Delivery care by obstetric nurses in maternity hospitals

This study aimed to assess whether nurses’ presence in delivery care in maternity hospitals linked to the Rede Cegonha program promotes access to best obstetric practices during labor and delivery. We conducted an evaluative study in 2017 in all 606 SUS maternity hospitals that joined this strategic policy in all Brazilian states. We collected data from maternity hospital managers and puerperae. The analysis was performed at two levels: hospital with or without a nurse in delivery care; and professionals that attended vaginal delivery, whether doctors or nurses. We used best practices and interventions for vaginal deliveries and cesarean section rates as dependent variables. We included 5.016 subjects for analyses of vaginal deliveries and 9.692 to calculate cesarean section rates. Multiple regressions were adjusted for geographic region, maternity hospital size, and puerperae skin color and parity. Maternity hospitals with nurses in delivery care used more the partograph and less oxytocin, lithotomy, episiotomy, and cesarean section. Deliveries attended by nurses had more frequent use of the partograph and a lower likelihood of lithotomy and episiotomy. The inclusion of nurses in vaginal delivery care has successfully brought women closer to a more physiological and respectful delivery.


Delivery care by obstetric nurses in maternity hospitals linked to the Rede Cegonha, Brazil -2017 introduction
The Brazilian delivery institutionalization process occurred in the mid-twentieth century, with doctor-centered care and removing women's autonomy. Thus, delivery shifted from a natural process to become a pathological event, with increasing interventions 1,2 , including cesarean section, which, in 2009, has become the primary delivery route in the country 3 .
In the 1980s, the Ministry of Health, professionals, and women's organizations started a movement seeking delivery care models to ensure women's right to reproductive planning, quality care during pregnancy, delivery, and the puerperium 4 . Besides safe delivery, with women-centered care, guaranteeing the Unified Health System (SUS) principles, which address the comprehensiveness, universality, equality, and organization of the different maternal and child care levels 5 .
In 2011, the Ministry of Health established the Rede Cegonha (RC) 6 , a public policy that seeks to implement a delivery and birth care model based on humanization principles. The provision of training and qualification of obstetric nurses and midwives was expanded, Normal Birth Centers (CPN) were established, with environments conducive to good obstetric practices, promoting these professionals' inclusion in vaginal delivery care without dystocia.
The strategies adopted by the RC are firmly based on scientific evidence that shows the benefits of delivery care by nurses and midwives, both for women and their children 7,8 , leading to increased satisfaction, without prejudice to perinatal indicators 9 .
This study aimed to assess whether nurses' presence in delivery care in maternity hospitals of the Rede Cegonha promotes pregnant women's access to good obstetric care practices during labor and delivery.

Methods
This paper uses data from the "Evaluation of Delivery and Birth Care in Maternity Hospitals of the Rede Cegonha", whose data collection was based on the Participatory Rapid Estimation (PRE) 10 technique recommended by the Pan American Health Organization (PAHO).
All 606 public and mixed hospitals (private covenanted to the SUS) with an RC action plan in 2015 and represented almost half of the births that year in the country, according to the Live Births Information System (SINASC), were included in the RC assessment.
The fieldwork was performed between December 2016 and October 2017, including the application of five specific electronic forms for each data source, using three different data collection methods: 1-personal interview with key informants: managers, health professionals, and puerperae, to verify their perception of the delivery and birth care and management model. 2-document analysis: to obtain information on the management of services, teamwork processes, and organization of care, check the availability of process indicators and delivery and birth care results; 3-on-site observation, which aimed to evaluate care processes and the conditions of infrastructure, physical plant, equipment, materials, supplies and the obstetric and neonatal bed count in the hospital 11 .
In the RC assessment, all puerperae with delivery during the evaluative study were included, except for those with severe mental disorders, who did not understand Portuguese, deaf, or admitted to the Intermediate Unit and Intensive Care Unit in the postpartum period. The sample was selected in two stages. The first consisted of hospitals, and the second of puerperae. Further details on the study's methodology are available in Vilella et al. 11 .
This study only considered women with term vaginal delivery (≥ 37 gestational weeks) and fetus/newborn with birth weight ≥ 2500g. In the analysis of good practices and interventions concerning the "professional who attended childbirth", women who could not confirm the professional who attended them during delivery and those who referred to professionals other than doctors or nurses were excluded. All births in the studied maternity hospitals were included in the comparative analysis of cesarean sections' frequency.
Two analysis methods were employed. In the first, women were compared considering the delivery model adopted in the maternity hospital as the exposure variable, with or without the nurse/ midwife's direct action in delivery care. In the second, the comparison was made by including the professional who attended the delivery as an exposure variable, whether a doctor or nurse/ midwife.
As outcomes, we included variables related to good practices during labor (provision of liquids and food, moving around, non-pharmacological pain relief methods, partograph use); good practices to the newborn (breastfeeding in the delivery room and immediate skin-to-skin contact); and interventions during labor (IV serum administration, oxytocin, analgesia, and amniotomy) and delivery (lithotomy, Kristeller's maneuver, episiotomy, and cesarean section). For the analysis of the cesarean section, all births that occurred in those maternity hospitals were included. The maternity hospital's size was extracted from the instrument applied to managers, and the other variables from the questionnaire applied to mothers, with "yes" and "no" answer options.
We used Pearson's chi-square test (χ 2 ) to compare the proportions of women's characteristics according to the delivery model adopted at the hospital (with or without nurses in delivery care) and professionals who assisted them during delivery (doctors or nurses). We used two logistic regression models to test the association between the delivery model adopted at the hospital and the professional providing delivery care and the good practices and interventions in assisting labor, delivery, and birth.
The significance level was set at 5%, and we used odds ratio (OR) estimates considering the effect of the sample design. ORs were adjusted for geographic macro-region, location, maternity size, skin color, and parity of the puerperae. For the analyses, we used a calibration factor by the inverse of the probability of inclusion of each puerperae to guarantee that the distribution of the sampled mothers corresponded to the distribution of deliveries in the maternity hospitals analyzed in 2017, considering the cluster effect in all analyses. SPSS 20.0 and Microsoft Excel version 2007 were used in the analyses.
The study was approved by the Human Research Ethics Committee of the Federal University of Maranhão and the Sérgio Arouca National School of Public Health, on December 14, 2016. All women were informed of the informed consent form's content and received a copy if they agreed to participate in the evaluation. Due care was taken to ensure data secrecy and confidentiality.

Results
A total of 10,665 puerperae interviewed in the immediate postpartum period. Five hundred seventy-one maternity hospitals remained after applying the exclusion criteria. A total of 5,016 mothers were included in the comparative analysis according to the maternity delivery model. Concerning the analysis according to the professionals who attended the delivery, women who could not confirm the professionals who attended them (9.2%) and those who reported professionals other than doctors or nurses (0.9%) were also excluded, leaving out 4,504 women. Finally, all 9,692 births in the maternity hospitals studied were included only to calculate the cesarean rate.
The distribution of births attended by nurses in Brazil differs regarding the macro-regions and by location and volume of births in the maternity hospitals. In the maternity hospitals evaluated, 30% of vaginal births were attended by nurses, reaching 46% in the North of the country. On the other hand, the Midwest and South regions had the lowest frequencies, 19.6% and 18.5%. Delivery care by nurses was also more implemented in inland region cities than metropolitan regions and maternity hospitals with ≥ 3,000 births per year, compared to smaller ones (Graph 1).
The similar profile of the women interviewed stands out, whether comparing by type of professional or by the type of maternity delivery model. Approximately 23% of women were teenagers, and 42% had completed high school or more. Only parity distinguished them regarding the type of professional who attended delivery, and primiparous women were the most attended by doctors. Concerning the maternity delivery model, in nurse-assisted deliveries, brown skin color and living with a partner were more frequently reported (Table 1).
Good practices during labor, delivery, and with the newborn were more frequent in the 309 maternity hospitals attended by nurses. There was a more significant provision of liquids and food, more encouragement for the pregnant woman's movement, use of at least one non-pharmacological pain relief method, more important skinto-skin contact between the newborn and the mother, and more breastfeeding in the delivery room (Table 2).
A similar result was found in the comparison by professionals who attended the delivery, with greater adherence to good practices and fewer obstetric interventions by nurses than by doctors. Amniotomy, which was not differentiated by maternity hospital delivery model, was also significantly less adopted by nurses (Table 2). Some interventions remain in the routine of the maternity hospitals studied. However, they were less used in maternity hospitals with direct nurse participation in delivery care. These hospitals had a lower frequency of IV serum administration, oxytocin use, and analgesia during labor. The use of the lithotomy position, Kristeller's maneuver, and episiotomy was also less frequent during delivery. The proportion of cesarean surgery was high in the maternity hospitals evaluated (48.2%). On the other hand, it was lower in those whose model included nurses in vaginal delivery care (Table 2).
In the multiple analysis, maternity hospitals with the inclusion of obstetric nursing in the vaginal delivery care had greater partograph use (OR=1.9 CI 1.3-2.7) and lower use of oxytocin (OR=0.7 CI 0, 5-0.9), lithotomy (OR=0.6 CI 0.4-0.9), and episiotomy (OR=0.7 CI 0.5-0.9) than those without the inclusion of obstetric nursing, even after adjusting for confounding variables. When analyzing all births, cesarean sections were also less performed in these maternity hospitals (OR=0.8 CI 0.7-0.9) ( Table 3).
When considering the type of professional who attended the delivery (doctor or nurse), the result of the multiple analysis indicated a more significant completion of partograph (OR=1.5 CI 1.2-1.9) and less use of lithotomy (OR=0.7 CI 0.5-0.8) and episiotomy (OR=0.4 CI 0.3-0.6) by nurses, a result similar to the comparison according to the maternity delivery model (Table 4).

Discussion
The World Health Organization (WHO) has designated 2020 as the "Year of the Nurse and the Obstetric/Midwife Nurse", a recognition designed to draw the world's attention to nurses/ midwives' essential role in transforming the health of women and babies 12  Almost a third of vaginal deliveries were attended by nurses, in contrast to the 16.2% found in SUS maternity hospitals in the 2011/12 study Nascer no Brasil 13 . Comparing the Rede Cegonha with public hospitals and mixed results from the 2011/12 Nascer no Brasil study, Leal et al. 14 observed a decline in interventions during labor and delivery and an upward trend in good practices, regardless of the care-providing professionals, suggesting a positive effect of this public policy to achieve the expected results, which aim respectful, safe, and dignified delivery/birth based on best scientific evidence.
Despite the advances identified, the rate of vaginal deliveries was low, and obstetric nurses' delivery care was still insufficient. The Brazilian legislation on obstetric nursing provides for care for usual-risk delivery from admission to discharge and the identification of dystocia and decision-making until the transfer of care to the medical professional 15 . These findings reveal barriers and difficulties identifying obstetric nursing, which can be attributed to the still insufficient investment in the training of these professionals, the low hiring by public hospitals, even less by private hospitals, unattractive salaries, obstetricians' resistance to collaborative work, and the federal and regional professional medicine councils' disputes [16][17][18][19] .
The role of the Brazilian nurses/midwives is significantly different from the experiences of European health systems, such as the United Kingdom, the Netherlands, Germany, and other countries such as Australia, where nurses/ midwives play an important role in monitoring pregnancy and delivery, and physicians are only responsible for assisting surgical or complicated deliveries 7,20-24 . The care model proposed by the RC proposes changes in the health care of women and children, assuring good practices and safety in delivery and birth care as a guideline. One of the actions is to encourage greater participation by nurses in prenatal care, delivery, and birth, including funding from the National Residency Program in Obstetric Nursing (PRONAENF) and courses to improve obstetric nurses across the country. These RC guidelines increased care to women by the obstetricians/obstetrical nursing, contributing to a more significant provision of good practices in delivery care, and reduced interventions 6,25 .
The proportion of nurses in conducting vaginal delivery was heterogeneous and depended on the characteristics of maternity hospitals. While having the lowest volume of deliveries in the country, almost half of vaginal deliveries in the Northern region were attended by nurses, which was also observed in the inland region's cities, with almost 40%. These data point out that the likelihood of women being serviced by nurses at the time of delivery increases with distance from large urban centers and capitals. A probable hypothesis for such a finding would be the smaller supply of doctors in these areas.
A study carried out on the distribution of doctors in the country pointed out the North region with the lowest ratio of doctors/inhabitants, 1.12 doctors per 100,000 inhabitants, while this ratio was 2.81 in the Southeast. Several elements are assumed, and the most likely is the attraction exerted by the economic factor 26 . In contrast, in  the Southeast region, nurses' participation in deliveries is growing even with a comprehensive provision of doctors, linked to the obstetric model's change, as recommended by the MS 6 and the WHO 12 .
We found that the percentage of deliveries attended by nurses was higher in larger hospitals. Prioritizing hospitals due to the higher concentration of deliveries has been a practice of public policies established in the last decade by the Ministry of Health in order to reach the largest number of professionals and women by providing support and incentives for the qualification of delivery and birth care, introducing a collaborative delivery model, including obstetric nurses, per the national guidelines 6 . We can also consider these large hospitals' relevance as centers that radiate public policies, as they are regional and SUS reference centers.
A previous study on RC evaluation did not observe a declining rate of cesarean sections in women with public delivery payment, which remained around 43% in the 2011-2017 period 14 .
Although this was not a direct objective of RC, it is an intervention, often unnecessary, with a higher likelihood of complications for women and fetuses 27,28 . On the other hand, we found a lower likelihood of cesarean delivery for women who had their deliveries in maternity hospitals adopting a delivery model with nurses in vaginal delivery care. Betrán et al. 29 show that the excessive use of cesarean sections does not show benefits to women and can increase the risk of harm. They also point out that midwife-spearheaded care has been associated with higher proportions of physiological births, safer results, and lower costs.
Studies on nursing contribution to reducing cesarean rates show results that vary from stable to declining rates [30][31][32] . According to Johantgen et al. 30 , these differences are often attributed to selection bias, assuming that women who "need" cesarean delivery are more likely to be attended by obstetricians. At the same time, those who decide to deliver with fewer interventions are more likely to be attended by nurses or midwives.
In contrast, Edmonds et al. 32 evidenced variations from 8.3% to 48.0% in cesarean rates among 3,031 women cared for by nurses and doctors in labor and delivery. These variations were not associated with the studied maternal conditions (nulliparous, term, single fetus, and cephalic presentation), but the care provided to women by nurses. The authors considered this finding a potential nursing factor, which can influence the probability of cesarean section.
Moreover, a study conducted by Hildingsson et al. 33 observed that the presence of a qualified nurse reduces anxiety and fear during the delivery and birth process. We believe that studies on this theme should be encouraged, providing better evidence on this care's effects. In this study, we explain the benefits of nurses' presence in delivery care, showing the advantages of a delivery model including these professionals.
It is noteworthy that 9.2% of the puerperae do not know which professionals attended their vaginal delivery. Introducing oneself to the parturient, informing one's name and function, is one of the items highlighted in the RC "obstetrics reception" guideline. Only 58.9% of the respondents answered that all or most of the professionals introduced themselves to them during hospitalization, which classified the set of maternity hospitals as "partially adequate", according to the RC 34 evaluation matrix.
Other good practices in labor and delivery care were also essential. The use of the partograph at the individual and maternity hospital level was greater with nurses, which is an indicator of the provision of care and recording of their practices because it provides more security to parturients, promotes the exchange of information between professionals, and records care history. In this sense, it is a technological device to strengthen information systems in maternity hospitals and at the national level, which allows monitoring and evaluating the RC and the progress of sexual and reproductive health care of women in the country, regarding Goal 3.7 of the SDG agenda 35 .
The RC consists of guidelines that encourage the adoption of good practices in labor and delivery, such as non-pharmacological pain relief methods, breastfeeding, and the strengthening of the mother-baby bond. Although it lost statistical significance after adjusting for confounding variables, we highlight a likely more effective breastfeeding practice in the delivery room and skinto-skin contact in hospitals with nurses working in the birth process, confirming these professionals' efforts to comply with RC guidelines.
Concerning avoidable practices in labor and delivery, oxytocin use as a routine, the lithotomic position, and episiotomy were less performed in nurses' presence in delivery care. This result is consistent with other studies that show fewer of such nurses' interventions [22][23][24]30 . However, the continuation of these practices has unfavorable implications and reinforces the need to improve hospital management to guarantee the safety and quality of maternal and neonatal care in maternity hospitals, ensuring equal access for women and improving the economic impact on the health system [36][37][38][39] .
There is an urgent need to strengthen and invest in the health system and policies aimed at women that support health and reproductive rights and expand the quality of maternal and neonatal care, which also includes the valorization of health workers, improved hospital structure and information systems, and a more inclusive relationship, giving women a voice 40 .

Conclusions
The results presented here confirm nursing care as less interventionist in delivery, besides exposing medical professionals' advance of good practices compared to a similar study conducted in Brazil in 2011/12 13 . They suggest a positive influence of the collaborative work between nurses and obstetricians in obtaining better results and on the humanization of care in labor and delivery for usual-risk women and their newborns.
Global strategies point to practice betterment initiatives to improve nurses/midwives' performance through training, education, and formation to influence delivery. However, analyses of these initiatives point to the need to face professional barriers related to gender inequalities and economic, social, and cultural issues regarding recognizing the performance and professionalization of nurses and midwives that provide them with a lower socio-cultural status 7,8,39,41,42 . Collaborations SGN Gama contributed to the concept and the entire process of writing the paper. She collected the bibliography, was the primary writer, especially the discussion, and did the last review. EF Viellas and CKRT Silva participated in the paper's design and methodology, data analysis and interpretation, and paper writing. ET Medina, AJ Angulo-Tuesta e SD Silva participated in data analysis and interpretation and paper writing. YRP Santos created the variables, ran all the tables, generated the chart, and contributed to the paper's methodology. AP Esteves-Pereira performed a critical review of the entire manuscript, including the method of data analysis. All authors evaluated and approved the final version.