Table 1.
Description and quality assessments of included studies
| Study | Inclusion criteria | Participants (% P0 1 ) | Caregivers | Study design | Data source | Duration of observation time after birth | Analyses stratified for parity | Study population representative | Quality |
|---|---|---|---|---|---|---|---|---|---|
| Amelink-Verburg et al.[24] |
All women under midwifery care and with an intended home birth in the Netherlands during 01.01.2001-31.12.2003 |
N = 168,618 |
Primary level midwifes |
Prospective |
The Dutch Midwifery Perinatal Database (LVR1) |
2 h after the birth of the placenta |
No |
Data from LVR1 covers 95% of midwifery practices. |
Medium |
| (Parity not described) | |||||||||
| Anderson et al.[22] |
All Nurse-midwifery practices in the USA during 1987-1991 |
N = 11,084 |
Independent midwives |
Retrospective |
Data collection forms from the midwives |
..”early postpartum period” |
No |
66% of midwifery practices participated. |
Medium |
| (Parity not described) | |||||||||
| BECG2[1] |
All NHS trusts providing intrapartum care at home in England (UK) during April 2008-April 2010 |
N = 16,840 |
National Health Service midwives |
Prospective |
Data collection forms from midwives and hospitals |
48 h postpartum |
Yes |
97% of trusts providing home birth services participated. (Home births attended by independent midwives in the region were not included) |
Good |
| (27.2%) | |||||||||
| Blix et al.[25] |
All planned home births in Norway during 01.01.1990-31.12.2007 |
N = 1631 |
Independent midwives |
Retrospective |
Midwives’ patient files |
5 days postpartum |
Yes |
Unclear, probably >70% of all planned home births during the study period |
Medium |
| (22.6% ) | |||||||||
| Davies et al.[23] |
All women in the North Regional Health Authority area (UK) who planned for a home birth and expected to deliver in 1993 |
N = 177 |
National Health Service midwives |
Prospective |
Data collection forms from midwives, women and GP’s |
Not described |
Partly |
Unclear, probably were all planned home births attended by NHS midwives included. |
Medium |
| (9.1%) | |||||||||
| Hansen and et al.[27] |
All home births assisted by midwives employed by the local health authorities in the Municipality of Copenhagen (Denmark) during 1980-1982 |
N = 102 |
Midwives employed at Hvidovre Hospital |
Retrospective |
Hospital patient files |
Not described |
No |
All planned home births assisted by midwives employed by the local health authorities were included. (Home births attended by independent midwives in the region were not included) |
Medium |
| (about 50%) | |||||||||
| Howe [17] |
All home births attended by a registered midwife in the south-west of Western Australia during 01.01.1983-31.12.1986 |
N = 165 |
Independent midwives |
Retrospective |
Midwifery registers |
Not described |
Partly |
All midwives participated |
Medium |
| (31.5%) | |||||||||
| Hutton et al.[18] |
All home births attended by Ontario midwives during 01.04.2003-31.03.2006 (Canada) |
N = 6,692 |
Certified midwives who are required to submit all data to a regional database |
Retrospective |
The Ontario Ministry of Health Database |
Not described |
Partly |
All planned home births were included |
Medium |
| (34.3%) | |||||||||
| Johnson and Daviss [4] |
All home births involving certified professional midwives across the USA and Canada during 01.01.2000-31.12.2000 |
N = 5,418 |
Independent midwives |
Prospective |
Data collection forms from the midwives |
Not described |
No |
73% of the midwives asked, participated. <1% of the women declined participation |
Medium |
| (31.2%) | |||||||||
| Janssen et al.[19] |
All planned home births attended by regulated midwives in British Columbia (Canada) during 01.01.1998-31.12.1999 |
N = 797 |
Regulated midwives |
Prospective |
Data collection forms |
Not described |
No |
>99% of the data collection forms were received |
Medium |
| (about 47%) | |||||||||
| Lindgren et al.[26] |
All planned home births in Sweden during 01.01.1992-31.07.2005 |
N = 1,025 |
Independent midwives |
Retrospective |
Data collection forms to the mothers |
…”shortly after planned home birth” |
Yes |
99% of the women asked, agreed to participate. Unclear if all home births were identified. |
Medium |
| (23.8%) | |||||||||
| McMurtrie et al.[15] |
The first 100 booked home births at the St. George Homebirth Program during Nov 2005-March 2009 in New South Wales (Australia) |
n = 70 attempted home births |
Midwives employed at St George Hospital |
Prospective |
Databases at the birth centre |
Not described |
No |
All planned homebirths were included. (Home births attended by independent midwives in the region were not included) |
Medium |
| (Parity not described) | |||||||||
| Murphy et al.[21] |
All nurse-midwifery practices providing home birth services in the USA during Dec 1994-Dec 1995 |
N = 1,221 |
Independent midwives |
Prospective |
Data collection forms from the midwives, data from hospital files |
Not described |
Partly |
64% of midwifery practices participated. 20% of women transferred to hospital were lost-to-follow-up |
Medium |
| (22.0%) | |||||||||
| Parratt et al.[16] |
All planned home births in Victoria (Australia) during 1995-1998 |
N = 419 |
Independent midwives |
Retrospective |
Midwives’ patient files |
Not described |
No |
50-60 births were not included |
Medium |
| (about 31%) | |||||||||
| Tyson [20] | All planned midwife-attended home births in Toronto (Canada) during Jan 1983-Jul 1988 | N = 1,001 |
Independent midwives | Retrospective | Midwives’ patient files | 4 days postpartum | Yes | All midwives participated | Medium |
| (Parity not described) |
1P0 = nulliparous women. 2Birthplace in England Collaborative Group.