For healthy women with low-risk pregnancies the incidence of adverse perinatal outcomes is low in all birth settings. However, for those who are nulliparous the risk of an adverse outcome appears to be higher for births planned at home, a national cohort study has found. There was also a high intrapartum transfer rate.

Over a two-year period nearly 65,000 women ‘attended by an NHS midwife during labour in the planned place of birth’ were enrolled in the Birthplace in England national prospective cohort study. Place of birth was categorised as obstetric unit, alongside midwifery unit (midwife-led unit on a hospital site with an obstetric unit) freestanding midwifery unit and home birth.

Not surprisingly women planning an obstetric unit birth differed from those planning a home birth. The latter were on average older, more parous, more affluent and educated, more likely to be white and speak English fluently. Even though low-risk obstetric cases were being surveyed, 20% of women planning an obstetric unit delivery had one or more complicating conditions at the start of labour compared with ≤ 7% in the other settings.

The chance of having a normal labour, defined as spontaneous, unaugmented, without epidural/spinal anaesthesia and with a normal vaginal delivery was lowest in the obstetric unit.

There were 250 recorded incidences of adverse perinatal outcomes (4.3/1,000 births). Perinatal deaths accounted for 13% of this total. Overall adverse outcome rates were uniform across the four places of birth.

However, when women with complications were removed (leaving just the lowest risk cases for analysis) adverse outcomes were more common in those who planned home births, odds ratio (OR) 1.59 (95% CI: 1.01-2.52) compared with obstetric unit births.

In the sub-group analysis by parity, the odds of an adverse outcome for nulliparous women were higher for planned home births than for planned obstetric unit births adjusted OR 1.75 (95% CI: 1.07-2.86). When only women with no complications at the start of labour were considered adjusted OR was 2.80 (95% CI: 1.59-4.92).

Transfer rates from all three non-obstetric unit settings were higher for nulliparous women: freestanding midwifery unit (36%); alongside midwifery unit (40%); and home (45%) than for multiparous women (9%, 13% and 12% respectively).

So, where is the best place to give birth? The answer appears to be that it depends on the presence of pre-existing risk factors and maternal preference. The outcomes in the midwifery units were very reassuring. However, the significantly higher figures for adverse perinatal outcomes and transfer during or just after labour from home to an obstetric unit in nulliparous cases opting for a home birth is food for thought.

Dr Chris Barclay