Planned Out-of-Hospital Birth - PowerPoint Presentation

Slide1

Planned Out-of-Hospital Birth

Draft Coverage Guidance for EbGS Consideration

April 4, 2019

Slide2

Overview

Update to 2015 coverage guidance process

February 2019: General background

Today’s meeting: Review of evidence

June EbGS meeting: Review of guidelines and discussion of draft coverage recommendations

Slide3

Background: Out-of-Hospital Birth in Oregon

In 2017, planned out-of-hospital births accounted for 1,758 (4.0%) of the 44,160 births in Oregon

60% of out-of-hospital births in the mother’s home

40% in freestanding birth centers

In addition to the CDC birth certificate question on intention of home birth (planned or unplanned), Oregon House Bill 2380 (2011) adds additional information:

“Did you go into labor planning to deliver at home or at a freestanding birthing center?”

Slide4

Background: Out-of-Hospital Birth in Oregon

Planned Attendant for Planned Out-of-Hospital Births, 2017

Attendant

Number

Percent

Medical Doctors and Doctors of Osteopathy (MD/DO)

0

0%

Naturopathic Physicians (ND)23814%Certified Nurse Midwives (CNM)54831%Licensed Direct-Entry Midwives (LDM)79445%Unlicensed Direct-Entry Midwives1368%Other422%TOTAL1,758100%

Other constitutes other family members, friends

Planned hospital births, 79% MD/DO attended, 21% attended by CNM

Eligible for OHA reimbursement for OOHB: MD/DO, ND, CNM, LDM

Slide5

Background: Out-of-Hospital Birth in Oregon

Training and Licensing Requirements for Birth Attendants

Unlicensed Direct-Entry Midwives

Licensed Direct-Entry Midwives (LDEMs)

Certified Nurse Midwives (CNMs)

Naturopathic Doctors (NDs)

Osteopathic or Allopathic Doctors (DO/MDs)

Accredited Training Program

 √**or equiv. trainingUndergraduate Degree √√√Graduate Degree √√√Postgraduate Training 

√

Settings

Home

Home,

birth center

Home,

birth center, hospital

Home,

birth center

Hospital (mostly)

Licensing Board

None

Board of Direct-Entry Midwifery

Oregon State Board of Nursing

Board of Naturopathic Medicine

Oregon Medical Board

Slide6

Oregon Public Health Data

Case review of 8 fetal and early neonatal deaths among planned out-of-hospital births in 2012

4 fetal, 4 early neonatal

6 of the 8 would not meet current HERC low-risk criteria for coverage of out-of-hospital birth

> 41 weeks gestational age (4)

Twin pregnancy (2)

Morbid obesity (1)

No or inadequate prenatal care (2)

Pre-existing or pregnancy-related maternal disease (2)

Slide7

Oregon Public Health Data

2012-2017 review of planned out-of-hospital births

“Low-risk” women: single gestation, no cesarean history, at 37 to 41 weeks gestation

Demographics, geography, transfer rates

Transfers counted as planned out-of-hospital births

Perinatal death: death at ≥ 20 weeks to first 6 days of life

Birth attendant type not reported

Slide8

Oregon Public Health Data

1.7 per 1,000 pregnancies for planned in-hospital birth

2.0 per 1,000 pregnancies for planned out-of-hospital birth

Slide9

Evidence Review

Slide10

Scope Statement

Population

Pregnant women

Interventions

Planned out-of-hospital birth (home or birth center)

Comparators

Planned birth in a hospital

Slide11

Scope Statement

Critical Outcomes

Delivery mode (cesarean, operative vaginal delivery, spontaneous vaginal delivery)

Perinatal mortality

Serious neonatal morbidity (e.g., seizures, NICU admission, low Apgar scores, hypoxic ischemic encephalopathy, sepsis)

Serious maternal harm (e.g., postpartum hemorrhage, serious infection, mortality)

Important Outcomes

Breastfeeding

Slide12

Key Questions

What is the comparative effectiveness of planned out-of-hospital birth compared to hospital birth?

Does the comparative effectiveness of planned out-of-hospital birth vary by:

Patient characteristics (demographics)

Risk factors (pregnancy or pre-pregnancy) and comorbidities

Setting (including home, out-of-hospital birth center)

Location (U.S. vs. non-U.S.)

Slide13

Key Questions

What are the harms of planned out-of-hospital birth compared to hospital birth?

Do the harms of planned out-of-hospital birth vary by:

Patient characteristics (demographics)

Risk factors (pregnancy or pre-pregnancy) and comorbidities

Setting (including home, out-of-hospital birth center)

Location (U.S. vs. non-U.S.)

Provider characteristics

Contextual questions, guidelines, and policies scheduled for June meeting

Slide14

Findings

Slide15

Evidence Sources

42

comparative observational studies

7 U.S. of good (1), fair (2), poor (4) methodological quality

35 Non-U.S. of good (5), fair (9), poor (21) methodological quality

14 noncomparative observational studies, all poor methodological quality

7 U.S. studies

6 non-U.S. studies

1 used combination of Canadian and U.S. data

Slide16

Evidence Findings- Overview

U.S. comparative data

U.S. noncomparative data

GRADE table with U.S. and non-U.S. comparisons

Subgroup findings

Limitations

Slide17

Mode of Delivery: U.S. Studies

2 studies

Snowden et al., 2015 (Oregon, 2012 to 2013)

Good-quality study

Planned home or birth center births (n = 3,804) compared to planned hospital (n = 75,923)

Thornton et al., 2017 (43 states, 2006 to 2011)

Poor-quality study

Planned birth center (n = 8,776) compared to planned hospital delivery (n = 2,527)

Slide18

Mode of Delivery: U.S. Studies

Snowden et al.,

2015

Planned home or birth center vs. planned hospital

NNT

Nonoperative Vaginal Delivery

93.8% vs. 71.9%

Adjusted risk difference 27.5, 95% CI 24.9 to 30.2, p < 0.001

3 95% CI, 3 to 4Cesarean Delivery5.3% vs. 24.7% Adjusted risk difference -24, 95% CI -26.6 to -21.4, p < 0.001495% CI, 3 to 4Thornton et al., 2017Planned birth center vs. planned hospitalNNT/NNHCesarean Delivery4.14% vs. 4.99% aOR 0.63, 95% CI 0.50 to 0.79, p = not reportedNNT117NNH: number needed to harm; NNT: number needed to treat

Slide19

Perinatal or Neonatal Mortality: U.S.

4 studies

Grunebaum, McCullough, Arabin et al., 2017 (U.S., 2008 to 2012)

Poor-quality study; planned home birth (CNM or other midwife) vs. hospital (midwife)

Grunebaum, McCullough, Sapra et al., 2017 (U.S., 2009 to 2013 )

Fair-quality study; all planned home births (n = 96,815) compared to hospital birth (midwife) (n = 1,077,197)

Snowden et al., 2015 (Oregon, 2012 to 2013)

Good-quality study; planned home or birth center births (n = 3,804) compared to planned hospital (n = 75,923)

Thornton et al., 2017 (43 states, 2006 to 2011)Poor-quality study; planned birth center (n = 8,776) compared to planned hospital delivery (n = 2,527)

Slide20

Perinatal or Neonatal Mortality: U.S.

Grunebaum, McCullough, Arabin et al., 2017

Grunebaum, McCullough, Sapra et al., 2017

Snowden, 2015

Planned home vs. planned hospital – attended by

a midwife

NNH

Neonatal death

0 to 27 days 2012 to 2013: 0.16% vs. 0.06%, aOR 2.87 (1.10 to 4.30), adjusted risk difference 0.077, 95% CI 0.013 to 0.1401,298 (714 to 7,692)2009 to 2013: 0.12% vs. 0.03%, standardized mortality ratio (SMR) 4.13, 95% CI 3.38 to 4.881,1112008 to 2012: 0.127% vs. 0.035%RR 3.62, 95% CI 3 to 4.41,087

Slide21

Perinatal or Neonatal Mortality: US

Standardized mortality ratio (SMR):

Observed deaths in a study population

Expected deaths in a standard population

>1: increased risk of death ; < 1: protective

For

Grunebaum

, McCullough,

Sapra et al., 2017, standard population is hospital deliveries attended by a midwifeReport controlling for age, parity, post-term datesMore an adjusted risk ratio than SMR

Slide22

Perinatal or Neonatal Mortality: U.S.

Snowden et al.,

2015

Planned home or birth center vs. planned hospital

NNH

≥ 20 weeks, intrapartum, through 27 days of life

0.39% vs. 0.18%

Adjusted

risk difference 0.158, 95% CI 0.055 to 0.261, p = 0.003633 (383 to 1,818)Thornton et al., 2017Planned birth center vs. planned hospitalNNHIntrapartum or newborn death0.03% vs. 0.04%OR 0.86, 95% CI 0.09 to 8.3, p = 0.99Calculated RR 0.96, 95% CI 0.54 to 1.70Not applicable

Slide23

Neonatal Morbidity: U.S.

3

studies

Li et al., 2017 (South Carolina, 2003 to 2013)

Poor-quality study

Planned home (n = 661) vs. planned birth center (n = 1,233) vs. planned clinic (n = 9) vs. planned hospital (n =547,523)

Snowden et al., 2015 (Oregon, 2012 to 2013)

Good-quality study

Planned home or birth center births (n = 3,804) compared to planned hospital (n = 75,923)Thornton et al., 2017 (43 states, 2006 to 2011)Poor-quality study Planned birth center (n = 8,776) compared to planned hospital delivery (n = 2,527)

Slide24

Neonatal Morbidity: U.S.

Snowden et al.,

2015

Li

et al.

, 2017

Planned home or birth

center vs. planned hospital

NNH or NNT NICU Admission1.7% vs. 2.9%Adjusted risk difference: 0.85 less, 95% CI -1.57 to -0.14, NNT = 117Apgar < 4Apgar < 7(Both at 5 minutes)0.6% vs. 0.4%Adjusted risk difference: 0.18 more, 95% CI 0.00 to 0.372.3% vs. 1.8%Adjusted risk difference: 0.50 more, 95% CI 0.07 to 0.93Not applicableNNH =200107 to 1,428Neonatal Seizures 0.13% vs. 0.04%Adjusted risk difference; 0.07 more, 95% CI 0.02 to 0.13Seizures: 0.05% vs. 0.02%; Calculated RR 2.2, 0.3 to 15.77Convulsions: 0.37% vs. 0.23%; Calculated RR 1.57, 95% CI 0.74 to 3.2NNH = 1,428Not applicable

Slide25

Neonatal Morbidity: U.S.

Snowden et al.,

2015

Li

et al.

, 2017

Planned home or birth

center vs. planned hospital

NNH or NNT Ventilator Support3.8% vs. 3.3%Adjusted risk difference 0.97 to 1.05aOR 1.36, 95% CI 1.14 to 1.62NNH = 95 to 113

Slide26

Neonatal Morbidity: U.S.

Thornton et al.,

2017

Planned birth

center vs. planned hospital

NNH or NNT

Apgar score 3 to 7

(at

5 minutes)aOR 1.60, 95% CI 0.82 to 3.16Not applicable Composite(mortality, hypoxic injury, low Apgar, seizure, respiration support, and meconium aspiration syndrome)aOR 1.44, 95% CI 0.95 to 1.82Not applicable Short-term positive pressure ventilation < 10 minutesaOR 1.31, 95% CI 0.95 to 1.82Not applicable

Slide27

Maternal Harm: U.S.

2 studies

Snowden et al., 2015 (Oregon, 2012 to 2013)

Good-quality study

Planned home or birth center births (n = 3,804) compared to planned hospital (n = 75,923)

Thornton et al., 2017 (43 states, 2006 to 2011)

Poor-quality study

Planned birth center (n = 8,776) compared to planned hospital delivery (n = 2,527)

Slide28

Maternal Harm: U.S. Studies

Snowden et al.,

2015

Planned home or birth center vs. planned hospital

NNT/NNH

Third or fourth degree perineal

laceration

0.9% vs. 1.3% adjusted risk difference 0.54 less, 95% CI −0.98 to −0.11, p = 0.02NNT: 185 (102 to 909)Blood transfusion0.6% vs. 0.4%adjusted risk difference 0.27 more, 95% CI 0.08 to 0.46, p = 0.006NNH: 370(217 to 1,250) Thornton et al., 2017Planned birth center vs. planned hospitalNNT/NNHPostpartum Hemorrhage (no volume definition)4.63% vs. 6.18%aOR 1.25, 95% CI 0.99 to 1.58Not applicable

Slide29

Breastfeeding: U.S.

2 studies

MacDorman & Declercq, 2016 (U.S., 2004 to 2014)

Poor-quality study

Planned home (n = 29,021) vs. planned birth center (n = 18,047) vs. planned hospital birth (n = 3,773,115)

Thornton et al., 2017 (43 states, 2006 to 2011)

Poor-quality study

Planned birth center (n = 8,776) compared to planned hospital delivery (n = 2,527)

Slide30

Breastfeeding: U.S. Studies

MacDorman & Declercq,

2016

Planned home vs. planned birth center vs. planned hospital

NNT/NNH

Breastfeeding initiation

97.9% vs. 98.1% vs. 80.8%

NNT

4 to 5Thornton, 2017Planned birth center vs. planned hospitalNNTBreastfeeding at discharge94.5% vs. 72.2%aOR 9.12, 95% CI 7.45 to 11.164

Slide31

Evidence: Noncomparative Studies

High levels of heterogeneity in the outcomes reported

Many studies excluded high-risk populations

High rates of unassisted vaginal deliveries across studies

Studies generally reported greater risks of maternal and neonatal harms for nulliparous women and women with a previous history of cesarean or instrument-assisted vaginal delivery

Breech presentation, multiple gestation, and high-risk maternal conditions (e.g., hypertension) were associated with higher risks of adverse neonatal events

Slide32

GRADE Table Findings

GRADE table includes only fair- to good-quality studies

U.S.-based evidence reported separately from non-U.S

Strength of evidence reflects all studies

Slide33

GRADE Table

Planned home or birth center vs. Planned hospital

U.S. Studies

Non-U.S. Studies

Nonoperative Vaginal Delivery

1 study

93.8% vs.

71.9%RD= 27.5 moreNNT = 3aOR 5.63, 95% CI 4.84 to 6.553 studies81% to 92.8% vs. 64.7% to 86%RD = 3.0 to 19.0NNT range from 5 to 33aOR range from 1.57 to 3.61Cesarean Delivery 1 study5.3% vs. 24.7%RD = 19.4 moreNNT = 5aOR = 0.18, 95% CI 0.16 to 0.222 studies2.8% to 4.0% vs. 11.1% to 11.7%RD range from 7.6 to 8.3 moreNNT range from 12 to 13aOR range from 0.31 to 0.76

Slide34

GRADE Table: Mode of Delivery

●●◌◌ Low confidence that planned out-of-hospital birth is associated with improved delivery mode outcome compared to planned in-hospital birth, based on 4 good- or fair-methodological-quality observational studies, 1 from U.S.

Slide35

GRADE Table

Planned home or birth center vs. Planned hospital

U.S. Studies

Non-U.S. Studies

Perinatal

or Neonatal Mortality

Neonatal death from 0 to 27 days (2 studies)

0.12% to 0.16% vs. 0.03% to 0.06%

RD from 0.09 to 0.1 moreNNH range from 1,111 to 1,000SMR 4.13, 95% CI 3.38 to 4.88; aOR 2.87, 95% CI 1.10 to 7.47Perinatal death (≥20 weeks through 27 days of life) (1 study)0.39% vs. 0.18%; RD = 0.15, NNH= 666; aOR 2.43, 95% CI 1.37 to 4.30Intrapartum or neonatal death from 0 to 7 days (2 studies)0.06% to 0.15% vs. 0.01% to 0.18% Adjusted odds ratios not statistically significant across both studies

Slide36

GRADE Table: Perinatal or Neonatal Mortality

●◌◌◌ Very low confidence that planned out-of-hospital birth is associated with increase perinatal or neonatal mortality, based on 4 good- or fair-quality observational studies, 2 from U.S.

Slide37

GRADE Table: Neonatal Morbidity

Planned home or birth center vs. Planned hospital

U.S. Studies

Non-U.S. Studies

Apgar

< 7 at 5 minutes

Apgar < 4 at 5 minutes

1 study (Oregon)

2.3% vs. 1.8% aOR = 1.31, 95% CI 1.04 to 1.661 study (Oregon)0.6% vs. 0.4% aOR = 1.56, 95% CI 0.98 to 2.471 study1.2% vs. 2.8%aOR 0.57, 95% CI 0.25 to 1.35 No non-U.S. studies reported this outcome

Slide38

GRADE Table: Neonatal Morbidity

Planned home or birth center birth vs. Planned hospital birth

U.S. Studies

Non-U.S. Studies

Neonatal

Seizures

1 study (Oregon)

0.13% vs. 0.04%

RD 0.06 to 0.07 (2 different methods)NNH 1,428 to 1,666aOR 3.60, 95% CI 1.36 to 9.50No non-U.S. studies reported this outcome

Slide39

GRADE Table: Neonatal Morbidity

Planned home or birth center birth vs. Planned hospital birth

U.S. Studies

Non-U.S. Studies

Birthplace Composite (stillbirth after onset of labor care, neonatal death 0 to 7 days, neonatal encephalopathy, meconium aspiration syndrome, brachial plexus injury, fractured humerus or clavicle)

No

U.S. studies reported this outcome

2 studies

0.05% to 3.5% for planned home or birth center birth vs. 0.05% to 4.4% for planned hospital birthAdjusted odds ratios not statistically significantly different across both studies

Slide40

GRADE Table: Neonatal Morbidity

Planned home or birth center birth vs. Planned hospital birth

U.S. Studies

Non-U.S. Studies

NICU Admission

1 study

1.7% vs. 2.9%

RD = 0.95

to 0.85 lessNNT = 105 to 117aOR 0.71, 95% CI 0.55 to 0.92No non-U.S. studies reported this outcome Ventilator Support 1 study3.8% vs. 3.3% RD: 0.97 to 1.05 moreNNH = 95 to 103aOR 1.36, 95% CI 1.14 to 1.62No non-U.S. studies reported this outcome

Slide41

GRADE Table: Neonatal Morbidity

●◌◌◌ Very low confidence that planned out-of-hospital birth associated with increased neonatal morbidity, based on 4 good- or fair-quality observational studies, 1 from U.S.

Slide42

GRADE Table: Maternal Harm

Planned home or birth center birth vs. Planned hospital birth

U.S. Studies

Non-U.S. Studies

Postpartum hemorrhage of ≥ 1000 ml

No U.S. studies of this outcome

4 studies

0.38% to 29.2% vs. 1.04% to 39.9%

Range of finding across 3 with analysisaOR =0.88, 95% CI 0.52 to 1.47aRR 1.10, 95% CI 0.67 to 1.79aOR =0.4, 95% CI 0.3 to 0.6Blood transfusion1 study0.6% vs. 0.4% RD 0.27 to 0.28 moreNNH = 357 to 370aOR 1.91, 95% CI 1.25 to 2.931 studyHome 0.5% vs. Hospital 1.2%aOR 0.72, 95% CI 0.47 to 1.12Birth center 0.6% vs. Hospital 1.2%RD 0.6 less, NNT =166aOR 0.48, 95% CI 0.32 to 0.73

Slide43

GRADE Table: Maternal Harm

Estimate of Effect for Outcome/

Confidence in Estimate

U.S. Studies

Non-U.S. Studies

Third or fourth degree perineal laceration

1 study

0.9% vs. 1.3%

RD to 0.45 to 0.54 lessNNT = 185, adjusted 222aOR 0.69 (95% CI 0.49 to 0.98) 2 studies1.9% to 4.9% vs. 3.2% to 4.2% RD 0.45 to 0.54 lessNNT = 76 to 166 aOR 0.77, 95% CI 0.57 to 1.05 to aOR 0.90, 95% CI 0.56 to 1.45

Slide44

GRADE Table: Maternal Harm

●◌◌◌ Very low confidence that planned out-of-hospital birth associated with less risk of third or fourth degree lacerations compared to planned in-hospital birth, evidence on risk of blood transfusion was conflicting (increased in U.S, similar or less from non-U.S. studies), and no U.S. findings on post-partum hemorrhage based on 5 good- or fair-quality observational studies, 1 from U.S.)

Slide45

GRADE Table: Breastfeeding

Estimate of Effect for Outcome/

Confidence in Estimate

U.S. Studies

Non-U.S. Studies

No U.S. studies with long-term outcomes identified

Exclusive breastfeeding at 6 months postpartum

4% to 22% for planned home or birth center birth vs. 1% to 9% for planned hospital birth

ARD = 3 to 13NNT = 7 to 33aOR 2.24, 95% CI 1.14 to 4.03●◌◌◌ (very low confidence that exclusive breastfeeding at 6 months was more common for planned home or birth center birth compared to planned home birth, based on 1 fair-quality observational study)

Slide46

Subgroup Considerations

Slide47

2015 CG High-Risk Groups: Prior Cesarean Delivery

2 U.S. Studies

Grunebaum, McCullough, Sapra et al., 2017 (fair quality); Tilden et al., 2017 (fair quality)

Grunebaum, McCullough, Sapra et al., 2017

Planned home vs. planned hospital

NNH

Neonatal death

0 to 27 days

Women with a prior cesarean deliverySMR 8.33, 95% CI 2.59 to 14.07Not able to calculate

Slide48

2015 CG High-Risk Groups: Prior Cesarean Delivery

Tilden, 2017

Planned home or birth

center vs. planned hospital

NNH or NNT

NICU

Admission

1.11% vs. 3.10%, aOR 0.40, 95% CI 0.29 to 0.57, p < 0.001

NNT =50Apgar < 4Apgar < 7(both at 5 minutes)0.73% vs. 0.40%, aOR 1.77, 95% CI 1.12 to 2.79, p = 0.0164.42% vs. 2.68%, aOR 1.62, 95% CI 1.35 to 1.96, p < 0.001NNH = 303NNH = 57Neonatal Seizures 0.19% vs. 0.02%, aOR 8.53, 95% CI 2.87 to 25.4, p = 0.003NNH = 588Ventilator Support 0.38% vs. 0.29%, aOR 1.36, 95% CI 0.75 to 2.46, p = 0.31Not applicableNeonatal Mortality0.13% vs. 0.08%, aOR 2.10, 95% CI 0.73 to 6.05, p > 0.05Not applicable

Slide49

2015 CG High-Risk Groups: Breech Position

Fair-quality study conducted in the U.S. (Grunebaum, McCullough, Sapra et al., 2017)

Breech presentation: SMR 8.14, 95% 2.17 to 14.11

Cephalic presentation: SMR 3.61, 95% CI 2.92 to 4.31

Slide50

Additional Subpopulations: Parity, Gestational Age, Maternal Age

2 studies

Grunebaum, McCullough, Sapra et al., 2017, U.S., 2009 to 2013

Fair-quality

Neonatal death (0 to 27 days) all planned home vs. hospital (midwife attended)

Grunebaum, McCullough, Arabin et al., 2017, U.S., 2008 to 2012

Poor-quality

Neonatal death (0 to 27 days) planned home birth group (attended by a certified nurse or other midwife) vs. hospital (midwife attended)

Slide51

Additional Subpopulations

Parity

Maternal

Age

Gestational

Age

Nulliparous

Multiparous

<35>35≥ 37 to < 41≥ 412009 to 2013All planned home vs. hospital (midwife attended)SMR 6.06, 95% CI 4.30 to 7.83SMR 3.49, 95% CI 2.68 to 4.30SMR 3.72, 95% CI 2.95 to 4.50SMR 5.11, 95% CI 3.19 to 7.03SMR 3.37, 95% CI 2.95 to 4.50SMR 4.78, 95% CI 3.43 to 6.122008 to 2012Planned home (CNM or other midwife attended) vs. hospital (midwife attended)RR 5.34, 95% CI 3.9 to 7.4RR 3.1, 95% CI 2.4 to 4RR 3.9, 95% CI 2.8 to 4.4RR 4.1, 95% CI 2.6 to 6.5

RR 3.25, 95% CI 2.6 to 4.2

RR 4.43, 95% CI 3.1 to 6.3

Slide52

Additional Subpopulations

Parity

Maternal

Age

Gestational

Age

Nulliparous

Multiparous

<35>35≥ 37 to < 41≥ 412009 to 2013All planned home vs. hospital (midwife attended)SMR 6.06, 95% CI 4.30 to 7.83SMR 3.49, 95% CI 2.68 to 4.30SMR 3.72, 95% CI 2.95 to 4.50SMR 5.11, 95% CI 3.19 to 7.03SMR 3.37, 95% CI 2.95 to 4.50SMR 4.78, 95% CI 3.43 to 6.122008 to 2012Planned home (CNM or other midwife attended) vs. hospital (midwife attended)

RR 5.34, 95% CI 3.9 to 7.4

RR 3.1, 95% CI 2.4 to 4

RR 3.9, 95% CI 2.8 to 4.4

RR 4.1, 95% CI 2.6 to 6.5

RR 3.25, 95% CI 2.6 to 4.2

RR 4.43, 95% CI 3.1 to 6.3

Slide53

Parity (England)

Birthplace Composite

Overall

Nulliparous

Multiparous

Planned Home

4.2%

9.3%

2.3%Planned Freestanding Birth Center3.5%4.5%2.7%Planned Obstetric (hospital)4.4%5.3%3.3%Effect Estimate Home vs. HospitalFBC vs. HospitalaOR 1.16, 95 % CI 0.76 to 1.77aOR 0.92, 95 % CI 0.58 to 1.46aOR 1.75, 95 % CI 1.07 to 2.86aOR 0.91, 95 % CI 0.52 to 1.60aOR 0.72, 95% CI 0.41 to 1.27aOR 0.91, 95% CI 0.46 to 1.80

Slide54

Parity (the Netherlands)

Neonatal Death

(0 to 7 days)

Nulliparous

Multiparous

Planned Home

0.05%

0.03% Planned Hospital- Midwife0.05% 0.03%Effect Estimate aOR 0.98, 95% CI 0.70 to 1.36aOR 1.07, 95% CI 0.70 to 1.65

Slide55

Subpopulations: Combinations

Fair-quality study from the U.S. (Grunebaum, McCullough, Sapra et al., 2017)

Increased risk of neonatal mortality (0 to 27 days) for several combinations of risk factors for planned home births (all providers) compared to planned hospital births attended by a midwife

Greatest risks among

Nulliparous women, 35 years and older: SMR 12.41, 95% CI 4.33 to 20.49

Nulliparous women at 41 weeks of pregnancy or later: SMR 9.57, 95% CI 5.84 to 13.30

Slide56

Evidence Review Limitations

Only observational studies identified

Strength of evidence, confidence of effect implications

Birth certificate data identifies planned home births

But planned home births transferred to hospital before delivery may be counted as hospital births (misattribution bias)

Only Oregon addresses transfer in birth certificate

Residual confounding in U.S. studies

System-level challenges

Slide57

Summary

Slide58

Overall Summary

Planned out-of-hospital birth is associated with favorable outcomes for women, but with increased risk of harm to newborns in U.S. studies

Comparisons to non-U.S. studies challenging because of differences in health care system delivery, outcome definitions, adjustments for confounding in non-U.S. studies, not available in U.S-based studies

Slide59

Summary: Maternal

Mode of delivery favors planned out-of-hospital birth

Higher rates of vaginal births, lower risk of cesarean section

Consistently greater for women planning home or birth center births compared to planned hospital births in U.S. and non-U.S.

Maternal harms were also more favorable in planned out-of-hospital settings

Lower rates of third or fourth degree perineal lacerations

Slide60

Summary: Infant

Perinatal or neonatal mortality higher for infants in planned out-of-hospital births compared to planned hospital births in U.S.-based studies of fair to good methodological quality

Magnitude of neonatal risk difference was greater in U.S. studies than international studies because neonatal death was less common across international studies

Slide61

Summary: Breastfeeding

Estimates of breastfeeding by planned place of birth tended to favor out-of-hospital settings, but the quality of evidence was very low and long term outcomes only available from international studies

Slide62

Overall Summary: Subgroups

Current evidence review affirms prior CG, increased risk of adverse infant outcomes for planned out-of-hospital birth for women with a prior cesarean delivery and breech presentation

Nulliparous women in U.S. found to have increased risk of neonatal death for planned out-of-hospital births

Non-U.S. studies also observed higher rates of adverse outcomes for nulliparous women, but direct comparisons between settings cannot be made

Additional groups at further increased risk, from findings of a single U.S. study: advanced maternal age, later gestational age, and combinations of parity and maternal age, gestational age

Slide63

Overall Summary: Subgroups

Nulliparous women in U.S. found to have increased risk of neonatal death for planned out-of-hospital births

Non-U.S. studies also observed higher rates of adverse outcomes for nulliparous or older women, but used different outcomes

Additional groups at further increased risk, from findings of a single U.S. study: advanced maternal age, later gestational age, and combinations of parity and maternal age, gestational age

No adjustment for medical conditions or known high-risk conditions (e.g., breech, prior cesarean delivery)

Included out-of-hospital attendants not covered by OHA

Slide64

Next Steps

Contextual Questions

What do applicable guidelines recommend as standards for consultation and referral or transfer of patients planning out-of-hospital birth?

What conditions require consultation?

What conditions require transfer?

What systems factors (e.g., coordination of out-of-hospital birth with consultants, hospitals, and emergency transportation) are associated with differential outcomes in out-of-hospital birth?

What is the rate of expected transfer to a hospital setting with a planned out-of-hospital birth?

What are example coverage criteria from other public and private payers?

Slide65

Discussion

[Draft coverage recommendation to be determined following the June EbGS meeting in which the relevant guidelines will also be reviewed.]

Slide66

Discussion

Values and Preferences

Slide67

Discussion

Resource Allocation

Slide68

Discussion

Other Considerations

Slide69

Discussion

Balance of Benefits and Harms

Slide70

Background: Counting OOHB

2003 CDC national birth certificate revision clarifies planned or unplanned intention for births in the home

Implemented over next 13 years

2017 birth certificate data (MacDorman & Declercq, 2018) out-of-hospital birth rate varies

Planned home births constitute >90% of home births in Alaska, Hawaii, Idaho, Indiana, Kansas, Maine, Missouri, Montana, Oregon, Pennsylvania, Utah, Vermont, Wisconsin, Wyoming

<51% in Alabama, Louisiana, New Jersey

Slide71

Background

Midwives have hospital admitting privileges in some countries (e.g., Canada, New Zealand, Norway), and that is not always the case in Oregon

In Oregon, hospitals can grant CNMs admitting privileges; this ability is not granted to licensed or unlicensed direct-entry midwives

Requirements for midwives to carry emergency medical supplies varies among countries, and sometimes varies within a country

These contextual differences may limit the generalizability of international findings to the U.S. and Oregon

Slide72

Evidence Limitations

Integration of home birth may account for variation in outcomes across studies from different countries

Regulations of out-of-hospital births, provider qualifications, and accessibility of emergency care may impact the safety of out-of-hospital births

Midwifery education and training requirements

Iceland: 2-year Master of Midwifery after completion of a BSc Nursing

The Netherlands: 4-year direct-entry baccalaureate program

England: 18-month postgraduate program after completion of BSc Nursing or 3-year direct-entry baccalaureate program

Washington: