Share on
INTRODUCTION
In recent years, in Italy, there has been an increase in the frequency of births occurring outside the hospital setting (at home or in maternity homes). Despite the national prevalence remaining low, the trend in 2020 and 2021 has shown an increase which may be due to the COVID-19 pandemic [1]. Between 2018 and 2022, the rate of out-of-hospital births increased by 87.5% (from 0.08% to 0.15%) [2, 3]. Specifically, the latest available data show that 0.04% of births occurred in “other” locations, with 0.11% occurring at home, with significant regional variations [3].
Globally, the practice of delivering out-of-hospital varies widely. The number of home births in the United States rose from 1.26% in 2020 to 1.41% in 2021 [4]. In Australia, 2% of births occurred in birth centres, 0.5% at home, and 0.7% in other contexts [5]. In Japan, 0.6% of births occurred in birth centres and 0.2% at home. In New Zealand, out-of-hospital births accounted for 3.4% of all births nationwide [6].
Among the European countries the Netherlands is an exception, with approximately 20% of births occurring at home, while in England, 63% of women give birth in midwifery-led birth centres (located either inside or outside hospitals) and 3% deliver at home. The prevalence of home births was 2.4% in Wales in 2019 and 2.2% in Iceland in 2012 [6]. Home births made up 1.4% of births in Scotland [7], between 1% and 2% in Denmark, 0.7 per 1,000 (0.07%) in Sweden, 1.5 per 1,000 (0.15%) in Norway, and less than 1% in Belgium in 2017. In Germany, approximately 2% of births took place outside of a hospital setting in 2010, while in France, the percentage was less than 1% in 2016 [6].
According to the World Health Organization (WHO), every woman should give birth in a place that is perceived as safe and capable of providing adequate, respectful, and timely care. For women with low-risk pregnancies, this place can be not only a hospital obstetrical unit (OU) but also a midwifery clinic, a birth centre, or their own home [8]. Additionally, the National Institute for Health and Care Excellence (NICE) [9] supports the idea that women who meet the criteria for low-risk pregnancies should be supported and encouraged to give birth at home or in midwifery-led birth centres. These settings offer a higher likelihood of spontaneous birth without unnecessary invasive interventions compared to hospitals, which are associated with a higher likelihood of interventions such as vacuum or forceps-assisted births, episiotomies, and caesarean sections (CS). Furthermore, there are no significant differences in neonatal health outcomes related to the planned place of birth in various settings [9]. The literature widely demonstrates that among low-risk pregnant women, the planned choice of birth setting has a minimal impact on adverse perinatal outcomes [10, 11]. Additionally, low-risk women who plan to deliver out-of-hospital are less exposed to invasive interventions (e.g., episiotomy, etc.) and severe morbidity during labour and delivery [12].
The American College of Nurse-Midwives supports home birth as a safe option for women who are essentially healthy at term with a singleton fetus, as the health outcomes for women and newborns from studies have been comparable or better than births that occur in the hospital [13]. Similar statements are supported by the Royal College of Obstetricians and Gynaecologists, and the Royal College of Midwives in a joint statement, emphasising a higher level of satisfaction with the birth experience for those who have chosen home birth [14]. The Australian College of Midwives also advocates home birth as a safe option for women with uncomplicated pregnancies [15].
In Italy, in addition to hospital OU delivery rooms, it is possible to choose to give birth in midwifery units (MUs) located inside or outside of hospitals, at private freestanding midwifery units (FMUs or maternity homes) managed by independent midwives, as well as at home. The services offered by MUs are part of the National Health Service (NHS), following a model similar to that adopted in Great Britain. In Italy, the established MUs are concentrated in the central-northern regions. There are three midwifery units inside hospitals recognised by the NHS located in Turin, Florence and Genoa; however, the Genoa unit has been closed since 2020. Additionally, various groups of independent midwives are actively involved in assisting births at private maternity homes and/or at home, with the costs being borne by the couple [12].
To our knowledge, there is only one Italian study on the midwifery care and maternal-neonatal outcomes of out-of-hospital births. The study included 1,099 women who chose to give birth outside of the hospital between 2014 and 2018, and illustrates that out-of-hospital birth can be considered a safe choice for women with low-risk pregnancies [16]. However, it is important that this option is adequately planned, monitored, regulated, and evaluated through healthcare control systems to ensure safe and effective care for both the mother and the newborn, as is done in hospitals [16].
Therefore, due to a lack of sound evidence from Italy and with the intention of implementing a surveillance system in the near future, a quantitative study was conducted with the aim of describing the characteristics of midwifery care and perinatal and maternal outcomes of women who gave birth in an out-of-hospital setting in the Lazio Region, Italy.
METHODS
Design
A cross-sectional study was carried out.
Participants and settings
Women eligible for out-of-hospital birth, in general, must meet the criteria for low obstetric risk. The women taken into consideration are all those who have sought the out-of-hospital birth at maternity homes and/or independent midwives. The professionals follow the selection criteria outlined in the national coordination guidelines and in the care protocol adopted by Lazio Region [17, 18]. Therefore, the study population was based on the physiological requirements, to be identified and met during pregnancy (e.g., absence of maternal and/or fetal pathology, absence of pregnancy-related conditions, single pregnancy, etc.) and at the onset of labor (single cephalic presentation, full-term pregnancy, normal fetal weight, premature rupture of the membranes <24 hours, regular fetal heart rate, normal maternal blood pressure, and temperature) [17, 18].
The study population was examined in the three-year period 2019-2021.
The inclusion criteria were:
- women who completed the process of planning an out-of-hospital birth (n=542).
- The exclusion criteria were:
- women who decided not to have an out-of-hospital birth;
- women who had planned an out-of-hospital birth but changed their decision during pregnancy for all the analyses (n=36);
- women who required transfer to a hospital before delivery for the analysis of out-of-hospital birth, neonatal outcomes, and the postpartum period (n=64).
The final total sample included 478 women who gave birth at home or in a maternity home (Figure 1).
Study instrument and data collection
An online form, including literature-based variables was built by three trained researcher midwives and was approved by a committee of independent midwives of Lazio Region which supported and disseminated the project.
The form was divided into four sections: socio-demographic characteristics, medical history and pregnancy, labour and delivery, and neonatal outcomes and postpartum period.
Thirty independent midwives, working in home or maternity home care settings in the Lazio Region, were trained to fill in the online form. Local coordination was established to ensure comprehensive data collection by the independent midwives.
Data analysis
Descriptive and inferential analyses were performed. The frequency and percentage of socio-demographic characteristics, medical history and prenatal care were determined.
A logistic regression was performed to analyze the factors associated with the continuation of out-of-hospital birth compared to those who had to discontinue the process.
Adjusted odds ratio (adjOR) and 95% confidence intervals (CI) were calculated.
The analyses were performed using STATA version 17 (StataCorp LP, College Station, TX, USA).
Ethical considerations
The design of the study was approved by the Ethics Committee of UniCamillus – Saint Camillus International University of Health and Medical Sciences, Rome, Italy. The midwives and the mothers were informed and agreed to the use of their anonymous data in accordance with Italian and European data protection legislation. No funding was received for this study. This manuscript was prepared in accordance with STROBE guidelines for observational studies [19].
RESULTS
In the years of the study period, 124 (21.5%) women in 2019, 174 (30.1%) in 2020, and 280 (48.4%) in 2021, totaling 578 women initiated the out-of-hospital birth process. Of these, 36 decided not to continue the process due to the onset of medical conditions that did not allow continuation; therefore, they were not included in the study. Among the remaining 542 women who initiated labour in the out-of-hospital setting, 64 were transferred to hospital due to the onset of obstetric risk conditions. A total of 478 women experienced labour and delivery in the out-of-hospital setting (Figure 1). Of these, 10 were transferred to hospital after completing the delivery.
Social and obstetric maternal characteristics (n=542 women)
Of the 542 women who initiated labour out-of-hospital, 87.0% were Italian, 70.5% had a university degree or higher education, and 79.3% were employed. The average age of the sample was 34.0 (standard deviation, SD±4.4) years.
About 61.3% of the women were multiparous, of whom 22.2% had chosen an out-of-hospital birth for their previous deliveries, and 10.8% had a history of CS. The median gestational age at the beginning of care by independent midwives was 26.0 (IQR 16-32) weeks, and the average number of visits during pregnancy with the chosen midwives was 5.9 (SD±2.9).
Birth settings and outcomes (n=542 women)
The average gestational age was 39.6 (SD±1.1) weeks. Transfer to a hospital or clinic pre-labour or during the first stage of labour was necessary in 64 cases (Table 1). In another 4 cases (0.7%), ambulance intervention was required, but no transfer was needed.
The principal reasons for transfer were: prolonged first or second stage of labour (n=25), fetal complications (n=14), and post-partum complications (n=8). After the transfer, the independent midwife stayed with the woman in 45 cases.
The prevalence of vaginal delivery was 94.1%. Among the 34 women who had a previous CS and attempted a vaginal birth (VBACs), 30 gave birth vaginally, 3 had a repeat CS, and one had an operative birth following transfer to the hospital.
Out-of-hospital births: characteristics (n=478 women)
Of the 478 women who gave birth in an out-of-hospital setting, 45.3% delivered in their own home and 54.7% in a maternity home. About 98% of the women had at least one support companion of the mother’s choice with them during labour and birth.
In 73.4% of cases women used a birthing pool during labour, 48.7% of whom also delivered in the birthing pool (Table 2). In 3.6% of cases, the placenta was subsequently delivered in the pool. Almost 95% of the women gave birth in a position of their own free choice. The perineum remained intact in 38.9% of women. An episiotomy was performed in one of 478 cases (0.2%). Among the spontaneous tears, 43.5% were first-degree tears, and 17.0% were second-degree tears. There were 2 cases (0.4%) of third-degree tears. Intrapartum emergencies occurred in 85 out of the 478 women who laboured and delivered in an out-of-hospital setting (17.8%). The most frequent emergency was maternal haemorrhage (n=47; 9.8%), including 38 cases (80.9%) with blood loss between 500 and 1,000 ml and 9 cases (19.1%) with blood loss over 1,000 ml. Other emergencies included 18 cases (3.8%) of meconium-stained amniotic fluid, 14 cases (2.9%) of shoulder dystocia, 9 cases (1.9%) of abnormal presentation, 3 cases (0.6%) of fetal heart rate abnormalities, 6 cases (1.2%) of uterine atony, 1 case (0.2%) of placental abruption, 1 case (0.2%) of tight nuchal cord, and 1 case (0.2%) of umbilical cord rupture.
Neonatal characteristics of the out-of-hospital births (n=478 babies)
The average birth weight was 3,429.3 (SD±4.1) g. The one minute Apgar score was 7 or above in 99.2% of cases (Table 2).
Cord clamping occurred after placental delivery in 95.6% of newborns, including 8.2% of cases where lotus birth was chosen. There were 2 cases (0.4%) of immediate cord clamping, one involving a newborn requiring ventilation and one involving cord rupture immediately after delivery.
No airway suctioning was performed for 462 newborns (96.7%) and no tactile stimulation was needed for 464 newborns (97.1%). However, 13 newborns required bag and mask ventilation despite having an Apgar score of 7 or above at 5 minutes. One case required neonatal cardiopulmonary resuscitation due to shoulder dystocia, and was followed by transfer to hospital.
Immediate skin-to-skin contact after birth occurred in 99.8% of mother-baby pairs. The duration of skin-to-skin contact was over 2 hours in 92.3% of cases. The first breastfeeding latch occurred within 2 hours of birth in 96.7% of newborns. Most of newborns (95.2%) had their first bath after 24 hours from birth. In most cases (71%), the first paediatric visit occurred after 6 hours from birth.
Postnatal period (n=478 women)
Seven women were transferred to a hospital or clinic during third stage, due to a retained placenta, and 3 women were transferred in the postpartum period (Table 1).
The average number of postpartum visits by independent midwife was 6.0 (SD±2.0), with the last visit occurring between 30 and 42 days after birth in 40.1% of cases.
149 women (31.1%) experienced post-birth issues that required attention. The most common difficulties were related to breastfeeding (20.9%). In 5.4% of cases problems with lacerations and perineal healing were reported, and 2.7% of women had issues of incontinence (Table 2).
About 96% of women were exclusively breastfeeding at one week post-delivery, and 94.6% were exclusively breastfeeding at one month.
Logistic regression model
The logistic regression model showed that, compared to primiparous, multiparous who had a previous spontaneous hospital birth (adjOR 9.7; CI 95% 4.33-21.68 P<0.001) and women who had a previous out-of-hospital birth (adjOR 24.2; CI 95% 3.23-181.48 P=0.002) were associated with the continuation of out-of-hospital birth. The socio-demographic characteristics (education, citizenship, age, marital status) and having had a previous caesarean section were not significantly associated with out-of-hospital birth (Table 3).
DISCUSSION
This is the second study reporting data on out-of-hospital births in Italy and the first that is representative of the Lazio Region. Similar to the first Italian study, the results appear representative for the low-risk pregnant women and confirm that the choice of giving birth outside the hospital is safe for this population [16].
Our regional data show a progressive annual increase in the number of out-of-hospital births, probably due to the COVID-19 pandemic. In fact, many studies have reported that the pandemic has led women to change the chosen place of delivery to ensure their desired birthing experience and to avoid the risk of exposure to COVID-19 during a hospital stay [20-22].
The average age, the high level of education (bachelor’s degree and higher), and occupational status are in line with the national study on out-of-hospital births [16] and the international literature [23, 24].
Consistent with the literature, multiparity is associated with out-of-hospital births [16]. Among multiparous women who chose the out-of-hospital setting for previous births, it is common to observe a continuation of this choice for subsequent deliveries. Additionally, based on national data from 2014-2018, 4.3% of women birthing out-of-hospital had a previous caesarean section [16], while in our study this frequency appears to be slightly more than twice as high, in line with the literature [25].
Regardless of the care setting, maternity services and healthcare professionals are responsible for ensuring that all care procedures are evidence-based, safe, and of high quality [26]. In the case of planned out-of-hospital births, ensuring a prompt transfer from home to the hospital represents a good practice when a potential risk is identified. Promoting high quality maternity care involves guaranteeing the transfer from home to hospital, which requires interprofessional collaboration, effective communication, and the implementation of standardised procedures to ensure personalised and safe care for the mother and baby [27]. In our study, the prevalence of transfers in the Lazio Region is 13.6%, in line with the literature [28, 29].
Consistent with previous studies [30-34], giving birth in an out-of-hospital setting is associated with a lower number of medical interventions (e.g., inductions, episiotomies, operative vaginal deliveries with vacuum extraction, and caesarean sections) and a lower rate of maternal complications (less instances of third and fourth-degree lacerations, haemorrhage, fever). In our study there were no cases of induced labour and a low number of medical interventions, while national data on hospital births, show a prevalence of 32.1% for induced labour, 4.24% for operative vaginal deliveries with vacuum extraction, and 31% for caesarean sections [3]. In this comparison, it is essential to emphasize how our sample is highly selected, as it comprises women with physiological pregnancies, a crucial requirement for giving birth in an out-of-hospital setting. Conversely, the hospital serves as the only alternative for women with high-risk conditions, despite the majority of pregnancies being physiological.
Regarding midwifery care, satisfactory results emerged in our study with respect to the utilisation of upright positions during birth, delayed cord clamping, the presence of a support person, immediate skin-to-skin contact to promote breastfeeding and prevent hypothermia, delayed bathing for 24 hours after birth, and continuity of care postpartum, all of which is in line with WHO recommendations on “intrapartum care for a positive childbirth experience” [35, 36]. In particular, as regard breastfeeding and companion of the mother’s choice, our study guaranteed this best practice compared to what happened in hospital setting during COVID-19 pandemic [37].
The practice of planned out-of-hospital birth requires unique skills, knowledge and methods related to the birthing process and midwifery. In out-of-hospital settings, birth occurs with respect to physiology, without routine interference, and with careful use of technologies appropriate for such an environment [38]. This aligns perfectly with the preferences and motivations that drive low-risk women to choose this setting [23]. Furthermore, the midwifery model of care leads to greater satisfaction among women [39-41], and its continuity during the postpartum period could bring long-term benefits, including increased breastfeeding [42, 43]. Our study shows that the prevalence of exclusive breastfeeding remains high at one week and one month after birth. In Italy, the rates of breastfeeding initiation and continuation vary by region, depending on education and socio-economic conditions [44]. Moreover, the use of breast milk substitutes is a widespread practice during the hospital stay as well as their prescription at discharge [45]. In the Italian context where breastfeeding prevalence is low, our results must be taken into consideration, as they show how respect for birth physiology and adoption of the midwifery model of care play a pivotal role in strengthening the breastfeeding skills of women and babies [45, 46]. In fact, regardless of the place of birth, competent health personnel able to meet women’s needs is a key factor for the satisfaction of the childbirth experience [47].
The strength of the study is the completeness and accuracy of data collection, covering 95% of out-of-hospital births in Lazio Region. Moreover, all groups of midwives involved in out-of-hospital assistance in the Lazio Region participated in this study. Trained midwives collected data using a coded questionnaire, ensuring data quality and completeness.
This descriptive study estimated how many women, having initiated the pathway, decide to forego it, and how many women are transferred due to problems during pregnancy, labor and delivery. These are useful indicators for comparing experiences. All these aspects can be a starting point for establishing an epidemiological surveillance system.
This study also has some limitations. First, the findings are not generalisable at a national level due to the involvement of only one region. Data collection relied on local coordination for completing the records due to the absence of a structured monitoring system. Possible implications for future research would include additional indicators in the medium to long term postpartum period to study the outcomes in terms of public health. Finally, it would be useful to replicate and implement the present investigation in other Italian regions in order to achieve a comprehensive national epidemiological overview.
CONCLUSIONS
This study confirms that a planned out-of-hospital birth is a safe, adequate, appropriate, and effective choice for low-risk pregnant women. However, it is essential to emphasise that the safety, quality, and appropriateness of such births depend on a regulatory system that includes careful planning and appropriate monitoring. Additionally, in view of the possible increase of out-of-hospital births, and in order to maintain the high standards of safety and quality, it would be desirable to implement a national surveillance system for monitoring maternal and neonatal outcomes of out-of-hospital births. A number of countries have adopted specific tools to support the continuous improvement and delivery of quality maternity care, for example the creation and maintenance of datasets such as the MANA Statistic Project (Midwives Alliance North America) in the United States [12].
Figures and tables
Figure 1. Flow diagram of study population in the Lazio Region, Italy.
| BIRTH SETTINGS AND OUTCOMES | n | % |
|---|---|---|
| Out-of-hospital birth | ||
| Yes | 478 | 88.2 |
| No | 64 | 11.8 |
| Place of birth | ||
| Home | 216 | 39.9 |
| Maternity house | 262 | 48.3 |
| Hospital | 61 | 11.3 |
| Clinic | 3 | 0.6 |
| Mode of delivery | ||
| Spontaneous vaginal birth | 510 | 94.1 |
| Operative delivery | 6 | 1.1 |
| Caesarean section | 26 | 4.8 |
| Mode of delivery for attempted VBACs | ||
| Spontaneous vaginal birth | 30 | 88.2 |
| Operative delivery | 1 | 2.9 |
| Caesarean section | 3 | 8.8 |
| Gestational age at birth (weeks) | mean 39.6 (SD±1.1) | |
| Transfer to hospital or clinic | ||
| Yes | 74 | 13.6 |
| No | 464 | 85.6 |
| No, but ambulance request | 4 | 0.7 |
| Timing of transfer | ||
| Pre-labour | 10 | 13.5 |
| Prodromal stage | 14 | 18.9 |
| Active labour | 32 | 43.2 |
| Second stage | 8 | 10.8 |
| Third stage | 7 | 9.5 |
| Immediate postpartum | 3 | 4.1 |
| Reasons for transfer | ||
| Prolonged 1st or 2nd stage of labour | 25 | 34.7 |
| Fetal complications | 14 | 19.4 |
| Post-partum complications | 8 | 11.1 |
| Other | 25 | 34.8 |
| Presence of the freelance midwife in the hospital | ||
| Yes | 45 | 60.8 |
| No | 29 | 39.2 |
| SD: standard deviation. | ||
| BIRTH CHARACTERISTICS | n | % |
|---|---|---|
| Immersion in water | ||
| Labour | 351 | 73.4 |
| Delivery | 233 | 48.7 |
| Placenta delivery | 17 | 3.6 |
| Delivery position | ||
| Reclining | 26 | 5.4 |
| On all fours | 168 | 35.2 |
| Squatting | 156 | 32.6 |
| Other | 128 | 26.8 |
| Perineal outcomes | ||
| Intact perineum | 186 | 38.9 |
| 1st degree tear | 208 | 43.5 |
| 2nd degree tear | 81 | 17.0 |
| 3rd degree tear | 2 | 0.4 |
| 4th degree tear | 0 | 0.0 |
| Episiotomy | 1 | 0.2 |
| Intrapartum emergencies | ||
| Abnormal fetal heart rate | 3 | 0.6 |
| Malpresentation | 9 | 1.9 |
| Shoulder dystocia | 14 | 2.9 |
| Umbilical cord prolapse | 0 | 0.0 |
| Meconium-stained amniotic fluid | 18 | 3.8 |
| Haemorrhage | 47 | 9.8 |
| >500cc to 1,000cc | 38 | 80.9 |
| >1,000cc | 9 | 19.1 |
| Other | 9 | 1.9 |
| NEONATAL CHARACTERISTICS | ||
| Apgar score at 1 minute | mean 9 (SD±0.8) | |
| ≥7 | 474 | 99.2 |
| Apgar score at 5 minutes | mean 9.9 (SD±0.4) | |
| ≥7 | 477 | 99.8 |
| ≥9 | 471 | 98.6 |
| Umbilical cord clamping | ||
| Immediate | 2 | 0.4 |
| After 1 minute, before placenta delivery | 19 | 4.0 |
| After placenta delivery, within 12 hours | 342 | 71.6 |
| After 12 hours, before spontaneous detachment | 76 | 15.9 |
| Lotus birth | 39 | 8.2 |
| Stimulation/resuscitation interventions on the newborn | ||
| None, or tactile stimulation only (rubbing, drying, massage) | 464 | 97.1 |
| Resuscitation with bag and mask ventilation or laryngeal mask | 13 | 2.7 |
| Advanced resuscitation maneuvers (intubation, chest compressions, medications) | 1 | 0.2 |
| Skin to skin | ||
| No | 1 | 0.2 |
| ≤2 hours | 36 | 7.5 |
| >2 hours | 441 | 92.3 |
| Early initiation of breastfeeding | ||
| Within 30 minutes of birth | 305 | 63.8 |
| After 30 minutes and within 2 hours of birth | 157 | 32.9 |
| After 2 hours | 12 | 2.5 |
| Non-breastfeeding | 4 | 0.8 |
| POSTNATAL PERIOD | ||
| Concerns addressed | ||
| Breastfeeding | 100 | 20.9 |
| Laceration/suturing | 26 | 5.4 |
| Urinary incontinence | 13 | 2.7 |
| Neonatal care | 8 | 1.7 |
| Other | 34 | 7.1 |
| Breastfeeding at 7 days | ||
| Exclusive | 459 | 96.0 |
| Predominant | 9 | 1.9 |
| Complementary | 8 | 1.7 |
| Non-breastfeeding - formula feeding only | 2 | 0.4 |
| Breastfeeding at 1 month | ||
| Exclusive | 452 | 94.6 |
| Predominant | 11 | 2.3 |
| Complementary | 9 | 1.9 |
| Non-breastfeeding – formula feeding only | 6 | 1.3 |
| SD: standard deviation. | ||
| Variable | % (n/tot) | adjOR* | CI (95%) | P |
|---|---|---|---|---|
| Primiparous | 76.7 (161/210) | 1 | ||
| Multiparous with previous in hospital caesarean section | 80.7 (25/31) | 1.5 | 0.57-4.06 | 0.4 |
| Multiparous with previous vaginal hospital birth | 96.5 (219/227) | 9.7 | 4.33-21.68 | 0.000 |
| Multiparous with previous out-of-hospital birth | 98.7 (73/74) | 24.2 | 3.23-181.48 | 0.002 |
| *Adjusted for education, citizenship, age, marital status. | ||||
| adjOR: adjusted odds ratio. CI: confidence interval. | ||||
PDF
