Artigos

Introduction
Syphilis is caused by the Treponema pallidum bacterium, renowned for its invasiveness, it can be transmitted via sexual exposure or vertically from mother to child during pregnancy (in utero infection).1. The World Health Organization (WHO) estimated in 2016, an incidence of 376.4 million cases of curable sexually transmitted infectious (STI) in individuals aged 15 to 49 years, with syphilis being one of the main diseases, along with chlamydia, gonorrhea, and trichomoniasis, with a global prevalence of 0.5% 2.
Although studies show that the prevalence of syphilis has been decreasing globally over the last three decades, when analyzed by WHO region, the Region of the Americas shows increases from 0.7% among women in 2012 to 0.92% in 2016; and 0.7% to 0.91% among men in 2012 and 2016, respectively3. In Brazil, it has been mandatory to report acquired syphilis since 2011, with the southeast region presenting, from 2012 to 2021, an average incidence rate of 59.6 per 100000 inhabitants, second only to the south region (77.5 per 100000 inhabitants for the same period)4.
In September 2016, member countries of the Pan American Health Organization (PAHO) approved the Action Plan for the prevention and control of HIV infection and sexually transmitted infections, 2016-2021. The objective is to promote an accelerated, focused, more effective, innovative, and sustainable response, paving the way to the goal of ending the AIDS epidemic and sexually transmitted infections as public health problems in the Region of the Americas by 20305.
Syphilis is one of the STI that has a cure, in Brazil, since the health system is public, it is possible to take rapid tests and treatment for syphilis and other sexually transmitted infections such as HIV and Hepatitis at the health unit. If not treated in time, syphilis has serious health consequences and can even be fatal, especially in the tertiary phase of the disease. The treatment, also offered by the public network, is carried out with the antibiotic benzathine penicillin, a safe medicine. For the infection to be completely eliminated, the patient cannot abandon treatment before its end4.
Geoepidemiology has contributed to the geographic assessment of STIs and their possible associated factors6. Determinants associated with syphilis include social, health, and behavioral factors. Several studies address the influence of the social determinants of health, such as access to health services, knowledge about the disease, unprotected sexual relations, and demographic characteristics, such as age, race and the beginning of sexual life7,8. The variables related to the health/disease process include social and economic status, education, employment, housing, among others. Hence, studying the social determinants of health (SDH) associated with syphilis is important, especially in countries characterized by large economic and health disparities, such as Brazil9.
The Human Development Index (HDI), measured in three basic dimensions (income, education, and health), can influence the incidence of many diseases10. Likewise, the Gini Index measures the degree of economic inequality, which may reflect inequality in access to health services11. Rapid testing and primary treatment of syphilis, which are important control actions, take place within the scope of Primary Health Care (PHC), highlighting the importance of adequate coverage in health territories12,13.
The importance of analyzing the temporal trend of acquired syphilis in Minas Gerais (MG) arises from the scarcity of studies concerning this condition14. The state had an acquired syphilis incidence rate of 73.3 cases per 100,000 inhabitants in 20224. Only a few studies have investigated the spatial or temporal distribution of congenital syphilis in the state14, 15. Studying the spatial-temporal distribution of acquired syphilis is necessary to guide the development of actions to eliminate the disease, as it is a surveillance tool capable of contributing to the establishment of action guidelines in priority geographic areas for preventive interventions16.
Given the gap in this methodological approach to this important problem in Minas Gerais, the objective of this study was to analyze the temporal and spatial heterogeneities of the incidence rates of acquired syphilis in the state, from 2011 to 2021, as well as its association with indicators that reflect the socioeconomic reality and access to health throughout the state. This being the first study to use this approach with acquired syphilis.


Methods
Study region
This is a population-based ecological study that addresses the temporal trend and spatial distribution of the occurrence of acquired syphilis in the municipalities of Minas Gerais, reported between 2011 and 2021, in addition to an evaluation of the influence of contextual factors on the observed disparities.
Minas Gerais is one of the 27 states in Brazil, located in the Southeast region of the country, with an area of 586,521.24 km2, ranking fourth among the largest states in terms of territorial size17. It is the second most populous state in Brazil, with an estimated population of 21 million inhabitants in 2022 and the largest number of municipalities, according to IBGE17.
The João Pinheiro Foundation (FJP), a state agency linked to the State Secretariat for Planning and Management of the State of Minas Gerais, conducts research and, in collaboration with other administrative units proposes state actions through public policies to support state agencies. In 1992, the FJP established the Planning Regions (RP), considering IBGE's regionalization into meso and micro-regions and urban influence areas, grouping municipalities with similar socioeconomic characteristics.
The units of analysis were the 853 municipalities grouped into 10 Planning Regions (PRs), according to the João Pinheiro Foundation (Fundação João Pinheiro – FJP), (number of municipalities), Figure 1a: Alto Paranaíba (31 municipalities), Central (157), Centro-Oeste (56), Jequitinhonha/Mucuri (66), Mata (143), Noroeste (18), Norte (90), Rio Doce (102), Sul de Minas (155), and Triângulo (35).

Data Sources

Data at the municipal level and by year of report of confirmed cases of acquired syphilis were obtained from the Health Surveillance Website of the Minas Gerais State Department of Health (Secretaria Estadual de Saúde de Minas Gerais – SES-MG), through the health information tabulator (TABNET, DATASUS technology) in the field of compulsory reporting conditions (general individual report). The study included cases reported and confirmed, through aggregated data according to the scales of year and municipality of residence, of syphilis acquired in the 853 municipalities of the state of Minas Gerais, from 2011 to 2021. The year 2011 was the first full year in which reporting of acquired syphilis became mandatory (mandatory from mid-2010) and the data were extracted in May 2022.
Crude incidence rates were calculated according to recommendations, in which the number of cases of acquired syphilis, in a given year of diagnosis and place of residence, was divided by the total population of individuals residing in the same place in the same year. Other variables extracted from Health Surveillance Website of the Minas Gerais were the number of confirmed reported cases, age range 10 years to over 80 years race/color, education, and sex, according to the municipality of residence and year of report, as well as the estimated resident population according to year/municipality.
Data from Family Health Strategy (FHS) coverage were obtained from the Ministry of Health database, e-Gestor, using the annual average coverage for each municipality. Human Development Index (HDI) and Gini Index data, according to the 2010 census, were obtained from the Atlas of Human Development in Brazil (Atlas do Desenvolvimento Humano no Brasil – PNUD). The cartographic base was used to formulate the maps, that is, the digital mesh made available by the Brazilian Institute of Geography and Statistics (IBGE), in the State of Minas Gerais, subdivided into its 853 municipalities.

Statistical Analysis

The incidence rates were age-standardized, using the direct method, seeking to eliminate the effects of the diversity of the age structure, which could lead to a bias in the estimates. To calculate the age-standardized rates, the age structure of the population residing in Minas Gerais, according to the 2010 census, was used. For the descriptive analyses, the crude and age-standardized rates of cases of acquired syphilis were calculated.
The spatial units of analysis were the state's municipalities of residence. Maps were constructed annually and during the analysis period of age-standardized and crude rates. The possible spatial dependence between age-standardized incidence rates was assessed using the Global Moran Indices and Local Moran Spatial Association Indicators (LISA)18. The municipalities were considered neighbors due to their shared border, using the Queen neighborhood matrix.
To detect the areas with the highest, lowest, and intermediate values of acquired syphilis incidence rates, maps were formulated with indicators obtained from the Moran scatter diagram (BoxMap). The indices are classified according to the quadrants of scatter plots between the standardized values of the rates in each area and the average of the standardized values of the rates of its neighbors. The quadrant whose municipalities have rates above the state average, just as their neighbors have an average above the expected, receives the classification of high/high, that is, both the municipalities and their neighbors have high incidence rates; in quadrant two (low/low), both municipalities and their neighbors are classified with low incidence rates; in quadrant three (high/low), the municipalities are classified with high rates and the neighbors with low rates; and in quadrant four (low/high), municipalities are classified with low rates and neighbors with high incidence rates19,20.
The Moran Index, which ranges from -1 to +1, is a measure of spatial autocorrelation, which assesses whether or not similar values are geographically close to each other in a geographic data set. Values close to +1 indicate positive spatial autocorrelation, whereas the opposite occurs with values close to -120. The univariate Moran Index is used to evaluate the spatial autocorrelation of a single variable in a set of geographic data21, while the bivariate Moran Index is an extension of the univariate index and evaluates the spatial autocorrelation of two variables simultaneously22.
The change over time was evaluated using generalized additive models (GAM Models), for each outcome (acquired syphilis), with a probability distribution of case counts in each year of reporting, using Poisson, and the offset being the natural logarithm of the at-risk population (resident population in each year).
Generalized Additive Models (GAMs) are a simplification of generalized linear models (GLMs), where a non-parametric smoothing function is applied to a term in the linear predictor, such as the year of reporting. Such a function is built based on points close to a given value (neighborhood). The smoothed curve then allows us to describe the shape, and even reveal possible non-linearities in the relationships studied, since it does not present the rigid structure of a parametric function23,24.
A smooth function (spline) was applied to the reporting year. The result and interpretation of the GAM model was obtained by graph analysis, enabling the evaluation of changes in rates and the assessment of the risk of occurrence over time. An interaction term between the PR of residence and the year in which annual incidence rates were reported, was also evaluated.
For the identification of spatiotemporal clusters, the data were analyzed using the Software for the Spatial, Temporal, and Space-time Scan Statistics (SaTScan), version 9.4.4, available at http://www.satscan.org25.
A conditional autoregressive model (CAR Model) was adjusted to estimate the risks of acquired syphilis incidence in the municipalities of Minas Gerais, from 2017 to 2019, compared to the risk throughout the state, after adjusting for spatial dependence. The response variable was the observed number of cases in the period in each municipality, with Poisson probability distribution. The linear predictor was formed by the sum of an offset term and structured and unstructured random spatial terms26. The term offset was defined as the natural logarithm of the number of cases expected in each municipality during the period, obtained by multiplying the age-standardized syphilis incidence rate with the average population exposed during the period of each municipality.
With the hypothesis that syphilis incidence rates among neighboring municipalities are more similar than rates among more distant municipalities, the spatially structured term was defined with a CAR prior with Gaussian distribution. In turn, the spatially unstructured term is related to unobserved variables, which are independent and identically distributed, with Gaussian priors of zero mean and major variance (prioris). The estimated risks were considered significant if the probability mass was consistently greater than zero; in this work, the probability of the risk being greater than 1 (one), greater than 0.70 was considered. Effects (risks) were adjusted using a Bayesian approach and parameter inference was based on the Integrated Nested Laplace Approximations (INLA) deterministic method27.
After extraction, the official databases were stored and organized in Microsoft Excel 2019. The analyses were carried out using the R statistical software28, through the INLA, mgcv, ggplot2, tidyverse, and ggthemes packages. The thematic maps were built in the QGIS software version 3.2829, using the cartographic base (digital mesh) of Minas Gerais (data gathered from IBGE). The spatial autocorrelation maps were produced using the GeoDa software, version 1.1830.
Ethical considerations
This study used secondary data. All information presented is available in the public domain, with no identification of participants, as the data is aggregated by municipality. Therefore, it was not necessary to submit the study to the Research Ethics Committee, according to the National Health Council’s Resolution No. 510, dated April 7, 2016.

Results
From 2011 to 2021, 83,432 confirmed cases of acquired syphilis were reported, with the majority being male (64.82%), pardo (44.55%), with complete secondary education (16.64%), and in the age range of 20 to 29 years (36.92%). It is important to highlight the high number of reports with the educational level completed as ignored (42.51%).
The comparison between crude and standardized rates showed no major difference over time, as shown in Figure 1 (b). There was a sharp increase in the incidence rate until 2019 (68.96/100,000 inhabitants), a decrease in 2020 (56.96/100,000 inhabitants), and a subsequent increase back to the same previous level (70.77/100,000 inhabitants). The annual distribution of incidence rates by Planning Region (Figure 2 - c) shows that the Central region had the highest rates. Among the other regions, the Triangle, Alto Paranaíba, and Centro-Oeste also stood out. The regions that presented the lowest rates were the Norte, Noroeste, and Jequitinhonha- Mucuri.

Fig. 1

The estimated smoothed curve for the risk of syphilis occurring over time, obtained through the Generalized Additive Model (GAM), using the Poisson distribution for the response variable number of cases, presented a better fit, compared to the Negative Binomial. It was observed that, from 2015 onwards, the point of intersection of the horizontal red line that represents no effect or risk (Figure 2 - a), a more pronounced increase was found in the risk of occurrence of acquired syphilis in the state of Minas Gerais. When analyzing by PR, most regions showed similar growth from 2015 onwards, while others showed a downward trend from 2020 onwards, such as the Norte and Noroeste regions (Figure 2b - center and bottom). According to the estimates of the fixed effect of PR, significant differences were found in all




PRs for the Central region (PR reference), with the Norte region showing the lowest risk. The estimated smoothed curve for the risk of syphilis occurring over time, obtained through the Generalized Additive Model (GAM), using the Poisson distribution for the response variable number of cases, presented a better fit, compared to the Negative Binomial. It was observed that, from 2015 onwards, the point of intersection of the horizontal red line that represents no effect or risk (Figure 2 - a), a more pronounced increase was found in the risk of occurrence of acquired syphilis in the state of Minas Gerais. When analyzing by PR, most regions showed similar growth from 2015 onwards, while others showed a downward trend from 2020 onwards, such as the Norte and Noroeste regions (Figure 2b - center and bottom). According to the estimates of the fixed effect of PR, significant differences were found in all PRs for the Central region (PR reference), with the Norte region showing the lowest risk.

Figure 2: Smoothed curves of the effect of time on the risk of the occurrence of syphilis according to purely temporal GAM models for the entire State (a) and considering an interaction term with the Planning Region (b- center and bottom), Minas Gerais, 2011 to 2020.

Fig. 2

The analysis of the spatial distributions of annual incidence rates between municipalities (Figure 3) showed, in general, that incidences increased in magnitude and were dispersed throughout the state during the period. A low incidence of acquired syphilis cases was also observed during the study period, especially in the municipalities of the Norte region.
It is possible to observe that there was an increase in the incidence of reported and
confirmed cases of acquired syphilis, beginning in the Central region and spreading throughout the state over the years. In 2020, there was a decrease in incidence, mainly in the Norte region, as identified by the blank areas on the map.



Figure 3: Spatial distribution of annual incidence rates of acquired syphilis in Minas Gerais, from 2011 a 2021.

Fig. 1

When analyzing the distributions of HDI and Gini index between the PRs in 2010, it was observed that there is a concentration of municipalities with higher HDI in the Central region and surrounding areas, whereas the Gini index proved to be more predominant in the Norteern region of the map (Figure 4a and Figure 4b). Next, there is a spatial overview of the distributions of the average FHS coverage in Minas Gerais from 2011 to 2021. One can see that the Central, Alto Paranaíba, Noroeste, and Triângulo regions have areas with lower coverage than the others. The Spearman correlation showed a correlation only between the FHS incidence
and coverage (rho=-0,250, p<0,001); no significant correlation (p<0,05) was found for HDI and Gini.
Based on the established correlation, it was observed that the Moran Bivariate (Figure 4f) showed many areas in the Norte region with a low/high ratio, low incidence and high coverage of FHS. While in the Central and Triangle regions, some regions showed a high/low ratio, high incidence and low coverage.
The analysis of spatiotemporal clusters with a high risk of syphilis occurrence was conducted using the Satscan software, considering scanning circle radii of up to 50% of the total population shows a cluster primarily involving the Central, Mata, and Central-West regions; for the only significant cluster, the estimated relative risk was 3.80, in the time interval from 2017 to 2021 (Figure 5b). The municipalities that presented a risk greater than one, considering the period from 2017 to 2019, were: Betim, Contagem, Ibirité, Ribeirão das Neves, Santa Luzia, and Sete Lagoas, all in the Central Region (Figure 5d).

FRig. 5

Discussion

To the best of our knowledge, this is the first study that analyzed the temporal and spatial heterogeneity of the incidence rates of acquired syphilis in the state of Minas Gerais. The results showed a predominance of male cases, pardo race, high school education, and age range from 20 to 29 years. Incidence rates in the trend analysis since 2011 showed a sharp increase in 2015, followed by a drop in 2020. The spatial distribution, considering the planning regions, showed that the Central, Alto Paranaíba, Centro-Oeste, and Triangle regions presented the highest incidence rates, with an estimated risk greater than one in seven municipalities in the Central region.
The Norte and Noroeste regions showed the lowest incidence in the assessed period, and the risk of incidence for syphilis for these regions was also lower when compared to the Central region. A correlation was found between lower incidence rates of acquired syphilis and higher FHS coverage, according to the bivariate Spearman and Moran correlation. The other indicators, HDI and the Gini Index, did not correlate with the incidence rates of acquired syphilis.
In Brazil, from 2011 to 2021, most reported cases of acquired syphilis were concentrated in males (60.6%) and in the 20 to 29 age group (35.6%), with complete secondary education (20.14%) and pardo (36.40%)4. In general, the results found in relation to demographic and educational stratifications in this study are in line with information published for the country4 (Epidemiological Bulletin by the Ministry of Health - MS (2023)). In this study and in the Ministry's data, the number of fields ignored for educational level (missing) was high, outweighing the fields with information.
Minas Gerais, as shown in Figure 1b, followed the national trend. The increase in the incidence of acquired syphilis in the states of the country, particularly between 2012 and 2018, is related to factors such as access to rapid testing, misinformation, decreased use of condoms, reduced use of benzathine penicillin in primary healthcare, and shortages of the drug31.
In Minas Gerais, the implementation of rapid tests began in 2016 through the Joint Technical Note of 01/2016, starting with some municipalities, and expanding to all municipalities as of 2018. In 2021, of the 853 municipalities, 516 (60, 5%) were in the process of training and implementing rapid testing, 78 (9.1%) of municipalities had not started the process, and 259 (30.4%) of the municipalities had already completed training to conduct testing in Primary Care32. Therefore, the implementation of rapid tests may have contributed to the increase in the number of notifications in recent years.
The coronavirus disease 2019 (COVID-19) has had a tremendous impact on health and economic systems around the world, exposing serious flaws in health systems, and producing a greater impact related to poverty, hitting the marginalized population hardest 31,33.
In the Norteern states of Brazil (Amazonas, Pará, Acre, Roraima, Rondônia, Amapá, and Tocantins), there was a significant drop in reports, nearly 60%, for acquired, gestational, and congenital syphilis, with a significant drop in acquired syphilis reports in all states evaluated in 202034.
Brazil presented a drop of approximately 33% in diagnostic and treatment procedures for syphilis in the first seven months of 2020, the year the pandemic began, as compared to the average for the first seven months of the previous four years35. In Minas Gerais, the drops in incidence were approximately 20% between 2019 and 2020. Considering the confirmed cases, the present study showed a 12% decline for acquired syphilis in the first-year pandemic.
One study emphasized regional differences in Minas Gerais, with emphasis on the Norte and Jequitinhonha-Mucuri PRs, mainly in relation to the Central region. These differences have been mounting over the years, as these regions have increased their contribution in poverty and population vulnerability indicators, based on specific indices, and contribute little to the GDP of Minas Gerais36.
It is known that STIs have a close relationship with social inequalities, as seen a review showing that poverty and conditions that make the population even more vulnerable are significantly associated with the occurrence of the disease, especially syphilis in pregnant women7.
It is also important to consider that, even under the hypothesis of underreporting in regions that demonstrated low incidence, as shown by a study by Souto-Marchand37, diseases that are traditionally linked to poverty, such as tuberculosis, leprosy, dengue, congenital syphilis, and Chagas, there were more incidents in municipalities with lower poverty rates. In other words, the highest rates of incidence of these diseases were found in cities that have a higher demographic density, a higher HDI, better infrastructure and urban support, and a lower percentage of extreme poverty, but which still present significant social inequality in its territory.
Although our study did not identify a direct correlation with the Gini Index, some studies have demonstrated a significant positive correlation with the incidence of syphilis, particularly gestational syphilis, suggesting that greater income inequalities are associated with higher disease incidence38. However, the Moran map revealed that municipalities in the northern region have high Gini Index values but low incidence rates, while also showing high primary care coverage. This suggests the protective role of primary care in mitigating the impact of social inequality.
The importance of the FHS in detecting and controlling diseases and health promotion actions is clear. In the study by Nunes et al.39, in the municipalities of Goiás, an inverse relationship was also detected between FHS coverage and incidence, since a significant increase was observed in cases of syphilis (congenital and in pregnant women) in municipalities presenting FHS coverage percentages of lower than 75% (p<0.001), which suggests failures in preventing the vertical transmission of the disease, especially in regions with lower FHS coverage.
In our study, this inverse relationship between incidence and coverage was also evidenced. The Central region, in turn, showed a greater number of areas with low coverage and high incidence, as it is a region with a greater flow of people, goods, and services, and is associated with a lower PHC coverage, which may be one of the causes that can explain the higher incidence in this region, even though it has a higher average HDI than the Norte and Jequitinhonha-Mucuri regions.
The Norte region presented a high FHS coverage and a low incidence of acquired syphilis, which suggests that, in this region, high primary care coverage may have contributed to the low incidence of the disease, since early detection and treatment carried out in health units would be responsible for breaking the transmission chain. It is necessary to improve federal and state programs, as well as funding, that encourage the expansion of primary care coverage by municipalities, aiming to reduce these regional disparities.
The Family Health Strategy (FHS) plays a fundamental role in promoting health and preventing diseases in Minas Gerais. Studies indicate that the FHS has contributed to reducing inequalities in access to healthcare services, especially among low-income populations. For example, research has shown that poorer households have higher visitation rates by FHS teams, demonstrating equity in the coverage of these services in the state40.
In addition, a study that evaluated the services offered by primary care for testing and treatment for syphilis, showed that the group of municipalities with the greatest reduction in vertical transmission showed a higher median of rapid tests and benzathine penicillin, which corroborates the perception that increasing access to primary care is an important strategy to face the syphilis epidemic in the country 41.
As demonstrated by a study by Santos et al.42 larger cities, with better HDIs, do not offer better equity in health care and care related to syphilis. PHC services tend to be more commonly used by people with greater social vulnerability, suggesting that cities with better HDIs show a greater disorganization in relation to syphilis control policies and that this relationship should be further investigated.
It is important to mention that the timely approach to diagnoses and the application of benzathine penicillin in PHC are currently the main strategies for combating the Syphilis epidemic in an attempt to break the transmission chain of this important disease.
One-third of the population in the Americas Region continues to face barriers hindering access to healthcare, with significant disparities in the availability of health personnel and access to essential medicines and technologies. The Pan American Health Organization (PAHO) emphasized the need to prioritize primary health care (PHC) to advance toward universal health coverage in the Americas43.
The fragility of primary health care, linked to population size, may have favored the growth of the acquired syphilis epidemic in Brazilian municipalities, and the increase in acquired syphilis trends was associated with partial coverage of primary health care teams42.
Therefore, the present study’s results suggest that, although Minas Gerais presents regional disparities regarding the distribution of syphilis, socioeconomic indicators, and access to health, it is important that the territories be adequately covered from a medical care point of view, especially with regard to basic care, being able to mitigate the impact of disease transmission in territories that have low economic and social development.
One limitation of this study is the fact that the acquired syphilis report form is a report/conclusion form used for various diseases; therefore, it is not specific to acquired syphilis, as is the case with forms for congenital syphilis and in pregnant women, which limits access to information on testing, or other data related to the diagnosis. Therefore, the lack of a specific form with specific and mandatory fields may have contributed to a high number of fields completed as ‘ignored’. Furthermore, the data used in this study refers to the data that was collected and confirmed. As such, there may have been a much larger number of ignored fields when considering the total number of reports of confirmed and unconfirmed cases. Another limitation is possible regional differences in coverage, scope and quality of information may have impacted the results found.
Regarding spatial analysis, as proposed by Lawson, the possible influence of the edge effect on the results found cannot be ruled out, considering the delimited planning regions, since these effects occur when the study area is delimited by administrative or geographic borders that do not impede the flow or movement of people beyond these limits to access resources or services44.
Despite the limitations listed, the results of this study are essential for understanding the behavior of acquired syphilis in Minas Gerais, in view of the scarcity of studies on this disease, especially when spatial analysis is applied, contributing to the future formulation of policies aimed at reducing the number of cases of acquired syphilis in Minas Gerais, in view of its regional peculiarities and access to basic health services.





References
1. World Health Organization (WHO). Global guidance on criteria and processes for validation: elimination of mother-to-child transmission of HIV, syphilis and hepatitis B virus. Geneva: World Health Organization, 2021. Disponível em: https://www.who.int/publications/i/item/9789240039360. Acesso em: 12 jun. 2024
2. World Health Organization (WHO). Guidelines for the Treatment of Treponema pallidum (syphilis). Geneva: WHO, 2016.
3. Pan American Health Organization, Pan American Health Organization. Epidemiological Review of Syphilis in the Americas, December 2021 [Internet]. Washington, D.C.: PAHO; 2022. Disponível em: https://iris.paho.org/handle/10665.2/56085.
4. Brasil. Ministério da Saúde (MS). Boletim Epidemiológico de Sífilis. Número Especial | Out. 2023 - versão eletrônica.
5. Pan American Health Organization (PAHO). Plan of Action for the prevention and control of HIV and sexually transmitted infections 2016 – 2021. Washington, D.C., 2017.
6. Rodrigues-Júnior AL, Ruffino-Netto A, Castilho EAD. Spatial distribution of the human development index, HIV infection and AIDS-Tuberculosis comorbidity: Brazil, 1982 - 2007. Rev Bras Epidemiol 2014; 17(suppl 2):204-215.
7. Costa MAC. Determinantes sociais da sífilis no Brasil - Uma revisão de literatura [dissertação]. Palmas: Universidade Federal do Tocantins, Programa de Pós-Graduação em Ciências da Saúde; 2019.
8. Mota GS. Determinantes sociais da saúde e fatores de risco para sífilis em mulheres de uma comunidade rural [dissertação]. Salvador: Escola de Enfermagem, Universidade Federal da Bahia; 2021.
9. Paiva S de S, Pedrosa NL, Galvão MTG. Análise espacial da AIDS e os determinantes sociais de saúde. Rev Bras Epidemiol 2019; 22:e190032.
10. Programa das Nações Unidas Para o Desenvolvimento (PNUD). Atlas do desenvolvimento humano no Brasil Brasília: 2023
11. Barreto ML. Desigualdades em Saúde: uma perspectiva global. Cien Saude Colet 2017; 22(7):2097-2108.
12. Roncalli AG, Rosendo TMSDS, Santos MMD, Lopes AKB, Lima KCDE. Efeito da cobertura de testes rápidos na atenção básica sobre a sífilis em gestantes no Brasil. Rev de Saúde Pública 2021; 55:94.
13. Figueiredo DCMM, Figueiredo AMF, Souza TKB, Tavares G, Vianna RPT. Relação entre oferta de diagnóstico e tratamento da sífilis na atenção básica sobre a incidência de sífilis gestacional e congênita. Cad Saude Publica 2020; 36(3):e00074519.
14. Alves PIC, Scatena LM, Haas VJ, Castro S de S. Evolução temporal e caracterização dos casos de sífilis congênita em Minas Gerais, Brasil, 2007-2015. Cien Saude Colet 2020; 25(8):2949-2960.
15. Amorim EKR, Matozinhos FP, Araújo LA, Silva TPR da. Tendência dos casos de sífilis gestacional e congênita em Minas Gerais, 2009-2019: um estudo ecológico. Epidem Serv Saúde 2021; 30(4):e2021128.
16. Ramos R de S P da S, Ramos VP. Análise espacial como ferramenta de identificação de áreas prioritárias de intervenção para prevenção da sífilis. Cien Saude Colet 2021; 26(suppl 2):3733-3742.
17. Instituto Brasileiro de Geografia e Estatística (IBGE). Malhas Territorias, 2023.
18. Bailey T, Gatrell AC. Interactive Spatial Data Analysis. Longman Group Limited, England, 1995.
19. Teixeira LO, Belarmino V, Gonçalves CV, Mendoza-Sassi RA. Tendência temporal e distribuição espacial da sífilis congênita no estado do Rio Grande do Sul entre 2001 e 2012. Cien Saude Colet 2018; 23(8):2587-2597.
20. Cliff AD, Ord JK. Spatial processes: Models and applications. Pion Limited, London, 1981.
21. Anselin L. Local indicators of spatial association - LISA. Geographical Analysis 1995; 27:91-115.
22. Getis A, Ord JK. A bivariate Moran's I for evaluating the colocation of arbitrarily shaped areas. J Reg Sci 1992; 32(4):443-459.
23. Latorre MRDO, Cardoso MRA. Análise de séries temporais em epidemiologia. Rev Bras Epidemiol 2001; 4(3).
24. Conceição GMS, Saldiva PHN, Singer JM. Modelos GLM e GAM para análise da associação. Rev Bras Epidemiol 2001; 4(3).
25. SaTScan™. Software for the spatial, temporal and space-time scan statistics. SaTScan v9.4.4 [Internet]. 2023.
26. Besag J, York J, Mollié A. Bayesian image restoration, with two applications in spatial statistics. Ann Inst Stat Math 1991; 43:1-20
27. Rue H, Martino S, Chopin N. Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations. J R Stat Soc 2009; 71:319-392.
28. R: A language and environment for statistical computing [computer program]. Vol. 2018. Vienna, Austria: R Foundation for Statistical Computing; 2021. Retrieved from https://www.R-project.org/.
29. QGIS [computer program]. Version 3.28. Developed by QGIS Development Team. Open Source Geospatial Foundation; 2021. Disponível em: https://www.qgis.org/.
30. GeoDa [computer program]. Version 1.18. Urbana, IL: University of Illinois, Urbana-Champaign; 2012.
31. Freitas CHS de M, Forte FDS, Galvão MHR, Coelho AA, Roncalli AG, Dias SMF. Inequalities in access to HIV and syphilis tests in prenatal care in Brazil. Cad Saúde Pública [Internet]. 2019;35(6):e00170918.
32. Minas Gerais. Secretaria Estado de Saúde de Minas Gerais. Nota Técnica Conjunta SVEAST/SRAS/SAPS/SES-MG. N 01/2016. Implantação dos testes rápidos nas Unidades Básicas de Saúde de Minas Gerais.
33. Ehrenberg N, Ehrenberg JP, Fontes G, Gyapong M, Rocha EMM, Steinmann P, Utzinger J, Zhou X-N, de Savigny D. Neglected tropical diseases as a barometer for progress in health systems in times of COVID-19. BMJ Global Health 2021; 6(4):e004709.
34. Brito CVB, Formigosa C de AC, Neto OSM. Impacto da COVID-19 em doenças de notificação compulsória no Norte do Brasil. Rev Bras Prom Saúde 2022; 35:1-11.
35. Furlam TO, Pereira CCA, Frio GS, Machado CJ. Efeito colateral da pandemia de Covid-19 no Brasil sobre o número de procedimentos diagnósticos e de tratamento da sífilis. Rev Bras Estudos População 2022; 39:1-15. e0184
36. Santos GR, Pales RC. Estratégias de desenvolvimento em Minas Gerais: uma análise comparada das macrorregiões de planejamento. Cad do Desenvolvimento, Rio de Janeiro 2014; 9(14):163-185.
37. Souto-Marchand AS. Doenças infecciosas e suas correlações com indicadores socioeconômicos e demográficos: estudo ecológico em diferentes estados brasileiros [tese]. Rio de Janeiro: Instituto Oswaldo Cruz, Fundação Oswaldo Cruz; 2017.
38. Santos RS, Leite IFC, Barros DCO. Desigualdades no acesso a testes de HIV e sífilis durante o pré-natal no Brasil: análise de fatores individuais e contextuais. Cad Saúde Pública [Internet]. 2019 [cited 2025 Jan 6];35(7):e00099418. Available from: https://cadernos.ensp.fiocruz.br/ojs/index.php/csp/article/view/7034/15329
39. Nunes PS, Guimarães RA, Rosado LEP, Marinho TA, Aquino EC, Turchi MD. Tendência temporal e distribuição espacial da sífilis gestacional e congênita em Goiás, 2007-2017: um estudo ecológico. Epidemiol Serv Saude 2021; 30(1):e2019371
40. Andrade MV, Noronha K, Barbosa ACQ, Rocha TAH, Silva NC da, Calazans JA, et al. A equidade na cobertura da Estratégia Saúde da Família em Minas Gerais, Brasil. Cad Saúde Pública [Internet]. 2015Jun;31(6):1175–87.
41. Figueiredo DCMM de, Figueiredo AM de, Souza TKB de, Tavares G, Vianna RP de T. Relação entre oferta de diagnóstico e tratamento da sífilis na atenção básica sobre a incidência de sífilis gestacional e congênita. Cad Saúde Pública [Internet]. 2020;36(3):e00074519. Available from: https://doi.org/10.1590/0102-311X00074519
42. Santos MMD, Rosendo TMSDS, Lopes AKB, Roncalli AG, Lima KCDE. Weaknesses in primary health care favor the growth of acquired syphilis. PLoS Negl Trop Dis 2021; 15(2):e0009085.
43. Organización Panamericana de la Salud. Salud universal en el siglo XXI: 40 años de Alma-Ata. Resumen Ejecutivo [Internet]. Washington, DC: OPS; 2019 [citado em 6 jan.2025].Disponível:https://iris.paho.org/bitstream/handle/10665.2/63087/9789275329474_spa.pdf?sequence=1&isAllowed=y.
44. Lawson, AB. Statistical Methods in Spatial Epidemiology. John Wiley & Sons, 2013.