Unveiling spatial clusters of systemic sclerosis mortality in Spain: A comprehensive geographical analysis

1 ORIGINAL RESEARCH ARTICLE Unveiling Spatial Clusters of Systemic Sclerosis Mortality in Spain: A Comprehensive Geographical Analysis Revelando Clústeres Espaciales de Mortalidad por Esclerodermia Sistémica en España: Un Análisis Geográfico Exhaustivo Authors: Lucia Cayuela (1) [email protected]om José-Juan Pereyra-Rodríguez (2,3) [email protected] Paz Collado Ramos (4) [email protected]om Nuria Garvín Grande (4) [email protected] Aurelio Cayuela (5) [email protected]om Institutions: 1. Department of Internal Medicine, Hospital Severo Ochoa, Leganés, Spain 2. Department of Dermatology, Virgen del Rocío University Hospital, Seville, Spain. 3. Department of Medicine, University of Seville, Seville, Spain. 4. Rheumatology Department, Transitional Care Clinic, Hospital Universitario Severo Ochoa, Madrid, Spain. 5. Unit of Public Health, Prevention and Health Promotion. South Seville Health Management Area. Seville, Spain Correspondence: José-Juan Pereyra-Rodríguez . Department of Dermatology, Virgen del Rocío University Hospital & Department of Medicine, University of Seville, Seville, Spain. E-mail: [email protected] 2 Title: Unveiling Spatial Clusters of Systemic Sclerosis Mortality in Spain: A Comprehensive Geographical Analysis Abstract: Objectives: This study examined the spatial patterns of systemic sclerosis mortality in Spain from 2003 to 2022, identifying provincial-level clusters and sex-specific differences to explore potential underlying factors. Methods: Mortality and population data (2003-2022) were sourced from the National Institute of Statistics. Age-standardized mortality rates were calculated, and spatial patterns were analyzed using standardized mortality ratios and smoothed relative risks via the Besag, York, and Mollié model, with statistical inference performed using the Integrated Nested Laplace Approximation (INLA) technique. Spatial clustering was assessed using Tango’s and Kulldorff’s tests. Results: A total of 2,016 SSc deaths were recorded, revealing a marked gender disparity. SSc mortality rates showed a consistent annual increase of 2.1% for both sexes, with women experiencing rates approximately 3.4 times higher than men. Age-specific analysis demonstrated higher mortality rates among women across all age groups, with an average sex ratio of 2.9, and rates increased with age for both sexes. Spatial analysis identified significant clusters of elevated SSc mortality in northwestern Spain, primarily in León and Asturias, with additional clusters extending to nearby provinces. These regions, characterized by silica mining and related industries, exhibited notable variations between men and women in the specific provinces affected. Conclusions: This study identifies significant geographic and sex-based disparities in SSc mortality across Spain, with prominent clusters in the north. The results highlight the potential impact of environmental and occupational exposures on disease outcomes, emphasizing the need for targeted public health interventions in high-risk areas. Keywords: Systemic Sclerosis, mortality, spatial analysis, clusters, gender disparities, environmental factors, occupational exposures, Spain 3 Título: Revelando Clústeres Espaciales de Mortalidad por Esclerodermia Sistémica en España: Un Análisis Geográfico Exhaustivo Resumen Objetivos: Este estudio examinó los patrones espaciales de mortalidad por esclerodermia sistémica en España desde 2003 hasta 2022, identificando clústeres a nivel provincial y diferencias por sexo para explorar posibles factores subyacentes. Métodos: Se obtuvieron datos de mortalidad y población (2003-2022) del Instituto Nacional de Estadística. Se calcularon las tasas de mortalidad estandarizadas por edad y se analizaron los patrones espaciales utilizando ratios de mortalidad estandarizados y riesgos relativos suavizados mediante el modelo de Besag, York y Mollié, con inferencia estadística realizada utilizando la técnica de Aproximación de Laplace Anidada Integrada (INLA). Se evaluó la agrupación espacial utilizando las pruebas de Tango y Kulldorff. Resultados: Se registraron un total de 2016 muertes por ES, revelando una marcada disparidad de género. Las tasas de mortalidad por ES mostraron un aumento anual constante del 2.1% para ambos sexos, con tasas en mujeres aproximadamente 3.4 veces más altas que en hombres. El análisis por edad mostró tasas de mortalidad más altas entre las mujeres en todos los grupos de edad, con una proporción media de sexo de 2.9, y las tasas aumentaron con la edad para ambos sexos. El análisis espacial identificó clústeres significativos de mortalidad elevada por ES en el noroeste de España, principalmente en León y Asturias, con clústeres adicionales que se extienden a provincias cercanas. Estas regiones, caracterizadas por la minería de sílice y las industrias relacionadas, exhibieron variaciones notables entre hombres y mujeres en las provincias específicas afectadas. Conclusiones: Este estudio identifica disparidades geográficas y basadas en el sexo significativas en la mortalidad por ES en toda España, con clústeres prominentes en el norte. Los resultados resaltan el posible impacto de las exposiciones ambientales y ocupacionales en los resultados de la enfermedad, enfatizando la necesidad de intervenciones de salud pública específicas en áreas de alto riesgo. Palabras clave: Esclerodermia sistémica, mortalidad, análisis espacial, clústeres, disparidades de género, factores ambientales, exposiciones ocupacionales, España 4 Introduction Systemic sclerosis (SSc) is a chronic autoimmune disease that is characterized by the presence of widespread fibrosis, vascular abnormalities and the presence of autoantibodies. Among the various complications associated with this condition, interstitial lung disease, which affects the tissue and space surrounding the lung's air sacs, represents a significant cause of mortality in patients with SSc1. Although SSc is relatively rare, it has a significant impact on patients' quality of life and places a considerable burden on healthcare systems due to its systemic nature2. SSc is characterized by significant geographic and ethnic disparities in incidence, prevalence, survival, and mortality. These variations suggest a complex interplay of genetic, environmental, and socioeconomic factors. Beyond genetic predisposition, geographical location, economic status, education, well-being, social support, and treatment adherence can all influence disease outcomes1,3. Epidemiological studies consistently demonstrate a north-south gradient in Europe, with higher rates in southern regions4. Additionally, specific geographical clusters of SSc cases have been identified in several countries5–10. These disparities suggest a potential role for environmental factors and genetic backgrounds in disease susceptibility. While the exact cause remains unknown, it is hypothesized that environmental triggers may initiate SSc in genetically predisposed individuals Recent research indicates a global increase in SSc cases, especially in the Americas and Oceania3,11. While SSc mortality has declined since the 2000s, it remains high. A recent US study found women's mortality increased until 1999 and then decreased, while men's remained stable until 2001 and then declined12. In Spain, SSc mortality rose from 1980 to 2019, especially for women13. Despite these insights, there has been no investigation into the geographical patterns of SSc mortality within Spain. Understanding the geographical distribution of SSc mortality is crucial, given the potential influence of environmental and genetic factors on disease outcomes. This study aims to address the current gap by examining the spatial patterns of SSc mortality in Spain from 2003 to 2022. By analysing provinciallevel variations and sex-specific differences. While this analysis can provide insights into regional disparities, it is important to recognize that the identification of potential causes remains tentative, as geographical studies of this nature can only suggest associations rather than establish direct causality. The findings of this study will contribute to a better understanding of geographical disparities in SScrelated deaths in Spain and may inform targeted public health interventions and resource allocation. 5 Methods Data sources Mortality and population data for Spain from 2003 to 2022 were obtained from the National Institute of Statistics. This dataset included annual counts of deaths stratified by sex, age group, and province of residence, as well as corresponding population figures. Deaths identified as the underlying cause of death attributed to SSc were recorded using the code from the International Classification of Diseases, Tenth Revision (ICD-10), specifically code M34. Statistical Analysis: Age-Standardized Mortality Rates: We calculated age-standardized mortality rates (ASMRs) using the direct method, referencing the Standard European Population, and expressed them per 1,000,000 inhabitants. Temporal trends in mortality were assessed using the Joinpoint Regression Program (Version 5.2.0.0). Provincial Spatial Patterns of SSc Mortality: To analyse the spatial patterns of SSc mortality at the provincial level, we computed the standardized mortality ratio (SMR), smoothed relative risk (RR), and posterior probabilities (PP) of RR exceeding 1 for the period 2003–2022. Age- and sex-specific SSc mortality rates were first calculated at the national level and then adjusted for each province's demographic structure through indirect standardization to estimate expected cases. The SMR was determined as the ratio of observed to expected death counts. To account for variability in SMR, we applied spatial smoothing using the Besag, York, & Mollié (BYM) model, a Bayesian hierarchical model that incorporates random effects for spatial contiguity and nonspatial heterogeneity14. The BYM model was fitted using the Integrated Nested Laplace Approximation (INLA) technique, implemented in the R-INLA package15. Excess Mortality Assessment: Excess mortality was evaluated based on PP thresholds9 for RR greater than 1: specifically, 0.05, 0.20, 0.80, and 0.95. Provinces with PP > 0.80 were classified as at risk, those with PP between 0.20 and 0.80 were categorized as having moderate evidence of excess risk, and provinces with PP < 0.20 were considered low risk. Neighbourhood structures for spatial analysis were defined using the 'queen criterion,' wherein neighbours are units that share a common edge or vertex. Statistical Tests and Software: 6 Analyses and maps were generated using GeoDa software. Overdispersion was assessed using the Chi-square test and the Potthoff-Whittinghill test. Global spatial clustering was examined with Tango's test, while local spatial clustering was investigated using Kulldorff and Nagarwalla's likelihood ratio test, with a 15% window size. Clusters yielding p-values < 0.05 were considered statistically significant16. Ethics Statement This study was conducted in accordance with the principles of the Declaration of Helsinki and adhered to Good Clinical Practice standards. We utilized anonymized, aggregate data provided by administrative health systems through regional governmental authorities and the National Institute of Statistics database. As the study was observational and involved no identifiable personal information, informed consent and ethics committee approval were not required, in line with Spanish legal provisions. To ensure the accuracy and transparency of the research, we followed the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER)23 and adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)24 guidelines. Results From 2003 to 2022, there were 2,016 deaths due to SSc in Spain, with a notable gender disparity: 407 deaths among men and 1,609 among women. Figure 1 shows the joinpoint analysis of SSc mortality trends by sex over the 20-year period. Both men and women showed a consistent upward trend, with an annual percentage change (APC) of 2.1% (confidence Interval: 1.3;2.9) for each sex. ASMRs increased from 0.53 to 1.15 per million in men and from 2.71 to 3.90 per million in women. Women had consistently higher death rates, with an average female-to-male ratio of 3.4, ranging from 2.1 in 2020 to 6.4 in 2006. Figure 2 shows the age- and sex-specific mortality rates used to estimate the expected number of provincial cases. Women had higher mortality rates than men in all age groups, with an average sex ratio of 2.9. Mortality rates increased with age for both sexes. 7 Figure 3 shows the SMRs for the 52 Spanish provinces. Among men, SMRs ranged from 0 to 3.2, with an average of 1.1, while in women, they varied from 0.2 to 2.7, with a mean of 1.0. Generally, northern Spanish provinces had higher SMRs for both sexes, although there were exceptions, such as Huelva in men and Malaga, Granada, and Santa Cruz de Tenerife in women. Figure 4 presents the RR and PP of RR > 1 for both sexes. High SSc-associated mortality RRs were concentrated in northwestern Spain, particularly in León and Asturias, with a PP above 0.8 for both sexes. Elevated risks were also observed in men from Palencia and Lugo, and in women living in Barcelona, Girona, Gipuzkoa, Lleida, Zaragoza (Northeast), Malaga, Ceuta (South), the Balearic Islands, and Santa Cruz de Tenerife. Figure 5 presents the findings from the Kulldorff cluster analysis, illustrating provincial clusters of systemic sclerosis (SSc) mortality. The clusters are ordered by their respective Kulldorff statistic, with cluster centers indicated by darker shades of grey. In the analysis for women, a prominent cluster of elevated SSc mortality was identified in north-western Spain, specifically concentrated in the provinces of León and Asturias. For men, a comparable cluster was observed, centered in Asturias and encompassing the provinces of León, Lugo, and Palencia. Additionally, another notable cluster, also centered in León, extended to include the provinces of Asturias, Salamanca, Orense, Lugo, Palencia, Zamora, Valladolid, Cantabria, and Burgos, all situated in the north-western region of Spain. Discussion This study analyses SSc mortality in Spain from 2003 to 2022, revealing increasing rates, gender disparities and regional variations. Our findings are consistent with previous research13,17 and highlight the complex interplay of genetic, diagnostic practices, environmental exposures, and disease outcomes. 8 Our study reveals a persistent increase in SSc mortality rates, contrasting with findings from population-based studies in the US18 and Denmark19, which reported a decline from the early 2000s to 2015. These differences may be attributed to variations in healthcare systems, diagnostic practices, and environmental exposures between regions. The increase in SSc mortality observed in Spain could be partially explained by advancements in diagnostic criteria and heightened disease awareness. For example, the adoption of the ACR-EULAR 2013 diagnostic criteria likely led to increased case identification, similar to trends seen in Sweden20. Such improvements in diagnostic practices could contribute to the rising mortality rates by uncovering previously undiagnosed cases rather than indicating a true increase in disease incidence. This trend contrasts with the general improvement in SSc outcomes reported in other developed countries21. Although the ten-year survival rate for SSc patients in Spain exceeds 70%, mortality remains a concern due to complications like interstitial lung disease, which adversely affects prognosis. In Spain, lung and cardiac involvement are primary causes of death, with peripheral vascular disease and coexisting malignancies also contributing22. Gender disparities in SSc mortality are well-documented, with women being significantly more affected than men. Our study confirms previous research showing higher ASMRs in women compared to men12,13,23. This disparity is likely due to the higher prevalence of SSc in women and the possibility of competing causes of mortality in men. The parallel upward trend in SSc mortality rates for both sexes in our results indicates that factors influencing these trends are affecting both genders equally, underscoring the need for continued surveillance and the development of targeted treatment strategies to address the rising mortality rates across both sexes. While mortality studies do not allow direct inferences about incidence, higher mortality rates generally suggest higher incidence. Although the exact pathogenesis of SSc remains unclear, environmental and occupational triggers, such as silica exposure in mining activities, may play a significant role in developing SSc, particularly in males24. Assessing geographical clusters of high mortality can be a crucial first step in identifying possible causes. 9 Geographic clusters of SSc mortality reveal regional disparities shaped by environmental and socioeconomic factors. Studies have consistently shown that higher mortality rates are associated with industrial pollution, hazardous waste sites, and poor socioeconomic conditions. Research from the U.S.8, Massachusetts5, Italy25, and Canada7 highlights elevated SSc mortality in areas with significant industrial pollution and limited healthcare access. Additionally, spatio-temporal clustering of SSc cases underscores the influence of environmental factors on disease distribution26. Silica and solvents are suspected contributors to SSc. Silica exposure is commonly associated with mining but is also significant in construction, agriculture, and manufacturing24. The rise in high-silica materials in artificial stone products has led to clusters of silica-related diseases, including SSc. Recent studies also suggest that exposure to heavy metals like antimony, cadmium, and lead might play a role in SSc development27. Our spatial analysis aligns with findings from a previous study by the Spanish Network for Epidemiological Research on Rare Diseases (REpIER), which examined the geographical distribution of mortality from "rare diseases," including SSc, in Spain from 1999 to 201328. We identified clusters of elevated SSc mortality in northern Spanish provinces like León and Asturias, regions known for silica mining and glassmaking. This suggests that silica exposure may significantly impact SSc outcomes. Given the established health risks of silica inhalation, further research is needed to explore its potential link to SSc mortality. Additionally, the northwest of Spain, including León and Asturias, has a history of heavy metal pollution from coal mining and metallurgy. This raises concerns about the possible role of heavy metals in SSc development and mortality, though direct causal links require further investigation. High SSc mortality rates in the province of Huelva, particularly among men, may also be related to mining activities. Previous studies have highlighted geographical differences in SSc phenotypes, clinical presentation, management, and prognosis4. These factors may influence mortality risk and contribute to the observed regional patterns in SSc mortality. Variability in healthcare quality may also contribute to these disparities. Addressing these disparities requires training healthcare professionals in the latest guidelines, improving information sharing between facilities, and effectively integrating care services. These measures will help standardize practices, reduce treatment variability, and enhance outcomes for SSc patients across provinces. 16 Figure 2: Age- and Sex-Specific Systemic Sclerosis Mortality Rates at the National Level (2003–2022) 17 Figure 3: Systemic Sclerosis Standardized Mortality Ratios (SMRs) by Sex for the 52 Provinces of Spain (2003–2022) 18 Figure 4: Relative Risk (RR) and Posterior Probability (PP) of RR > 1 for Systemic Sclerosis Mortality by Sex in Spain (2003–2022) 19 Figure 5: Significant Clusters of Elevated Systemic Sclerosis Mortality Rates by Sex Identified Using the Kulldorff Method in Spain (2003–2022) 20 Suppl1. STROBE Statement—checklist of items that should be included in reports of observational studies Item No. Recommendation Page No. Relevant text from manuscript Title and abstract 1 (a) Indicate the study’s design with a commonly used term in the title or the abstract 2-3 The title accurately reflects the study's scope and objectives. (b) Provide in the abstract an informative and balanced summary of what was done and what was found 2-3 The abstract provides a concise overview of the study's objectives, methods, results, and conclusions Introduction Background/rationale 2 Explain the scientific background and rationale for the investigation being reported 4 The introduction provides a clear and concise overview of systemic sclerosis, including its prevalence, mortality, and known risk factors. The study's rationale is well-established, highlighting the need to investigate geographical patterns of SSc mortality in Spain Objectives 3 State specific objectives, including any prespecified hypotheses 4 The specific objectives of the study are clearly stated Methods Study design 4 Present key elements of study design early in the paper 5 The study design is described as an ecological study, using mortality data from administrative health systems Setting 5 Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection 5 The study setting is clearly defined as Spain 21 Participants 6 (a) Cohort study—Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up Case-control study—Give the eligibility criteria, and the sources and methods of case ascertainment and control selection. Give the rationale for the choice of cases and controls Cross-sectional study—Give the eligibility criteria, and the sources and methods of selection of participants 5 The study population consists of individuals who died from systemic sclerosis in Spain between 2003 and 2022 (b) Cohort study—For matched studies, give matching criteria and number of exposed and unexposed Case-control study—For matched studies, give matching criteria and the number of controls per case Variables 7 Clearly define all outcomes, exposures, predictors, potential confounders, and effect modifiers. Give diagnostic criteria, if applicable 5 The key variables, including mortality, population data, age, sex, and province of residence, are identified. Data sources/ measurement 8* For each variable of interest, give sources of data and details of methods of assessment (measurement). Describe comparability of assessment methods if there is more than one group 5 The sources of mortality and population data are specified as the National Institute of Statistics Bias 9 Describe any efforts to address potential sources of bias 6 we followed the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) Study size 10 Explain how the study size was arrived at 5 All cases of mortality from this cause have been included Quantitative variables 11 Explain how quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen and why 5 The variable analyzed is mortality, which is the one included 12 (a) Describe all statistical methods, including those used to control for confounding 6 22 Statistical methods (b) Describe any methods used to examine subgroups and interactions The statistical methods used for data analysis, including age-standardized mortality rates, spatial analysis techniques, and statistical tests, are clearly outlined. (c) Explain how missing data were addressed (d) Cohort study—If applicable, explain how loss to follow-up was addressed Case-control study—If applicable, explain how matching of cases and controls was addressed Cross-sectional study—If applicable, describe analytical methods taking account of sampling strategy (e) Describe any sensitivity analyses Results Participants 13* (a) Report numbers of individuals at each stage of study—eg numbers potentially eligible, examined for eligibility, confirmed eligible, included in the study, completing follow-up, and analysed 6 2,016 deaths due to SSc in Spain, with a notable gender disparity: 407 deaths among men and 1,609 among women. All have been included in the analysis (b) Give reasons for non-participation at each stage (c) Consider use of a flow diagram Descriptive data 14* (a) Give characteristics of study participants (eg demographic, clinical, social) and information on exposures and potential confounders 6-7 Descriptive statistics regarding the number of SSc deaths, age distribution, sex distribution, and temporal trends are presented. (b) Indicate number of participants with missing data for each variable of interest (c) Cohort study—Summarise follow-up time (eg, average and total amount) Outcome data 15* Cohort study—Report numbers of outcome events or summary measures over time 6 Case-control study—Report numbers in each exposure category, or summary measures of exposure Cross-sectional study—Report numbers of outcome events or summary measures 23 Main results 16 (a) Give unadjusted estimates and, if applicable, confounder-adjusted estimates and their precision (eg, 95% confidence interval). Make clear which confounders were adjusted for and why they were included 6-7 Fig 1- 5 : The key findings of the study, including the identification of spatial clusters, gender disparities, and associations with specific regions, are clearly presented. (b) Report category boundaries when continuous variables were categorized (c) If relevant, consider translating estimates of relative risk into absolute risk for a meaningful time period Other analyses 17 Report other analyses done—eg analyses of subgroups and interactions, and sensitivity analyses 7 The results of statistical tests, such as Tango's and Kulldorff's tests, are reported. Discussion Key results 18 Summarise key results with reference to study objectives 8 The key findings of the study are summarized. Limitations 19 Discuss limitations of the study, taking into account sources of potential bias or imprecision. Discuss both direction and magnitude of any potential bias 10 The strengths and limitations of the study are discussed, including the ecological nature of the analysis and potential biases in mortality data. Interpretation 20 Give a cautious overall interpretation of results considering objectives, limitations, multiplicity of analyses, results from similar studies, and other relevant evidence 8-11 The findings are interpreted in the context of existing literature and the study's objectives. Generalisability 21 Discuss the generalisability (external validity) of the study results 10 The generalizability of the findings to other populations is discussed Other information 24 Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on which the present article is based 11 Funding: None *Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies. Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting. The STROBE checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/, Annals of Internal Medicine at http://www.annals.org/, and Epidemiology at http://www.epidem.com/). Information on the STROBE Initiative is available at www.strobe-statement.org.