Background
Planned home births have leveled off in the United States in recent years after a significant rise starting in the mid-2000s. Planned home births in the United States are associated with increased patient-risk profiles. Multiple studies concluded that, compared with hospital births, absolute and relative risks of perinatal mortality and morbidity in US planned home births are significantly increased.
Objective
To explore the safety of birth in the United States by comparing the neonatal mortality outcomes of 2 locations, hospital birth and home birth, by 4 types of attendants: hospital midwife; certified nurse-midwife at home; direct-entry (“other”) midwife at home; and attendant at home not identified, using the most recent US Centers for Disease Control and Prevention natality data on neonatal mortality for planned home births in the United States. Outcomes are presented as absolute risks (neonatal mortality per 10,000 live births) and as relative risks of neonatal mortality (hospital-certified nurse-midwife odds ratio, 1) overall, and for recognized risk factors.
Study Design
We used the most current US Centers for Disease and Prevention Control Linked Birth and Infant Death Records for 2010–2017 to assess neonatal mortality (neonatal death days 0–27 after birth) for single, term (37+ weeks), normal-weight ( >2499 g) infants for planned home births and hospital births by birth attendants: hospital-certified nurse-midwives, home-certified nurse-midwives, home other midwives (eg, lay or direct-entry midwives), and other home birth attendant not identified.
Results
The neonatal mortality for US hospital midwife-attended births was 3.27 per 10,000 live births, 13.66 per 10,000 live births for all planned home births, and 27.98 per 10,000 live births for unintended/unplanned home births. Planned home births attended by direct-entry midwives and by certified nurse-midwives had a significantly elevated absolute and relative neonatal mortality risk compared with certified nurse-midwife–attended hospital births (hospital-certified nurse-midwife: 3.27/10,000 live births odds ratio, 1; home birth direct-entry midwives: neonatal mortality 12.44/10,000 live births, odds ratio, 3.81, 95% confidence interval, 3.12–4.65, P <.0001; home birth–certified nurse-midwife: neonatal mortality 9.48/10,000 live births, odds ratio, 2.90, 95% confidence interval, 2.90; P <.0001). These differences increased further when patients were stratified for recognized risk factors.
Conclusion
The safety of birth in the United States varies by location and attendant. Compared with US hospital births attended by a certified nurse-midwife, planned US home births for all types of attendants are a less safe setting of birth, especially when recognized risk factors are taken into account. The type of midwife attending US planned home birth appears to have no differential effect on decreasing the absolute and relative risk of neonatal mortality of planned home birth, because the difference in outcomes of US planned home births attended by direct-entry midwives or by certified nurse-midwives is not statistically significant.
Planned home births have leveled off in the United States in the past 4 years after a significant rise starting in the mid-2000s ( Figure 1 ). The American College of Obstetricians and Gynecologists (ACOG) committee opinion on planned home birth stated that “… planned home birth is associated with a twofold to threefold increased risk of neonatal death when compared with planned hospital birth ….” Planned midwife-attended home births in the United States are associated with increased patient-risk profiles, and multiple studies concluded that, compared with hospital births, absolute and relative risks of perinatal mortality and morbidity in US planned home births are significantly increased.
Why was this study conducted?
To compare neonatal mortality at planned home births with that of hospital births attended by certified nurse-midwives.
Key findings
Planned US home births, whether attended by direct-entry midwives or by certified nurse-midwives, had a significantly greater risk of neonatal mortality compared with certified nurse-midwife-attended hospital births (neonatal mortality: home births direct-entry midwives: 12.44/10,000 live births, odds ratio, 3.81, P <.0001; home births certified nurse-midwife: 9.48/10,000, odds ratio, 2.90, P <.0001; hospital certified nurse-midwives: 3.27/10,000 live births odds ratio 1).
What does this add to what is known?
The type of midwife attending US planned home birth has no significant differential effect on the increased absolute and relative risk of neonatal mortality of planned US home birth.
The causes of the increased neonatal mortality at US planned home births are likely the location and the failure to select women at low risk of neonatal mortality.
The objective of this study was to analyze the safety of US intended births by location and attendant, using the most recent data on neonatal mortality for planned home births as indicated in the US Centers for Disease Control and Prevention (CDC) Linked Birth/Infant Death records. Our goal was to explore the safety of birth by comparing the outcomes of 2 locations, hospital birth and planned home birth, by 4 types of attendants: hospital-certified midwife; home-certified nurse-midwife; home “other” (also known as direct-entry type) midwife; and attendant at home not identified. Outcomes are presented as absolute risks (neonatal mortality per 10,000 live births), and as relative risks of neonatal mortality (hospital certified nurse-midwife odds ratio, 1) overall, and for recognized risk factors.
Materials and Methods
We used the most current CDC Linked Birth and Infant Death Records for 2010–2017, to assess neonatal mortality (neonatal death on days 0–27 after birth) for single, term (37+ weeks), normal-weight (>2499 g) infants for home births indicated in birthplace as intended (interchangeably known as planned), and hospital births by birth attendants: hospital-certified nurse-midwives, planned home birth–certified nurse-midwives, planned home birth direct-entry midwives (midwives without the level training of certified nurse-midwives), and not identified other home attendants. The CDC natality database categorizes home births as either intended, not intended, or unknown if intended. In this study, we used the categories of intended interchangeably with planned, and not intended interchangeably with unplanned. Hospital deliveries by certified nurse-midwives served as comparison. Home births categorized as not intended were excluded from this study, as were home births categorized as unknown whether intended (most of the unknown were from the State of California, where all home births are categorized as unknown whether intended), and births where the birth attendant was listed as “unknown or not stated.”
This dataset (linked file) contains detailed information for the approximately 4 million births in the United States each year for which a birth certificate is created, including birth setting, birth attendant, and neonatal mortality, and is generally the preferred source for infant and neonatal mortality in the United States.
Period-linked files use all births in a year as the denominator and all deaths in a year as the numerator, regardless of when the birth occurred (eg, if the birth was in late 2015 and the neonatal death occurred in 2016, that death is counted in the 2016 numerator). Studies using linked US birth and death records are considered reliable and have been used in numerous studies.
The 2010–2017 period-linked birth and infant deaths dataset was analyzed to examine neonatal mortality (defined as the death of a live-born neonate days 0–27 of life in term (≥37 weeks), normal size (birthweight of ≥2500 g), singleton births by birth setting (planned home and hospital) and attendant: hospital-certified nurse-midwife, planned home birth–certified nurse-midwife, planned home birth by direct-entry midwife, and planned home birth attendant not identified. We also examined the relative and absolute risks associated with delivery by setting and attendant with birth by hospital certified nurse-midwives as comparison. The odds ratios and 95% confidence intervals (CIs) were computed here: https://www.medcalc.org/calc/odds_ratio.php . We calculated neonatal mortality overall and for frequently occurring common risk factors (≥35 years maternal age, ≥41 weeks gestational age, nulliparity). Our study used nonidentifiable data from a publicly available dataset and was not considered human subjects research and therefore did not require review by Zucker Medical School at Hofstra/Northwell Institutional Review Board.
Results
Between 2010 and 2017, there were 195,026 home births and 2,280,044 hospital midwife-attended births in the United States. In total, 177,156 (87%) home births were categorized as intended/planned home births and 17,870 (9.2%) were categorized as unintended/unplanned home births. For the purpose of comparative analysis, unintended/unplanned home births were excluded from the study. The majority of the 177,156 planned home births were attended by direct-entry midwives (50.4%; n=89,247), followed by certified nurse-midwives (28.6%; n=50,658) and other attendants not identified (21%; n=37,251). Table 1 shows the characteristics of women for each of the types of birth attendant.
| Hospital midwives | Home-certified nurse-midwife | Home other (direct-entry) midwife | Attendant not identified | All intended | |
|---|---|---|---|---|---|
| Maternal age, y | |||||
| <35 | 1,993,522 (87.4) | 39,009 (77.0) | 70,373 (78.9) | 28,851 (77.5) | 138,233 (78.0) |
| ≥35 | 286,522 (12.6) | 11,649 (23.0) | 18,874 (21.1) | 8,400 (22.5) | 38,923 (22.0) |
| Gestational age, wk | |||||
| <41 | 1,814,277 (78.6) | 3,9513 (78.0) | 69,683 (78.1) | 29,172 (78.3) | 138,368 (78.1) |
| ≥41 | 465,676 (20.4) | 11,145 (22.0) | 19,564 (21.9) | 8,079 (21.7) | 38,788 (21.9) |
| Parity | |||||
| 0 | 885,581 (38.8) | 10,601 (20.9) | 17,564 (19.7) | 6,063 (16.3) | 34.228 (19.3) |
| >0 | 1,394,463 (61.7) | 4,0057 (79.1) | 71,683 (80.3) | 31,188 (83.7) | 142,928 (80.7) |
| Multiparity | |||||
| <5 | 2,149,892 (94.3) | 40,719 (80.4) | 70,469 (79.0) | 25,275 (67.9) | 136,463 (77.0) |
| ≥5 | 130,152 (5.7) | 9,939 (19.6) | 18,778 (21.0) | 11, 976 (32.1) | 40,693 (23.0) |
The neonatal mortality rates were significantly elevated for all planned home births: 13.66 per 10,000 live births (242/177,156; odds ratio, 4.19; 95% CI, 3.62–4.84, P <.0001) and for unintended/unplanned home births: 27.98 per 10,000 live births (50/17,870; odds ratio, 8.58; 95% CI, 6.44–11.43, P <.0001), compared with hospital-certified nurse-midwife-attended births: 3.27 per 10,000 live births (odds ratio, 1: 745/2,280,044) ( Table 2 ).
| Location and attendant | Neonatal mortality per 10,000 live births (n/all) | OR (95% CI) | P value |
|---|---|---|---|
| Hospital midwife | 3.27 (745/2,280,044) | 1 | |
| Intended home birth–certified nurse-midwife | 9.48 (48/50,658) | 2.90 (2.17–3.89) | <.0001 |
| Intended home birth other (direct-entry) midwife | 12.44 (111/89,247) | 3.81 (3.12–4.65) | <.0001 |
| Intended home birth someone else | 22.28 (83/37,251) | 6.83 (5.44–8.57) | <.0001 |
| All intended home births | 13.66 (242/177,156) | 4.19 (3.62–4.84) | <.0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
