Hemförlossningar i Sverige 1992-2005 : Förlossningsutfall och kvinnors erfarenheter

Abstract

This is the first national survey of planned home births in Sweden. We aimed to study women who had given birth or planned to give birth at home over the period from 1st of January 1992 to 30th of June 2005.

In study I, data from the Swedish Medical Birth Register for the period 1992 to 2001 for 352 women who gave birth outside hospital were compared with data from 1760 women who had given birth in hospitals. Women who gave birth at home differed from those who gave birth in the hospital in various aspects. Women with an out-of-hospital birth had a higher number of children, a higher level of education but a lower combined family income and were employed to a lesser degree than those who gave birth in hospital. The women with an out-of-hospital birth were more frequently born in a European country other than Sweden but were less frequently born outside Europe. The prevalence of home birth based on the register code for out-of-hospital births was 0.38 per thousand births.

In study II, a total of 757 women who had planned a home birth over the period 1st of January 1992 to 30th of June 2005 were identified through midwives who assist home deliveries, announcements and through a home birth association. Of the women who were asked to participate 99 percent consented. Data were collected through questionnaires. A total of 100 women who had altogether 141 planned home births were not found in the medical birth register. The prevalence of planned homebirths was 0.95 per thousand based on data for all women who had been identified through home birth midwives and announcements and 0.85 per thousand based on register data for those women. Transfer to hospital during or immediately after birth occurred in 12.5 percent of planned home births (Study II). The most common reasons for transfer were slow progress (46%) and the midwife who had been booked not being available at the onset of labour (14%). One in four primiparas completed their planned home birth in hospital, the most common reason for transfer among primiparas being that a different midwife than the one the woman had seen for check-ups during the pregnancy assisted at the birth. Factors that influenced transfer were a previous caesarean section (multiparas) and, for both primiparas and multiparas, the pregnancy being post-term.

In Study III data from the Swedish Medical Birth Register were compared for 897 planned homebirths who consented to participate in the study and have given their personal identity code, and 11,341 hospital births (control group). The criterion for selection of the control group was single births in gestational weeks 37-42 with spontaneous onset of labour. Women who had planned home births more frequently had spontaneous vaginal deliveries and less frequently serious perineal or vaginal tears.

In study IV, (the same sample as in study II) perceived risks in connection with place of birth had been considered by three-quarters of the women. The women dealt with the risk of themselves or their babies being harmed or dying in the birth through their own preparation, discussing principally with their partners and home birth midwifes. The women avoided discussing risks with conventional healthcare staff.

Conclusions: In Sweden, between 1992 and 2005, around 100 women each year planned a home birth. One out of four primiparas was transferred to hospital during or immediately after their planned home homebirth and the total transfer rate was 12.5 percent. The women considered risks associated with a home birth but avoided discussing them with medical staff. The neonatal mortality in the home birth group was 2.2 per thousand compared to 0.6 in the hospital group. Women who planned a home birth more frequently had a spontaneous vaginal delivery with fewer serious perineal or vaginal tears after the birth compared to women who gave birth in the hospital.