The process of implementation of emergency care units in Brazil

ABSTRACT OBJECTIVE To analyze the process of implementation of emergency care units in Brazil. METHODS We have carried out a documentary analysis, with interviews with twenty-four state urgency coordinators and a panel of experts. We have analyzed issues related to policy background and trajectory, players involved in the implementation, expansion process, advances, limits, and implementation difficulties, and state coordination capacity. We have used the theoretical framework of the analysis of the strategic conduct of the Giddens theory of structuration. RESULTS Emergency care units have been implemented after 2007, initially in the Southeast region, and 446 emergency care units were present in all Brazilian regions in 2016. Currently, 620 emergency care units are under construction, which indicates expectation of expansion. Federal funding was a strong driver for the implementation. The states have planned their emergency care units, but the existence of direct negotiation between municipalities and the Union has contributed with the significant number of emergency care units that have been built but that do not work. In relation to the urgency network, there is tension with the hospital because of the lack of beds in the country, which generates hospitalizations in the emergency care unit. The management of emergency care units is predominantly municipal, and most of the emergency care units are located outside the capitals and classified as Size III. The main challenges identified were: under-funding and difficulty in recruiting physicians. CONCLUSIONS The emergency care unit has the merit of having technological resources and being architecturally differentiated, but it will only succeed within an urgency network. Federal induction has generated contradictory responses, since not all states consider the emergency care unit a priority. The strengthening of the state management has been identified as a challenge for the implementation of the urgency network.


INTRODUCTION
In the early 2000s, the Ministry of Health set up a national urgency care policy, with the implementation of new components such as emergency mobile services (SAMU) and emergency care units (UPA). The pre-hospital implementation was a good decision, as national 9,12,15 and international experiences 2,21 show the positive impact of this care. Chronologically, the implementation of the pre-hospital components occurred separately 15 , in three different moments 10 : until 2002 -initial regulation, 2003-2008 -emphasis on the SAMU, after 2009 -emphasis on the UPA. In 2011, the need for an urgency network was defined, with regionalization and reorganization of pre-existing services 13 .
The UPA, the main fixed component of pre-hospital urgency, are intermediate units between primary care and hospital emergencies. They are classified in three different sizes, according to the population covered, the physical area, the number of available beds, the management of persons, and the capacity to care 14 .
The ordinances edited over the years have provided for UPA that are strategically integrated into urgency care networks 6 . The necessary coexistence with the SAMU and the requirement to expand the primary health care coverage with the Family Health Strategy (FHS) are conditionalities that seek to strengthen the network vision and impel managers to invest in other components of the urgency network. This is how the intended resolution of the UPA is demarcated, differentiating them from the early emergency rooms, producers of "emergency appointments" that are little resolutive 6 .
A national research 3 has identified that all UPA classify risk and are structurally well-equipped. The main challenges found included the professional management and the "hospitalization" of patients 3 . Rio de Janeiro, the first state to implement UPA before the federal ordinances and creator of the name, built its park of units counting on a convergence of interests that has allowed its rapid expansion 9 . This state rapidly expanded the UPA without effective organization of the urgency network 7 . Another research has evidenced the limits in the management of the health work, expressed in the types of hiring, setting, and qualification of the professionals in the UPA 11 with implications in the quality of the care. Despite the difficulties presented in the studies 7,9,11 , the UPA represented an increase in access to medical appointments and complementary tests, especially in less complex cases 7 .
The aim of this research was to analyze the implementation process of the UPA in Brazil, identifying the rules and resources that facilitated its implementation, the obstacles and facilitators in this process, the influence of the agents of the different governmental spheres in the implantation of these services, and the proposed expansion of the UPA.

METHODS
We carried out an analysis of the official legislation, documents, and websites of the municipalities, states, and federation, as well as interviews with key players and a panel of experts.
We interviewed state emergency care coordinators who participated in a panel of experts. These players were chosen because of their importance for the integral assistance to urgencies, the negotiation of pacts and goals, and the cooperation between federative entities. In total, 24 Brazilian state coordinators agreed to participate. The interviews were based on a semistructured guide and took place in 2013 and 2014. The panel of experts collectively addressed the same subjects of the interviews, providing the opportunity to share experiences. The panel took place in two days (16 hours) at the end of 2013 and was attended by the urgency and emergency general coordinator of the Ministry of Health and three members of the Brazilian Emergency Cooperation Network (RBCE), a key institution in the formulation and implementation of the policy of urgency, in addition to the interviewed players. The material of the panel and interviews were analyzed together, considering: policy background and trajectory, players involved in the implementation, expansion process, advances, limits, and implementation difficulties, and state coordination capacity. The documents made up the rules and resources available for the action of the players involved. The analysis plan aimed to identify the structural circumstances available for the action of managers and the strategies that differentiated the states, characterizing the implementation of the UPA in Brazil.
We used the theoretical framework of the analysis of the strategic conduct of the Giddens Theory of Structuration 5 , according to which social practices can be understood as skillful procedures, methods, or techniques performed by social agents, using structural rules and resources 5 . Structural dimensions (the power base) are either facilitative or restrictive, and agents access them to manipulate and influence social interactions. There are three types of structuration, correlating the cognitive capacities of the agents and the structural characteristics: meaning, which is enabled by communication from an interpretive plan, domination, which depends on the mobilization of allocative and authoritarian resources, and legitimacy, exercised from the sanction of standards 5 .

RESULTS AND DISCUSSION
We analyzed the ordinances regarding the formulation of the Urgency Policy. The Ministry of Health provided a report that informed the sphere responsible for the management of the UPA and we consulted the webpage of the Strategic Management Support Room of the Ministry of Health, which are the main sources of descriptive quantitative data.

History of the Fixed Pre-hospital Care
For decades, the fixed pre-hospital care was mainly assumed by municipal managers who performed this care role in different ways. Large urban centers were the precursors, with units open to spontaneous demand, but with predominant structure and equipment for low complexity care.
We highlight the emergency care services (SPA), the immediate care units (UAI), the medical emergency services (PAM), the municipal urgency and emergency centers (CMUM), and the regional health centers (CRS). The outpatient medical care units (AMA) of the municipality of São Paulo represented one of the alternatives prior to the UPA. The AMA aimed to broaden the access of the population with acute conditions of low and medium complexity to integrate them into primary care 19 . However, we identified a predominance of units with unsatisfactory integration 19 .
Before the UPA, there was a network for fixed pre-hospital that was not regulated by the Government, which fulfilled its role without systematic evaluation. Most of the time, these units were not expressive in structural terms, did not classify risk, and produced little resolutive appointments.
Although they are not pre-hospital units, small hospitals eventually had this role, especially in small municipalities and in the countryside. Unable to fulfill their function as an inpatient unit, because of deficiencies in the structure and human resources, in some places, small hospitals only provided care of spontaneous demand, confirming their idleness 20 .
The federal resource and the rule for structural definition were inductors for the UPA. However, municipal and state agents had already defined their strategies to set up the fixed pre-hospital urgency.

Implementation and Expansion of the UPA
The implementation of the UPA began with an ordinance in 2008. The state of Rio de Janeiro was the pioneer in this process, implementing the first UPA in the country in 2007, before Table 1 shows the expansion of the APU in Brazil between 2011 and 2016. We chose this period as it characterizes the beginning of federal induction from the Growth Acceleration Program (PAC). Up to 2011, the UPA were concentrated in the Southeast region. In 2016, we have an expressive number of UPA throughout the country. Of the 485 UPA in 2011 (sum of those built and under construction), 446 are ready and in operation since 2016. By 2016 there was still an expectation of expansion -620 UPA were under construction -, despite the 132 that were built but which did not work. São Paulo has the largest park of UPA in operation, under construction, and not working. Proportionally, the North and Midwest regions have the largest number of UPA not working, considering the total in operation. The North was conservative in the implementation and expansion of UPA, with the exception of the state of Pará.
According to the interviewees, some factors explain this scenario of varied expansion in Brazil: divergence between political interests and technical criteria, financial or operational difficulties at the municipal level, such as hiring doctors, lack of funding resources with federal payment, difficulty in the regionalization of municipal UPA, projects of UPA that later proved to be a mistake from the lack of planning, and lack of credibility of the UPA and state support.
There was unanimity about the lack of federal resources for implementation and costing, the latter theoretically shared between state and municipality.
The PAC 1 and PAC 2 were important sources of financing for the expansion of the UPA. In PAC 2, some municipalities were selected regardless of the regional urgency plans. According to the state coordinators, the attempt to regionalize the urgency was compromised when the state was removed from the expansion process, as in the following statement: "I think it was an interference that produced some expectations... the municipality says I want my own UPA... it weakened the technical part a bit... the policy that financed PAC 2 defined the expansion" (E14).
Other states were better able to act as coordinator of municipal demands: "I received several letters from the lower house requesting an UPA for the municipality... I always gave the same answer... the criterion is technical, the need of the region" ( E9).
Federal induction also produced the transformation of services previously financed by states or municipalities into UPA. "The SPA (emergency care services) are 100% funded by the state. They will receive federal resource when they become UPA" (E3).
There was a coordinator who was unfavorable to the implementation of the UPA component in the urgency network, which did not prevent its implementation. Others recognized the suitability and purpose of the UPA. "I think the UPA is fundamental for the network" (E22).
The state of São Paulo illustrates a peculiarity, since its park of UPA is located outside the capital, evidencing the independence of the municipal level.
State managers were emphatic in stating that the expansion of the number of UPA, more than the initial implementation, was a federal ambition. "I think we are in a very good number. Not their expansion, but making them work is the biggest challenge with the articulation with the municipal system in which they are inserted" (E16).
The UPA, unlike SAMU 15 , did not respond to a substantive care gap for the urgency network. Its implementation depended more on the federal financial incentive, which, because it was expensive to the municipalities, produced some frustrating capacities of responses, in the conception of the managers.
The municipal manager relied on the federal entity for the expansion of the UPA and most of the state managers felt powerless in the execution of their state urgency plans.

The Structural Conditions for the Care
The main characteristic of the UPA is the adequacy of its structure, the comfort of the facilities, and the suitability of the equipment. For the first time, a component of the SUS was proposed with great exigency in the structural criteria. This was an expressive differential, according to the interviewees, in relation to the existing emergency care park. "It evolved from the model of ER that had no resolutiveness, physical space, qualified team, material, or medication" (E7).
There was reference to the sufficiency of laboratory diagnostic resources and access to imaging resources, as observed in a national research 3 . However, the maintenance of the UPA eventually became a problem. "Lack of maintenance of the equipment ends up undermining the physical structure" (E16).
Information systems were mentioned as a structural problem. "The UPA are not computerized. They make manual reports of the care type" (E13).
However, the most striking structural issue was the lack of professionals. "The bottleneck is the human resources" (E16).
This problem had been identified nationally 3 . The low qualification of the professionals was exposed by the fact that 34% of the physicians did not have any qualification to act in the UPA. In addition, the size of the number of professionals fell short of the demand 3 . Another study has identified selection and fixation problems, predominance of young and inexperienced professionals, high turnover of physicians, and UPA seen as temporary work 11 .
The problems indicated by these studies were recognized by the interviewees, who also highlighted the incipient qualification and issues with the type of employment of professionals. "Physicians come from the FHS with no experience. Difficulty of training" (E4). "You have public workers from the municipality, borrowed from the state, those hired, it is a mixture" (E7).
The federal entity was the protagonist of a proposal of a structurally differentiated unit. However, none of the three government levels was able to positively equate the management of persons. Table 1 presents and analyzes how UPA worked together with the primary care and the hospital. We highlight the following functions for the UPA: to complement primary care, to complement the hospital emergency, and to function as an inpatient unit.

The UPA and other Components of the Emergency Network
The fixed pre-hospital component was regulated as an integral part of the primary care. As its implantation mainly occurred to unburden the hospitals, many managers understood the UPA as complementary to the hospital, as already evidenced 7 .
Most of the care services were classified as blue or green, that is, of low severity, as already identified in local 7,9,16 and national studies 3 . There is no contradiction in the predominance of less severe urgencies in pre-hospital units. However, as we have already indicated 7,16 , the interviewees stated that the expectation of professionals is to treat primarily critical patients. The preponderance of less critical patients was understood as a deficiency of the primary care.
Considering that the FHS coverage in the country reached 61% in 2015, the interviewees evaluated that the UPA were not effectively related to primary care.
Some of them argued that the UPA would have less purpose if the primary care were better structured. "Dispensable if there is good primary care coverage" (E9).
Despite the impact of insufficient primary care in the UPA, the impact of the lack of hospitals is more serious, which has generated some distortions. "Some managers think that the UPA solves the shortage of hospitals" (E10).
Two serious problems can be identified in the Brazilian hospital park: a concentration of beds in small hospitals and a tendency to reduce the already undersized number of beds. The number of beds in the SUS decreased (10.5%) from 2005 to 2014 4 . In 2014, the SUS had 1.56 beds per 1,000 inhabitants, which is less than the parameter suggested by the Ministry of Health (MH) of 2.5 to three beds per 1,000 inhabitants, and which is still quite conservative 4 . "Instead of opening doors, policy should qualify them and increase beds" (E16).
Despite the mandatory hospital referral, the difficulty of hospitalizing patients was considered the greatest challenge for the UPA.
The permanence of patients for more than 24 hours in the UPA because of the lack of beds impacts the quality of care 3 . A transfer that is longer than four hours for a hospital bed aggravates the condition of the patient 8 . This long stay is a relevant issue, which reflects the difficulties of the hospital network and we highlighted it in 1, when indicating the risk of hospitalization of patients in the UPA.
In addition to the absolute deficit of beds, most of these beds are in small hospitals, with low resolution capacity and low occupancy rates. However, these beds are accounted for, despite the technical insufficiency to use them 4 . The problem is not restricted to smaller hospitals. It also reaches hospitals with more than one hundred beds, which are the main referrals for UPA. The managers knew this reality well. "We have problems with length of stay in large hospitals. We need to train the head on duty, clinical coordinators, break feuds" (E11).
The low supply of reference beds for the UPA has generated distortions. Some managers treated the more complex beds in the red room as hospital beds.
Strategies to address the problem were identified, which ranged from differentiated financial transfer up to the strengthening of the management of small and philanthropic hospitals. "There were contracts with philanthropic network beds. Resources for new beds and qualified beds" (E11).
At the federal level, initiatives to address the hospital problem are still frustrated. The consequence is overcrowding in hospital emergencies of the major hospitals of reference. This overcrowding generates pressure from the entrance 1 and UPA may be essential because it is another open access option. Another component of overcrowding 1 , the capacity of a hospital to absorb the patient from the emergency, releasing beds, impacts the UPA, which needs this bed.

Situation identified in the UPA Evaluation of interviewees Our analysis
To meet low and medium severity urgencies complementarily to primary care The quantitative and qualitative insufficiency of the primary care produces this pattern of care. The UPA has its function in the low severity care as it works 24 hours, complementing primary care. However, it should not be implanted to replace primary care.
"UPA meet outpatient clinic situations in most cases" (E5) They indicate that the noninvestment in primary care has been replaced by UPA.
"Patients of the UPA should be in PC. Government should decide to invest in PC" (E22) To be complementary to hospital emergency The UPA were implemented to replace hospital emergency rooms and, in smaller municipalities, to replace inefficient hospitals.
The meeting of the low-risk demand in pre-hospital units may benefit the hospital emergency that should prioritize the more complex cases. However, that is not the purpose of the UPA.
"UPA are close to hospitals to replace the ER of the hospital" (E8) To work as an inpatient unit The interviewees are against the "hospitalization" in the UPA. However, they admit that some managers count the beds of the red room as hospital beds, including ICU. The structural difficulties of primary and hospital care did not favor the configuration of an urgency network to which the UPA was integrated.

The Management of the UPA and the Interaction between Federated Entities
Box 2 illustrates how municipal and state players mobilized resources, legitimized, and signified the UPA as components of the urgency network, building new contexts 5 , in which the UPA can be implemented from state plans in agreement with the municipality or from autonomous municipal decision. We could see that the significance of the UPA as a structuring modality did not correspond to the expectation of a component that was integrated into the urgency network, but it meant a new access to the system. It was legitimized when it was identified as a fundamental component capable of fulfilling one of its functions, which is to be complementary to primary care. The domination structure showed that the mobilization of material and authoritative capacities generated a pattern of implementation that was sometimes more political than technical, with little state governance.
Regarding the management entity of the UPA, most of the UPA in the Brazilian regions are under municipal management, except in the Northeast. Therefore, the main institutional player of the management was the Municipal Health Department, as already evidenced 3 . The explanation has been agreements made directly between municipalities and the Ministry of Health. "We have UPA that come from the network and UPA that the mayor goes to the MH and says: 'I want an UPA' and the UPA comes. " (E15) Even the planned UPA are often not under state management by manager choice. "The state wants to assume what is from the state, the hospitals. " (E15) Box 2. Analysis of the structuring types and action of federative entities according to the scenario found and the statements of managers. There was report of transfer of management from the municipality to the state. One explanation was the "bankruptcy" of the municipalities in assuming the management of the UPA, being necessary to return these units to state management.

Category of the structuring theory Scenario found Statements of managers
We could see the predominance of UPA located in the countryside (Table 2). A national study 3 indicates the predominance of UPA in populous municipalities. With the current concentration in the countryside, we can infer a change in this concentration of UPA in large municipalities. The North and Midwest regions have more balance in UPA located in the capital and in the countryside. Most UPA are size III, but there is a national balance between sizes. The North region has more size II UPA, and the Northeast has more size I UPA (Table 2).
Regionalization is another fundamental strategy for the countryside where the municipalities do not meet the population criterion of 50,000 inhabitants. The sharing of resources has been a challenge. "Regional UPA do not exist because only one municipality receives resources despite the agreements made" (E8).
Another challenge is to reconcile this component with those before the UPA, with broad popular acceptance, regardless of effectiveness. "We have small hospitals in municipalities that should not exist. Maybe turn them into UPA. But they can't close the hospital in the municipality with 10,000 or 20,000 inhabitants" (E22).
The state manager builds his or her urgency plans and the municipal manager autonomously defines the priorities, but he or she suffers political influence in the defense of local interests 17,18 .
States had different opinions on the importance of UPA for their networks, and municipalities made their choice, but they often did not have the resources to maintain it.
Box 3 presents complementary statements on the categories analyzed.

FINAL CONSIDERATIONS
Currently, we face a project of attention to urgencies that seeks to be inclusive; however, its implementation occurred in a fragmented way and with fragile articulation among its components.
The federal normative and financial induction generated contradictory responses, in which some states consider UPA as a priority and others do not. The current number of UPA under construction needs to be reviewed to prevent the increase of UPA that do not work. The 620 UPA under construction should be analyzed in the management spaces provided for in the SUS, as the conducting groups, including the federal entity in the negotiation after planning and review of the expansion.

Illustrative statements Analytical synthesis
Implantation and expansion of emergency care units (UPA) "The value that the ministry of health and the state will give ... this cost is still very high for the municipality" (E13) Federal underfunding "The state is not interested. I think the pre-hospital service it has is enough and one more component, the UPA, will not change much" (E9) Rejection of the state "I see the UPA as an evolved model of the emergency care that had no resolutiveness. From physical space, team qualification, equipment, medication, the UPA model is more appropriate" (E7) Acceptance of the state Structural conditions for care "Well structured, they improved the working conditions of professionals" (E5) "We have UPA that are much better than the urgency of the hospital" (E24) Good structural conditions "Lack of pediatrician, serious problem" (E13) Lack of medical professionals The UPA and other components of the emergency network "In the countryside, small municipalities have 100% of coverage of the FHS, but the doctors attends once, twice a week" (E6) Difficulty in relation to primary care "The state wanted to have ICU beds in the UPA" (E5) Lack of hospital beds "Mirror Hospital -metropolitan reference hospital with SH or philanthropic hospital that integrate each other to define pharmacy, bed rotation, etc." (E11) Strategies to cope with difficulties in hospital management Management of UPA and interaction between federated entities "Regional UPA does not make sense. All municipalities have a small hospital that has to solve more than the UPA. The municipality ceases to invest in a service for it to invest in another regional hospital that does not have great resolutiveness. It would be different if it were a specialized regional hospital that would solve more than the municipality is capable" (E9) Difficulty of agreement between municipalities "The municipality asked for an UPA in 2009, but gave up and returned the appeal. The state is resisting assuming it because the pre-hospital park it has is enough" (E6) Difficulty of agreement between municipality and state FHS: Family Health Strategy; ICU: intensive treatment unit; SH: small hospital