Homebirth is a topic that often elicits a strong emotional reaction from health professionals and women alike. It is a subject that, after being deeply out of favour, now has a positive evidence base that demonstrates low risk women having their second or subsequent baby at home is as safe and is associated with better maternal outcomes than giving birth in hospital (Birthplace in England Collaborative Group et al, 2011). It also continues to gain support in England from influential national reports, such as Changing Childbirth (Department of Health, 1993) and Better Births (National Maternity Review, 2016). While homebirth may now be an acceptable choice, in 2016, only 2.3% of women in England and Wales pursued this as an option; a figure that is unchanged since 2012 and has in fact reduced since 2.9% of women opted for a homebirth in 2007, the highest rate in 30 years (Office for National Statistics, 2016).
Homebirth has been seen as a middle class aspiration (Kaufert and Robinson, 2004) and some of the highest UK homebirth rates demonstrate this association. The Northampton homebirth team achieved a 9.6% homebirth rate (although at the time there was no birth centre option available) (O'Connell et al, 2012), and South Northamptonshire is among the 20 least deprived local authorities in England (Nove et al, 2008). However, Southwark stands out as having a 6.6% homebirth rate while being one of the top 20 most deprived local authorities in England (Nove et al, 2008). This rate was largely influenced by active promotion by King's College Hospital and in particular by the Albany Midwifery Practice (Homer et al, 2017).
Women give birth within a cultural context and the creation of a new ‘cultural norm’ of birth at home was considered key to increasing the Albany homebirth rate to 43.5% (Reed 2015; Homer et al, 2017). Telling homebirth stories in antenatal groups and through online social networking was believed to build relationships and transform the hospital birthing cultural norm and influence birth choices (Howes, 2013; Fage-Butler, 2017). The support of partner, family and friends is also considered integral to supporting women's choice.
The Birthplace in England study (2011) provided the catalyst for a long desired dedicated homebirth service at Bir mingham Women's and Children's NHS Foundation Trust, which has more than 8000 births per year. The local Clinical Commissioning Group (CCG) secured £750 000 for a Service development and Improvement Project to set up a homebirth service and pilot it over three years. The aim was to increase the homebirth rate to 3% of all births, but also to improve choice and clinical outcomes for low-risk women, expand the capacity in the hospital and to become a self-funding service by the end of the pilot.
Before the setup of the homebirth service, provision for homebirth rested with the four traditional community teams who maintained a 24/7 on-call service for women. This proved unsatisfactory as community midwives were part of the hospital escalation policy and were called into cover the hospital whenever activity was high. This resulted in loss of confidence in the availability of the service and in homebirth midwifery skills and fewer women had their babies at home. In 2013 there were only 23 (0.3%) planned homebirths.
Process
Staffing and training
The official start date of the service was 1 April 2014 although referrals and care commenced from December 2013 while the team was still being established. The original estimation of staff required to achieve the anticipated homebirth target was revised and by the end of the pilot the team comprised a band 7 team leader, band 6 midwives, band 3 maternity support workers (MSW) and band 5 midwives on 8-week placements.
The MSWs commenced weekly assessment based-training to become competent in clinical areas. They also completed a 2-year (day release) foundation degree in order to be able to attend low risk homebirths as second birth attendants, which also qualified them to undertake other skilled tasks such as phlebotomy, newborn bloodspot screening and breastfeeding support.
There was a weekly homebirth service meeting for shared learning to discuss complex cases and care plans of women who planned a homebirth outside guidance as well as new referrals and women who were 36 weeks' gestation or more and planning a homebirth.
Trust mandatory training, home-orientated skills and drills sessions were regularly undertaken on a rotating monthly programme that included shoulder dystocia, postpartum haemorrhage, breech presentation, cord prolapse, neonatal resuscitation and maternal collapse. Trust specialists were also asked to teach on relevant issues such as managing anaemia and perineal suturing.
Referrals and promotion
Referrals originated from a wide variety of sources. The majority came through community midwives by telephone, email, or in person, although women also self-referred, typically with a subsequent pregnancy or after personal recommendation, and a few referrals were in response to promotional information.
Various strategies were employed in promoting the homebirth service. Community midwives had a key role in this and were supported by an alongside study (Henshall et al, 2016) that promoted a brief, focused discussion about birthplace options around the 16-week gestation appointment, and provision of supporting literature. This was backed up by applying ‘Recommended Place of Birth’ stickers to pregnancy notes: homebirth if women were low-risk and multiparous, birth centre if low-risk and nulliparous, and delivery suite if high-risk.
Other means were tried of increasing the number of referrals of low risk multiparous women. These included a phone call around the time of booking to inform women about the homebirth service and offering home visits to discuss the birth plan at 36 weeks' gestation. Neither was considered to be an effective use of time as they resulted in few women birthing at home.
Monthly drop-in tea parties were widely advertised and were a place to meet the team as well as learn more about homebirth, including from women who had received intrapartum care from the team. Parents appeared to particularly appreciate hearing the birth stories and receiving peer advice and reassurance. It also seemed to be significant for some to be able to return in the postnatal period to support others, even when things had not gone to plan for themselves.
Other ways of promoting the homebirth service included speaking on local radio; Facebook (including advertising the homebirth service tea parties); giving leaflets to women birthing in the birth centre on discharge to increase awareness for the next birth; sending promotional leaflets and posters to all GP surgeries within the CCG; placing video clips of families and staff discussing the practicalities, safety and experience of homebirth on the Trust website and YouTube; and organising team visits to mums and babies/toddler groups held at children's and leisure centres.
Homebirth service care
Women who requested referral to the homebirth service were contacted by a midwife from the team to arrange a home visit to discuss homebirth. This could be at any time in the pregnancy, from booking to the day of the birth. Women who did eventually go on to plan a homebirth or home assessment usually also received antenatal and postnatal care by a member of their named homebirth service.
Intrapartum care for women booked for homebirth was provided by one of the homebirth service midwives. There was a daily rota for births, formed of two 12-hour on-call shifts. The on-call midwife was supported at night by a second homebirth service midwife, a band 5 midwife or a qualified MSW from the team. During the day, a community midwife would sometimes act as second midwife if a second birth attendant from the homebirth service was not available.
Homebirth outside guidance
The focus of the service was on offering home as a place of birth choice to low-risk women; however, some women either referred to the homebirth service with existing risk factors or developed risk factors during the pregnancy and still wanted a homebirth. These additional risks were explained and, as for all women referred to the homebirth service team, support was given so that an informed decision could be made about the planned place of birth. Care plans for women who requested homebirth outside guidance were developed by the named midwife, in conjunction with the family, an obstetrician and senior midwife. These plans included a summary of advice and details specific plans for antenatal, intrapartum and postnatal care. Intrapartum care plans always included provisions to allow two midwives to attend the birth. It occasionally took considerable time to develop a care plan to which all parties could agree.
Findings
Targets
The overall target was 3% homebirths of the total births at the Trust, starting from a 0.3% homebirth rate in 2013, before the pilot. The target was for a 1% increase per annum (Table 1).
| Year | Target | Achieved | % |
|---|---|---|---|
| 1 (2014/15) | 80 | 88 | 1.1 |
| 2 (2015/16) | 160 | 98 | 1.2 |
| 3 (2016/17) | 240 | 116 | 1.4 |
The first year target was achieved, but the figures increased slowly over the following 2 years and the subsequent two targets were not attained.
Overall care
Data on all women referred to the homebirth service were gathered each month, to document the care they received (Table 2).
| April 2014 to March 2015 | April 2015 to March 2016 | April 2016 to March 2017 | TOTAL | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N | M | Total | N | M | Total | N | M | Total | N | M | Total | |
| Total referrals | 60 | 144 | 204 | 78 | 230 | 308 | 84 | 290 | 374 | 222 | 664 | 886 |
| Did not book | 13 | 36 | 49 | 19 | 80 | 99 | 23 | 107 | 130 | 55 | 223 | 278 |
| Transferred in pregnancy | 18 | 26 | 44 | 27 | 59 | 86 | 26 | 67 | 93 | 71 | 152 | 223 |
| Intrapartum care at home | 29 | 82 | 111 | 32 | 91 | 123 | 35 | 116 | 151 | 96 | 289 | 385 |
| Homebirths achieved | 16 | 72 | 88 | 17 | 81 | 98 | 15 | 101 | 116 | 48 | 254 | 302 |
| Intrapartum transfers | 11 | 5 | 16 | 9 | 5 | 14 | 16 | 5 | 21 | 36 | 15 | 51 |
| Homebirth with postnatal transfer | 2 | 7 | 9 | 6 | 5 | 11 | 8 | 14 | 22 | 16 | 26 | 42 |
| Home assessment | 2 | 5 | 7 | 6 | 5 | 11 | 4 | 10 | 14 | 12 | 20 | 32 |
| Homebirth outside guidance | 1 | 9 | 10 | 2 | 19 | 21 | 4 | 17 | 21 | 7 | 45 | 52 |
N: nulliparous; M: multiparous
There was a total of 886 homebirth referrals received between April 2014–March 2017, with an average of 25 per month. One-quarter (n=222; 25%) of the referrals were from nulliparous women.
Of the women referred to the homebirth service, just under one-third (n=278) decided not to have a homebirth, and a further 223 who had planned to have homebirths transferred back to hospital during pregnancy, primarily due to the development of risk factors. This resulted in just over one-third of women (n=302) who had originally requested a referral for homebirth actually achieving a homebirth.
There were 385 women who received intrapartum care and 302 who birthed their baby at home. This included 15 planned homebirths of babies born before the arrival of the midwife. There were also 17 women who chose home assessment in latent or early phase of labour and then gave birth in hospital.
Referrals
Where possible, reasons were obtained as to why women did not book with the homebirth service (Figure 1). One-quarter (n=77; 25%) of referrals were from women with higher risk factors who were advised, and agreed, to give birth in the hospital. A further quarter (n=80; 26%) of women changed their minds after the initial discussion with homebirth service midwife. There were also 28 (9%) women who decided not to birth at home due to family reasons or concerns.
Transfers from home to hospital
In total, there were 93 peripartum transfers during labour and up to 6 hours postnatal from home to hospital; 51 of these before giving birth (Tables 3 and 4). There were 18 women (33%) who transferred to hospital for delay in labour and 13 (25%) transfers for fetal distress. The transfer rate for nulliparous women was 54% (n=52) and the rate for multiparous women was 14% (n=41), an overall transfer rate of 24% (n=93). This was slightly higher than the Birthplace in England study (2011), which had a transfer rate of 45% and 12% respectively, although this only included transfers immediately after the birth.
| April 2014–March 2017 | TOTAL | % |
|---|---|---|
| Fetal distress | 13 | 25 |
| Delay in first stage of labour | 10 | 17 |
| Delay in second stage of labour | 8 | 16 |
| Antepartum haemorrhage | 8 | 16 |
| Meconium stained liquor | 7 | 14 |
| Suboptimal maternal observations | 3 | 6 |
| Breech presentation | 2 | 4 |
| Prolonged ruptured membranes (>24 hrs) | 1 | 2 |
| TOTAL | 51 | 100 |
| April 2014 to March 2017 | TOTAL | % |
|---|---|---|
| Suboptimal neonatal observations | 12 | 29 |
| Perineal tear | 9 | 21 |
| Low APGAR score | 5 | 12 |
| Failed neonatal pulse oximetry screening | 5 | 12 |
| Retained placenta | 5 | 12 |
| Postpartum haemorrhage | 4 | 10 |
| Meconium neonatal observations | 1 | 2 |
| Maternal hypertension | 1 | 2 |
| TOTAL | 42 | 100 |
There were 5 babies with low APGAR scores, all of whom were resuscitated in accordance with Resuscitation Council Guidelines (2015) and transferred in for observation despite good APGAR scores later. The most frequent reason for postnatal transfer (n=12, 29%) was suboptimal neonatal observations when babies were transferred to the alongside birth centre, delivery suite, transitional care or the neonatal unit for observation. In total, 18 babies were admitted to the neonatal unit.
There were 2 unexpected breech babies born at home: one had a low APGAR score and was successfully resuscitated, and the other had suboptimal observations at 1 hour old. Both were transferred in for observation, but neither was admitted to the neonatal unit.
Homebirth outside guidance
There were 52 women who birthed at home outside guidance (Table 5). The most frequent reason for homebirth being outside guidance was raised body mass index (BMI) (n=11; 21%) followed by estimated fetal weight plotting over the 90th customised centile (n=6, 12%) and polyhydramnios (n=6, 12%).
| April 2014 to March 2017 | Nulliparous | Multiparous | TOTAL |
|---|---|---|---|
| Body mass index (BMI) >35 | 0 | 11 | 11 |
| Estimated fetal weight >90th centile | 0 | 6 | 6 |
| Polyhydramnios | 1 | 5 | 6 |
| Group B Strep positive screening | 1 | 3 | 4 |
| Prolonged rupture of membranes >24 hrs | 2 | 2 | 4 |
| Previous postpartum haemorrhage >500mls | 0 | 4 | 4 |
| Declined all screening in pregnancy | 1 | 2 | 3 |
| Previous caesarean section | 0 | 3 | 3 |
| Suboptimal fetal growth | 0 | 2 | 2 |
| Declined induction for post-maturity | 1 | 1 | 2 |
| Previous shoulder dystocia | 0 | 2 | 2 |
| Cholestasis | 0 | 2 | 2 |
| Epileptic | 0 | 1 | 1 |
| Maternal age over 40 years | 1 | 0 | 1 |
| Low lying placenta | 0 | 1 | 1 |
| TOTAL | 7 | 44 | 52 |
Discussion
The homebirth service pilot achieved a five-fold increase in the Trust homebirth rate with good outcomes and no adverse incidents. There was high satisfaction from the families booked with the homebirth service whether or not they achieved a homebirth, evidenced by Friends and Family Test results and other feedback. Alongside the positives, there were two particular challenges during the course of the pilot.
Firstly, the targeted homebirth rate was not achieved and only grew marginally between years two and three (1.3–1.4%), despite the huge effort put into homebirth service promotion. One reason for this was that homebirth was previously considered an unsafe option. While this has been challenged for low risk women with normal pregnancies (Birthplace in England Collaborative Group et al, 2011; de Jonge et al, 2013), the popular perception of birthing at home is still negative, which suggests that the culture or social acceptability of homebirth still needs to be addressed (Coxon et al, 2014; Fage-Butler, 2017).
Home assessment was offered was offered for women who were unsure about homebirth, meaning that the decision about where to birth could be made in labour. Unlike some other studies, however (Brintworth and Sandall, 2013; O'Connell et al, 2012), there was limited uptake of this service. This may be due to the lack of continuity of care in the intrapartum period, as births were attended by the homebirth service midwife on-call rather than necessarily the named midwife. Further research on this topic is needed.
While a good number of referrals were made to the homebirth service, there was a high attrition rate of nearly two-thirds, which meant that higher numbers were required in order to achieve the targeted number of homebirths. To achieve 20 homebirths per month, approximately 60 referrals were needed: one-third would not chose a homebirth and a further third would transfer in pregnancy usually due to development of risk factors. It also meant that a significant amount of time was spent caring for women who did not receive intrapartum care.
Secondly, achieving and maintaining appropriate staffing levels was challenging. The MSWs were not trained and ready to be the second birth attendant until halfway through the pilot, which was longer than expected, and there was a high staff turn-over. Some MSWs found that the role was not what they had envisaged, or struggled with the academic training, while for others it opened opportunities: two trained to become midwives after completing the course.
There were also some challenges in recruiting and retaining midwives as they needed to live locally, work flexibly and undertake regular on-calls. While the positive benefits of caseload midwifery have been documented (Newton et al, 2014; Homer et al, 2017), the reality for some appeared to be challenging, particularly for those with family commitments.
Next steps
Following an options appraisal at the end of the homebirth service pilot, as agreed by the Trust and the CCG, homebirth service midwives were allocated to one of the four community teams in order to provide additional support, as they were then counted in the community numbers. The challenge will be to become part of the community teams while retaining a separate identity as a homebirth service and to continue to grow the service.
Summary
The homebirth service pilot has increased the number of planned homebirths from 0.3–1.4% in 3 years and provided 1.9% of women in the pilot year with intrapartum care. It has not met the CCG pilot target of 3% of all births at the Trust, however, which means that the homebirth service has not achieved self-funding status. This may have implications for future funding.
The peripartum transfer rate from home to hospital was 54% for primiparous and 14% for multiparous women, similar to the study by the Birthplace in England Collaborative Group et al (2011). Outcomes for women who received intrapartum care were very positive, with appropriate transfers and no serious incidents. MSWs have also been positively integrated as second attendants at low-risk homebirths.
Approximately one-third of the women who initially wanted a homebirth achieved one, although many changed their minds or were recommended to have a hospital birth due to identified risk factors. This has meant that while there has been excellent feedback for those who have used the homebirth service, significantly more women have expressed an interest in having a homebirth than have achieved intrapartum care at home.
The team has worked hard to promote the homebirth service and have increased the number of women birthing at home. It is, however, ongoing work that will take longer than a 3-year pilot to change the culture of childbirth, and further research on this topic is needed.