Cervical cancer prevention in Pernambuco: improvements for whom? Inequity scenario in the state of the Northeast Region

,


Introduction
In Brazil, cervical cancer is the third kind of cancer with highest incidence among women, apart from non-melanoma skin cancer. 1 At the same time, it represents one of the highest prevention potential and cure, as long as it is early diagnosed.
The high prevention potential and cure might be justified by the fact that it is a slow-evolutional disease, with well delimited stages and easily detected in terms of alteration even in its early stage, which promotes early diagnosis and effective treatment. 2,3 In Brazil, cervical cancer epidemiological context is very heterogenous. When analyzing the regions in the country, the highest incidence is in the North (26.24/100,000), along with the Northeast (16.10/100,000) and the Midwest (12.35/100,000) regions. Both the South (12.60/100,000) and Southeast (8.61/100,000) present lower incidences. 1 Recent studies 4 analyzed the over time tendency on cervical cancer mortality nationwide and calculated its projections up to 2030, pointing out that the South, Southeast and Midwest demonstrated a significant tendency of decreasing, thus, the Northeast presented in 10 years (1996 through 2006) a rising trend, followed by a stability. About projections for 2030, an opposite national tendency of mortality reduction for this type of cancer, in the North and Northeast regions will follow the highest rates, maintaining the inequality in the country.
The investment to reduce the occurrence and the mortality rates on cervical cancer began in 1998 with the creation of the Programa Nacional de Combate ao Câncer de Colo Uterino (National Combat on Cervical Cancer Program). Among the actions of the program, one of the most effective strategies for its control is cancer screening and its precursor lesions by performing the cervical cytopathology tests. 5 The Health Ministry along with the Instituto Nacional de Controle de Câncer (INCA) (Nation Institute of Cancer Control), emphasizes the necessity of this exam to be performed annually, prioritizing 25-to-64-year-old women, but after two consecutive years of negative exams, and it can be performed every three years. 6 The World Health Organization says that achieving a coverage of 80%, associated with treatment in the early stages resulting in a reduction of up to 90% in the occurrence rates of cervical cancer. 7 In order to strengthen the health system, the Pacto pela Saúde (Pact for Health) program was released in 2006, in which a setting of commitments were organized in the federative entities defining its three guidelines: Pacto pela Vida, em Defesa do SUS e de Gestão (Pact for life in defense of the Public Health System and the Management). In the first guideline, to control cervical cancer was included as one of the agreed priorities. 8 In 2011, the Federal Decree no. 7,508/2011 institutionalizes the Contrato Organizativo da Ação Pública da Saúde (COAP) (Public Health Organizational Action Contract) and reinforces collaboration agreements stablished among the federal entities in view to organize and integrate actions and health services through agreed indicator and goals of assessing criteria performance. 9 In that perspective, in 2014, the Secretaria Estadual de Saúde de Pernambuco (SES/PE) (State Health Secretary of Pernambuco) released a Caderno de Metas e Resultados dos Indicadores do Pacto/COAP (Goals and Results Report of the Pact Indicators/COAP), and undertakes 64 indicators as priorities, among them is: "the reason for cervical cancer cytopathological exams should be performed on women aged between 25 to 64 years old as well as in the female population in the same age group". The referred indicator is universal and contributes to the access of preventive cervical cancer examination assessment, which allows to identify situations that demands interventions, besides subsidizing policy planning, management and assessment aiming for this specific group. 10 In 2008-2011 the Plano Estadual de Saúde (PES) (State Health Plan), as well as in 2011-2015 in the State of Pernambuco asserts its commitment by registering the goals for planning, promoting, coordinating, supporting and supervising in a state level, to guarantee the reviewed and programmed actions focusing on women's health, upon enforcement of the guidelines which included female morbitity and mortality reduction, especially by inevitable causes, as well as the control on cervical and breast cancer. 11,12 Given the magnitude problem of cervical cancer and the commitments agreed on by the State of Pernambuco, it is important to present how the access to preventive cervical cancer examination outlines over time in the State.

Women's access to preventive examination: an analytical look
In order to analyze the access of preventive examination in the State of Pernambuco, a quantitative approach study was carried out considering the cytopathological exam coverage in the period of 2002 to 2015.
The State of Pernambuco with 9.3 millions inhabitants is located on the Northest region of Brazil. It is extended in a territory of more than 98,000 km² long, with 184 cities and towns including the insular district State of Fernando de Noronha. 13 The Health Regions in the State are distributed into 4 macroregions ( Figure 1).
These regions are geographically made up with bordering cities and towns that have social, economical and cultural similaries, share communication and transport networks, whose goal is to integrate the organization, into planning and executing health actions and services. The macroregions are territorial organizations composed by the Health Regions with the objective of offering medium and high complexity services, that is, procedures or actions that require higher technological complexity and overload the capacity of a single region. 13 An evaluative indicator coverage was used: "the relationship between the number of cytopathological cervical examinations in women aged 25 to 64 years old, in a certain city and year, having as a denominator of female population in the same age group of 25 to 64 years old, in the same place and year divided by 3. 14 Since this is a thirteen-year historical series about a procedure that had, during this period, its registration on different information systems (Siscolo, Siscan, SIA), there was an option to use calculation methods and available and elaborated database by the Health Ministry at SUS (Public Health System) 5 Computer Department in order to avoid biases. By doing so, the chronologies of the following programs were analyzed: Data tabulation was done by Tabnet software and afterwards were exported and processed in an electronic Excel ® spreadsheet, organized and consolidated in the only database indicator. Statistic measures were produced: coefficient regression of the over ime trend of the regression line and the significance test (F Test) for all the cities and towns in this State.
To check the over time trend of the cervical cancer coverage evaluative indicator, a simple linear regression model was applied by the formula of: Y= α+βX, where Y is a dependent variation, represented by the indicator, and X is the independent variable, represented by the year of the indicator coverage assessment. To test the significance (pvalue), the cut-off point was 0.05 of the regression analysis, either for the F-Test and the T-Test, was confirmed by the respective confidence intervals (CI95%). 16    these, 79 cities and towns were below 0.2. In 2015, there was an important increase in the preventine coverage in 76 cities and towns with high /too high coverage (above 0.6) and 44 were classified as median coverage (0.4 to 0.6). Nevertheless, 25 cities and towns still present a low or very low coverage (less than 0.4), which reinforces the necessity of paying more attention to these locations, despite the improvements in the State scenario ( Figure 3).
When analyzing the annual trend for the last fourteen years (Figure 4), it was possible to notice that from 184 cities and towns including insular district of Fernando de Noronha, 173 (94,0%) obtained a increased trend of growth, statiscally significant, in terms of cytopathological examination coverage along its historical series. However, the situation in 12 cities and towns present a concerning negative growth trend.
It is noticeable that from the regression analysis and its hypothetical statistic tests (Tests F and t), there is an increasing growth in the State on cytopathological coverage over the last fourteen years (Figure 4), with a significant growth value (p<0.05).

Reflections on inequality in accessing and the challenges for the coverage extension
As observed in the findings of this study, between 2002 and 2010, the indicator had grown and the regions of the triennial coverage mean were equated, despite still falling short of the coverage recommended by World Health Organization. 7 Upon the following triennial (2011 to 2013), a gradual decline on some regions is evident, as well as the difference among them.
The increased coverage observed for the first triennials might have been driven by the expansion to access the Atenção Primária à Saúde (APS) (Primary Health Care) in the country, 17  On the other hand, such initiatives, even though related to firm commitments at State level, 11,12 they were not enough to continue the coverage expansion for years ahead, which demonstrates a problematic scenario, especially in the outback of Pernambuco, the region where its coverages were the worst over the years. Petrolina (VIII) and Ouricuri (IX) regions are the most negative regions since they continued to be classified as having the lowest coverage in the State. The context regarding VIII region is the most concerned one, due to fact that it corresponds to a macroregion, and therefore, is responsible for a more structured service for the cities and towns. Moreover, it is stated in the database systems that this region is one of the few in the State to offer all the procedures for cervical cancer care. 18 It is noteable to point out the isolated expansion of cytopathological coverage is not sufficient to reduce the morbity and mortality indicators in cervical cancer. The quality of the material collected and its subsequent analysis are fundamental in providing a reliable result, which avoids false-negative result. In this matter, according to Santos et al., 19 during the period of 2006 and 2009, Pernambuco was among the States with the highest percentage of unsatisfying samples. The quality, as the collecting and the preventive laboratorial analysis are crucial for a precise diagnosis in order to guarantee adequate follow up for the women. 20 The situation in the State is fragile when associating the findings in this study to the obtained results from Lira`s research, 21 which points out the highest incidence of atypias in the cytopathological examinations were concentrated in the regions located in the outback of the State.
The decreased indicator tendency and several patterns among the health regions in the State, might be interpreted based on the issues that were highlighted in previous studies, as for the difficulty in having access to APS and/or the resistence of women to perform the exam with fear, shame or embarassment, besides the lack of knowledge regarding how the procedure is important. 22,23 Although, it is necessary to note that these factors are concentrated on women in vulnerable conditions.
The social inequities are related both to a low cytopathological coverage and the highest mortality rates. 24,25,26 When verifying the Índice de Desenvolvimento Humano (IDH) 27 in the State and the Health Region, it is noteable that Pernambuco is classified in an intermediate range in development, and nearly half of the region was classified having the lowest index. Regarding to Petrolina (VIII) and Ouricuri (IX) regions, a low coverage convergence of HDI was observed in these regions. Although, the VIII region demonstrates an intermediate HDI, detailing the index by the city and town, observing that most of them are below the HDI. This comparison reinforces the relation between social inequality and health access.
Another aspect to take into account, is the fact that this study is based on secondary database. These sources may present an overloaded or underreported data due to inadequate feeding from the information systems. Such fragility might result in the indicators values that may not relate to the real situation. The result in Afogados da Ingazeira (X), for example, points out the necessity for further detailed research in this region to understand atypia (coverage above 1.0) that was observed in the past years.
Despite positive result in some cities analyzed in the last year, when compared to the first scenario, it is noteable that the best results were concentrated in cities and towns closer to the capital (Metropolitan and Florest regions). Girianelli et al. 28 study reinforces the understanding about the inequality of people who live distant from large urban centers to have access to health. The authors affirm that in the Northeast, mortality rates on cervical cancer are decreasing only in its capitals, while there was an increase in the number of mortality in the towns in the countryside.
Another factor that can also contribute to the low coverage established by the WHO is the procedure performed in the private health system, as pointed out by a previous study developed by Brio-Silva et al. 2 This study implies that the research development contemplate on the coverage calculation estimates and the supplementary health participation should also be taken into consideration. The State of Pernambuco, during the period of 2002 to 2015, increased approximately, in populational terms, from 8.0 to 9 millions inhabitants. In regard to women, the growth was 11.8% and the age range of 25 to 64 years old increased to 24.0%. 13 Although, the increase of female population of this age group, the State also had a general tendency for the expansion coverage indicator in the State, from 0.18 in 2002 to 0.53 in 2015, which represents an increase in the access to preventive examination. Nevertheless, the indicator settings are very heterogenious when relating to smaller territorial levels.
Despite the recommendations and the relevance of the indicator, the present study indicates that all regions at sometime or in all times present less coverage than expected, which implies the need to identify the factors that define this scenario so that the expansion may widen for everyone to have access to preventive examination.
The situation observed in Pernambuco, especially in the outback, is a great concern and requires the development of strategies that enable changes in this scenario. It is also suggested the development of the studies with a qualitative or mixed approach that can contribute to a better understanding of the dynamics in the Health Regions in Pernambuco, improving the dimension of factors related to a low cytopathological coverage.

Authors' contribution
Brito Silva KS and Leite AFB contributed to the concept and planning of this work, the analysis and interpretation of the data. Silva DMC also contributed to the analysis and interpretation of the data. Tanaka OY, Louvison M and Bezerra AFB contributed to the critical review of the content. All authors approved the final version of the manuscript.