Introduction

Chronic Pancreatitis (CP) is a syndromic entity with diverse clinical manifestations associated with potentially debilitating and severe outcomes, including diabetes mellitus, exocrine insufficiency, and pancreatic cancer [1]. The progressive nature of the inflammatory process underlying CP causes significant clinical damage and requires prolonged follow-up, recurrent hospitalization, and complex surgical approaches [2]. Therefore, the long-term consequences of CP invariably result in poor quality of life and significant socioeconomic burden [3, 4].

In the United States, benign pancreatic diseases, including CP, resulted in an annual cost of 4.9 billion in 2018 [5]. However, although the total number of hospital discharges and the length of hospital stay with a primary diagnosis of CP decreased considerably in the United States based on a nationwide database, the cost associated with these hospitalizations almost tripled between 1997 and 2014 [6]. Although epidemiological studies concerning CP are scarce, it is important to call attention to the fact that the burden of chronic noncommunicable diseases can become particularly heavy on low and middle-income countries, as their prevalence continue to increase [7, 8]. Developing and highly populated countries are associated with social inequities and environmental and societal changes that can affect disease development and the associated outcomes [9]. In fact, inequalities in healthcare access are present even in high-income nations, as exposed to the recent COVID-19 pandemic [10].

Most studies on CP have been case series, and only a few epidemiological studies have been published to date. The highest age-standardized incidence rates have been reported in the USA, Canada, and Greenland, whereas Djibouti, Madagascar, and South Sudan have the lowest incidence rates [11]. Nevertheless, considering the chronic course of the disease, longitudinal large-scale epidemiological studies on CP are essential to determine the actual incidence and prevalence rates. In addition, analysis of temporal trends may provide information on the relationships between risk factors, regional characteristics, urbanization patterns, and clusters of people affected by CP, as well as potential new factors [12, 13].

Establishing the epidemiology of CP and associated mortality rates can be challenging. Substantiating the definitive diagnosis of CP may take years owing to the diverse clinical presentations and variable course of disease progression [14]. However, a relatively recent proposal for defining CP based on pathogenic mechanisms reinforces the importance of discovering complex gene-environment interactions in the etiopathogenesis of CP to provide insights into individual patients [15]. CP is usually associated with exposure to environmental and behavioral risks such as alcohol consumption and tobacco smoking [4]. Nevertheless, alcohol consumption and smoking are relevant risk factors for several diseases and deaths, in addition to representing leading factors related to long-term disability worldwide [16].

Currently, limited information is available concerning CP in Brazil, the 5th largest country in the world and a continental country with more than 5500 municipalities spread across five distinct macroregions. A better understanding of the epidemiology of CP in Brazil can reveal important aspects of the complex relationships between different patterns of urbanization, geographic diversity, and inequalities within the country, which may also be present in other parts of the world. Therefore, this ecological study aimed to analyze the geographic distribution and time trends of CP in Brazil during an eleven-year study from 2009 to 2019. In particular, different Brazilian macroregions were analyzed, considering the attributable burden of CP by age group, sex, and possible associated risk factors, most notably alcohol and tobacco. Since social disparities and health conditions are intrinsically linked and are a matter of great concern worldwide, descriptive data about the demographics of populations, their patterns of associations; their relationship with social indicators were included in this study, as they may contribute to identifying population health risks.

Methods

Ethical Considerations

Ethical approval was waived by the local Ethics Committee of the University Hospital Clementino Fraga Filho of the Federal University of Rio de Janeiro, given the retrospective nature of the study, which investigated an anonymously archived population-based health and disease registry.

Data Source

Data from the Health Informatics Department of the Brazilian Ministry of Health (DATASUS) (http://www2.datasus.gov.br/DATASUS) were retrospectively retrieved from the hospitalization records of patients with the primary diagnosis of chronic pancreatitis (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code of EC, ICD-10 codes: K86.0 and K86.1) upon hospital admission. DATASUS is a high-quality population-based open-access registry system that provides information about the Brazilian Unified Health System (SUS/Sistema Único de Saúde), including hospital admissions, surgical procedures and discharges, in-hospital mortality, and demographic variables. In-hospital lethality was considered as death due to any reason during the hospital stay. DATASUS is present in all municipalities (the smallest administrative territorial units) of the country through its regional offices and is responsible for defining the standards for capturing and transferring health information from the SUS state and municipal managers to the unified databank of the Brazilian Ministry of Health. As DATASUS covers approximately the entire Brazilian population, we assumed that this registry reflects the actual hospitalization numbers for CP in the country.

Estimates of the resident population, demographics, and socio-economic indices were retrieved from the Brazilian Institute of Geography and Statistics (IBGE), the official government agency responsible for collecting and providing geographic, environmental, and statistical data for each of the 5570 Brazilian municipalities.

Tobacco and alcohol use data were retrieved from VIGITEL (National System of Surveillance of Chronic Diseases). VIGITEL represents an essential tool for obtaining complementary information on general health conditions of the population, anonymously archived, including smoking and alcohol use [17]. In parallel to the analysis of DATASUS-derived data, we analyzed data obtained from the VIGITEL system and noted changes in the rates of tobacco and alcohol use, during the same period of time. The platform classifies the individuals regarding tobacco and alcohol use based on the following criteria: tobacco use is considered positive for every smoker regardless of the frequency and intensity of smoking habits; alcohol abuse is considered positive according to the frequency of alcohol intake (four or more doses for women or five or more doses for men) of alcoholic beverages on the same occasion within the last 30 days [17,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Ethical approval was not required because all the data were anonymized.

Study Design, Population, and Variables

An ecological study was conducted to investigate the temporal trends of hospitalizations due to CP in Brazil by analyzing the geographic distribution of the disease, associated demographic variables, and temporal changes in the related risk factors. Ecological studies are observational investigations that allow for temporal and geographic analyses at the population level to estimate the rates of a particular disease and its temporal trends.

Data were retrospectively collected from the DATASUS platform between January 2009 and December 2019. The study period was selected on the basis of the most recent and consistently available data. The IBGE describes the Brazilian territory in five macroregions according to geographic location: North, Northeast, Central-West, Southeast, and South. The macroregions comprise 26 federated states, the Federal District and 5570 municipalities. The DATASUS data related to each municipality or macroregion were accessed. The data included age, sex, town (municipality) of residence, duration of hospitalization, and in-hospital death. Patients were analyzed according to sex and age. Age groups were stratified as follows: < 20 years, 20–49 years, 50–69 years, and 70 years. The results of the exploratory analysis of hospitalization rates per 100,000 inhabitants and the in-hospital lethality of each municipality were included in a worksheet to plot maps depicting their distribution in the country.

The analysis of each geographic region and the comparison of each state between the metropolitan and interior areas allowed a detailed study of chronic pancreatitis admissions and lethality rates in the country.

Statistical Analysis

Similar to our previous studies, the statistical analysis determined CP hospitalization rates after correcting for the total number of available hospital beds and division by regional population annually during the study period [20, 21]. The correction was performed because of the dynamic changes in the number of available hospital beds. For such purpose we used the following formula:

$${\text{Adjusted}}\;{\text{admission}}\;{\text{ rate}}\;\left( {{\text{per}}\;10^{5} } \right)\; = \;\frac{{{\text{admission}}\, {\text{rate}}\; \left( {{\text{per}}\, 10^{5} } \right)}}{{{\text{proportion}}\; {\text{of}}\; {\text{available}}\; {\text{public}}\; {\text{hospital}}\; {\text{beds}}}}.$$

Estimates of the resident population were obtained from the IBGE projections. In-hospital mortality rates were evaluated as percentages. Simple linear regression was used to estimate temporal trends in CP hospitalization and lethality by age and sex using Microsoft Excel Software (Microsoft Excel for Mac 2011, Version 14.4.9, 2010; Microsoft Corp, Redmond, Wash). Statistical analyses were performed using the statistical software package IBM SPSS for Windows (Version 20; SPSS Inc, Chicago, IL). Exploratory procedures with a quantitative approach were applied to the data using IBM SPSS software for Windows (version 20, SPSS Inc.). Descriptive statistics and graphical displays are summarized in Tabwin 3.2 (Tab for Windows 3.2, a free software that allows the organization of multiple applications into grouped tabs, available at http://www2.datasus.gov.br/DATASUS).

Results

Hospitalizations for CP

Admission rates for CP in Brazil, adjusted to available hospital beds, decreased by nearly half (− 54.68%) from 6.4 to 2.9 per 100,000 inhabitants, similarly in both sexes (− 53% in males, from 8.3 to 3.9%, − 56.5% in females, and 4.6 to 2.0) from 2009 to 2019. A total of 64,609 admissions were retrieved, with 41,059 in males (63.54%) and 23,550 in females (36.46%) (Fig. 1).

Fig. 1
figure 1

Chronic Pancreatitis (CP) adjusted admissions rates by sex in Brazil, from 2009 to 2019. Rates were adjusted according to the total available hospital beds

Full size image

Throughout the study period, hospitalization numbers were higher in patients aged 70 years or older, intermediate in those 50–69 years old, and lower in those group of 20–39 years old in both sexes. Very few cases were registered in patients < 20 years of age (data not shown). Considering sex, admission rates of CP in males were higher in all age groups, with hospitalizations in males older than 70 years showing more than twice the rates in the youngest group. Of note, male-to-female ratios concerning hospitalizations decreased steadily with aging, numbers equaling in for people older than 70 years (20–49y:2.29; 50–69y:1.87; ≥ 70y:1.19), in 2009. In 2019, however, the male-to-female hospitalization ratios showed a slight increase in patients aged between 50 and 69 years old, before decreasing in the older age group (20–49y: 2.1; 50–69y:2.45; ≥ 70y:1.27). Taking sex together and considering the study period, patients aged 70 years or older showed the most significant reduction in admissions (− 64%) compared with other age groups. Considering sex, the greatest reduction in admission rates during the study period was observed among females 70 years or older (− 65.3%). The smallest reduction in admission rates was observed in males from 50 to 69 years old (− 51.49%). The first 5-year records mainly drove this difference, since the data from the last six years (2014–2019) showed similar rates of hospitalization for men in both age groups. (Fig. 2).

Fig. 2
figure 2

Chronic Pancreatitis (CP) age-standardized admissions rates by sex in Brazil, from 2009 to 2019. A Male and B female

Full size image

Lethality Rates for CP

Discordant with the reduction in the rates of hospital admissions, the estimated in-hospital lethality of CP remained stable from 2009 to 2019, ranging between 5 and 6%. Regarding sex, the lethality was identical for males and females (5%) in 2009 and 2019. Nevertheless, the lethality rates were quite different among age groups, increasing in proportion to age in both males and females (Fig. 3).

Fig. 3
figure 3

Chronic Pancreatitis (CP) age-standardized in-hospital lethality rates by gender in Brazil, from 2009 to 2019. A Male and B female

Full size image

For male patients aged 20–49 years, lethality rates ranged between 3 and 4%, whereas lethality in those aged 50–69 years ranged from 6 to 8% during the study period. The highest mortality rates among males were registered in those aged 70 years or older, ranging from 12 to 17%. In females, lethality rates in patients aged 20–49 years ranged from 2 to 4%, whereas in those aged 50–69 years, lethality rates ranged from 4 to 6%. Similarly, the highest lethality rates in women were registered in those 70 years or older, ranging from 10 to 15% (Fig. 3). Further lethality analysis among age groups of the same sex at the end of the study revealed that the rates for both men and females older than 70 years were more than triple the rates of the youngest age groups, under 49 years (3.66 and 3.0, male and female ratios, respectively).

Geographic Distribution of CP

The geographic distribution of CP was obtained through database analysis using DATASUS. All 5565 municipalities were included, distributed in 27 federative units and gathered in five macroregions (north, northeast, southeast, south, and central–west).

All five macroregions of the country showed remarkable decreases in hospitalization numbers, especially the North and Central-West regions (− 77%). The highest decrease in admission rates was noticed in the Northeast (− 70%), followed by the South (− 59%) and the Southeast (− 42%) regions (Table 1, Supplementary Figure S1).

Table 1 Distribution of chronic pancreatitis hospitalization and lethality rates in Brazil, from 2009 to 2019, according to geographic regions
Full size table

Considering Brazil as a unit, in-hospital lethality rates remained similar from 2009 to 2019. All regions demonstrated a decrease in mortality, except for the south. Notably, the South showed the lowest lethality rates in 2009 (4.3%), but the highest in 2019 (8.3%), an increase of + 104% (Table 1, Supplementary Figure S2).

Due to the continental dimensions and the significant economic, geographic, and demographic disparities in Brazil, we utilized methods similar to those used in previous studies by our group. This model classifies municipalities according to population and demographic density as rural small (< 50,000 or < 80 inhabitants/km), rural medium (≥ 50,000 or ≥ 80 inhabitants/km, even if the population is < 50,000 inhabitants), and true urban centers (> 100,000 inhabitants). Hospital admission and lethality rate data were collected considering municipality size and its association with metropolitan regions.

Irrespective of size, all municipalities integrated into metropolitan regions showed decreased CP hospitalizations during the study. Although both integrated and not integrated municipalities showed reductions in CP admission rates, the highest decrease was noted in municipalities integrated into the metropolitan region: from 6.0 to 2.6 per 100,000 (− 56%) versus 6.3 to 3.2 per 100,000 (− 50%). With regard to demographics, large municipalities registered a higher decrease in admission numbers (− 54%) than small (− 50%) and medium-sized towns (− 45%).

Despite the CP lethality rates remaining stable in Brazil (from 5.3 to 5.4%), municipalities presented different trends. The most significant increase in CP lethality rates (+ 10%) was observed in municipalities integrated into metropolitan regions regardless of size. The most significant increase in lethality was registered in small municipalities integrated into metropolitan regions (+ 124%). However, the same did not occur in municipalities that were not integrated into metropolitan regions, which showed a decrease (− 7%). This reduction was mainly driven by the fact that large and medium-sized municipalities were not integrated into metropolitan regions, while medium-sized municipalities showed an increased rate (+ 18%) (Table 2).

Table 2 Distribution of chronic pancreatitis hospitalization and lethality rates in Brazil, from 2009 to 2019, according to urban and rural demographics data
Full size table

Considering the geographic macroregions, the South and Southeast display the country’s highest gross domestic product (GDP) and Human Development Index (HDI), in contrast to the North and Northeast, where the poorest states are located. Nevertheless, during the period analyzed, the North and Northeast had the lowest adjusted number of hospital admissions for CP, coinciding with the greatest increases in HDI and GDP in the country and the greatest reduction in smoking and alcohol consumption (Table 3).

Table 3 Alcohol, tobacco, HDI and GPD in Brazil, from 2009 to 2019, according to geographic regions
Full size table

Discussion

In the last decade, most publications focusing on CP epidemiology have originated from countries located in Western Europe, North America, and East Asia. Few studies have assessed trends in hospital admissions for CP, and variable rates have been reported [13, 13,23,24,25]. Most studies were based on small populations with unequal exposure to risk factors, and different CP classifications were often applied. In addition, changes in the frequency of CP causes may occur over time, namely the contribution of a dynamically modifying environment, justifying the need for longitudinal studies [26,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28]. To the best of your knowledge, this is the first study that investigated over an eleven-year period, the temporal trends of hospital admissions, lethality, and geographic distribution of CP in Brazil. The results of this study showed that CP rates in males were higher in all age groups, as in all patients ≥ 70y, but the overall hospitalizations decreased by nearly half in both sexes, between 2009 and 2019. The greatest reduction in admissions was observed among patients ≥ 70y. Nevertheless, CP in-hospital lethality rates remained stable and low for both sexes, while patients ≥ 70y showed the highest lethality. However, the greatest increase in CP lethality rates was detected in municipalities integrated into metropolises, which was mainly driven by small-sized municipalities.

Currently, the therapeutic management of CP is mainly directed towards symptom control; however, disease reversal or cure remains unavailable [29]. With disease progression, comorbidities and associated complications of CP have become increasingly common, invariably resulting in hospital admissions and greater mortality in older patients. Up to 80% of patients with CP develop exocrine pancreatic dysfunction and as many as 40 to 60% require at least one endoscopic procedure or surgical intervention over the course of the disease [26, 27, 30]. The age-stratified analysis in this study showed that the number of hospitalizations for CP was much higher for those aged 70 years or older, intermediate for those aged 50–69 years, and lower for those under 50 years old. Elderly people account for 14.3% of the Brazilian population, an increase of 500% in 40 years and the forecast is of continuous raising in the decades to come [31,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33]. The higher prevalence of CP in males than in females is in accordance with other studies [34, 35]. The predominance of CP among males in all age groups is probably related to lifestyle characteristics (a complex combination of social, cultural, and environmental circumstances) that determine higher exposure to environmental and social risks such as tobacco and alcohol. Interestingly, considering the differences in admissions between sexes, although male rates were noticeably higher in all age groups, the smallest difference was observed for patients aged 70 years or older, suggesting that after many years of disease, serious health damage affects males and females more similarly [36, 37].

During the period analyzed in this study, the in-hospital mortality rates remained stable for both males and females, in all age groups. However, the lethality rates were twice as high as those in previous reports from other origins, which showed variable results ranging between 1 and 4% [38, 39]. While such differences could denote a lower quality of healthcare in Brazil compared to other more developed countries, data are difficult to compare directly because of the different methodologies of the studies. In addition, the specific features of healthcare systems, including the availability of hospital beds, and outpatient capacity and demand, certainly differ. Although the stability of lethality rates observed in the current study may indicate the absence of progress in the healthcare system, the overall decrease in hospitalizations due to CP may reflect recent changes in the criteria for hospital admission, at least in part due to the slight but gradual reduction in the number of available beds in the public system. These changes cause a progressive selective pressure, resulting in the admission of the most severe cases with greater risks and complications. As expected, in-hospital lethality rates were higher among patients older than 70 years. Elderly patients usually have more comorbidities and associated complications, and the mortality rate observed in this study may not be directly related to CP. In addition to the risks related to environmental exposure and behavior, including smoking and alcohol overuse, cardiovascular complications, liver cirrhosis, and extrapancreatic malignancies are common causes of death among patients with CP [40, 41]. Higher rates of deaths among patients with CP surgically treated have been associated with preoperative and postoperative continuous moderate or heavy alcohol consumption, heavy smoking, age ≥ 50 years, comorbidities, unemployment, disability, insulin-dependent diabetes, pancreatic exocrine insufficiency and low body mass index [42, 43]. Moreover, the association between CP and pancreatic cancer is well-established, with a lifetime cumulative risk of approximately 4–5%, which may further increase in association with smoking and diabetes mellitus and is associated with poorer outcomes [42,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44].

Regarding geographic distribution, previous studies by our group concerning the epidemiology of other non-communicable diseases in Brazil (gastrointestinal cancers and Inflammatory Bowel Disease) displayed a similar pattern of distribution [45, 46]. In the current study, the highest adjusted rates of admissions for CP were observed in the more developed regions of the Southeast and South, which together account for more than 60% of the Brazilian gross domestic product (GDP) [33]. Similarly, the Central-West region, third in the observed adjusted admission rates for CP, has been an important growing sector in Brazil over the last decade [20]. The Federal District, located in the Central-West, followed by states in the South and the Southeast, displays the country's highest GDP and Human Development Index (HDI). In contrast, the North and Northeast Regions, where the poorest states are located, had the lowest adjusted number of hospital admissions for CP. This appears to reveal the association of CP with the GDP and HDI of the nation. In fact, the greatest increases in HDI and GDP observed in this study coincided with the greatest reductions in smoking and alcohol consumption.

Considering urbanization and conurbation patterns, all types of municipalities (large, medium, or small) integrated or not integrated into a metropolis displayed a trend for reduction in CP admissions. The greatest reductions were observed in municipalities integrated into metropolitan regions, where the urban network promoted public transport integration, access to work, primary healthcare units, and tertiary hospitals. However, the greater number of hospitalizations observed in municipalities not integrated into metropolitan regions may reflect the urban–rural migratory process that has been taking place in Brazil in recent decades, in addition to the limited local urban net. In analyzing the Brazilian migratory scenario of the twenty-first century, two relevant movements should be highlighted: metropolitan migratory dispersion, defined by the return of people from the southeast towards the northeast, and the urban–rural migratory movement, characterized by intra-state migratory flow involving non-metropolitan towns. Consequently, improvements in health access and notifications are expected. Therefore, the process of metropolitan deconcentration and the resulting creation of industrial centers in rural regions may have contributed to this scenario [47, 48]. Tertiary and high-cost health units are characteristic of large urban centers where more complex cases of CP are expected. Nonetheless, the increase in in-hospital CP lethality identified in municipalities integrated into metropolitan regions was mainly driven by small-sized municipalities. A similar tendency was observed in a previous study [49]. As patients with CP suffering from long-term disease and associated complications may require diagnostic and therapeutic tools and resources that are unavailable in small towns, the unified health system provides access to more complex medical procedures, mainly concentrated in large urban centers. However, despite admission to a metropolitan area, mortality was associated with the town of residence.

A reduction in CP in-hospital lethality was noted in all macroregions of the country, except in the south, where it practically doubled. The social and economic influence of tobacco in that region, comprising more than 90% of the tobacco farms in the country, the presence of large metropolises with the highest smoking frequency, and the increase in regional alcohol consumption may have contributed to the disproportionate increase observed in the South. In contrast, the in-hospital mortality reduction associated with CP observed in the Northeast and North may be related to the greatest drop in tobacco consumption in these regions and the improvement in the health network leveraged by GDP growth and HDI. Taken together, these data appear to reinforce the importance of environmental and behavioral factors in disease development and, in particular, underscores socio-economic aspects underlying the pathogenesis of CP.

Diet, tobacco use, regional lifestyle habits, and personal preferences are also factors associated with patterns of alcohol use. Over the last 30 years, global adult per capita alcohol consumption has increased from 45 to 47%, and is forecasted to continue until the next decade [50]. Alcoholic beverages (mainly heavy drinkers) are associated with higher cancer risk. Different drinking patterns, such as frequency of drinking, usual quantity of daily drinking, and especially heavy drinking, have all been associated with an increased risk [51, 52]. Although CP is considered a multifactorial disease, the amount and duration of alcohol consumption are critical determinants that increase the risk of pancreatitis. Despite the significant reduction in admissions due to CP (-57%) observed in this study, the overall alcohol consumption showed a timid reduction during the same period (− 0.3%). However, a positive association was observed between male CP admissions and prevalence of alcohol abuse. In addition, the retrieved data indicated a progressive reduction in alcohol consumption as age advances, which may have contributed to the decrease in hospitalizations for CP recorded in older patients. Smoking is a major risk factor for several diseases including CP and pancreatic cancer. Among the 87 risk factors, tobacco is considered the number one for disability-adjusted life years in males and the seventh for females [23, 26, 29, 53].

In recent decades, the reduction in smoking in Brazil has surpassed that observed in other sociodemographic, more developed countries. Between 1990 and 2005, and 2005 to 2015, Brazil was rated among the 13 countries with the greatest drop in tobacco consumption worldwide. Despite this, Brazil still ranks among the top ten countries with the highest indices of tobacco smoking [16, 54]. Nonetheless, the temporal trends of the reduction in tobacco consumption and the number of admissions for CP in Brazil appear to support a cause-and-effect association (− 35.26% in men and: − 38.40% in women). We may also interpret that the success of the smoking prevention policies implemented in the country may have directly contributed to the reduction in admissions due to CP observed in this study.

Although the findings of this study provide new insights into CP time trends and geographic distribution in Brazil, several limitations must be acknowledged. First, the retrospective nature of the study restricted the information contained in the database to all the variables analyzed. Moreover, the database did not contain information on disease staging, comorbidities, or therapy. In addition, a potential discrepancy in the available data may result from different healthcare quality and notifications in the country's less developed regions. However, the limitations and likelihood of an ecological fallacy [45] in this study were greatly diminished owing to the clarity and simplicity of the data, which were recorded individually by a DATASUS-registered physician into a single electronic database. In addition, the use of municipalities may compensate for potential losses due to underreporting or deficiencies in the registry because the system covers the whole country in its minimal administrative units, encompassing official data.

Conclusion

This temporal analysis of CP in a large, developing, and populous country such as Brazil revealed great variability and may contribute to a better understanding of CP and its progression. The overall trend of a decreasing number of hospitalizations related to CP, while maintaining stability in the associated lethality, appears to indicate relatively recent changes in risk factors and public health care. The improvement in some regional socioeconomic indices, the significant reduction in smoking, the counter-migratory process towards the northeast and from urban to rural areas, and industrial decentralization appear to constitute important factors in understanding such changes. Better distribution of more complex specialized services, greater investment in primary healthcare and preventive measures, and a more integrated and universalized health system must be prioritized to improve the country’s health system.