Impact statement
This systematic review provides synthesized evidence about the mental health of vulnerable populations who have experienced fire or drought. Experiences of depression, posttraumatic stress disorder, anxiety and anger were common, influenced by limited access to mental health services, loss of community and loss of income. Few publications disaggregate data for vulnerable individuals, and multiple measures of mental health are used in this body of literature, limiting our understanding of how the mental health consequences of natural hazards intersect with different forms of vulnerability. Nevertheless, the data indicate that the mental health needs of vulnerable members of the community warrant specific consideration following natural hazards.
Introduction
Natural hazards are intensifying globally, and climate change is a major contributor to this (WMO, 2021). Natural hazards have great economic costs (Kousky, Reference Kousky2014), and affected populations can experience negative health consequences (Noji, Reference Noji2000). Drought and wildfires/bushfires (hereafter, fires), which often co-occur, are among the natural hazards expected to increase in frequency and intensity in the coming years. It is predicted that there will be a global increase in extreme fires of 14% by 2030, 30% by the end of 2050 and 50% by the end of the current century (United Nations Environment Programme, 2022). Similarly, by 2100, economic losses due to drought may become five times higher than current levels (European Commission Joint Research Centre et al., Reference Cammalleri, Naumann, Mentaschi, Formetta, Forzieri, Gosling, Bisselink, De Roo and Feyen2020).
In addition to their significant economic, social, environmental and political impacts (Middlemann, Reference Middlemann2007), fire and drought can have major impacts on the health and well-being of the people who are directly affected by them (World Health Organization, 2022a, b). In an early review, Laugharne and colleagues found that people directly affected by fire, as well as their close family, are at an increased risk of adverse psychological effects, including traumatic stress and depression (Laugharne et al., Reference Laugharne, Van de Watt and Janca2011). Exposure to fires is associated with lasting psychological impacts including depression, posttraumatic stress, suicidality and increased drug and alcohol use (McFarlane et al., Reference McFarlane, Clayer and Bookless1997; Finlay et al., Reference Finlay, Moffat, Gazzard, Baker and Murray2012). Similarly, experiencing drought negatively influences mental health in complex and diverse ways (Vins et al., Reference Vins, Bell, Saha and Hess2015), and is implicated in contributing to mental distress and suicidality (Austin et al., Reference Austin, Handley, Kiem, Rich, Lewin, Askland, Askarimarnani, Perkins and Kelly2018). A recent analysis of the Australian Rural Mental Health Study, a longitudinal study of 1,800 households across rural and remote New South Wales that examines the determinants of mental health as influenced by individual, family and community factors, suggests that while mental distress might abate after about three years of drought exposure, general life satisfaction and ability to maintain good health can continue to decline over time (Luong et al., Reference Luong, Handley, Austin, Kiem, Rich and Kelly2021).
Certain populations have social or physical vulnerabilities that contribute to poor health and well-being and have implications disaster context (Tierney, Reference Tierney, Daniels, Kettl and Kunreuther2006; Blaikie et al., Reference Blaikie, Wisner and Cannon2014). Building on the definition by Waisel (Reference Waisel2013), for the purpose of this review we define vulnerable populations as including people who are members of ethnic minority groups, are at least 60 years as defined by the World Health Organization, occupy a low socioeconomic position, have a chronic medical condition, are bereaved of a spouse, or reside in rural/remote areas (Waisel, Reference Waisel2013). There is substantial evidence from diverse settings indicating that members of vulnerable groups are at an elevated risk of poor physical health outcomes after experiencing fire/drought (Stanke et al., Reference Stanke, Kerac, Prudhomme, Medlock and Murray2013; Kondo et al., Reference Kondo, De Roos, White, Heilman, Mockrin, Gross-Davis and Burstyn2019; Walter et al., Reference Walter, Schneider-Futschik, Knibbs and Irving2020; Haikerwal et al., Reference Haikerwal, Doyle, Wark, Irving and Cheong2021). For example, a study focusing on drought mortality from 2000 to 2019 in Brazil showed that excess mortality risk attributable to extreme drought exposure was 0.99%; however, it increased to 2.28% for children, 1.57% in the elderly and 3.19% in women aged 65–74 years – showing that these vulnerable groups had an elevated risk of mortality compared to non-vulnerable groups (Salvador et al., Reference Salvador, Vicedo-Cabrera, Libonati, Russo, Garcia, Belem, Gimeno and Nieto2022).
While the physical health impacts of fire/drought on vulnerable groups are well documented, less is known about the mental health challenges faced by these populations after natural hazards. Further, the mental health impact of hazards cannot be fully understood when examined in isolation from other individual and social factors (Weldon, Reference Weldon and Mazur2008) that influence mental well-being. Thus, an understanding of the distinct factors influencing the mental health of vulnerable groups after experiencing fire/drought can inform emerging research priorities and is important for the development of effective interventions to support recovery. In consideration of the increasing threat posed by drought and fire globally, the aim of this systematic review is to describe the global literature examining the mental health of vulnerable populations after experiences of drought or fire and to identify knowledge gaps.
Methods
Database searches
This systematic review followed the ‘Preferred Items for Systematic Review and Meta-Analyses’ (PRISMA-) guidelines (Page et al., Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann, Mulrow, Shamseer, Tetzlaff, Akl, Brennan, Chou, Glanville, Grimshaw, Hróbjartsson, Lalu, Li, Loder, Mayo-Wilson, McDonald, LA, Stewart, Thomas, Tricco, Welch, Whiting and Moher2021). On November 19, 2021, searches were conducted in four different databases: Ovid MEDLINE (Medical Literature Analysis and Retrieval System Online), EMBASE (Excerpta Medica Database), CINAHL (Cumulative Index to Nursing and Allied Health Literature) and Ovid PsycInfo (APA PsycINFO). The search was repeated on March 22, 2022, to identify the most updated literature. The OSF registered protocol (10.17605/OSF.IO/SQEMC) and full protocol (Makleff et al., Reference Makleff, Varshney, Krishna, Romero and Fisher2022) describe the methodology in detail. This study was initially registered as a scoping review based on published guidance (Munn et al., Reference Munn, Peters, Stern, Tufanaru, McArthur and Aromataris2018). However, based on the robustness of our methods that fulfill the PRISMA guidelines (Page et al., Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann, Mulrow, Shamseer, Tetzlaff, Akl, Brennan, Chou, Glanville, Grimshaw, Hróbjartsson, Lalu, Li, Loder, Mayo-Wilson, McDonald, LA, Stewart, Thomas, Tricco, Welch, Whiting and Moher2021), including quality appraisal by two researchers, and a precise research question, this review is more appropriately described as a systematic review and is presented as such in this paper.
We focused on bushfire and drought in this review because these forms of natural hazard are becoming increasingly common in geographic regions across the globe due to similar reasons, such as high temperatures, low humidity and strong winds, may co-occur, are exacerbated by escalating climate change, and have significant socioeconomic and health impacts (Middlemann, Reference Middlemann2007; Richardson et al., Reference Richardson, Black, Irving, Matear, Monselesan, Risbey, Squire and Tozer2022; World Health Organization, 2022a, b). The search strategy utilized a combination of database-specific subject headings and free text terms that cover three concept areas: (a) bushfires, wildfires and natural disasters; (b) mental health and well-being; and (c) disadvantaged and vulnerable populations. Search terms were inclusive to cover qualitative approaches including grounded theory, focus groups, phenomenology and interviews; and quantitative methodologies including cohort designs, cross-sectional studies and case–control studies. There were no restrictions on dates of publication. Only English-language studies were included. The MEDLINE (Ovid) search strategy is provided in the protocol (Makleff et al., Reference Makleff, Varshney, Krishna, Romero and Fisher2022).
Screening process
Using Covidence (Veritas Health Innovation, 2017), we removed all duplicate articles. Next, three research team members screened articles for eligibility based on title, abstract and keyword. Two members of the research team independently assessed the full text of all remaining articles to determine eligibility for inclusion in the review. Articles were included if they fulfilled the following criteria: (a) were original research (excluded reviews, editorials and commentaries), (b) were written in English, (c) were conducted in a setting with people affected by drought/fire (fires with a natural cause, such as bushfires/wildfires), (d) included at least one adult participant from a vulnerable group, and (e) provided findings regarding mental health outcomes for vulnerable participants. There were no restrictions on the country of study or study design for original, peer-reviewed research studies; mixed methods studies were eligible for inclusion if they also fulfilled the inclusion criteria.
Data extraction
We extracted the following data on study characteristics: year of data collection, year of fire/drought occurrence, year of publication, type of hazard, location of study, study design and description, study objectives/aims, vulnerable population characteristics, and key mental health findings for (a) the entire sample and (b) the vulnerable population in the study. We first summarized data separately based on type of hazard (fire/drought) and type of study (quantitative/qualitative). Next, we pooled and summarized the following data: type of hazard, study design, study location (country), year of publication and study sample size. We synthesized additional findings relating to study methodology, types of vulnerabilities and mental health findings qualitatively.
Quality assessment
All included studies were assessed for methodological quality using the Joanna Briggs Institute (JBI) critical appraisal tools, focusing on the extent to which the study had addressed the possibility of bias (Joanna Briggs Institute, 2020). Two members of the research team scored each paper; any discrepancies in scoring were discussed by the team to finalize the scoring. A numeric score was calculated for each paper in the review based on the total number of “yes” or “no”/"unclear" metrics of the JBI checklist (an “unclear” was assigned the same score as a “no”), as has been conducted in prior reviews (Bowring et al., Reference Bowring, Veronese, Doyle, Stoove and Hellard2016; Xu et al., Reference Xu, Chen and Wang2017). Based on the JBI critical appraisal tools for each study design, qualitative studies were assessed on a ten-item scale, cohort studies on an eleven-item scale and cross-sectional studies on an eight-item scale. Quality assessment scores were compared across studies, and mean assessment scores with standard deviation were analyzed by study design. Following Adalbert et al. (Reference Adalbert, Varshney, Tobin and Pajaro2021) and to provide comparisons, relative scores were depicted graphically to illustrate the percent value of each study relative to the others. These analyses elucidate the general quality of evidence of the existing literature and highlight the methodological strengths and weaknesses of the studies included in the systematic review.
Results
Screening of studies
Searches from all databases produced a total of 3,401 articles, and after removal of duplicates, 2,098 articles remained. With the removal of 2,066 articles after screening by title/abstract, 32 articles underwent full-text analysis, of which 18 were ultimately deemed eligible for inclusion in this review (see Figure 1).
Figure 1. Process of screening articles for this scoping review based on the PRISMA 2020 flow diagram (Page et al., Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann, Mulrow, Shamseer, Tetzlaff, Akl, Brennan, Chou, Glanville, Grimshaw, Hróbjartsson, Lalu, Li, Loder, Mayo-Wilson, McDonald, LA, Stewart, Thomas, Tricco, Welch, Whiting and Moher2021).
Study characteristics
Of the 18 studies included in this review, nine were in settings of drought and nine in settings of fire. Studies were conducted between 2006 and 2022 in Australia (n = 11), the United States (n = 3), Greece (n = 2), Iran (n = 1) and Canada (n = 1). The number of drought- and fire-affected individuals by study ranged from 23 to 5,012 (unspecified in Hayati et al., Reference Hayati, Yazdanpanah and Karbalaee2010). The pooled total was 15,959 participants across all 18 studies. All but two studies in the review (Parslow et al., Reference Parslow, Jorm and Christensen2006; Scher and Ellwanger, Reference Scher and Ellwanger2009) included participants living in a rural or remote area, and ten of the studies focused exclusively on rural or remote residents. Other common vulnerabilities in the included papers were occupying a low socioeconomic position, experiencing chronic health conditions and/or mental health problems prior to the hazard event, having a low educational attainment, belonging to an ethnic minority group and being unemployed.
Measures of mental health varied by study. Qualitative studies, which used either individual interviews, surveys with open-ended questions, or focus groups, relied on participants’ accounts of their experiences of different aspects of mental health. Most quantitative studies used self-report surveys and symptom checklists to assess mental health outcomes. Some studies used standardized measures such as the Trauma Screening Questionnaire (TSQ) (Brewin et al., Reference Brewin, Rose, Andrews, Green, Tata, McEvedy, Turner and Foa2002), Symptom Checklist 90-Revised (SCL-90-R) (Derogatis and Savitz, Reference Derogatis and Savitz1999), Kessler Psychological Distress Scale (K6), Kessler 10-L (K10) (Kessler et al., Reference Kessler, Barker, Colpe, Epstein, Gfroerer, Hiripi, Howes, Normand, Manderscheid and Walters2003), Impact of Events Scale-Revised (IES-R) (Weiss and Marmar, Reference Weiss and Marmar1997) and the PTSD Symptoms Checklist (PCL-5) (Blevins et al., Reference Blevins, Weathers, Davis, Witte and Domino2015) to determine the prevalence of symptoms of anxiety, distress, posttraumatic stress and anger. Other measures, including the Sense of Community Index (Chavis et al., Reference Chavis, Lee and Acosta1986) and the Sense of Place (Shamai, Reference Shamai1991) scale, examined social support and community circumstances.
Data collection for included studies occurred at different time points relative to the hazard event. Studies occurred during the hazard event (n = 3; only for drought), in the same year as the event (n = 4), six months later (n = 3), one year later (n = 1), five years later (n = 1), in the year between multiple hazard events (n = 1), or they did not specify the time point relative to the hazard event (n = 4). One study (Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022) collected data twice: two years after the fire and then six years after the event. Qualitative studies included in the review are listed in Table 1, quantitative studies in Table 2 and pooled characteristics for the studies are shown in Table 3.
Table 1. Qualitative study findings
Table 2. Quantitative study findings
Table 3. Pooled study findings (total studies = 18)
Quality assessments
Complete quality assessment critical appraisal checklist scorings are in Supplementary Tables S1–S3. Studies generally ranged from moderate to high quality overall. Qualitative studies (n = 6) had a mean of 73.3% of possible points (SD = 12.1; Range = 60–90%). Cross-sectional studies (n = 8) had a mean of 78.1% of possible points (SD = 17.4; Range = 50–100%). Cohort studies (n = 4) had a mean of 72.7% of possible points (SD = 19.61; Range = 45.4–90.9%). The most common methodological flaws identified across studies using the JBI criteria were inconsistencies identifying and addressing confounding factors, not reporting details about participants who were lost to follow-up (in cohort studies) and a lack of representation of participant perspectives (in qualitative studies). Quality assessments for all studies are depicted in Figure 2.
Figure 2. Quality assessment scores by study design.
Outcomes in fire-affected settings
Two qualitative studies (shown in Table 1), both in the United States, were conducted with participants who had experienced fire in the 2017 Northern California Wildfires and the multiple fires between 2012–2020 in Okanogan, Washington. In one (Domínguez and Yeh, Reference Domínguez and Yeh2020), approximately one-third of the sample was deemed vulnerable due to one of the following factors: low income, being an ethnic minority, identifying as a sexual/gender minority, or having low educational attainment. In the other (Humphreys et al., Reference Humphreys, Walker, Bratman and Errett2022), all participants were vulnerable as the entire sample resided rurally, but further vulnerabilities were not clearly defined. In both studies, vulnerable individuals reported anger, cynicism, a perceived lack of support due to marginalization and an increased susceptibility to physical and mental health problems after experiencing fire.
Seven quantitative studies were conducted with participants who experienced fire (shown in Table 2). In three studies, all participants were vulnerable because they resided in a rural area. In four studies, between one-third and two-thirds of participants were classified as vulnerable due to unemployment, experiencing prior health or mental health conditions, belonging to an ethnic minority group (including being Indigenous), or having low educational attainment. Because the characteristics of vulnerability are not mutually exclusive, for these studies it was not possible to determine the total number of vulnerable participants in the study. People experiencing vulnerability often had mental health problems after the event, with prevalence estimates up to 36% for PTSD (Parslow et al., Reference Parslow, Jorm and Christensen2006; Austin et al., Reference Austin, Handley, Kiem, Rich, Lewin, Askland, Askarimarnani, Perkins and Kelly2018; Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021; Cowlishaw et al., Reference Cowlishaw, Metcalf, Varker, Stone, Molyneaux, Gibbs, Block, Harms, MacDougall, Gallagher, Bryant, Lawrence-Wood, Kellett, O’Donnell and Forbes2021; Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022), 10% for anger (Cowlishaw et al., Reference Cowlishaw, Metcalf, Varker, Stone, Molyneaux, Gibbs, Block, Harms, MacDougall, Gallagher, Bryant, Lawrence-Wood, Kellett, O’Donnell and Forbes2021), 15% for anxiety (Scher and Ellwanger, Reference Scher and Ellwanger2009; Papanikolaou et al., Reference Papanikolaou, Adamis and Mellon2011; Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021) and 15% for depression (Scher and Ellwanger, Reference Scher and Ellwanger2009; Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021); rates of psychosis and paranoia were noted to be high in one study, but numbers were unspecified (Papanikolaou et al., Reference Papanikolaou, Adamis and Mellon2011). In a study in Greece, those with lower educational attainment had a higher risk of developing somatization symptoms, not further described (Papanikolaou et al., Reference Papanikolaou, Adamis and Mellon2011). Indigenous people in Fort McMurray, Alberta, Canada, were shown to have more severe symptoms of mental illness, such as depression and anxiety, than those from other racial groups (Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021).
Two of the quantitative studies examined mental health at different points in time. An Australian study (Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022), which investigated the long-term impact of the bushfire-instigated Hazelwood mine fire, indicated that traumatic symptoms because of the fires not only lasted years after the event but also increased over time. The study found that younger participants (average age of 25 years, compared to groups with an average age of 45 and 65 years old) reported higher levels of ongoing distress in response to their exposure to the fire, even six years after the event, at the second round of data collection. These higher levels of stress likely occurred due to subsequent fire, as opposed to solely an increase in levels of stress occurring over time (Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022). Further, media coverage about hazards, as well as similar fire- and smoke-related events locally, served as triggers elevating traumatic stress symptoms among participants. A Canadian study found that a prior history of mental health problems along with experiencing financial stress increased the odds of developing, or having more severe symptoms of, PTSD, depression, insomnia, anxiety and drug/alcohol dependency for those affected by fire (Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021).
While most of the studies in the review did not compare between vulnerable mental health experiences and those in the general population, some fire-related studies employed other forms of comparison. Two studies in Greece compared those exposed and unexposed to the fires (Papanikolaou et al., Reference Papanikolaou, Adamis and Mellon2011a Reference Papanikolaou, Leon, Kyriopoulos, Levett and Pallis,b). A prospective cohort study in Australia focused on the mental health impacts immediately after a bushfire and compared these with the same participants four years later (Parslow et al., Reference Parslow, Jorm and Christensen2006). Another Australian study stratified data based on age and the level of exposure to fire among participants in the Hazelwood mine fire study (Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022).
Outcomes in drought-affected settings
Across the four qualitative studies conducted in settings affected by drought (shown in Table 1), all participants were deemed to be vulnerable as they resided in rural settings. Other aspects of vulnerability in these studies included low socioeconomic position and being Indigenous. Mental health problems such as depression, addiction, anxiety and suicidal thoughts and behavior were described by participants in these studies, who discussed the loss of community, and resultant decline in social interaction, as contributing to their mental health problems. Lost income, alongside a need to work longer hours, and in addition to drought-related worry, detracted from well-being; participants described a lack of accessible mental health services as a barrier to recovery.
Five quantitative studies, all conducted in Australia, focused on drought (shown in Table 2). The proportion of vulnerable participants in these studies ranged from 18% to 100%, with one study not reporting the number of participants classified as vulnerable (Parslow et al., Reference Parslow, Jorm and Christensen2006). The most frequent types of vulnerability in these studies were living in a rural/remote area, having a low level of education and being unemployed or having a low income. Worry, distress and overall poor mental health (which was not further defined in one study (Powers et al., Reference Powers, Dobson, Berry, Graves, Hanigan and Loxton2015) were shown to be higher among certain vulnerable populations. For example, for drought-affected participants, living in very remote areas was associated with each of the following factors: a high likelihood of having symptoms of mental illness (Austin et al., Reference Austin, Handley, Kiem, Rich, Lewin, Askland, Askarimarnani, Perkins and Kelly2018), low levels of well-being and high worry (Kelly et al., Reference Kelly, Lewin, Stain, Coleman, Fitzgerald, Perkins, Carr, Fragar, Fuller, Lyle and Beard2011., Reference Kelly, Lewin, Stain, Coleman, Fitzgerald, Perkins, Carr, Fragar, Fuller, Lyle and Beard2010; Stain et al., Reference Stain, Kelly, Carr, Lewin, Fitzgerald and Fragar2011). Drought-affected individuals experiencing financial/food insecurities and social isolation due to their remote or regional locations tended to have poor mental health outcomes. For example, Friel et al. (Reference Friel, Berry, Dinh, O’Brien and Walls2014) found that higher levels of psychological distress were significantly associated with experiencing financial stress and food insecurity. Another study found that unemployed individuals were at four times higher risk to develop mental illness compared to their employed peers (Austin et al., Reference Austin, Handley, Kiem, Rich, Lewin, Askland, Askarimarnani, Perkins and Kelly2018).
Discussion
This systematic review contributes to our understanding of the mental health outcomes and experiences of vulnerable groups affected by natural hazards, specifically fire and drought. It provides evidence, primarily from high-income countries, that vulnerable individuals affected by drought or fires are more likely to experience anxiety, depression and general distress than less vulnerable groups. Common experiences reported by vulnerable individuals affected by drought included worry and, at worst, suicidality. Those affected by fire reported symptoms of PTSD and anger and there was some evidence of increased risk of psychosis. Time scale is a possible explanation for the differences in mental health outcomes after drought compared to bushfire. Drought generally occurs over long time periods, with prolonged stressors that can gradually contribute to mental health consequences such as suicidality (Padhy et al., Reference Padhy, Sarkar, Panigrahi and Paul2015), while bushfires are rapid-onset events and may trigger different mental health consequences through different mechanisms (Askland et al., Reference Askland, Shannon, Chiong, Lockart, Maguire, Rich and Groizard2022; Zhang et al., Reference Zhang, Workman, Russell, Williamson, Pan and Reifels2022).
Notably, mental health problems arising after these natural hazards were described as exacerbating already existing mental health, physical health and socioeconomic challenges. For example, poor mental health prior to the disaster event increased the odds of developing financial problems and mental illness post-hazard (Parslow et al., Reference Parslow, Jorm and Christensen2006; Belleville et al., Reference Belleville, Ouellet, Lebel, Ghosh, Morin, Bouchard, Guay, Bergeron, Campbell and MacMaster2021; Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022; Humphreys et al., Reference Humphreys, Walker, Bratman and Errett2022). Similarly, those who had PTSD prior to hazard events tended to have worse mental health outcomes after the event (Domínguez and Yeh, Reference Domínguez and Yeh2020). Other research can help understand why vulnerabilities such as poverty or hazard experiences can exacerbate poor mental health. Most mental health problems, including depression and suicidality, are underpinned psychologically by experiences of being entrapped and feeling powerless and unable to escape; this may be common in some forms of vulnerability, such as poverty and interpersonal violence (Fisher et al., Reference Fisher, McKelvie and Rees2020). These feelings can be made worse if the situation is intrinsically humiliating, as in the case of being marginalized or discriminated against, rejected, having a sense of subjective incompetence compared to others, or having fewer capabilities or resources (Fisher et al., Reference Fisher, McKelvie and Rees2020). Further, there is evidence that helps understand why natural hazards may exacerbate poor mental health. Direct trauma and physical danger, as well as indirect damage to personal environment, livelihood and property, have been identified as aspects that may exacerbate mental health conditions in the case of bushfires (Zhang et al., Reference Zhang, Workman, Russell, Williamson, Pan and Reifels2022). Factors that may exacerbate physical and emotional challenges after drought include the emergence of air pollution, a loss of access to fresh water and compromised agricultural production with concurrent damage to peoples’ livelihoods (Vins et al., Reference Vins, Bell, Saha and Hess2015).
The most common type of vulnerability in the included studies was residing in a rural setting. This reflects that fire and drought are more common in rural and remote than urban settings, where farmers and Indigenous and First Nations communities, among others, often already face socioeconomic challenges. These communities rely on drought- or fire-affected lands for their livelihoods and their mental well-being. Together, the findings across studies indicate that the mental health needs of the most vulnerable members in a community warrant specific consideration following a natural hazard. This review highlights a lack of examination in the current literature of the intersectionality of vulnerability factors and the potential compounding effect that these may have on mental health after natural hazard exposure. As social characteristics cannot be understood separately from each other (Weldon, Reference Weldon and Mazur2008), an intersectional lens helps examine how overlapping forms of marginalization and discrimination can impact the lives of individuals (Victoria Government, 2021). Intersectionality encompasses the layering of individual characteristics such as age, location, ethnicity and gender (Walker et al., Reference Walker, Reed and Fletcher2021). Historically used in gender and racial justice movements, intersectionality has also been applied to understanding how climate change and climate hazards can exacerbate existing inequalities (Kaijser and Kronsell, Reference Kaijser and Kronsell2014; Thompson-Hall et al., Reference Thompson-Hall, Carr and Pascual2016; Walker et al., Reference Walker, Reed and Fletcher2021). The dearth of studies in this review that explicitly examine vulnerability, let alone the intersection between different types of vulnerability, emphasizes the need to adopt an intersectional lens when studying post-hazard mental health outcomes in the future. This is important because overlapping vulnerabilities across the lifespan influence not only the hazard experience but also the ability to recover and rebuild from the hazard.
The findings have implications for psychologically informed responses across the spectrum from promotion of mental health and prevention of mental health problems to early intervention and treatment. These include non-health-sector and health-sector actions and the need for these to be culturally safe and explicitly inclusive of members of vulnerable populations. The findings also have implications for community-centred hazard response. During and immediately after a hazard, it is important to consider strategies to promote inclusive community cohesion and peer-to-peer recovery activities to address immediate safety and survival needs (Chang, Reference Chang2010; Ludin et al., Reference Ludin, Rohaizat and Arbon2019). Clear communication and transparency from local and state authorities about emergency and recovery services and resources and inclusive recovery planning, in which diverse community members are represented, is an approach that can promote trust and reduce frustration (Rosenberg et al., Reference Rosenberg, Errett and Eisenman2022).
Loss of income-generating work and property, including homes, farm infrastructure, stock and crops, contributes to despondency and hopelessness after fire and drought. Programs and resources to address the urgent priorities of food and financial insecurity, and emergency housing, tailored to ensure that vulnerable populations have equity of access, are vital. Provision of job-acquisition support programs and other training programs to improve livelihoods will also remain valuable. An example of such a program is the Catchment Management Authority Drought Employment in Victoria, Australia, where those affected by drought are employed in public-good and environmental projects while concurrently developing new skills that will help them become employable in other spheres (Victoria State Government, 2021).
A strong case has been articulated for policies that include social and financial support for services aimed at reducing health inequities and structural vulnerabilities throughout the various phases of a disaster – from pre-disaster planning phase to the chronic posttraumatic reestablishment phase (McFarlane and Williams, Reference McFarlane and Williams2012, Finucane et al., Reference Finucane, Acosta, Wicker and Whipkey2020). Social policies to reduce inequities, for example by improving income, can address some of the underlying contributors to poor mental health and thus have the potential to indirectly improve mental health. For example, a longitudinal study of flood survivors in Germany showed that financial support, alongside supportive counseling, was associated with lower levels of mental health strain among vulnerable individuals (Daniel and Michaela, Reference Daniel and Michaela2021).
In terms of health sector approaches, the findings suggest that strategies to improve mental health need to consider the structural barriers impeding access to mental health services such as stigma, lack of affordability and limited availability of service providers, particularly in rural areas, as has been suggested elsewhere (Cosgrave et al., Reference Cosgrave, Hussain and Maple2015; Morgan et al., Reference Morgan, Wright and Reavley2021). Inclusion of primary care practitioners in identifying people with poor mental health and delivering mental health services, particularly in rural and remote areas, is one approach that can be considered to increase access to mental health support (McFarlane and Williams, Reference McFarlane and Williams2012).
Longer-term investments in supporting mental health are also recommended. Drought and fires can have lasting negative impacts on vulnerable populations and repeated hazard events can amplify these experiences (Vins et al., Reference Vins, Bell, Saha and Hess2015; Cianconi et al., Reference Cianconi, Betrò and Janiri2020). Adverse mental health impacts are evident years after a hazard (Raker et al., Reference Raker, Lowe, Arcaya, Johnson, Rhodes and Waters2019) and long-term support services are needed that include mental health support for hazard-affected communities, particularly for vulnerable members of these communities (Wilson-Genderson et al., Reference Wilson-Genderson, Heid and Pruchno2018). To enhance relevance and acceptability, such services should be co-designed with communities and consider cultural safety.
Many people seek psychologically informed practical assistance rather than specific psychological services, and some identify a need for crisis counseling services focused on mental health (Jogia et al., Reference Jogia, Kulatunga, Yates and Wedawatta2014). However, mental health services alone will be insufficient. It is critical that investment also be placed into programs and interventions that prioritize mental health promotion, as well as those which seek to address the underlying risk factors for mental health problems, such as financial insecurity, domestic violence and discrimination (Oram et al., Reference Oram, Khalifeh and Howard2016; Vargas et al., Reference Vargas, Huey and Miranda2020; Virgolino et al., Reference Virgolino, Costa, Santos, Pereira, Antunes, Ambrósio, Heitor and Vaz Carneiro2022).
The strengths of this review are that the search strategy was designed by a specialist information analyst, the protocol was pre-published (Makleff et al., Reference Makleff, Varshney, Krishna, Romero and Fisher2022) and it followed standard guidelines. In addition, it included a quality assessment process that allows for an interpretation of the findings taking study quality into account. We acknowledge the limitation that the search was restricted to studies published in English and relevant studies published in other languages might have been missed. Further, our search terms may have missed relevant papers that examine indirect aspects of drought or fire experiences. Nevertheless, we believe that the strengths of the study outweigh its limitations and that it provides an accurate account of the state of knowledge in this field.
There are methodological strengths and limitations in this body of evidence. First, a subset of included studies only had a partial focus on vulnerable individuals and had limited comparison of the mental health experiences of vulnerable individuals and the general population. Second, the heterogeneity in measures precluded meta-analysis and we are unable to estimate the prevalence of mental health outcomes in vulnerable populations after drought or fire with precision. Third, quality ratings varied among studies corresponding to the various study designs, methods of recruitment, use of instruments and efforts to minimize possible biases. One of the main detractors of quality identified through the appraisal process was a lack of consideration of potential confounding variables. Last, while this review did include findings from different cultural and national contexts, most studies were conducted in Australia and other high-income countries. It is possible that some of the findings are not relevant to low- and middle-income countries, which may have fewer resources to support affected populations. In terms of methodological strengths, while vulnerable groups and the general population were not compared in any papers in the review, four studies (Parslow et al., Reference Parslow, Jorm and Christensen2006; Papanikolaou et al., Reference Papanikolaou, Adamis and Mellon2011a Reference Papanikolaou, Leon, Kyriopoulos, Levett and Pallis,b; Carroll et al., Reference Carroll, Campbell, Smith, Gao, Maybery, Berger, Brown, Allgood, Broder, Ikin, McFarlane, Sim, Walker and Abramson2022) did use other forms of comparison (exposed vs. unexposed, over time and by level of exposure) to examine mental health outcomes.
To strengthen the body of evidence in this burgeoning field, future research could focus on intersectional experiences of vulnerability and examine potential confounders that may have influenced mental health experiences. In addition, studies should aim, as appropriate for their research questions, to incorporate standardized measures that have been tested for their reliability and validity to allow for comparison of data beyond the particular study (Boynton and Greenhalgh, Reference Boynton and Greenhalgh2004; Boateng et al., Reference Boateng, Neilands, Frongillo, Melgar-Quiñonez and Young2018). Key factors to consider in the selection of measures would be formal validation against a gold standard diagnostic measure, comprehensibility for people of diverse literacies and ideally having been used in equivalent studies (Boynton and Greenhalgh, Reference Boynton and Greenhalgh2004; Boateng et al., Reference Boateng, Neilands, Frongillo, Melgar-Quiñonez and Young2018). It is beyond the scope of this paper to review individual measures though this would be a valuable area of future research activity.
Our review adds to evidence from prior systematic and scoping reviews about the mental health of people who have experienced a natural hazard (Laugharne et al., Reference Laugharne, Van de Watt and Janca2011; Finlay et al., Reference Finlay, Moffat, Gazzard, Baker and Murray2012; Vins et al., Reference Vins, Bell, Saha and Hess2015) by synthesizing the available evidence about the mental health of vulnerable communities who have experienced fire and drought. Based on our findings, this focus on vulnerability has relevance for the mental health of farming, rural, and Indigenous and First Nations communities that depend on the land for their livelihoods, who live in settings that are experiencing catastrophic fires and extended drought more frequently.
Conclusion
This systematic review contributes to a more comprehensive understanding of the mental health consequences of natural hazards among vulnerable communities. The evidence indicates that many members of vulnerable groups experience mental health problems after exposure to drought and fire, including PTSD, depression, anxiety, suicidality, overuse of alcohol and anger. We found that limited access to mental health services, isolation and loss of community and income were drivers of mental health problems in these communities.
This review highlights the importance of improving the evidence base about mental health in vulnerable communities affected by natural hazards by including standardized measures and comparison groups. Further, there is a gap in studies that examine the role of intersectional vulnerabilities and systematically disaggregate data to allow for analysis of the particular mental health experiences of vulnerable communities after disaster. Future studies that draw on these approaches to examine the mental health effects of drought and fire on vulnerable individuals will help ensure that programs are informed by an understanding of the unique needs of these communities.
Findings have relevance for post-disaster efforts and can be used to inform policies and programs to help vulnerable groups build their resilience against hazards and prepare for, respond to, and recover from disasters. In conclusion, the mental health of vulnerable individuals and communities recovering from natural hazards must be considered and addressed as part of holistic recovery efforts aiming to improve health and well-being in the context of structural disadvantage.
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/gmh.2023.13.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/gmh.2023.13.
Data availability statement
Data available within the article or its supplementary materials.
Acknowledgements
We acknowledge the valuable contribution to this research of the Monash University–led Fire to Flourish (2021-2026) program team and partners.
Author contribution
S.M., R.N.K., L.R., J.F. conceptualized the study; the research strategy for the Fire to Flourish project overall was supported by R.W. L.R. developed the search strategy and conducted the search. K.V., S.M., K.V., R.N.K. contributed equally to data extraction and quality assessment. K.V., S.M., R.N.K. contributed equally to writing the paper, with further contributions from L.R., J.W., R.W., J.F. R.V. developed all tables and figures, with further contributions from S.M., R.W., J.W., J.F. All authors reviewed and approved the final manuscript.
Financial support
Cornerstone funding for Fire to Flourish, which supported this review, is provided by the Paul Ramsay Foundation and Metal Manufactures Pty Ltd., with additional philanthropic support from the Lowy Foundation. The funders had no influence on any aspect of the protocol development or the implementation of the review. JF is supported by the Finkel Professorial Fellowship, which receives funding from the Finkel Family Foundation.
Competing interest
The authors declare none.
Open access
Comments
2 November 2022
Global Mental Health
Editorial Director
Dear Editorial Committee,
We are pleased to submit the attached manuscript entitled “Impact of experiencing a drought or bushfire on the mental health of vulnerable groups: a scoping review” for consideration for publication. The objective of this scoping review is to describe the mental health outcomes of vulnerable populations after droughts and bushfires/wildfires and identify gaps in the literature. This falls within the journal’s remit to contribute to the emerging field of global mental health.
We confirm that neither the manuscript nor any parts of its content are currently under consideration or published in another journal. All authors have approved the manuscript and agree with its submission to Global Mental Health: Cambridge Prisms. None of the authors report a conflict of interest.
The word count is 4668 words excluding abstract, references, and supplemental tables.
Best wishes,
Professor Jane Fisher
JANE FISHER AO, BSc (Hons), PhD, MAPS, FCCLP, FCHP
Finkel Professor of Global Health
Head, Division of Social Sciences
Immediate Past President International Marcé Society for Perinatal Mental Health
Global and Women’s Health
Public Health and Preventive Medicine
Monash University
Level 4, 553 St Kilda Rd
Melbourne VIC 3004
Australia
E: [email protected]