Understanding Associations between Disasters and Sustainability, Resilience, and Poverty: An Empirical Study of the Last Two Decades

Abstract

This study investigates the impact of disasters on sustainability, resilience, and poverty, using data from the “Sustainable Development Report” and the Emergency Events Database (EM-DAT) from 2000 to 2023. Regression models assessed the effects of disasters, deaths, injuries, affected individuals, and economic damage on normalized values of the dependent variables with lag periods of one, two, and three years of independent variables. The results reveal that disasters consistently negatively impact sustainability and resilience, highlighting the need for robust disaster risk reduction strategies and resilient infrastructure. Higher mortality rates significantly hindered development, emphasizing the importance of improving early warning systems, emergency preparedness, and healthcare infrastructure. While injuries and the number of affected individuals did not show significant associations, economic damage was positively associated with resilience, suggesting that financial losses might drive recovery investments. Additionally, disasters were found to exacerbate poverty levels over time with significant associations in the two and three-year lag models. This study also uncovered significant regional disparities with lower resilience, sustainability, and higher poverty levels in certain regions compared to others. Higher-income groups demonstrated better resilience and lower poverty levels. These findings underscore the necessity for targeted, region-specific strategies to enhance resilience, reduce poverty, and support sustainable development, leveraging post-disaster recovery phases for long-term improvement.

1. Introduction

For over thirty years, member countries of the United Nations (UN) have been dedicated to advancing sustainable development, which aims to improve human well-being while safeguarding the environment. This commitment began in June 1992 at the Earth Summit held in Rio de Janeiro, Brazil, where representatives from more than 178 countries convened and endorsed Agenda 21 [1,2]. Subsequently, in September 2000, the Millennium Declaration was adopted at the Millennium Summit, outlining eight Millennium Development Goals (MDGs) aimed at alleviating extreme poverty by 2015 [3]. The momentum continued in June 2012 at the United Nations Conference on Sustainable Development (Rio + 20), where member states gathered again in Rio de Janeiro [4]. During this summit, they embraced “The Future We Want” document, marking the inception of the Sustainable Development Goals (SDGs). Finally, in September 2015, the 2030 Agenda for Sustainable Development was formally adopted at the UN Sustainable Development Summit, encompassing the 17 SDGs (as follows) aimed at addressing global challenges comprehensively: 1. No poverty; 2. Zero hunger; 3. Good health and well-being; 4. Quality education; 5. Gender equality; 6. Clean water and sanitation; 7. Affordable and clean energy; 8. Decent work and economic growth; 9. Industry, innovation, and infrastructure; 10. Reduced inequalities; 11. Sustainable cities and communities; 12. Responsible consumption and production; 13. Climate action; 14. Life below water; 15. Life on land; 16. Peace, justice, and strong institutions; 17. Partnerships for the goals [5,6].

From 1 January 2016, the global community embarked on the implementation of the 2030 Agenda for Sustainable Development, a comprehensive framework comprising 17 SDGs aimed at tackling pressing global issues over a 15-year period. Since then, annual progress on the SDGs has been documented through the UN Secretary-General’s SDG Progress Reports, commencing in 2016 and continuing through the latest release in 2024 [7,8,9,10,11,12,13,14]. Additionally, every four years, the General Assembly conducts a quadrennial review of SDG achievements, the results of which are detailed in the Global Sustainable Development Report. Previous editions of this report were published in 2019 and 2023 [5,15,16].

According to the 2024 progress assessment, global efforts toward achieving the 2030 Agenda for Sustainable Development are falling significantly short [17]. The report highlights that only seventeen percent of the SDG targets are currently on track, while nearly half are showing minimal to moderate progress. Alarmingly, progress on more than a third of the targets has either stalled or regressed. Among the 135 targets with trend data, approximately 30% indicate marginal progress, 18% show moderate progress, 18% have stagnated, and 17% have regressed below the baseline levels set in 2015. Of particular concern are SDG Goal 1, which aims to eradicate poverty in all its forms by 2030, and Goal 2, which focuses on urgent action to combat climate change and its impact—both of these goals are notably off track [17]. The initial positive momentum observed in critical SDG indicators has slowed since 2019, largely due to global challenges such as the COVID-19 pandemic, ongoing conflicts, geopolitical tensions, trade disputes, and escalating climate-related issues. These multifaceted challenges, compounded by systemic economic weaknesses and historical inequalities, disproportionately impact developing nations and vulnerable communities, underscoring significant gaps in global solidarity [17].

The global drive to eliminate extreme poverty has encountered substantial setbacks due to the COVID-19 pandemic and other major disruptions spanning 2020 to 2022 [17]. These challenges have reversed progress by approximately three years, marking the first increase in poverty levels in decades. Recovery efforts have been uneven, with low-income countries bearing the brunt of the impact, rendering the goal of eradicating poverty by 2030 increasingly improbable. By 2022, approximately 9% of the world’s population—equivalent to 712 million people—were living in extreme poverty. If current trends persist, this number is projected to reach 590 million by 2030. Economic losses attributed to disasters have remained persistently high, averaging over USD 115 billion annually between 2015 and 2022. Substantial investments are urgently required to enhance social protection coverage for children and expand essential services, given that current global government expenditures on these vital areas hover around 50% [17].

The “2023 Global Assessment Report (GAR) Special Report on Mapping Resilience for Sustainable Development Goals” underscores that disasters are increasingly reversing global development progress at unprecedented rates [18]. Developed over a two-year period by a team of 90 experts and presented at the UN High-Level Political Forum, the report highlights the severe impact of escalating disaster risks and various shocks on global economic stability and sustainable development efforts. A significant focus of the report is on climate-related disasters such as droughts, which are now causing water stress for two billion people and elevating the risk of crop failures by 80% in regions like sub-Saharan Africa and Southeast Asia. Additionally, it warns that rising temperatures could halve labor productivity at 34 °C and potentially lead to the loss of 80 million jobs if global warming exceeds 1.5 °C. The GAR stresses the critical need for investments in resilience measures and advocates for a comprehensive approach to managing interconnected risks to effectively achieve the SDGs. This call for action underscores the urgency of addressing these challenges to safeguard global development progress and enhance global resilience [18].

Disaster risk reduction (DRR) plays a crucial role in advancing sustainable social and economic development, a recognition underscored in several key global frameworks. The Yokohama Strategy and Plan of Action for a Safer World, established in 1994, was among the first international initiatives to emphasize the integral link between sustainable development and DRR [19]. Since then, this connection has been consistently reinforced through pivotal global agreements, including the Millennium Development Goals (MDGs), the Johannesburg Plan of Implementation (2002), the Hyogo Framework for Action (2005–2015), “The Future We Want” outcome document from Rio + 20 (2012), the Sendai Framework for Disaster Risk Reduction (2016), and the 2030 Agenda for Sustainable Development (2015) [19]. Targets within SDG 11 (sustainable cities and communities) and SDG 9 (industry, innovation, and infrastructure) specifically highlight the interdependence between disaster risk reduction efforts and SDGs. These frameworks collectively stress the imperative of integrating resilience-building measures into development strategies to mitigate disaster risks and foster sustainable progress worldwide.

To counteract concerning trends, nations and communities must establish robust systems capable of preventing or effectively managing risks. The ability to manage risks and recover from disasters in a transformative manner is pivotal for building resilience. As emphasized, “If the SDGs are to be achieved, it is vital that resilience is built into societies and governance models; otherwise, poverty and inequality will persist” [18]. Central to these efforts is the integration of the 2030 Agenda for Sustainable Development with the Sendai Framework for Disaster Risk Reduction 2015–2030, which embeds disaster risk reduction and climate change adaptation across all sectors of sustainable development [20]. This integrated approach not only strengthens global resilience but also promotes coherent policy actions and enhances regional cooperation to manage shared risks effectively. Furthermore, proactive disaster risk reduction has demonstrated greater cost-effectiveness compared to focusing solely on response and recovery efforts. Policymakers are increasingly advocating for the integration of disaster risk management into development strategies with support from organizations like the Economic and Social Commission for Asia and the Pacific (ESCAP) to assist nations in enhancing their resilience to natural hazards.

Yamaguchi et al. (2023) conducted a bibliometric analysis of 312 scientific publications on SDG reviews between 2015 and 2022, seven years after the SDGs were established [21]. Their findings provide insights into the current status and trends in SDG-related research, particularly literature reviews addressing the 2030 Agenda. Key findings include a rise in SDG publications, diversification in research areas, and the prominence of SDG 9 (artificial intelligence) based on citation analysis. The study also identified gaps in research integration and highlighted the need for interdisciplinary approaches and robust methodologies to assess SDG progress effectively. Future research should develop frameworks to address these gaps and promote practical, interdisciplinary, and systems-thinking approaches to achieving the 2030 Agenda.

In a study by Mishra et al. (2023) analyzing 12,176 articles published between 2015 and 2022, it was found that SDG-related research is heavily concentrated in developed countries with significant emphasis on general sustainability, environmental sciences, and green technology [22]. The study highlights disparities in participation with wealthier countries contributing more to SDG research. Additionally, research is highly regionalized with developed nations focusing on education, cities, and climate, while developing countries prioritize poverty, hunger, and gender equality. Despite progress, gaps in collaboration and concentration of research in a few countries remain critical concerns. The study emphasizes the need for comprehensive, interdisciplinary assessments of SDGs to address trade-offs and interdependencies among goals, ensuring the achievement of the 2030 Agenda.

The literature reviewed by Yamaguchi et al. (2023) and Mishra et al. (2023) highlights that SDG-related research is predominantly concentrated in developed countries with a strong focus on sustainability and technology [21,22]. However, a significant gap exists in systematically examining the association between disaster impacts and SDG progress, particularly in poverty reduction and community resilience. Understanding disaster impacts is essential because disaster risk reduction (DRR) is crucial for achieving the SDGs. The Sendai Framework and the 2030 Agenda emphasize the need to integrate DRR into SDG strategies to build resilience, reduce poverty, and ensure sustainable development globally. Addressing this gap is vital for developing comprehensive strategies that align DRR with SDGs [19]. This study aims to empirically explore the impact of disasters on sustainable development, resilience, and poverty levels over the past two decades, and to validate the following hypotheses:

H1: 

There is a negative association between the impact of disasters and the progress toward achieving the SDGs.

H2: 

There is a negative association between the impact of disasters and the level of community resilience.

H3: 

There is a negative association between the impact of disasters and the prevalence of poverty.

This study offers a novel contribution by systematically examining the associations between disaster impacts and progress toward achieving the SDGs, particularly in poverty reduction and community resilience. While prior research has focused on sustainability and technology in developed countries, this study addresses a significant gap by integrating disaster risk reduction (DRR) with SDG strategies. The empirical analysis provides valuable insights into how disasters influence national sustainability, resilience, and poverty, guiding policy formulation and implementation of SDG goals. The findings will benefit young researchers, practitioners, policymakers, and public officials as they seek to understand and address disaster impacts on SDG progress.

2. Materials and Methods

2.1. Study Area

This study focuses on 166 countries, which covered seven regions, namely, Eastern Europe and Central Asia, East and South Asia, Latin America and the Caribbean, Middle East and North Africa, Oceania, Organisation for Economic Co-operation and Development (OECD) members, and Sub-Saharan Africa (Figure 1). In a regional breakdown of countries, 23 countries (14%) are located in Eastern Europe and Central Asia, 19 countries (11%) in East and South Asia, 23 countries (14%) in Latin America and the Caribbean, 16 countries (10%) in the Middle East and North Africa, 38 countries (23%) are OECD member countries, 2 countries (1%) in Oceania, and 45 countries (27%) in Sub-Saharan Africa.

The global Sustainable Development Index stands at 66.69 out of a maximum of 100 points. Among the regions, OECD member countries score the highest with 77.81, followed by Eastern Europe and Central Asia at 71.76, Latin America and the Caribbean at 70.18, East and South Asia at 67.20, the Middle East and North Africa at 67.08, Sub-Saharan Africa at 53.02, and Oceania at 52.71 [23]. However, Sub-Saharan Africa and Oceania record lower scores of 53.02 and 52.71, respectively, both falling below the global average of 66.68.

The Sustainable Development Index scores vary significantly based on a country’s income level. For the fiscal year 2025, high-income countries—defined as those with a gross national income (GNI) per capita exceeding USD 14,005—achieve the highest average score of 78.16. They are followed by upper-middle-income countries with GNI per capita between USD 4516 and USD 14,005, scoring 71.73. Lower-middle-income countries, whose GNI per capita ranges from USD 1146 to USD 4515, have a score of 63.40. Low-income countries, defined by the World Bank Atlas method as having a GNI per capita of USD 1145 or less in 2023, record the lowest score at 50.70. These data underscore the correlation between a country’s economic status and its sustainable development performance [17,23,24].

2.2. Data

The study utilized two datasets: The first dataset, the SDG Index from the Sustainable Development Solutions Network (SDSN), focuses on sustainable development and was sourced from the “Sustainable Development Report 2024”, provided by the SDG Transformation Center, complete with indicators [23]. The data were publicly available online at the SDG Transformation Center website, where we downloaded the “Sustainable Development Report 2024”, dataset, and the accompanying methodology paper.

The SDG Index is calculated through a series of steps designed to measure progress toward each of the 17 SDGs [23,25]. First, relevant indicators are selected across social, economic, environmental, and institutional dimensions. These indicators are then normalized to a common scale (typically 0 to 100), where 100 indicates that the SDG target has been fully achieved. Next, the indicators are weighted based on their significance, either equally or according to expert input, and the scores are aggregated to produce individual SDG scores. These scores are averaged to generate the overall SDG Index score for each country or region with higher scores, indicating better progress. Countries are ranked based on their index scores, allowing for global comparisons. The SDG Index is updated annually to reflect the most recent data, ensuring it remains a comprehensive tool for tracking global progress toward sustainable development.

The “Sustainable Development Report 2024” leverages the SDG Index to gauge how well each UN member state is progressing toward the SDGs. This index assigns each country a score from 0 to 100, reflecting how close they are to achieving their SDG targets based on the latest available data. A perfect score of 100 indicates that a country has fully met all its SDG targets, while any score below 100 shows how much further it needs to go. Due to ongoing updates in methodology and indicators, it is important to note that comparing SDG Index scores across different reports is not straightforward. However, to help track progress over time, the report includes a recalculated time series of the SDG Index using this year’s methods and indicators applied to data from previous years [23]. The time series data encompasses SDI scores spanning from 2000 to 2022 for each country, totaling 4140 observations.

The report uses a simple color-coded arrow system to show how different targets are faring: Green arrows point to targets on track to be met by 2030; orange arrows indicate no significant progress; yellow suggests moderate improvements, and red arrows warn of deterioration. If a previously met target is now declining, it is marked with a red arrow as well. For a global perspective, the report assesses the status of SDG targets using a population-weighted average, recognizing indicators as on track only if they show consistent progress both in 2015 and in the latest year. This methodical yet accessible approach helps us understand where the world stands in its quest to achieve sustainable development [23].

The second dataset was the Emergency Events Database (EM-DAT), which consists of a detailed archive of global disaster data covering natural and technological hazards from 2000 to 2023 [26]. The data were publicly available on the EM-DAT website (public.emdat.be) through a user account and were downloaded during the study period. EM-DAT defines a disaster as an event that overwhelms local capabilities, meeting one or more of the following criteria: at least 10 fatalities, 100 or more people affected, a call for international aid, or a declaration of a state of emergency. The database thoroughly documents human impact data, which includes the total number of deaths (including long-term missing persons) and a detailed count of those affected—encompassing the injured, homeless, and other impacted individuals. It also records the total estimated economic damages from disasters, calculated in US dollars at the time of the event, and adjusts these figures for inflation using the Consumer Price Index (CPI) from the Organization for Economic Cooperation and Development (OECD), ensuring the data remains relevant over time.

It should be noted that the EM-DAT database—while the only comprehensive, free-access disaster loss database with effective global coverage—has limitations due to varying sources and reporting standards worldwide [27,28]. Three types of data quality issues can be considered: EM-DAT faces limitations, including missing disaster events, events with incomplete impact data, and discrepancies in documented attributes compared to other sources [29]. The main limitations include over- or under-reporting of certain types of risk (risk bias), gaps in historical records (time bias), reliance on direct or indirect financial losses (accounting bias), focus on high-intensity events (threshold bias), and over-focus on densely populated or more accessible areas (geographic bias) [27,28].

2.3. Methods

This study employed regression models to analyze the impact of various factors on sustainability, resilience, and poverty, utilizing data from the “Sustainable Development Report 2024” and the Emergency Events Database (EM-DAT) covering the period from 2000 to 2023. The dependent variables were the normalized logarithmic values of sustainability, resilience, and poverty. Independent variables included the logarithmic transformations of disasters, deaths, injuries, affected individuals, and economic damage.

The complex relationship between disasters, sustainability, resilience, and poverty is increasingly evident in scholarly discussions. The association between disasters and sustainable development, resilience, and poverty is multifaceted [30,31,32]. Disasters, whether natural or man-made, often worsen existing vulnerabilities, hampering sustainable development and deepening poverty [28,31,33]. Marginalized populations bear the brunt of these disasters, trapped in cycles of poverty due to their heightened exposure and limited recovery support [28,34]. The Intergovernmental Panel on Climate Change (IPCC) Working Group I (WGI) Sixth Assessment Report (AR6): “Climate Change 2021, The Physical Science Basis” highlights that climate change is escalating the frequency and severity of disasters, disproportionately impacting vulnerable communities and disrupting sustainable development [35]. Effective governance, fair resource distribution, and comprehensive risk management are essential for building resilience and breaking the poverty cycle.

The World Bank Group’s “COP26 Climate Brief” reveals that natural disasters push millions into poverty annually and advocates for increased investment in climate resilience, which can provide significant economic returns despite current shortfalls in adaptation efforts [36]. Tierney [37] argues that disasters stem from social and institutional frameworks that prioritize economic growth over risk reduction, making a compelling case for integrating resilience and addressing social inequalities to mitigate disaster risks and foster long-term development [7,8,9,10,11,12,13,14,15,16,31]. This underscores the need for a holistic approach that combines sustainable practices, poverty alleviation, and resilience-building to effectively manage disaster risks and support sustainable development [30]. Sustainable development and disaster risk reduction are intrinsically linked, as a lack of sustainable practices can exacerbate disaster risk and impact [30,31,32]. Therefore, achieving sustainable development requires robust strategies for enhancing resilience and mitigating the impact of disasters to prevent poverty exacerbation and promote long-term development [30,31,32].

Based on our literature review, nine regression models were developed to capture the delayed effects of disasters on the dependent variables over different time lags: Sustainability Models (Models 1, 2, 3) examined the impact on sustainability scores with one-year, two-year, and three-year lags, respectively; Resilience Models (Models 4, 5, 6) assessed the impact on resilience scores with one-year, two-year, and three-year lags, respectively; and Poverty Models (Models 7, 8, 9) analyzed the impact on poverty scores with one-year, two-year, and three-year lags, respectively.

We used the following regression specification for each model:

Y_it = β₀ + β₁ log (Disasters_it) + β₂ log (Deaths_it) + β₃ log (Injuries_it) + β₄ log (Affected_it) +
β₅ log (Damage_it) + γ Region_i + δ IncomeGroup_i + ϵ_it,

(1)

where Y_it represents the normalized logarithmic value of the dependent variable (sustainability, resilience, or poverty) for country _i at time _t. The normalization of dependent variables for sustainability, resilience, and poverty can be achieved using the min–max normalization method, which scales the data to a range between 0 and 1. The formula for min-max normalization is Normalized Value = (X − X_min)/(X_max − X_min), where X is the original value of the variable, X_min is the minimum value of the variable in the dataset, and X_max is the maximum value of the variable in the dataset.

The logarithmic transformations of the independent variables were included to address skewness in the data and to linearize relationships. Regional dummy variables (γ Region_i) and income group dummy variables (δ IncomeGroup_i) were included to control for regional and income group effects, respectively, while ϵ_it represents the error term. Ordinary least squares (OLS) regression was used to estimate the coefficients of the models. The goodness of fit for each model was evaluated using R-squared (R²) and adjusted R-squared (adj. R²) values, as well as the root-mean-square error (RMSE). The statistical significance of the coefficients was assessed using t-tests with p-values indicating the level of significance.

In this study, sustainability was assessed using SDG Index scores, which measure a country’s progress toward achieving the SDGs. The index assigns each country a score from 0 to 100 with 100 indicating that all SDG targets have been fully met. Scores below 100 highlight the gap remaining to achieve these goals. Resilience was evaluated using the SDG 11 index, which aggregates indicators related to urban inclusivity, safety, resilience, environmental sustainability, and governance. A higher SDG 11 score reflects stronger performance in developing cities and human settlements that are inclusive, safe, resilient, and sustainable. Poverty was measured using the SDG 1 index, which tracks progress toward eradicating poverty in all its forms by examining indicators such as the proportion of the population living below the international poverty line, access to basic services, and the coverage of social protection systems. A higher SDG 1 score indicates substantial progress in reducing poverty and improving living standards, underscoring the effectiveness of poverty eradication efforts.

The independent variables include disaster impacts measured by the number of disasters, total deaths, injuries, affected population, and total economic damage. According to the EM-DAT documentation, disasters are counted based on significant events that meet EM-DAT’s inclusion criteria, such as causing 10 or more deaths, affecting 100 or more people, or leading to a declaration of a state of emergency. Total deaths are measured as the sum of all fatalities, including both deceased and missing persons. Injuries are defined as the number of people who sustained physical injuries, trauma, or illness requiring immediate medical assistance due to the disaster. The affected population is measured as the total number of individuals requiring immediate assistance due to the disaster. Total economic damage is calculated as the value of all economic losses directly or indirectly caused by the disaster, reported in thousands of US dollars (‘000 USD’) relative to the start year, and is unadjusted for inflation.

To capture the delayed effects of disasters, we introduced lagged variables in the regression models. Models 1, 4, and 7 included one-year lagged independent variables, Models 2, 5, and 8 included two-year lagged variables, and Models 3, 6, and 9 included three-year lagged variables. This approach allowed us to observe how the impacts of disasters evolve over time. The residuals versus fitted values plots were examined to check for any patterns or heteroscedasticity in the residuals, ensuring the validity of the model assumptions. The results from these models were used to understand the complex interplay between disasters and their impacts on sustainability, resilience, and poverty. The findings were interpreted to provide insights into policy implications and strategies for enhancing resilience and sustainable development in the face of disasters.

3. Results

The findings of this study are presented in two primary areas: the distribution of disasters by income group and geographic region, and their subsequent impacts on sustainability, resilience, and poverty.

3.1. Disasters by Income Group

This study presents an in-depth look at the distribution and impact of disasters classified by technological, natural, and climate-related events across various global income groups. The data span from 2000 to 2022 and involve 187 countries grouped into high, upper-middle, lower-middle, and low income (

Table 1. Distribution of disasters and their impacts by income group. The table displays the number of countries, total disasters, technological disasters, natural disasters, climate-related disasters, floods, deaths, injured individuals, affected individuals, and economic damage across high, upper-middle, lower-middle, and low-income groups. The percentage contributions of each income group to the total number of disasters and impacts are also provided. Developed by the authors.

Income Group	No. of Countries	Disasters	Technological Disasters	Natural Disasters	Climate-Related Disasters	Flood	Deaths	Injured	Affected	Damage (USD)
High income	57	3175	741	2434	2540	762	276,221	349,505	137,132,979	3,010,456,154
Upper-middle income	50	5212	2134	3078	3214	1383	506,224	3,397,433	2,152,934,471	1,050,554,510
Lower-middle income	54	4953	2098	2855	2979	1243	782,149	2,548,648	1,832,986,150	354,563,886
Low income	26	2020	673	1347	1382	599	147,360	1,028,339	437,830,014	35,767,987
Total	187	15,360	5646	9714	10,115	3987	1,711,954	7,323,925	4,560,883,614	4,451,342,537
High income	30%	21%	13%	25%	25%	19%	16%	5%	3%	68%
Upper-middle income	27%	34%	38%	32%	32%	35%	30%	46%	47%	24%
Lower-middle income	29%	32%	37%	29%	29%	31%	46%	35%	40%	8%
Low income	14%	13%	12%	14%	14%	15%	9%	14%	10%	1%
Total	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%

, Figure 2).

In high-income countries, a total of 3175 disasters were reported, accounting for 21% of global technological disasters and 13% of natural disasters. These countries also faced 25% of the climate-related disasters, impacting over 137 million people. Despite these large numbers, the human toll was relatively lower with 276,221 deaths and 349,505 injuries. However, the economic impact was severe with damages reaching approximately USD 3.01 trillion—68% of total damages recorded globally.

Upper-middle-income countries experienced a heavier burden with 5212 disasters, comprising 34% of technological and 38% of natural disasters. The human cost was particularly high with 506,224 deaths and 3,397,433 injuries. Economically, these countries suffered considerable losses totaling USD 1.05 trillion, which represented 24% of the total damages.

Lower-middle-income countries faced 4953 disasters, constituting 32% of technological and 37% of natural disasters. They recorded the highest casualty numbers with 782,149 deaths and 2,548,648 injuries, yet their economic damages were lower compared to other groups, totaling USD 354 billion or 8% of the total damages.

Low-income countries reported 2020 disasters, making up 13% of technological and 12% of natural disasters. Despite their vulnerability, these countries saw fewer deaths (147,360) and injuries (1,028,339) compared to other groups, and their economic damages were the least, amounting to roughly USD 35.7 billion or just 1% of total damages.

The analysis highlights stark disparities in how different income groups handle and are affected by disasters. Upper and lower middle-income countries, despite fewer resources, endure a greater human toll, whereas high-income countries, though they experience fewer disasters, face massive economic losses. This pattern suggests significant differences in asset values, infrastructure resilience, and disaster preparedness across income groups. The findings emphasize the critical need for disaster risk reduction strategies that are tailored to the specific vulnerabilities and capacities of each income bracket, aiming to reduce both human and economic losses from disasters.

3.2. Disasters by World’s Regions

Table 2. Distribution of disasters and their impacts by region. The table presents the number of countries, total disasters, technological disasters, natural disasters, climate-related disasters, floods, deaths, injured individuals, affected individuals, and economic damage for different regions: Eastern Europe and Central Asia (EECA), East and South Asia (ESA), Latin America and the Caribbean (LAC), Middle East and North Africa (MENA), OECD countries, Oceania, and Sub-Saharan Africa (SSA). The percentage contributions of each region to the total number of disasters and impacts are also provided. Developed by the authors.

Income Group	No. of Countries	Disasters	Technological Disasters	Natural Disasters	Climate-Related Disasters	Flood	Deaths	Injured Individuals	Affected Individuals	Damage (USD)
EECA	23	1122	359	763	805	364	92,422	125,313	56,787,019	56,195,289
ESA	20	4695	1856	2839	2981	1246	795,977	2,301,547	3,574,793,160	1,088,014,744
LAC	29	1639	471	1168	1200	544	274,805	2,745,414	170,688,633	128,089,839
MENA	17	1075	693	382	418	197	86,573	643,488	53,530,149	76,578,170
OECD	38	3131	729	2402	2468	723	264,829	546,134	169,943,299	3,066,967,576
Oceania	11	193	7	186	187	36	2151	9053	7,331,256	2,889,606
SSA	49	3505	1531	1974	2056	877	195,197	952,976	527,810,098	32,607,313
Total	187	15,360	5646	9714	10,115	3987	1,711,954	7,323,925	4,560,883,614	4,451,342,537
EECA	12%	7%	6%	8%	8%	9%	5%	2%	1%	1%
ESA	11%	31%	33%	29%	29%	31%	46%	31%	78%	24%
LAC	16%	11%	8%	12%	12%	14%	16%	37%	4%	3%
MENA	9%	7%	12%	4%	4%	5%	5%	9%	1%	2%
OECD	20%	20%	13%	25%	24%	18%	15%	7%	4%	69%
Oceania	6%	1%	0%	2%	2%	1%	0%	0%	0%	0%
SSA	26%	23%	27%	20%	20%	22%	11%	13%	12%	1%
Total	100%	100%	100%	100%	100%	100%	100%	100%	100%	100%

presents a detailed examination of the impact of disasters across various global regions, delineated by income groups and types of disasters. This comprehensive review sheds light on how different regions cope with the challenges posed by natural and technological calamities, emphasizing the influence of geographical exposure, economic conditions, and disaster management capabilities (Table 2, Figure 3).

Eastern Europe and Central Asia (EECA) experienced a moderate number of disasters, totaling 1122 events, which constitutes 12% of the global total recorded in the dataset. Despite these numbers, the region reports relatively low economic damage and a smaller number of affected individuals. This could suggest that EECA has effective disaster response mechanisms in place or that the disasters are less severe. In stark contrast, East and South Asia (ESA) face a significant brunt of disasters, suffering from 31% of the global total, primarily in natural and climate-related categories. This region endures the highest number of affected individuals and substantial economic damage, highlighting its vulnerability and the severe impact of these events.

Latin America and the Caribbean (LAC) report 11% of global disasters, and they experience considerable human toll, including numerous injuries and deaths. However, the economic damage and the total number of affected individuals are lower, suggesting variable impacts across different disaster types. The Middle East and North Africa (MENA) encounter fewer disasters overall but have a higher proportion of technological disasters. The relatively moderate economic damage and number of affected people suggest that the impacts are manageable or are of lower severity.

OECD countries, while experiencing 20% of the recorded disasters, incur disproportionately high economic losses, making up 69% of the total damages reported. This indicates that high-value assets are likely at risk, and the financial stakes are significant in these developed areas. Oceania shows a minimal share of global disasters with low figures across all metrics, which might suggest lower incidence rates or underreporting. Sub-Saharan Africa (SSA) reports a significant share of global disasters, particularly in terms of climate-related disasters and floods, yet the economic impact and the number of people affected are relatively low, pointing to either lesser economic value at risk or gaps in data capture.

Collectively, the data from 187 countries documenting 15,360 disasters reveals profound global impacts with 1,711,954 deaths and 7,323,925 injuries. The substantial economic losses, totaling approximately USD 4.45 trillion, further underscore the severe repercussions of these disasters. The observed disparities in death tolls, injuries, and economic damages relative to the number of disasters across different regions highlight the varied severities of disasters and the effectiveness of disaster management. This dataset underscores the critical need for region-specific disaster risk reduction strategies that effectively mitigate both human and economic losses arising from disasters.

3.3. Disasters and Sustainability

In this study, we investigated the long-term effects of disasters on sustainability by employing regression models that analyzed outcomes over one-, two-, and three-year time lags. Each model used the normalized sustainability score as the dependent variable, allowing for meaningful comparisons across different contexts. The independent variables—log-transformed data on disasters, deaths, injuries, and economic damage—were chosen to linearize relationships and effectively manage data skewness.

Our findings, as detailed in

Table 3. Regression results for models analyzing the impact of disasters on sustainability, resilience, and poverty. The table presents the regression results for nine models examining the impact of disasters, deaths, injuries, affected individuals, and economic damage on normalized logarithmic values of sustainability (Models 1, 2, 3), resilience (Models 4, 5, 6), and poverty (Models 7, 8, 9) with one-year, two-year, and three-year lags, respectively. Coefficients (β) and t-statistics (in parentheses) are provided for each independent variable. The table also includes region and income group dummy variables with Oceania and low-income groups as the omitted categories. The R-squared (R²), adjusted R-squared (adj. R²), and RMSE values indicate the goodness of fit for each model. Developed by the authors.

Models	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Dependent Variable	Sustainability	Sustainability	Sustainability	Resilience	Resilience	Resilience	Poverty	Poverty	Poverty
Lagged Years	1	2	3	1	2	3	1	2	3
Disasters	−0.0181 **	−0.0207 ***	−0.0192 **	−0.0143 *	−0.0158 *	−0.0164 *	−0.0131	−0.0167 *	−0.0169 *
	(−3.04)	(−3.38)	(−3.09)	(−2.14)	(−2.29)	(−2.37)	(−1.64)	(−2.02)	(−2.10)
Deaths	−0.0181 ***	−0.0167 ***	−0.0168 ***	−0.0184 ***	−0.0188 ***	−0.0187 ***	−0.0122 **	−0.0110 **	−0.00893 **
	(−6.25)	(−5.60)	(−5.68)	(−5.82)	(−5.80)	(−5.83)	(−3.21)	(−2.76)	(−2.73)
Injured	0.000371	0.000971	0.00196	−0.00255	−0.00120	0.000509	−0.00286	−0.00236	−0.00492
	(0.21)	(0.54)	(1.08)	(−1.14)	(−0.50)	(0.21)	(−1.28)	(−1.02)	(−1.90)
Affected	−0.00223	−0.00222	−0.00232	0.0000402	−0.000342	−0.000794	−0.00173	−0.00119	0.00160
	(−1.52)	(−1.45)	(−1.52)	(0.02)	(−0.18)	(−0.41)	(−1.09)	(−0.73)	(1.01)
Damage	0.0106 ***	0.00996 ***	0.00894 ***	0.00744 ***	0.00657 ***	0.00610 **	0.00532 **	0.00509 **	0.00244
	(7.51)	(6.86)	(6.02)	(4.12)	(3.49)	(3.25)	(2.90)	(2.74)	(1.19)
Regions
1. Oceania (Omitted)	0	0	0	0	0	0	0	0	0
	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)
2. OECD members	0.0883	0.0909	0.0781	−0.121 ***	−0.111 ***	−0.111 ***	0.171	0.187	0.212
	(1.49)	(1.41)	(1.06)	(−8.38)	(−6.26)	(−5.92)	(1.12)	(1.20)	(1.39)
3. MENA	0.0393	0.0405	0.0263	−0.178 ***	−0.167 ***	−0.168 ***	0.259	0.271	0.299 *
	(0.66)	(0.62)	(0.36)	(−10.05)	(−8.17)	(−7.50)	(1.69)	(1.73)	(1.97)
4. LAC	0.0305	0.0301	0.0171	−0.159 ***	−0.146 ***	−0.151 ***	0.126	0.141	0.157
	(0.51)	(0.47)	(0.23)	(−11.80)	(−8.89)	(−8.37)	(0.82)	(0.91)	(1.03)
5. E. Europe and C. Asia	0.0381	0.0431	0.0273	−0.171 ***	−0.163 ***	−0.158 ***	0.160	0.177	0.196
	(0.64)	(0.67)	(0.37)	(−10.89)	(−8.61)	(−7.84)	(1.04)	(1.13)	(1.28)
6. East and South Asia	0.0832	0.0844	0.0741	−0.166 ***	−0.155 ***	−0.151 ***	0.180	0.198	0.220
	(1.41)	(1.32)	(1.01)	(−11.03)	(−8.48)	(−7.73)	(1.18)	(1.27)	(1.45)
7. Sub-Saharan Africa	−0.0368	−0.0364	−0.0517	−0.192 ***	−0.191 ***	−0.184 ***	−0.0847	−0.0675	−0.0747
	(−0.62)	(−0.56)	(−0.70)	(−8.83)	(−8.11)	(−7.30)	(−0.55)	(−0.43)	(−0.49)
Income groups
1. Low income	0	0	0	0	0	0	0	0	0
	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)	(.)
2. Upper-middle	0.167 ***	0.169 ***	0.163 ***	0.122 ***	0.118 ***	0.122 ***	0.328 ***	0.334 ***	0.319 ***
	(9.45)	(9.36)	(8.61)	(4.78)	(4.57)	(4.53)	(12.37)	(12.11)	(11.84)
3. Lower-middle	0.290 ***	0.290 ***	0.285 ***	0.299 ***	0.294 ***	0.298 ***	0.488 ***	0.493 ***	0.477 ***
	(15.45)	(15.03)	(14.28)	(12.13)	(11.49)	(11.48)	(17.67)	(17.23)	(17.99)
4. High income	0.384 ***	0.383 ***	0.380 ***	0.357 ***	0.349 ***	0.353 ***	0.594 ***	0.599 ***	0.599 ***
	(16.20)	(15.75)	(15.25)	(12.80)	(11.93)	(11.99)	(21.13)	(20.63)	(22.74)
Constant	0.287 ***	0.293 ***	0.320 ***	0.622 ***	0.634 ***	0.634 ***	0.255	0.232	0.227
	(4.50)	(4.26)	(4.13)	(17.68)	(17.59)	(16.56)	(1.63)	(1.45)	(1.46)
N	739	713	676	739	713	676	714	690	587
R2	0.809	0.807	0.808	0.715	0.706	0.702	0.839	0.840	0.846
adj. R2	0.805	0.803	0.804	0.710	0.700	0.696	0.836	0.837	0.843
RMSE	0.0829	0.0832	0.0825	0.102	0.104	0.105	0.107	0.107	0.107

* p < 0.05, ** p < 0.01, *** p < 0.001.

, consistently demonstrate how disasters impact sustainability across all models. In the One-Year Lag Model (Model 1), a one-unit increase in the logarithm of disasters led to a significant decrease of 0.0181 points in sustainability scores (p < 0.01), highlighting the immediate negative impact of disasters. Similarly, an increase in the logarithm of deaths reduced sustainability scores by 0.0181 points (p < 0.001), underscoring the profound effect of fatalities on sustainable development within a year. Conversely, the log of economic damages showed a positive relationship with sustainability, where a one-unit increase resulted in a 0.0106 point increase in scores (p < 0.001), likely reflecting the investments in recovery and infrastructure following significant financial losses. Interestingly, the number of injuries, as captured by log injuries, did not significantly affect sustainability, indicating that injuries alone might not influence broader sustainability outcomes.

The Two-Year and Three-Year Lag Models (Models 2 and 3) revealed persistent negative impacts from disasters and deaths, consistent with the findings from the one-year model. These results underscore the enduring detrimental effects of these factors on sustainability. Moreover, the positive influence of economic damage remained evident in both models, suggesting that substantial financial losses can catalyze investments that foster sustainable recovery initiatives.

The robustness of our models is corroborated by strong statistical metrics: The sample sizes were substantial (N = 739, 713, 676 for Models 1, 2, and 3, respectively), providing solid support for our findings. The R-squared values were consistently high (approximately 0.808 on average), indicating significant explanatory power. Furthermore, the adjusted R-squared values were stable, and the RMSEs were minimal, highlighting the models’ accuracy and reliability in predicting sustainability outcomes.

Overall, this comprehensive analysis reveals that the impacts of disasters on sustainability are intricate and multifaceted. While disasters and deaths invariably challenge sustainability efforts, highlighting the need for urgent policy interventions, the positive outcomes associated with economic damages suggest that disasters, though devastating, also present opportunities for substantial recovery and development. This nuanced understanding promotes a balanced approach in policymaking, emphasizing the importance of disaster preparedness and responsive recovery strategies that utilize post-disaster phases as catalysts for sustainable development.

3.4. Disasters and Resiliency

In this study, we explored the time-delayed effects of disasters on urban resilience, aligning our analysis with SDG 11, which promotes the creation of inclusive, safe, resilient, and sustainable cities and human settlements. We conducted regression analyses over one, two, and three-year periods, using normalized logarithmic values of resilience as the dependent variable. This approach helped ensure uniformity and comparability across various scales and contexts.

We utilized three distinct models—Models 4, 5, and 6—which corresponded to lag periods of one, two, and three years, respectively. Each model assessed the impact of disasters, deaths, injuries, the number of affected individuals, and economic damages on the logarithmic values of resilience. The results indicated a consistently significant negative correlation between the frequency or severity of disasters and resilience across all models with coefficients ranging from −0.0143 to −0.0164 (p < 0.05). This trend underscores the substantial adverse impact that disasters can have on urban resilience. Regarding mortality, the effect of deaths on resilience was strongly negative in all models with coefficients ranging from −0.0184 to −0.0188 (p < 0.001). This finding sheds light on how higher mortality rates significantly compromise resilience efforts in urban environments. Interestingly, the number of injuries did not significantly influence resilience with initial models even indicating a slight negative trend. This suggests that the impact of injuries on resilience might be mitigated by varying severities and the recovery outcomes associated with these injuries. Furthermore, the number of individuals affected by disasters did not show a significant impact on resilience, indicating that merely experiencing a disaster does not necessarily weaken the resilience capacities of individuals or communities. In contrast, economic damages had a positive correlation with resilience (coefficients ranging from 0.00610 to 0.00744, p < 0.01), suggesting that financial losses may drive investments in recovery and resilience-enhancing measures, potentially leading to a rebound or build-back-better effect in post-disaster scenarios.

Significant regional differences emerged with regions such as the Middle East and North Africa (MENA), Latin America and the Caribbean (LAC), Eastern Europe and Central Asia, East and South Asia, and Sub-Saharan Africa showing lower resilience scores compared to the omitted category, Oceania. Moreover, higher economic groups—upper-middle, lower-middle, and high-income—demonstrated significantly better resilience than low-income groups, highlighting the role of economic capacity in bolstering resilience.

The models demonstrated strong explanatory power with R-squared values ranging from 0.702 to 0.715, suggesting that they effectively captured a significant proportion of the variance in resilience. The RMSEs were consistently low, enhancing the reliability of the models. Overall, the findings reveal a complex interplay between the impacts of disasters and urban resilience, highlighting both vulnerabilities and strengths within urban settings. The beneficial impact of economic recovery following damages, along with the detrimental effects of physical harm, calls for nuanced strategies that leverage post-disaster recovery phases to enhance resilience. This study underscores the need for targeted resilience-building strategies, especially in economically disadvantaged regions, to fulfill the global objectives of sustainable urban development.

3.5. Disasters and Poverty

In this study, we investigated the effects of various factors on poverty, aligning our analysis with SDG 1, which aims to end poverty in all its forms. Utilizing regression analysis, we assessed the impacts of disasters, deaths, injuries, those affected by disasters, and economic damage on normalized logarithmic values of poverty across one-, two-, and three-year time lags. Our findings from Models 7, 8, and 9, which correspond to these respective periods, reveal consistent patterns and significant insights.

We observed that increases in disaster events consistently exacerbate poverty levels with statistically significant negative impacts noted in the longer lags of Models 8 and 9 (β = −0.0167 and β = −0.0169; p < 0.05). Similarly, higher mortality rates were strongly correlated with worse poverty outcomes across all models (β = −0.0122, β = −0.0110, β = −0.00893; p < 0.01), indicating that deaths from disasters significantly contribute to increasing poverty.

The impact of injuries and the extent to which individuals were affected by disasters were less consistent and did not reach statistical significance in most cases. This suggests a complex relationship with poverty that may be influenced by other mediating factors, such as the severity of injuries and the effectiveness of disaster response and recovery efforts.

Interestingly, economic damage showed a positive correlation with poverty reduction in the first two models (β = 0.00532 and β = 0.00509; p < 0.01), implying that in some scenarios, financial losses might lead to inflows of aid or funds for reconstruction that may temporarily alleviate poverty levels.

Significant regional variations were also observed. Most regions did not show significant differences in poverty outcomes compared to Oceania, the omitted category, but the MENA region in Model 9 and other areas displayed variations often reflecting regional vulnerabilities or the effectiveness of local economic policies. Income levels also played a crucial role; upper-middle, lower-middle, and high-income groups exhibited significantly lower poverty levels compared to low-income groups across all models (β values ranging from 0.319 to 0.599; p < 0.001), highlighting the protective effect of higher income against the exacerbating impacts of disasters on poverty. The models demonstrated strong explanatory power with R² values ranging from 0.839 to 0.846, suggesting a substantial proportion of the variance in poverty was effectively captured by the models. The adjusted R² values and RMSE remained consistent, enhancing the reliability and precision of our findings.

These results illustrate the complex interplay between disasters and poverty, revealing that while immediate impacts such as deaths exacerbate poverty, economic damages can sometimes activate resources that mitigate poverty, at least temporarily. The findings underscore the need for targeted poverty reduction strategies that consider regional disparities and income levels to bolster resilience against the poverty-exacerbating effects of disasters, supporting global efforts to achieve SDG 1.

The analysis of residuals against fitted values for the nine models—categorized into Sustainability Models (1, 2, and 3), Resilience Models (1, 2, and 3), and Poverty Models (1, 2, and 3)—indicates that the models generally fit the data well (Figure 4). The residuals for all models are mostly randomly dispersed around zero, suggesting that the assumptions of normality and homoscedasticity are reasonably met. While slight indications of heteroscedasticity are observed in Resilience Model 2 and Poverty Model 3, they are not pronounced enough to impact the overall model validity significantly. Thus, these results affirm the robustness of the models in studying the impacts on sustainability, resilience, and poverty.

4. Discussion

This study investigates the associations between disasters and sustainability, resilience, and poverty by using the SDI as a measure of sustainability. The SDI score ranges from 0 to 100, with 100 indicating full achievement of the SDGs. Resilience is assessed through Goal 11, which focuses on making cities inclusive, safe, resilient, and sustainable, while poverty is measured by Goal 1, which aims to eradicate poverty. We applied nine regression models to assess the effects of disasters, deaths, injuries, affected populations, and economic damages on these variables over one-, two-, and three-year lag periods.

The analysis revealed negative associations between disaster impacts and sustainability, resilience, and poverty levels, while injuries and the number of affected individuals did not significantly affect these variables. Although injuries and displacements are vital for immediate disaster response, their long-term impact on development indicators can be alleviated through recovery processes, which may reduce their statistical significance in broader analyses [38,39]. According to UN Deputy Secretary-General Amina J. Mohammed (2023), disasters can rapidly undo decades of sustainable development progress, severely impacting the most vulnerable populations. The inability to identify, prevent, and mitigate risks before they escalate into disasters not only jeopardizes the achievement of the SDGs but also disproportionately affects those already at the greatest risk.

The findings emphasize the critical importance of effective disaster risk reduction (DRR) strategies in achieving SDGs, reducing poverty, and building resilient communities. When the SDGs were adopted in 2015, two interlinking agreements were also established: the Sendai Framework for Disaster Risk Reduction 2015–2030 and the Paris Agreement on Climate Change (COP21) [30,40]. Over the past 30 years, DRR strategies have evolved from disaster management to resilience-based approaches and integrating climate change adaptation (CCA) with DRR is now essential for sustainable development. Transforming governance mechanisms across all levels is crucial for mitigating disaster and climate risks, ensuring safe growth, and fostering a resilient future [30].

Moreover, the study’s findings underscore the importance of early warning systems in mitigating disaster risks and reducing their potential impacts. The “Early Warnings for All” initiative, launched to protect everyone on Earth from hazardous weather, water, and climate events by 2027, underscores the necessity of these systems, particularly in the context of increasing extreme weather due to climate change [41]. Despite progress, significant gaps remain, especially in small island developing states and least developed countries, prompting a global call to action by UN Secretary-General António Guterres in 2022.

The study also reveals significant regional variations, with regions such as MENA, LAC, Eastern Europe and Central Asia, East and South Asia, and Sub-Saharan Africa showing lower resilience and sustainability scores compared to Oceania. These disparities highlight the need for region-specific disaster response and recovery strategies. Policymakers must tailor their approaches to address regional vulnerabilities while leveraging local strengths. Income group analysis showed that higher-income groups had significantly better resilience and lower poverty levels than low-income groups, underscoring the protective effect of economic capacity against disaster impacts. Enhancing economic growth in lower-income regions could be crucial for building resilience and reducing poverty.

These findings underscore the importance of regionalizing or localizing the SDGs to adapt them to the specific needs, priorities, and contexts of different regions [42]. While the SDGs offer a global framework for sustainable development, their successful implementation depends on understanding which goals are most relevant to local decision-makers and communities. Tailoring the SDGs to regional and local scales requires considering the unique environmental, social, economic, and institutional conditions of each area. Additionally, sub-national governance institutions play a vital role in translating these goals into actionable policies, necessitating capacity building and stronger governance mechanisms at the regional level. This localized approach ensures that the SDGs address the specific challenges and opportunities within each community, fostering more effective and inclusive sustainable development outcomes.

In conclusion, this study underscores the profound impact of disasters on sustainability, resilience, and poverty, highlighting the necessity of integrating DRR and CCA into sustainable development strategies. The findings reveal significant regional variations, emphasizing the need for localized approaches that consider specific environmental, social, and economic contexts. Adapting to climate change, strengthening governance, and tailoring SDG implementation to regional needs are critical for building resilient communities and achieving SDGs globally.

Based on the findings, targeted policy recommendations should focus on regions and countries with poor performance by strengthening DRR and CCA strategies tailored to local contexts. Investing in early warning systems, particularly in vulnerable areas, is crucial for minimizing disaster impacts. Additionally, promoting economic growth in low-income regions is essential for reducing poverty and enhancing resilience. Improving governance capacity at regional and local levels will ensure effective implementation of the SDGs, fostering inclusive and sustainable development outcomes across diverse regions.

5. Conclusions

This study demonstrates the significant impact of disasters on sustainability, resilience, and poverty, emphasizing the critical need for integrating DRR and CCA strategies into sustainable development efforts. The findings highlight regional disparities, underscoring the importance of tailored, localized approaches for implementing SDGs. Theoretically, this study contributes to understanding the interplay between disasters and development outcomes, while practically, it informs targeted policy interventions. However, this study’s limitations include its reliance on specific datasets and regional coverage. Future research should explore long-term impacts, cross-regional comparisons, and the integration of additional variables to enhance the robustness of the findings and provide more comprehensive policy recommendations.