Am. J. Prev. Med. 57 (2), 262–272. [ DOI ] [ PubMed ] [ Google Scholar ] Stroebe K, Kanis B, Richardson J, Oldersma F, Broer J, Greven F, Postmes T, 2021. Chronic disaster impact: the long-term psychological and physical health consequences of housing damage due to induced earthquakes. BMJ Open 11 (5), e040710. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Suglia SF, Duarte CS, Sandel MT, 2011. Housing quality, housing instability, and maternal mental health. J. Urban Health 88, 1105–1116. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Sui Y, Ettema D, Helbich M, 2022. Longitudinal associations between the neighborhood social, natural, and built environment and mental health: a systematic review with meta-analyses. Health Place 77, 102893. [ DOI ] [ PubMed ] [ Google Scholar ] Sullivan WC, Chang C-Y, 2011. Mental health and the built environment. Making Healthy Places: Designing and Building for Health, Well-Being, and Sustainability 106–116. [ Google Scholar ] Tito VR, Kazem H, Kadia S-O, Bernard B, 2024. A systematic review of mental health and climate change in the Philippines. Asian J. Psychiatry, 104191. [ DOI ] [ PubMed ] [ Google Scholar ] To P, Eboreime E, Agyapong VI, 2021. The impact of wildfires on mental health: a scoping review. Behav. Sci. 11 (9), 126. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Tupinier Martin F, Boudreault J, Campagna C, Lavigne É, Gamache P, Tandonnet M, Généreux M, Trottier S, Goupil-Sormany I, 2024. The relationship between hot temperatures and hospital admissions for psychosis in adults diagnosed with schizophrenia: a case-crossover study in Quebec, Canada [Article] Environ. Res. 246. 10.1016/j.envres.2024.118225. N.PAGN.PAG. [ DOI ] [ PubMed ] [ Google Scholar ] Ulrich RS, 1984. View through a window may influence recovery from surgery. Science 224 (4647), 420–421. [ DOI ] [ PubMed ] [ Google Scholar ] Venkataramanan V, Packman AI, Peters DR, Lopez D, McCuskey DJ, McDonald RI, Miller WM, Young SL, 2019. A systematic review of the human health and social well-being outcomes of green infrastructure for stormwater and flood management. J. Environ. Manag. 246, 868–880. [ DOI ] [ PubMed ] [ Google Scholar ] Webb EK, Stevens JS, Ely TD, Lebois LA, van Rooij SJH, Bruce SE, House SL, Beaudoin FL, An X, Neylan TC, 2024. Neighborhood resources associated with psychological trajectories and neural reactivity to reward after trauma. JAMA Psychiatry 81 (11), 1090–1100. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Wendelboe-Nelson C, Kelly S, Kennedy M, Cherrie JW, 2019. A scoping review mapping research on green space and associated mental health benefits. Int. J. Environ. Res. Publ. Health 16 (12), 2081. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] World Health Organization, 2020. Risk of Bias Assessment Instrument for Systematic Reviews Informing WHO Global Air Quality Guidelines. [ Google Scholar ] Xiong CZ, Hao HJ, Liong CW, Loon CW, Chenn KS, Kai IS, Husain M.b., Yaacob M.J.b., 2016. Prevalence of post traumatic stress disorder (PTSD) among flood victims in Malaysia: difference between Kuala Lumpur and Kelantan. Int. Med. J. 23 (2), 114–117. https://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2016-22316-001&site=ehost-live&scope=site&authtype=shib&custid=s8516548 . [ Google Scholar ] Xu Y, Wheeler SA, Zuo A, 2024. Drought and hotter temperature impacts on suicide: evidence from the Murray-Darling Basin, Australia. Climate Change Economics 15 (1), 2350024. [ Google Scholar ] Yang Y, Zhang Y, Sheng S, 2023. Residential greenness for mitigating impacts of extreme heat events on depression and supporting mental health. Front. Public Health 11, 1310410. 10.3389/fpubh.2023.1310410. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Zhang MX, VanLandingham M, Park YS, Anglewicz P, Abramson DM, 2021. Differences in post-disaster mental health among Vietnamese and African Americans living in adjacent urban communities flooded by Katrina. PLoS One 16 (8). 10.1371/journal.pone.0255303. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Zhang R, Zhang C-Q, Rhodes RE, 2021. The pathways linking objectively-measured greenspace exposure and mental health: a systematic review of observational studies. Environ. Res. 198, 111233. [ DOI ] [ PubMed ] [ Google Scholar ] Zhang W, Kinney P, Rich D, Sheridan S, Romeiko X, Dong G, Stern E, Du Z, Xiao J, Lawrence W, Lin Z, Hao Y, Lin S, 2020. How community vulnerability factors jointly affect multiple health outcomes after catastrophic storms. Environ. Int. 134. 10.1016/j.envint.2019.105285. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials ZHANGetal2026_Supplementary NIHMS2169307-supplement-ZHANGetal2026_Supplementary.doc (804.5KB, doc) Data Availability Statement No data was used for the research described in the article. ACTIONS View on publisher site PDF (1.3 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• When disaster strikes home: a scoping review linking the built environment to mental health outcomes following extreme environmental events.#360976
• Development of Post Traumatic Stress Disorder After Natural Disaster#995259
• Systematic review and meta-analysis of interventions for mental health in extreme weather events.#250984
• Effects of natural disasters on the cognitive state of older adults with dementia: a scoping review.#320420
• A disaster in the disaster: a study on post-traumatic stress symptoms among flood relief workers in Pakistan.#251020