Atenção Primária à Saúde: elementos de continuidade e mudanças na saúde do Distrito Federal

Primary Health Care (PHC) has a prominent position in the Brazilian governmental agenda. This study focuses on the health system in the Federal District and the initiatives to reorganize PHC, with the analysis of expenditure behavior in order to identify aspects of continuity and change during the period from 2005 to 2014. To achieve this purpose, documental research was carried out focusing on data from the Public Budget Information System. Drawing on historical neo-institutionalism, the results reveal contradiction between the discourse in defense of PHC and the maintenance of high level of expenditure with hospital services, thus confirming path dependence characteristic.


Introduction
As from the 1970s, the debate on Primary Health Care (PHC) is internationally intensified as a result of questionings about the organization of health care based on a specialized and interventionist hegemonic medical model, with the fragmentation of care and little impact on the improvement of the population's health condition (gioVanella;MenDonÇa, 2008). Another factor that stimulated the debate was the existing gap between health organizational conditions in developed and developing countries, drawing attention to the disparity in the access to health services (caStro;MacHaDo, 2010).
In Brazil, an option was made in the 1990s for the use of the term 'basic care'. According to Heimann and Mendonça (2005), this denomination was adopted in defense of a model of care that would differ from the preventive proposition and would approach the logic of social determinants of diseases. More recently, the Ministry of Health's publications have been using the terms 'basic care' and 'primary care' as synonyms.
The Brazilian Sanitary Reform introduced the notion of health as social right, i.e., it recognized health as a fundamental citizenship right and as being the responsibility of the State to provide it by means of social and economic policies. This notion, expressed in article 196 of the 1988 Constitution of the Republic of Brazil (BraSil, 1988), parted with the previous model of social protection based on meritocracy and advanced toward a universalist model.
According to Gragnolati, Lindelow and Couttolenc (2013), the principle of universality features health as a citizenship right when defined by the Federal Constitution as a right of all citizens and a duty of the State. In this sense, when declaring that the State has the duty to deliver this care to the entire population, universality was a key principle for the creation of the Unified Health System (Sistema Único de Saúde -SUS) in Brazil.
Since its creation, which was formally instituted by the 1988 Constitution, PHC represents the 'entrance door' to the system, with emphasis on the delivery of health services and the integrality of health actions (corDeiro, 2001). By means of the delivery of low and medium complexity services, i.e., by means of the prevention and promotion of health, it seeks to reduce the demand for high complexity services (specialized and more expensive) and to guarantee universal access to health (gUerra, 2013).
In Brazil, the 1990s was marked by important changes in the national PHC policy and by large investments to broaden the access to it. It can be stated that this transformation is related to changes in the federative arrangement instituted after the 1988 Constitution and the decentralization process as well as to the prominence given to PHC in the sectoral agenda (caStro; MacHaDo, 2010). The new federal arrangement in the health sector comprised advances in the movement of political-administrative decentralization, with emphasis on municipalization; in the creation of new health financing mechanisms; and in the progressive transfer of responsibilities concerning health actions and services to be carried out directly by states and especially by municipalities, with emphasis on outpatient services. Therefore, PHC received a prominent position in the federal agenda especially after the creation of the Family Health Program (FHP) in 1993/1994 -which was later transformed in the Family Health Strategy (FHS) -whose proposed model of care was directed to the protection and promotion of health through integral and continuing care, focusing on the family (caStro, 2015).
After the decentralization of SUS, there was the municipalization of PHC; this determined the advances in the process of universalization of the service, which was previously constituted by different organizational accumulations in the public network of the country's various regions (caStanHeira, 2014). According to Miclos, Calvo and Colussi (2015) the assessment of this level of care is strategic to identify the persistent fragilities that hinder the organization and operationalization in the direction of the desired resolubility of the service.
Having similar characteristics to those in the large metropolitan cities of the country, the Federal District health system is an important example for the analysis of the repercussions of those initiatives upon PHC reorganization.
In this context and considering the contribution of historical institutionalism, the objective of this study is to analyze the behavior of the expenditure with primary care in the Federal District, aiming to identify elements of continuity and change within this policy in the period from 2005 to 2014; and also to analyze the institutional characteristics of the health policy correlating it with the behavior of expenditure.

Neo-institutionalism in the analysis of public policies
According to Matias-Pereira (2012), public policies may be seen as institutions, understood as formal and informal rules that mold actors' behavior because they produce resources and establish incentives associated to specific behavior, affecting the intentions among actors of a given sector (PierSon, 2006;MatiaS-Pereira, 2012).
In this approach, public policy changes the way social actors are involved in the policy and, therefore, it should be seen beyond its formal institutions. Neo-institutionalism stresses the importance of the institutional factor to explain concrete political events, referring not only to the limitations of the decision-making process rationale (such as lack or excess of information) but also of the interference of general rules that in every society exert decisive influence upon individuals' actions (PierSon, 2006;FreY, 2000;SoUSa, 2014).
When taking the definition of institutions from Thelen and Steinmo (1992), these include formal rules, compliance procedures, standard operational practices that structure the relationship between individuals in the various unities of politics and economy, and also include formal rules and informal constraints related to behavioral codes and conventions in general (MarQUeS, 1997;SoUSa, 2014).
This theory leads to the understanding that not only do individuals or groups have relevant strength to influence public policies and their agendas, but they also highlight the role of formal and informal rules that govern institutions in the determination of their implementation, with a stress on the struggle for power and resources between social groups and on the mediation exerted by political and economic institutions that conduct public policies in a given direction for the benefit of a specific group (SoUSa, 2014).
The answers to the investigation questions related to neo-institutionalism lie in the intermediate structures, i.e., in the institutions that shape the political and social processes and mediate the relationship between State and society, and between economic structures and the behavior of individuals and groups (liMa;geraSSi, 2011). According to Thelen and Steinmo (1992), the historical neo-institutionalism seeks to understand the functioning of institutions with the purpose of deterring, retarding, consolidating, or accelerating social change processes, through the recuperation or reelaboration of historical-structural models. These authors consider institutions as carriers of legacies of socio-political and historical forces that shape them, juxtaposing different logic of political order, each with its own temporal feature. Institutional arrangements that compose a policy may emerge in different epochs.
In this context, institutions should be understood according to their main functions, i.e., agent and product of history, being at the same time responsible for shaping and restricting the performance of political actors and also for representing the result of strategies developed by these actors. In their theoretical approach, five levels of public policies analyses are determined: path dependence, institutional arrangements, coalitions, discourse communities, and ideas; the analysis in this study uses the perspective of path dependence MatiaS-Pereira, 2012). The term 'path dependence' is understood as a succession of historical points that present critical conjunctures and increasing benefits (Hall;taYlor, 2003).
In the field of public policies, according to North (1990), path dependence can be defined as the powerful influence of the past on current and future decisions. Pierson (2004) relates the concept to the idea that past events or a path may influence present political decisions (SoUSa, 2014). Lima, Machado and Gerassi (2011) state that the present is a result of past decisions and their consequences, not only of contemporary conditions. In the sphere of public policies, according to these authors, when a path is adopted, political and economic costs to have it changed are very high, due to the creation of rules and structures in the mobilization of different types of resources and the involvement of social actors, raising the chances of continuity and decreasing the possibilities of radical course changes.

The path of Primary Health Care policy in the Federal District
According to the study conducted by Göttems et al. (2009), the health system of the Federal District, with similar characteristics to those in the large metropolitan cities of the country, constitutes a relevant example for the analysis of the repercussions of initiatives on PHC reorganization.
The Federal District (DF) of Brazil is surrounded by municipalities of the states of Goiás and Minas Gerais, which compose the Integrated Economic Development Region (Ride/DF); in 2010, the estimate population of DF was 2.5 million inhabitants and the Ride-DF had 3.7 million inhabitants (iPea, 2013). The health system of DF has its own health care network comprising all complexity levels.
Public hospital beds belonging to the Secretariat of Health of DF (SES-DF) are equivalent to 83% of the total number of beds of SUS, which represent 70.4% of the total number available in DF (BraSil, 2014). Regarding PHC, there is an extensive regionalized physical network, with an average of one primary health unit per 25,000 inhabitants. However, within PHC coexist the traditional model, the Family Health Strategy (FHS), and the Program of Community Health Workers (PCHW); this hinders the proposed reorganization. Presently, the population coverage by FHS is 8.7%, and by PCHW is 15.4%, totalizing 24.1% according to data provided by SES-DF Regarding the health policy, Göttems et al. (2009) carried out a study of the path of Primary Health Care policy in DF, in the period from 1960 to 2007, drawing on the theoretical framework of historical neo-institutionalism.
From 1999 to 2007 the SES-DF was headed by three secretaries of health, with an average length of stay of 2.6 years. In the two last periods the implementation of FHS underwent operational changes, such as: compensation of 20% over the salary for civil servants of SES-DF in FHS teams; construction of 30 health units in the urban area for those teams to work in; hiring of other professionals in order to reach FHS full composition. PHC became the landmark of changes in the health policy; it was introduced in the agenda of debates; it became an object of political parties' disputes; and it justified public-private partnerships for the sector's management; however, it was not able to alter the functioning of PHC traditional units (gÖtteMS et al., 2009). With financial and political resources, massive investments were made in the construction of hospitals, followed by initiatives to strengthen PHC, which were discontinued and disarticulated in the medium and high complexity services network, with reduced offer of professional training, justified by the need to guarantee hospital care rear. According to the authors, the contradiction between the discourse in defense of PHC and the maintenance of goals for the construction of new hospitals at every new administration confirmed the strong path dependence characteristic.
Regarding the services network organization, despite innovations, SES-DF administration had traditional characteristics, such as: investments in hospitals as central locus of the health system; centralization of financial and human resources management; use of cutting-edge technology; and reinforcement of medical residency centered in specialty, to the detriment of the other professionals. Even when it was part of the health policy agenda, PHC was considered as accessory to hospital care, aiming to 'relieve' it (gÖtteMS et al., 2009).

Methodology
In this study the theoretical framework used was the historical institutionalism for it seeks to analyze politics by studying institutions as well as interests and actors. This theory considers that other factors influence political life, not only institutions. Adepts of this theory defend a social causality dependent on the path that has been followed, according to which the same active forces produce different results, because they are modified by properties inherited from the past in every local context (Hall;taYlor, 2003).
In order to analyze the institutional characteristics of the health policy relating it to the behavior of expenditure, this study adopted as strategy a qualitative approach and documental research. Data for the analysis were retrieved from the Public Health Budget Information System (Sistema de Informações sobre Orçamentos Públicos em Saúde -Siops). This system is a reference for the monitoring, inspection, and control of the application of resources related to public health actions and services. The analysis took into consideration the path referring to the temporal sequence of political processes and their impacts on current results, as well as institutional characteristics and the political path, which constitute categories that reflect the set of rules and agreements guiding health policy in various moments, thus contributing to the understanding of its present configuration.

Results and discussion
Data presented here were retrieved from Siops, research module -Data Informed by Federal Government, States and Federal District, and Municipalities, under option 'History of expenditure by sub-function, consolidated by institution, stage of expenditure', with the following search criteria: expenditure by sub-function, consolidated by stage of expenditure; Federal District; committed allocation.
Data in Siops are available from 2002 to 2014; but when searching by sub-function, by State, and by committed expenditure, only data in the period from 2004 to 2015 are generated. Two sets of calculations were produced: 1 -percentage of expenditure with PHC in relation to the total amount of expenditure with health; and 2 -percentage of expenditure in relation to the previous year, with increase for values higher than 100% and decrease for values lower than 100%. The objective was to verify PHC participation within Function Health and its evolution over time.

Participation of sub-function Primary Health Care in the Federal District in relation to Brazil
From the analysis of expenditure with PHC it was possible to identify relevant seasonality, without there being an annual spending pattern. While in 2005 almost 5% of the total amount of health expenditure was spent with PHC, in the years 2007 and 2008 less than 1% was spent in the DF. Thus, the average of values spent with PHC was only 2.33% in the period under analysis (2005 to 2014).  Table 2 shows the evolution of expenditure with sub-function PHC and with Function Health, for all the country's states, searched by committed expenditure. Two sets of calculations were produced: 1 -percentage of expenditure with PHC in relation to the total amount of expenditure; and 2 -percentage of expenditure in relation to the previous year, with increase for values higher than 100% and decrease for values lower than 100%. The objective was to verify PHC participation within Function Health and its evolution over time. From the analysis it was possible to verify that the average of expenditure with PHC in the analyzed period was 5.24%. When comparing the average percentage of PHC participation in the expenditure with health in DF (2.33%) in relation to the national average -average participation of States when analyzed as a whole -it was verified that the country's capital (DF) spending was less than half of the average; from this fact it was possible to verify the low values allocated to PHC. In some years, e.g. 2007 and 2008, the percentage was lower than 20% of the national average, demonstrating inconstancy in the planning and permanence of policies, with such seasonality in the spending of resources.   Tables 4A and 4B show the evolution of expenditure with the sub-functions belonging to Function Health for all the country's states, searched by committed expenditure. Two sets of calculations were produced: 1 -percentage of participation of expenditure in the sub-functions in relation to the total amount of expenditure with Function Health, and 2 -percentage of expenditure with PHC in relation to the previous year. Increases and decreases were considered in the same previous bases. The objective was to verify the participation of PHC and Hospital and Outpatient Care within the Function Health in the sphere of the country's states.   confirms what was verified in the study by Göttems et al. (2009), that there was contradiction between the discourse in defense of PHC and the maintenance of high expenditure with the sub-function Hospital and Outpatient Care, thus verifying the strong path dependence characteristic. From the analysis of spending it is possible to infer, regarding the services network organization, and despite innovations, that SES-DF administration made an option for traditional paths, such as the investment in hospitals.

Conclusions
The health system in DF, with similar characteristics to the large metropolitan cities, constitutes an important example for the analysis of the repercussions of initiatives concerning the reorganization of PHC. Regarding health management, relevant investments of financial and political resources were verified in the construction of hospitals.
It was also verified that initiatives to strengthen PHC were discontinued and disarticulated from the medium and high complexity services network, with reduced offer of professional training, justified by the need to guarantee hospital care rear.
The contradiction between the discourse in defense of PHC and the maintenance of goals for the construction of new hospitals presented to every new administration testifies the strong path dependence characteristic. Regarding the services network organization, despite innovations, the health management in the DF has followed traditional paths, such as: investments in hospitals as central locus of the health system; centralization of financial and human resources management; use of cutting-edge technology; and reinforcement of medical residency centered in specialty in detriment of other professionals.
The percentage of PHC participation in the expenditure with health in DF (2.33%) is lower than half the average participation of the states when analyzed as a whole (5.44%), reaching levels that represent the discontinuity of expenditure in PHC participation, as in the years 2007 and 2008 (0.81% and 0.41%), drastically decreasing in relation to the historic average.
It was also verified that the absolute increase of the global health expenditure in DF (207%) presented growth above the national average (180%); in relation to PHC, though, there was a decrease in the expenditure: in 2014 the application corresponded to 88.58% of spending in 2005. Contrary to political priorities, expenditure with Hospital and Outpatient Care presented an increase of 65.38% in the analyzed decennia, with 80% participation in the total amount of Related Sub-functions -contrary to the decrease verified in PHC.
Therefore, from the analysis of the information presented in the study it is possible to verify that there was contradiction between the discourse in defense of PHC and the maintenance of high spending with the sub-function Hospital and Outpatient Care, confirming the strong path dependence characteristics, by maintaining similar characteristics as those from the 1970s and 1980s, that is, hospital-centered natured. s