Labor and childbirth care in maternity facilities in Brazil’s North and Northeast regions: perceptions of the evaluators of the Stork Network Program

This article eevaluates delivery and birth care practices in maternity facilities in Brazil’s North and Northeast regions. We conducted a qualitative evaluation of 91 facilities in the North and 181 facilities in the Northeast. The data was collected using systematic observation by a team of 44 previously trained evaluators and recorded in a field diary. A thematic analysis of the collected data was performed, resulting in three core themes: challenges of collegial management; challenges for coping with obstetric violence; and the potential of the evaluation process for driving change. Advances were made in the implementation of good labor and childbirth care practices; however, some maternity facilities still reproduce hierarchical models without spaces for collegial management and accounts of obstetric violence were common. Health professionals used the presence of risk to justify the low level of adoption of good practices. However, the findings reveal progress towards the humanization of care. The results also show the potential of the evaluation process for driving change. Although progress has been made towards the adoption of the good practices recommended by the Stork Network Program both in the area of management and care delivery, many challenges remain in view of the dominance of a hierarchical management model associated with an interventionist approach to health care.


introduction
Humanized childbirth care is one of the pillars of the Programa Rede Cegonha (RC) or Stork Network Program, created in 2011 by the Ministry of Health with the aim of reducing maternal and neonatal mortality. 1,2 The RC shifts away from the logic of the hegemonic model of maternal care, characterized by high rates of cesarean sections, the medicalization of delivery and birth, unsafe abortion, and elevated rates of perinatal morbidity and mortality. The program aims to promote healthy labor and childbirth through the provision of women-centered care that guarantees privacy and autonomy and avoids unnecessary interventions. 3 These changes involve aspects related to work processes and more equal relations between managers and care staff, professional groups, and health professionals and patients. 4 However, the implementation of this model faces a number of challenges stemming from an organizational structure and logic that is centered on biomedical knowledge and power and hierarchy within and between professional groups and fails to recognize women's autonomy and subjectivity. 5 One of the key challenges of the RC is to combine strategies designed to promote changes in care practices with the systematic monitoring and evaluation of the country's maternity facilities based on the goals and guidelines of the RC. 1 To this end, between 2013 and 2015, the Ministry of Health conducted the first RC evaluation cycle including all maternity facilities encompassed by regional action plans 6 , followed by a second cycle undertaken between 2016 and 2017. 7,8 This study draws on the results of the second cycle.
Considering the deep inequalities in health service coverage and access in the country's North and Northeast regions and, consequently, the persistently high rates of maternal and infant mortality in comparison to the South and Southeast, the efforts of the RC in these regions have been more effective. Between 2010 and 2015 (one year before and five years after the creation of the RC), Brazil saw a reduction of 11.6% and 15.6% in the maternal mortality ratio and infant mortality rate, respectively. 9 This article evaluates delivery and birth care practices in maternity facilities located in Brazil's North and Northeast regions based on the perceptions of the RC evaluators in order to gain a deeper understanding of the challenges faced by the country's maternal health services.

Method
We conducted a qualitative evaluation of maternal health services as part of the project "Evaluation of labor and childbirth care management in Brazil's Unified Health System (SUS)". 7,8 The evaluation included the following types of public and private facilities (hospitals under contract with the SUS) located in health regions in the North and Northeast: 1) facilities performing at least 500 deliveries a year regardless of whether or not they were receiving financial incentives from the RC (n=245); 2) facilities performing less than 500 deliveries a year receiving financial incentives from the RC (n=15); and 3) facilities that do not perform deliveries, but receive funding for neonatal units (n=12). Eligible facilities were identified using the latest available data (2015) from the SUS's Hospital Information System.
The data were collected by a previously trained team of 44 evaluators (28 in the Northeast and 16 in the North) and 10 supervisors made up of health professionals with experience in maternal and infant health care and/or management. The supervisors provided support to the evaluators and coordinated the data collection process in conjunction with local and regional health managers.
The data were collected between December 2016 and June 2017. In each facility, on the first day of the evaluation, the evaluation team held a meeting with local and regional health managers (local government health managers, directors and heads of obstetric and neonatal services, and representatives from the health regions) to explain the aims of the evaluation and methods used. The evaluators worked in pairs and spent three days in each facility in the Northeast and five days in the North.
The data collection method adopted by this study was systematic observation. Observations were made during the evaluators' stays in the facilities based on a guide designed to capture the different perspectives of the subjects, including ambiguities and tension.
The guide contained questions related to the following aspects of labor and childbirth care practices: 1) general information about the ma-ternity facility; 2) use of protocols and indicators; 3) maternal and newborn care; 4) quality of information provision and welcoming of mothers and companions; 5) managers' stance towards the institutional model of labor and childbirth care and adoption of good practices; 6) staff participation in institutionalized or informal decision-making spaces; and 7) manager-care staff communication in relation to work processes and working conditions. All observations were recorded in a field diary.
We conducted a thematic analysis of the collected data 10 drawing on theoretical frames of reference for work processes, the humanization of care and implementation of good obstetric and neonatal practices. We addressed power relations, ways of organizing work process and working conditions. This study was approved by the Maranhão Federal University and Sergio Arouca National School of Public Health human research ethics committees (CAAE 56389713.5.3001.5240). The health facilities were identified using a code consisting of the letter N (North) or NE (Northeast), followed by the state and facility number.

results and discussion
The analysis of the field diaries generated three core themes: 1) Challenges of collegial management; 2) Challenges for coping with obstetric violence; and 3) The potential of the evaluation process for driving change.

challenges of collegial management
Our findings show that some facilities had implemented or were in the process of implementing the device "collegial management" and had shared decision-making spaces, revealing important progress towards making the changes to the management model envisioned by the RC.
"The institution has shared management spaces, where cases and work routines are discussed with the aim of improving care delivery. Cases are discussed to evaluate conduct. Staff regularly take training and refresher courses relevant to their area". (N3- 6) "There is no collegial management body, but it is in the process of organization. However, conversation circles are held on a monthly basis where all professional staff are invited and generally attend. There are training courses and professional staff always seek training opportunities, since they are encouraged and committed to improving work processes". (N6- 7) Weaknesses in collegial management processes were highlighted in the first RC evaluation cycle. Although spaces for communication such as steering groups, collegial management bodies and forums were recognized as mechanisms that strengthen shared decision-making, few have been put in place so far. In addition, political transition and lack of continuity in state and local government administrations and hospital management were highlighted as setbacks for the RC implementation process. 6 There were also cases of services without collegial management mechanisms and others that had mechanisms in place but reproduced the same centralized management model that fails to promote staff and patient participation. The following accounts illustrate the situations found by the evaluation team: "There are meetings between the hospital managers and unit coordinators to discuss work processes; however they don't include care staff on a regular basis. Only now and again and informally". (NE2- 20) "The decisions taken by the collegial bodies neither involve technicians nor patients. They seem to work by theme, like maternal mortality for example; and the other meetings are held by separate professional groups". (NE2- 23) Despite progress in collegial management, in general, the classical model based on division of labor between managers and workers prevails, separating management from care. In this approach, the planning and the organization of work processes rests with managers 11,12 and care workers are charged with the task of everyday care delivery. Staff, particularly those involved in care delivery, do not participate in decision-making. In addition, each professional group takes care of its own, reinforcing fragmentation and establishing a hierarchy of power.
The debate on the management of work processes in health care is by no means new and highlights the need to question the hegemonic management logic in health services, which restrains inventiveness and strengthens mechanisms of control and subjugation of professionals and patients. 13 The guideline Participatory and Shared Management states that management and care processes are inseparable because the management of work processes in health care cannot be understood as an administrative task separated from care practices. 14 How work processes are organized therefore directly influences care delivery. The subordination of care staff to managers in collegial management bodies jeopardizes communication, turning it into the mere transmission of decisions, reinforcing existing power relations within and between different professional groups.
Collegial management is a humanization device that views the management of work processes as a collective challenge. 13 This device has the potential to produce change, promoting the democratization of services by including new subjects and sharing responsibility for ways of managing and caring. 15 On its own, the implementation of humanization devices does not secure the participation of the different actors involved in the care process. 16 This means that the group that makes up the collegial management body needs to evaluate its own mode of operation and the work processes underpinning day-to-day practices.
The evaluators highlight that communication technologies such as email and WhatsApp® were used as tools to improve communication between managers and care staff.
"In general, we noticed good relations between managers and staff. The professionals made some very positive comments about managers, who are visibly active. Communication takes place openly via WhatsApp groups, email, meetings, training courses etc. Although there are no formal shared management spaces, professionals generally feel that the questions they raise are taken into consideration by managers in decision-making". (NE2- 26) Although WhatsApp® was used to facilitate the shared decision-making, it was also evident that it was used to reproduce traditional institutional dynamics, serving more as a channel to communicate decisions taken by managers than a space for collective decision-making.
"There are meetings between unit coordinators and some information is passed on to the professionals in the WhatsApp groups (mainly to the doctors), but they don't participate and are not briefed about the discussions held in the management meetings". (NE4- 3) Studies highlight that strengthening communication between and within groups is vital to democratizing health services. Effective communication strengthens subjects' capacity to exert mutual influence and assign new meanings to their work and day-to-day practices. Shared organization of work processes requires dialogue and sharing between professionals, leading to quality improvement and a clearer definition of shared responsibilities between professionals and teams. 5 One of the justifications given by managers for the low level of staff participation in meetings is shift patterns. This is a crucial issue to be considered in tackling the organizational structure and logic of hospitals, since, to ensure participatory decision making and shared responsibility for care, staff need to incorporate sharing into their everyday practices. Management and care models will not change unless work processes change, meaning that it is necessary to rethink organization. The horizontalization of care is an issue that needs to be addressed by the creation of referral teams. 17 Another issue highlighted by the evaluators, which is equally important, is staff turnover. In this respect, to promote staff health and well-being, retain health workers and improve the quality of care, it is important to value health work and workers by guaranteeing stable employment relationships and improving working terms and conditions. 18

challenges for coping with Obstetric violence
One of the main focuses of the RC is the promotion of strategies to reduce obstetric violence. An important step forward towards change in maternal and infant health care is the problematization of normalized practices and conduct that are not conducive to the humanization of care.
Misconceptions about obstetric violence, including legal aspects, make the criminalization of this type of violence infeasible, despite evidence of its practice. 19 Internationally, the term obstetric violence is associated with the violation of women's rights, being recognized as an important public health issue by the World Health Organization. 20 Obstetric violence, which includes abusive treatment, disrespect and neglect during childbirth, is widespread in health systems in various countries 21,22 . Studies in Brazil suggest that it is one of the dimensions of institutional violence 23 and gender violence 24 , highlighting that RC is a key strategy for tackling this problem.
This process, which includes acknowledging the problem and promoting changes in deep-rooted practices in health services, is slow and marked by advances and setbacks. The transformation towards good practices coexists with the situations of violence identified by the evaluators in facilities in both the North and Northeast regions.
"There are reports of obstetric violence from puerperal women and you can tell from their accounts that they are scared. Since everyone knows each other and it's the only hospital in the municipality (...), [they] are scared of reprisals". (N4-9) "Some of the puerperal women complained about the treatment. They were mistreated by nursing technicians who complained about their screaming and told them to "shut up" and lay still on the bed". (N4-11) "In cases of laceration or episiotomy, women have walk to the procedure room at the end of the corridor of cubicles improvised for stitching".
The findings also show that the diagnosis of "obstetric risk" justified the low level of adoption of good practices by professionals in referral hospitals for high-risk women. For both managers and staff, high-risk deliveries are seen as a medical procedure.
"There is a belief that high-risk deliveries are not included in the humanization model and all interventions are justified by the fact that it is a high-risk maternity facility". (NE2- 15) According to the health professionals, carrying out interventions that are not explained to or have not been requested by women is justified by the risk-safety principle. Conducting procedures without consent is not construed as violence, but rather a way of ensuring safe delivery. 25 This type of care gives preference to medical hegemony and technical care to the detriment of patients' rights.
Pregnancy risk assessments are conducted to prevent adverse outcomes and its utilization establishes different categories of pregnancies. According to Robles, 26 risk assessment is used to legitimize the use of hard technologies and medical interventions seen as a form of control over disease.
The institutionalization of abuse and disrespect for women's autonomy is normalized to save lives. Studies have shown the existence of mistreatment of women in maternity facilities in different settings. 27,28,29 The over-medicalization of the natural process of birth is an everyday practice, characterizing the obstetric violence, disrespect and abuse brought upon women by health professionals. 28 The idea that giving birth is a risk in itself, and that there is no such thing as a pregnancy without riskhence the term "normal-risk" -may be at the root of these abuses and explain the results found in this study.
Rooted in gender stereotypes, violence against women is present in medical training courses in Brazil and Latin America, which teach painful unnecessary normalized procedures that are incorporated into work processes. 30 According to Diniz (2016), 31 professionals are taught that parturient women do not have the right to make choices and that medical students' educational needs are more important than women's autonomy and integrity, trivializing the violation of women's rights, which is a hidden part of the curriculum.
The evaluators' suggest that in teaching hospitals, including university hospitals, practical classes are often placed before the health needs and rights of women.
"A lot of women reported feeling uncomfortable with the presence of residents and students. The labor and delivery rooms are always full of students. Some behave inappropriately, causing women to suffer". (NE2- 14) "The maternity facility receives students on supervised internships. With regard to medical internships, women undergo unnecessary interventions so that the residents/interns can learn the procedures, leading to an increase in the number of unnecessary episiotomies and cesareans and repetitive vaginal exams". (NE8-8) "Students do vaginal exams one after the other without asking permission. Women feel obliged to be an object of study. One woman said: 'because it's their work and I can't complain'". (NE2- 27) "The hospital receives students and sometime the doctors give classes to the undergraduate students in the conventional intermediate care unit". (NE4- 10) The examples of violence illustrated in the above accounts have been discussed in other studies without explicitly mentioning the term obstetric violence. One study on violence committed by health professionals in maternity facilities shows that, although most care staff acknowledge rude and disrespectful treatment, this behavior is often trivialized and regarded as "joking". The argument used to justify violent conduct such as putdowns, threats and reprimands is the need to use authority to manage day-to-day situations in health services. 28 Health professionals also highlighted that lack of facilities and human resources and overcrowding were other factors responsible for the violation of rights: "There was only one bathroom in the labor room, which was really dirty and had a queue of women in labor waiting to use it. The normal delivery rooms are in the surgical block and after delivery the women lay on stretchers on the two sides of the corridor because there are no beds in rooming-in".  "We can't say there are good practices, the maternity facility is routinely overcrowded. Women have no privacy whatsoever and are not allowed a companion, neither in the labor room nor in the delivery and postpartum rooms". (NE2-27) "There are not enough staff to deliver quality care. Understaffing prevents the adoption of various good practices because professionals have only a small amount of time to see each patient". (NE4-5) "Professionals from different sectors find it difficult to adopt good practices, mainly due to overcrowding and the fast pace of work. Care staff-patient relations are permeated with neglect and violence". (NE2-23) "The maternity facility doesn't allow companions throughout the entire period of labor, justified by overcrowding and lack of privacy". (NE9- 8) These and other factors, such as operating beyond capacity, understaffing and lack of ongoing training, lead to excessive workloads and a deterioration of working conditions.
Overcrowding can also lead to misunderstandings within and between health teams, between care staff and patients, and between patients and family members. This situation affected the quality of care delivered to patients and their families, including the right to a companion of choice, offering non-pharmacological pain relief methods and provision of information on mother and baby health status. These practices are elements of the women-friendly care approach adopted in accredited facilities participating in the Baby-Friendly Hospital Initiative created in 2014. 32 "They (pregnant women) don't receive information about the right to a companion, the ombudsman, or any right whatsoever. And little information is given about their health status. Many violations of women's rights were recorded, stemming from practices that denote power relations that depersonalize women, reducing them to an object of intervention. 25,27,31 In contrast, evaluators also found institutions whose managers and professionals were engaged in the process of implementing changes to promote non-violence and the defense of patients' rights. Within this process, practices that come close to the humanizing praxis coexist with more conventional approaches, expressing the tensions present in manager-care staff and care staff-patient/family relations.
"During the evaluation process, actions that show both management and staff commitment to good labor and childbirth care practices were evident, therefore converging towards the proposal of the Rede Cegonha". (NE2-13) "Companion present 24-hours a day. Non-pharmacological pain relief methods offered. Free to walk around. Drinks and food offered during labor. Skin-to-skin contact immediately after birth is a well-established routine in the maternity facility. (...) Welcoming with risk classification not in place. After admission, women are examined in the labor room without adequate privacy. There seems to be little integration between managers and staff ". (NE3-8) "The adoption of labor and childbirth care practices is already routine, [the staff] are surprised when women don't want [the practices], but always respect the patients' wishes. There are volunteer doulas who also support the implementation of good practices, empowering women during childbirth". (NE9- 14) "Managers are committed to the changes in the labor and childbirth care model". (N1-4) This reality reveals the complexity of the process of change driven by the RC and the need to ensure continuity. Despite significant advances, more investment in continuing professional training and development is required, together with increased dissemination of information to help women understand their rights.

the potential of the evaluation process for driving change
The findings from the second evaluation cycle reveal a number of advances triggered by the results of the first evaluation cycle.
"In the interview, the managers mentioned the previous evaluation [in 2015, during the first cycle], in which the maternity facility obtained various low scores. Thereafter, they sought to improve the indicators". (NE2- 13) These findings demonstrate transformative potential of the RC monitoring process. In the first cycle, the evaluators reported that the filling out of the evaluation instrument prompted discussions about the strengths and weaknesses of the service, leading to the development of actions to improve indicators. 6 One of the aims of the RC evaluation process is precisely to promote the development of a set of actions that foster changes in the health care management model adopted in SUS maternity facilities. 33 In the second cycle, managers were informed previously about the list of protocols and indicators that would be requested. The fact that they knew they were going to be evaluated led them to prepare for the evaluation and this process frequently spurred change.
In many cases, it was evident that the protocols provided to the evaluators had been prepared for the evaluation and were not actually in place in the facility.
"Some staff reported that the protocols were placed in the sectors on the day of our arrival in the hospital". (N4-42) Good practice protocols aim to promote changes in the conduct of health teams, encouraging the adoption of practices based on evidence and group discussions. However, the results show that practices varied from care professional to care professional and shift to shift: "Over the days on which the evaluation was conducted, it was evident that conduct varied according to the professional/shift". (N1-5) "Some inadvisable practices were performed, depending on the doctor on shift ". (NE4- 13) The findings also show that, although indicators were not always used, this evaluation provided managers with the opportunity to discuss and reappraise their use: "Indicators were not used because they didn't realize how they could be used to improve service delivery". (N4-4) "After the document analysis, the managers demonstrated commitment and an understanding of the importance of indicators for care planning". (N4-8) "Few indicators are tracked, but, making the most of the evaluation and the training aspect of the evaluation process, the managers suggested they would monitor the requested indicators". (N4- 29) Tracking indicators is important for the analysis of work process outcomes. It enables health teams to correct the direction and change track in order to improve care quality. Building an evaluative culture that cuts across the planning and management process is a major challenge for today's health systems. 34 The transformative power of evaluation was evident both in situations in which indicators were put together to present to the evaluation team and where the evaluation team helped managers to calculate them, proving to be key to actions directed at improving services.
Another positive element was the meetings held during the evaluation process. In some facilities, the participation of local and state government health managers in the evaluation process resulted in the scheduling of meetings and discussions and stronger support for and commitment to the RC.
"The managers of the maternity facility showed a desire for and commitment to change, reflected in the meetings that this evaluation prompted between service managers and local health managers. Discussions between state and local health managers regarding support for the implementation of the RC across the whole municipality were also scheduled, leading to better coordination between hospital care staff and health teams from the family health program". (NE2-4) "On the course during the evaluation visit, service managers and local government health managers agreed various types of support". (NE2-7) "Some agreements were made that will unfold after the evaluation process". (NE2- 19) The methodological approach adopted in the evaluation of the RC emphasized the inclusion of different actors, including managers at multiple levels. This approach not only validates and legitimizes the process, but also provides an opportunity for sharing and making decisions that have the potential to transform the country's health care management model.
The accounts presented here -produced from interviews with managers, staff and patients and the observation of practices in each facility-provide an evaluative look at the reality of maternity facilities in the North and Northeast of Brazil. The picture of theses settings provided by the descriptions recorded in the field diaries and their subsequent analysis, together with the results of the questionnaires, provide valuable insights into pressing issues facing maternal and infant care in Brazil.
One of the strengths of this qualitative study is that it permitted the analysis of the implementation of the RC in regions with the greatest needs, which have had persistently high rates of maternal and infant morbidity and mortality for centuries. We were unable to find other studies evaluating the management of women's and children's health care across all states in Brazil's North and Northeast regions after the creation of the RC.
One of the limitations of this study is that methodological differences between studies hamper the comparison of the results of the two evaluation cycles.

Final considerations
The findings reveal that progress has been made towards the adoption of the good practices recommended by the Rede Cegonha both in the area of management and care delivery. However, many challenges remain in view of the dominance of a hierarchical management model associated with an interventionist approach to health care. Thus, despite the advances, more investment is needed to promote the humanization of care.
Considering that this study was undertaken during the second RC evaluation cycle, efforts are needed to promote the effective transformation of health care management, especially in the following areas: creation of effective shared management spaces; training of professional staff and health managers; implementation of good practice protocols, and incorporation of evaluation processes into everyday practices in maternity facilities to enable the monitoring of the RC. Moreover, the findings of the first evaluation cycle show that institutional support plays a central role in the implementation of the guidelines of the RC. Strengthening this role is key to the consolidation of the changes needed to improve care quality.
It is important to stress the importance of developing actions in primary care settingsfrom antenatal through to labor and childbirth careto empower women and help them understand and take ownership of their rights. With appropriate training, primary health teams can play an important role in promoting women's protagonism and autonomy.
The results of the RC evaluation process can make a significant contribution to improving the management of maternal and infant health services. Further evaluation cycles should therefore be conducted to permit the ongoing monitoring of care quality.

collaborators
All persons designated as authors (ZC Lamy, LLM Gonçalves, RH Britto, MTSSB Alves, ME Koser, MS Martins, NP Leal, EBAF Thomaz) take public responsibility for the content of this article. We declare that we contributed equally to all stages of this work, including data analysis and interpretation, writing the article and final approval of the version to be submitted.