Linking fire incidence to chronic respiratory disease outcomes in Brazil’s Legal Amazon: an ecological study

Abstract

Introduction

Wildfire activity has intensified in the Brazilian Legal Amazon, increasing population-level exposure to air pollutants associated with respiratory morbidity and mortality.

Objective

To examine the association between wildfire occurrence and deaths and disability-adjusted life years (DALYs) from chronic respiratory diseases in the states of the Legal Amazon.

Methods

This ecological analytical study used secondary data from the Global Burden of Disease (GBD) database and the National Institute for Space Research (INPE – Queimadas Program). Annual absolute numbers of deaths and DALYs from chronic respiratory diseases were obtained for all ages and both sexes from 2011 to 2021. Annual wildfire foci counts were extracted from INPE. Associations between wildfire activity and health outcomes were assessed for each state using Spearman’s rank correlation (n = 11 annual observations per state).

Results

Pará presented the largest absolute burden of deaths and DALYs, followed by Maranhão and Mato Grosso, which also showed high wildfire activity. Most states exhibited increasing temporal trends in deaths and DALYs, while Maranhão showed declining deaths and Rondônia declines in both outcomes. A pronounced peak in wildfire activity and respiratory disease burden was observed in 2020. Significant positive correlations between wildfire foci and both deaths and DALYs were identified in Acre, Amazonas, Mato Grosso, and Rondônia, whereas no significant associations were observed in other states. Limitations: Findings are based on aggregated annual data and reflect ecological associations; causal inference and individual-level risk estimation are not possible.

Conclusion

Wildfire activity is heterogeneously associated with the population-level burden of chronic respiratory diseases across the Legal Amazon, underscoring the importance of region-specific environmental and public health responses.

Introduction

Extreme weather events have become increasingly intense and frequent as a consequence of global climate change. In South America, climatic phenomena such as El Niño have contributed to higher temperatures and prolonged dry periods, particularly in Brazil, creating environmental conditions that favor the occurrence of wildfires [1, 2]. Similar patterns have been observed in other regions of the world, including North America, Europe, and Oceania, where extreme climatic conditions have been associated with increased wildfire activity and adverse public health effects [3].

Wildfires represent a major source of air pollution and pose a substantial risk to respiratory health. Combustion processes release large amounts of carbon monoxide, nitrogen dioxide, and fine particulate matter, which can penetrate deeply into the respiratory tract and deposit in terminal bronchioles and alveoli [4]. These pollutants trigger inflammatory and oxidative responses that contribute to the development and exacerbation of chronic respiratory diseases, including asthma and chronic obstructive pulmonary disease (COPD), leading to increased respiratory morbidity and mortality [5–7].

The Legal Amazon is particularly vulnerable to these impacts due to the combined effects of extensive forest cover, deforestation, and biomass burning associated with agricultural expansion [8–9]. States within this region have experienced recurrent wildfire activity, with Pará consistently reporting the highest number of wildfire foci in Brazil, followed by Mato Grosso and Maranhão [10]. This persistent fire activity increases the likelihood of sustained population-level exposure to harmful air pollutants and places additional pressure on already strained health systems.

Evidence indicates that air pollution resulting from wildfires contributes to worsening respiratory outcomes, especially among populations with pre-existing chronic respiratory conditions [4, 7]. While the health effects of heat exposure and other climate-related stressors have been documented, wildfire-related air pollution represents a more direct and measurable pathway linking environmental change to chronic respiratory disease burden in the Amazon region [1, 7].

Despite the recognized role of wildfires as an environmental determinant of respiratory health, population-level assessments examining the temporal association between wildfire occurrence and chronic respiratory disease burden in the Brazilian Legal Amazon remain limited. In particular, few studies have evaluated how wildfire activity co-varies with mortality and disability-adjusted life years (DALYs) attributable to chronic respiratory diseases across different states and over extended time periods.

Accordingly, this study aimed to examine the association between wildfire occurrence and deaths and DALYs due to chronic respiratory diseases in the states of the Legal Amazon between 2011 and 2021.

Methods

The data used in this study are publicly available, and it is not possible to access identifying information about the participants. Furthermore, this study adheres to Resolution Number 510 of 2016 of the National Health Council (Brazil). This analytical study used data from two secondary sources: the Global Burden of Disease (GBD) database and the INPE – Queimadas Program. The GBD is a free, comprehensive, and systematic global health database that provides evidence for population health decision-making. Coordinated by the Institute for Health Metrics and Evaluation (IHME) at the University of Washington, the GBD applies standardized methods to estimate the burden of more than 300 diseases and injuries in 195 countries and territories [11].

Key metrics include Years of Life Lost (YLL) due to premature mortality, Years Lived with Disability (YLD), and DALYs, which combine YLL and YLD to provide an overall measure of disease burden. The GBD highlights health disparities, informs policymakers and health professionals in prioritizing interventions, and optimizes resource allocation. It is updated periodically, allowing comparisons over time and evaluation of public health interventions [11].

For this study, we selected the following filters from the GBD platform: GBD Estimate – “Cause of Death or Injury”; Measure – “Deaths” and “DALYs”; Metric – “Number”; Cause – “Chronic Respiratory Diseases.” We included all ages and both sexes across the nine states of the Legal Amazon (Acre, Amapá, Amazonas, Maranhão, Mato Grosso, Pará, Rondônia, Roraima, and Tocantins) between 2011 and 2021. The selected indicators were exported to Excel spreadsheets for data organization.

Because the GBD platform provides aggregated estimates and does not allow access to state-level age-stratified population data for the entire study period, recalculation of population-standardized rates was not feasible. Therefore, all analyses were conducted using absolute estimates (“numbers”) of deaths and DALYs, and no mortality or DALY rates were calculated. To improve interpretability and to contextualize differences in population size across states, mean population estimates for the period 2011–2021 were included solely as descriptive denominators and were not used to derive rates.

INPE’s mission (Queimadas Program) is to advance science and technology in space and Earth system sciences while providing unique products and services for Brazil [12]. Its Burned Areas Program monitors active fire outbreaks detected by reference satellites and compiles daily data into temporal series, enabling trend analysis over time [13]. Filters applied included “By State” followed by “Amapá,” “Amazonas,” “Maranhão,” “Mato Grosso,” “Pará,” “Rondônia,” “Roraima,” and “Tocantins,” within the tab “Statistics: States, Regions, and Brazilian Biomes.”

To evaluate the association between wildfire foci and health outcomes (deaths and DALYs) from chronic respiratory diseases, Spearman’s rank correlation coefficient (ρ) was used. This nonparametric test was selected because it does not assume normality and is appropriate for short time series [14]. Correlations were calculated using annual aggregated absolute estimates for each state from 2011 to 2021 (n = 11 annual observations per state).

Spearman’s rho coefficients and corresponding p-values were calculated for associations between wildfire foci and deaths, and between wildfire foci and DALYs. Statistical significance was set at p < 0.05. For descriptive interpretation of correlation strength, rho values were classified as weak (|ρ| < 0.40), moderate (|ρ| = 0.40–0.69), or strong (|ρ| ≥ 0.70).

Confidence intervals for Spearman’s rho were explored; however, given the short time series and the rank-based nature of the statistic, robust estimation of confidence intervals was not feasible across all states. Therefore, confidence intervals are not reported, and correlation estimates should be interpreted cautiously in light of sampling uncertainty. All analyses were conducted using SPSS version 20.0.

Temporal trends in deaths and DALYs were assessed descriptively. Percent change was calculated as the relative difference between values observed in the first (2011) and last (2021) years of the study period for each state. This approach was selected given the ecological nature of the analysis and the limited number of observations. No regression-based trend modeling was applied. Annual state-level data for deaths, DALYs, and wildfire foci are publicly available through the GBD and INPE – Queimadas Program platforms and can be directly accessed and reproduced using the filters described above; therefore, no supplementary dataset was generated.

All data were publicly available and de-identified, exempting the study from research ethics committee review, according to Brazilian National Health Council Resolution No. 510/2016 [15]. We focused on absolute numbers of deaths and DALYs to characterize the total population-level burden of chronic respiratory diseases across states. Although population-standardized rates are more appropriate for individual-level risk comparisons, absolute estimates are particularly relevant for public health planning, health system preparedness, and resource allocation. Population denominators were therefore used for contextual interpretation only, and no rate standardization was applied.

Results

Mortality and DALYs capture distinct but complementary dimensions of disease burden, with mortality reflecting fatal outcomes and DALYs integrating both premature death and years lived with disability. Considering these two components jointly allows a more comprehensive assessment of the spatial and temporal heterogeneity of chronic respiratory diseases in the Legal Amazon.

Pará presented the highest mean annual mortality from chronic respiratory diseases among the states of the Legal Amazon (1,834.49 deaths), followed by Maranhão (1,634.35) and Mato Grosso (902.59). In contrast, Roraima (72.21), Amapá (120.08), and Acre (228.08) exhibited the lowest mortality contributions, with values substantially lower than those observed in Pará, differing by up to approximately one order of magnitude (Fig. 1; Table 1). Intermediate levels of mortality were observed in Amazonas (618.95), Rondônia (416.00), and Tocantins (375.51). This wide gradient illustrates the marked heterogeneity of the mortality burden across the region, which is further highlighted by the choropleth map (Fig. 2), visually emphasizing the geographic concentration of higher burdens in Pará, Maranhão, and Mato Grosso, and lower burdens in Roraima, Amapá, and Acre.

Fig. 1.

State-level distribution of the burden of chronic respiratory diseases in the Legal Amazon (2011–2021). Legend – Panel A shows the mean annual absolute number of deaths from chronic respiratory diseases by state, and Panel B shows the mean annual absolute number of disability-adjusted life years (DALYs). Estimates represent averages for the period 2011–2021 and are presented for descriptive comparison only; no population-standardized rates were calculated

Table 1.

Mean annual absolute numbers of deaths and dalys from chronic respiratory diseases, mean population size, wildfire foci, and descriptive Temporal trends in the States of the legal Amazon (2011–2021)

State	Deathsa	DALYsa	Mean population (millions)b	Wildfire focic	Descriptive trendd
Acre	228.08	7,117.08	0.88	6,409.45	↑ Deaths (+ 41%), ↑ DALYs (+ 25%)
Amapá	120.08	4,543.64	0.86	1,757.18	↑ Deaths (+ 47%), ↑ DALYs (+ 24%)
Amazonas	618.95	23,597.97	4.20	11,067.27	↑ Deaths (+ 37%), ↑ DALYs (+ 24%)
Maranhão	1,634.35	52,069.15	7.10	21,597.18	↓ Deaths (− 1%), ↑ DALYs (+ 3%)
Mato Grosso	902.59	28,307.94	3.55	25,709.45	↑ Deaths (+ 39%), ↑ DALYs (+ 24%)
Pará	1,834.49	59,731.82	8.70	31,634.36	↑ Deaths (+ 30%), ↑ DALYs (+ 16%)
Rondônia	416.00	13,023.54	1.80	9,477.18	↓ Deaths (− 7%), ↓ DALYs (− 7%)
Roraima	72.21	3,020.31	0.65	2,043.18	↑ Deaths (+ 10%), ↑ DALYs (+ 10%)
Tocantins	375.51	11,460.86	1.60	12,907.27	↑ Deaths (+ 15%), ↑ DALYs (+ 7%)

^aDeaths and DALYs represent mean annual absolute numbers for the period 2011–2021, derived from the Global Burden of Disease (GBD) study

^bMean population represents the approximate average population size (millions) for each state during 2011–2021 and is provided for contextual interpretation only; no rates were calculated

^cWildfire foci represent the mean annual number of fire detections reported by the INPE Burned Areas Program

^dDescriptive trends correspond to percent change between 2011 and 2021; no regression-based trend modeling was applied. ↑ indicates a positive percent change and ↓ indicates a negative percent change between 2011 and 2021. These symbols represent descriptive temporal variation only and do not imply statistical significance

Fig. 2.

Spatial distribution of mortality and DALY burden from chronic respiratory diseases in the Legal Amazon (2011–2021)

Beyond mortality, DALYs revealed additional layers of heterogeneity in the overall disease burden. Pará also exhibited the highest mean annual DALY burden (59,731.82 DALYs), followed by Maranhão (52,069.15) and Mato Grosso (28,307.94), whereas Roraima (3,020.31), Amapá (4,543.64), and Acre (7,117.08) consistently presented the lowest values (Fig. 1; Table 1). While the ranking of states was broadly consistent between deaths and DALYs, the magnitude of DALY differences was proportionally larger, reflecting the additional contribution of non-fatal health loss to the overall burden. For example, Maranhão and Pará showed similar mortality levels but differed more markedly in DALYs, suggesting variation in the relative contribution of disability to total disease burden.

These state-level differences largely reflect variation in population size and wildfire activity rather than direct comparisons of individual risk, as no population-standardized rates were calculated. The presented estimates should therefore be interpreted as indicators of total burden rather than individual-level risk (Table 1). Overall, states with higher numbers of wildfire foci tended to present higher absolute numbers of both deaths and DALYs, although substantial heterogeneity was observed across the region. Figure 1 highlights a pronounced contrast between states with very high absolute burden (Pará, Maranhão, and Mato Grosso) and those with consistently low burden (Roraima, Amapá, and Acre) for both outcomes.

Temporal trends further highlighted differences between the fatal and non-fatal components of disease burden. Descriptive trend analysis indicated an overall increase in both deaths and DALYs in most states over the study period. These trends represent percent changes between 2011 and 2021 and should be interpreted as descriptive temporal patterns rather than modeled temporal effects. For deaths, increases ranged from + 10% in Roraima to + 47% in Amapá, while Maranhão exhibited a slight decline (− 1%). For DALYs, most states showed upward trends, with increases of + 25% in Acre, + 24% in Amapá, Amazonas, and Mato Grosso, + 16% in Pará, and + 7% in Tocantins. Maranhão showed a modest increase in DALYs (+ 3%) despite a slight decline in deaths, suggesting a growing contribution of non-fatal health loss to the overall burden. Rondônia was the only state to present consistent reductions in both deaths and DALYs (− 7% for each outcome).

In 2020, wildfire foci peaked in Mato Grosso, Pará, and Amazonas, coinciding with the highest annual estimates of both deaths and DALYs in most states. This temporal overlap suggests a potential synchrony between years of intense biomass burning and peaks in respiratory health burden. Rondônia and Roraima deviated from this pattern, with reductions observed in 2021. Despite the overall upward tendency, temporal trajectories were not uniform across states, indicating heterogeneous temporal responses in both fatal and non-fatal components of disease burden. Figure 3 illustrates the aggregated annual trajectory of deaths from chronic respiratory diseases in the Legal Amazon, highlighting a pronounced peak in 2020.

Fig. 3.

Annual deaths from chronic respiratory diseases in the Legal Amazon, 2011–2021. Legend – Annual absolute number of deaths from chronic respiratory diseases in the Legal Amazon between 2011 and 2021. Values represent aggregated annual estimates derived from the Global Burden of Disease (GBD) study. The dashed vertical line highlights the year 2020, when the highest number of deaths was observed during the study period. No rates or regression-based trend modeling were applied

Spearman correlation analysis identified statistically significant positive associations between wildfire foci and the absolute numbers of deaths and DALYs from chronic respiratory diseases in states with predominant Amazon forest cover, including Acre, Amazonas, Mato Grosso, and Rondônia (Tables 2 and 3). For DALYs specifically, strong correlations were observed in Rondônia (ρ = 0.93, p < 0.0001) and Mato Grosso (ρ = 0.85, p = 0.001), while moderate correlations were detected in Amazonas (ρ = 0.70, p = 0.01) and Acre (ρ = 0.68, p = 0.02). These associations suggest that, in these states, interannual fluctuations in wildfire activity were closely aligned not only with mortality but also with broader health losses attributable to disability. No statistically significant associations were observed in the remaining states. States with stronger correlations generally corresponded to those with extensive forest cover and recurrent biomass burning, whereas more populous and environmentally heterogeneous states exhibited weaker or absent associations.

Table 2.

Correlation coefficients are based on mean annual absolute numbers of deaths, DALYs, and wildfire foci, using n = 11 annual observations per state (2011–2021)

State	Coefficient (ρ)	p	Interpretation
Acre	0.72	0.01	Strong positive correlation
Amapá	-0.45	0.16	No correlation
Amazonas	0.65	0.02	Moderate correlation
Maranhão	-0.38	0.25	No correlation
Mato Grosso	0.88	0.001	Strong positive correlation
Pará	-0.12	0.72	No correlation
Rondônia	0.91	0.0001	Strong positive correlation
Roraima	0.15	0.66	No correlation
Tocantins	0.54	0.09	Weak correlation (Not sig.)

Correlations are based on n = 11 annual observations per state. Confidence intervals for Spearman’s rho were explored but are not reported due to limitations associated with short time series and rank-based estimation

Not sig. Not significant

Table 3.

Correlation coefficients are based on mean annual absolute numbers of deaths, DALYs, and wildfire foci, using n = 11 annual observations per state (2011–2021)

State	Coefficient (ρ)	p	Interpretation
Acre	0.68	0.02	Positive correlation
Amapá	-0.50	0.12	No correlation
Amazonas	0.70	0.01	Moderate correlation
Maranhão	-0.45	0.16	No correlation
Mato Grosso	0.85	0.001	Strong positive correlation
Pará	-0.10	0.77	No correlation
Rondônia	0.93	< 0.0001	Strong positive correlation
Roraima	0.20	0.56	No correlation
Tocantins	0.58	0.06	Weak correlation (Not sig.)

Correlations are based on n = 11 annual observations per state. Confidence intervals for Spearman’s rho were explored but are not reported due to limitations associated with short time series and rank-based estimation

Not sig. Not significant

Together, these findings indicate that wildfire-related respiratory burden in the Legal Amazon is not limited to fatal outcomes, but also involves substantial and spatially heterogeneous non-fatal health losses, as captured by DALYs.

Discussion

The findings of this study reinforce the population-level association between wildfire activity and the burden of chronic respiratory diseases in the Legal Amazon. Wildfires emit large amounts of fine particulate matter (PM₂.₅), carbon monoxide, and other air pollutants that adversely affect respiratory health. Prolonged exposure to these pollutants has been consistently associated with increased hospitalizations and mortality due to asthma, chronic obstructive pulmonary disease, and respiratory infections [16, 17].

Among the states of the Legal Amazon, Pará presented the highest absolute numbers of deaths and DALYs from chronic respiratory diseases, followed by Maranhão and Mato Grosso, whereas Roraima, Amapá, and Acre consistently exhibited the lowest absolute numbers for both outcomes. These differences primarily reflect variation in population size, demographic structure, and cumulative exposure context rather than direct comparisons of individual risk. The coexistence of elevated wildfire activity and higher absolute health burden is consistent with the temporal co-variation observed in the descriptive and correlation analyses [18].

Pará, which has experienced extensive deforestation and recurrent fire activity driven largely by agricultural expansion, represents a critical setting where environmental degradation and public health intersect. Large-scale forest conversion leads to substantial emissions of air pollutants, and continuous exposure to these emissions is a recognized contributor to respiratory disease exacerbation [19, 20]. Notably, despite presenting the largest absolute burden of deaths and DALYs, Pará did not exhibit a statistically significant correlation between wildfire foci and respiratory outcomes. This pattern indicates that high absolute disease burden does not necessarily correspond to stronger temporal association with wildfire activity at the state level, particularly in large and heterogeneous populations where multiple pollution sources coexist.

Mato Grosso, although located in Brazil’s Midwest region, contains extensive forest and savanna areas that are frequently affected by fire, often in the context of agricultural frontier expansion [21, 22]. In contrast to Pará, Mato Grosso exhibited strong positive correlations between wildfire activity and respiratory outcomes, indicating closer temporal alignment between annual wildfire foci and respiratory disease burden. Similarly, Rondônia, despite presenting lower absolute numbers of deaths and DALYs than the most populous states, showed the strongest correlations between wildfire activity and respiratory outcomes, suggesting a tighter temporal coupling between environmental exposure and health burden. Maranhão displayed a distinct pattern, characterized by elevated absolute estimates of deaths and DALYs alongside a slight declining temporal trend [23, 24].

Across most states, upward trends in the absolute numbers of deaths and DALYs were observed over the study period. These trends coincided with increases in wildfire activity and broader environmental and demographic changes during the same period [25–27]. Rondônia differed from other states by showing declining absolute trends while maintaining strong positive correlations between wildfire foci and respiratory burden. This combination indicates that, although the overall burden decreased, year-to-year fluctuations in wildfire activity remained temporally associated with respiratory health outcomes in this state.

The pronounced peak in wildfire activity and respiratory disease burden observed in 2020 requires careful interpretation. From a descriptive standpoint, 2020 represents a year in which both environmental and health indicators reached their highest levels during the study period. Environmentally, this year was characterized by severe drought conditions and intensified deforestation across the Amazon, which coincided with extensive wildfire activity and increased exposure to smoke and particulate matter [18, 28].

Concurrently, 2020 marked the first year of the COVID-19 pandemic, which substantially affected respiratory morbidity and mortality and disrupted healthcare utilization, diagnostic pathways, and reporting systems [29]. COVID-19–related mortality may have contributed to elevated respiratory death counts through direct effects, exacerbation of pre-existing chronic respiratory conditions, or indirect effects related to delayed access to care.

Given the ecological design and the use of aggregated annual data, it is not possible to disentangle wildfire-related effects from pandemic-related influences at the individual level. The observed 2020 peak should therefore be interpreted as reflecting overlapping population-level processes rather than a single dominant driver. Within this context, wildfire activity and pandemic-related factors represent concurrent stressors captured in the same temporal window.

Significant positive correlations identified in Acre, Amazonas, Mato Grosso, and Rondônia indicate consistent temporal co-variation between wildfire activity and respiratory disease burden in these states. These associations were observed in settings with substantial biomass availability and recurrent fire activity [16, 18]. Differences in population density, urbanization, and access to healthcare may further shape how these associations manifest at the state level [30].

Beyond environmental exposure, several contextual factors may contribute to the observed heterogeneity in the burden of chronic respiratory diseases and its temporal association with wildfire activity across the Legal Amazon. Demographic structure, including age distribution and the proportion of older adults, may influence baseline vulnerability to respiratory morbidity and mortality. Population density and settlement patterns also shape exposure profiles, as densely populated urban areas may experience overlapping sources of air pollution, while sparsely populated regions may be more directly affected by biomass burning emissions.

Socioeconomic conditions and regional inequalities in income, education, and housing quality further modulate susceptibility and adaptive capacity, potentially influencing both disease burden and access to timely care. In addition, disparities in healthcare access, infrastructure availability, and diagnostic capacity across states may affect hospitalization, mortality, and DALY estimates, as well as the detection and reporting of respiratory conditions. Variability in surveillance systems and health information infrastructure may therefore contribute to differences in observed trends and correlations at the state level.

In this context, particular attention should be given to highly vulnerable populations, including isolated and recently contacted Indigenous peoples (PIAVs), who may be disproportionately affected by wildfire-related exposures. These groups often live in close dependence on forest ecosystems, experience sustained exposure to smoke and fine particulate matter, and face substantial barriers to timely access to healthcare services. Respiratory insults associated with wildfire emissions may therefore have more severe and persistent consequences in these populations, particularly given their limited access to preventive care, diagnostic resources, and treatment.

Importantly, the use of aggregated population-level data likely underestimates the true burden of wildfire-related respiratory outcomes among these groups. Localized spikes in morbidity, subclinical respiratory impairment, and indirect health effects may not be adequately captured by routine surveillance systems, which often underrepresent territorially isolated communities. Future research should prioritize spatially disaggregated approaches and culturally sensitive monitoring strategies to better characterize the health impacts of wildfires on Indigenous and other socially marginalized populations in the Amazon.

Finally, regional disparities in environmental governance, wildfire prevention policies, and emergency response capacity may shape the intensity, duration, and health impact of wildfire events. Differences in the implementation of prevention and control strategies across states may partially explain why similar levels of wildfire activity are associated with distinct respiratory health patterns in different contexts. These structural and contextual factors highlight the importance of interpreting the findings within a broader socio-environmental framework rather than attributing observed patterns solely to wildfire exposure.

In contrast, the absence of clear associations in other states reflects heterogeneity in land-use patterns, climatic conditions, urbanization, and exposure profiles. Strong correlations in some states and absent associations in others should therefore be interpreted as differences in temporal coupling between wildfire activity and respiratory outcomes, rather than as differences in susceptibility alone.

An important consideration in interpreting these findings is the potential for ecological fallacy. Associations observed at the state level do not necessarily reflect individual-level relationships between wildfire exposure and respiratory health outcomes. This ecological analysis captures population-level patterns and temporal co-variation, but does not allow causal inference or estimation of individual risk. Additionally, the use of aggregated annual data with only 11 observations per state limits the assessment of seasonal dynamics, exposure lags, and short-term health effects.

Another important methodological consideration of this study is the reliance on absolute numbers rather than population-standardized rates. While rates would be more appropriate for comparing individual-level risk across states, our objective was to quantify the total burden of chronic respiratory diseases potentially associated with wildfire activity at the population level. Absolute estimates better reflect the cumulative pressure on health systems, emergency response capacity, and social protection structures, which is particularly relevant in a region marked by large territorial extension and demographic heterogeneity such as the Legal Amazon.

Finally, although reliance on absolute estimates partially reflects differences in population size and age distribution across states, the inclusion of population denominators and the use of DALYs derived from internally standardized GBD models support cautious interpretation focused on spatial patterns and temporal trends rather than direct comparisons of risk magnitude. Future studies integrating individual-level data, age-standardized rates, high-resolution exposure metrics, and multivariable analytical approaches will be necessary to further clarify the pathways linking wildfire activity and respiratory health outcomes in the Amazon region.

Conclusion

This study provides consistent evidence of an association between wildfire activity and the burden of chronic respiratory diseases in the Legal Amazon. States characterized by larger populations, intensive deforestation, and higher wildfire activity—such as Pará, Maranhão, and Mato Grosso—exhibited a greater absolute burden of mortality and DALYs, underscoring the public health relevance of wildfire-related air pollution in environmentally vulnerable regions.

These findings highlight the urgent need for coordinated public health and environmental policies that prioritize wildfire prevention, strengthen surveillance of air quality and respiratory health, and ensure timely access to healthcare for vulnerable populations. Integrated strategies combining environmental protection, sustainable land-use practices, and health system preparedness are essential to mitigate the long-term health impacts of wildfire-related pollution.

Future studies using high-resolution temporal data and analytical approaches capable of capturing delayed and cumulative health effects will further refine the understanding of these associations and support the development of evidence-based interventions.

Authors’ contributions

ACN, APS, ESM, FRPQ – Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Validation, Visualization, Writing (original draft, review, and editing).

Funding

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES PDPG, Edital 16/202288887.692321/2022-00).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The data used in this study are publicly available, and it is not possible to access identifying information about the participants. Furthermore, this study adheres to Resolution Number 510 of 2016 of the National Health Council (Brazil).

Consent for publication

Not applicable.

Competing interests

Prof. André Pontes-Silva, PhD, serves as Editor and Reviewer for the BMC Group. The authors declare no further competing interests.