Association between gestational weight gain and severe adverse birth outcomes in Washington State, US: A population-based retrospective cohort study, 2004-2013.

BACKGROUND: Suboptimal weight gain during pregnancy is a potentially modifiable risk factor. We aimed to investigate the association between suboptimal gestational weight gain and severe adverse birth outcomes by pre-pregnancy body mass index (BMI) categories, including obesity class I to III. METHODS AND
FINDINGS: We conducted a population-based study of pregnant women with singleton hospital births in Washington State, US, between 2004 and 2013. Optimal, low, and excess weight gain in each BMI category was calculated based on weight gain by gestational age as recommended by the American College of Obstetricians and Gynecologists and the Institute of Medicine. Primary composite outcomes were (1) maternal death and/or severe maternal morbidity (SMM) and (2) perinatal death and/or severe neonatal morbidity. Logistic regression was used to obtain adjusted odds ratios (AORs) and 95% confidence intervals. Overall, 722,839 women with information on pre-pregnancy BMI were included. Of these, 3.1% of women were underweight, 48.1% had normal pre-pregnancy BMI, 25.8% were overweight, and 23.0% were obese. Only 31.5% of women achieved optimal gestational weight gain. Women who had low weight gain were more likely to be African American and have Medicaid health insurance, while women with excess weight gain were more likely to be non-Hispanic white and younger than women with optimal weight gain in each pre-pregnancy BMI category. Compared with women who had optimal weight gain, those with low gestational weight gain had a higher rate of maternal death, 7.97 versus 2.63 per 100,000 (p = 0.027). In addition, low weight gain was associated with the composite adverse maternal outcome (death/SMM) in women with normal pre-pregnancy BMI and in overweight women (AOR 1.12, 95% CI 1.04-1.21, p = 0.004, and AOR 1.17, 95% CI 1.04-1.32, p = 0.009, respectively) compared to women in the same pre-pregnancy BMI category who had optimal weight gain. Similarly, excess gestational weight gain was associated with increased rates of death/SMM among women with normal pre-pregnancy BMI (AOR 1.20, 95% CI 1.12-1.28, p < 0.001) and obese women (AOR 1.12, 95% CI 1.01-1.23, p = 0.019). Low gestational weight gain was associated with perinatal death and severe neonatal morbidity regardless of pre-pregnancy BMI, including obesity classes I, II, and III, while excess weight gain was associated with severe neonatal morbidity only in women who were underweight or had normal BMI prior to pregnancy. Study limitations include the ascertainment of pre-pregnancy BMI using self-report, and lack of data availability for the most recent years.
CONCLUSIONS: In this study, we found that most women do not achieve optimal weight gain during pregnancy. Low weight gain was associated with increased risk of severe adverse birth outcomes, and in particular with maternal death and perinatal death. Excess gestational weight gain was associated with severe adverse birth outcomes, except for women who were overweight prior to pregnancy. Weight gain recommendations for this group may need to be reassessed. It is important to counsel women during pregnancy about specific risks associated with both low and excess weight gain.

Introduction

Weight gained during pregnancy is a potentially modifiable risk factor for adverse pregnancy outcomes [1,2]. In 2009, the Institute of Medicine (IOM) published recommendations for optimal weight gain during pregnancy for women with singleton pregnancies, based on pre-pregnancy body mass index (BMI) [3]. These recommendations have been adopted by the American College of Obstetricians and Gynecologists (ACOG) guidelines for clinical practice to potentially reduce adverse outcomes during pregnancy for both mother and baby [1].

The association between low and excess gestational weight gain and perinatal outcomes has been studied with respect to small for gestational age (SGA), large for gestational age (LGA), macrosomia, neonatal seizures, low Apgar score, neonatal intensive care unit admission, and infant death [4-9]. And while the suboptimal weight gain is associated with preterm delivery [5,10-13], the majority of studies on gestational weight gain have included only term pregnancies [8,9,14,15], which largely underestimates the overall association between suboptimal weight gain and adverse pregnancy outcomes.

The effects of gestational weight gain on severe maternal morbidity (SMM), perinatal death, and severe neonatal morbidity have been understudied. Prior reports were relatively small and showed an elevated risk of cesarean delivery, preeclampsia, and blood transfusions among women who had excess weight gain during pregnancy [5,14,15]. One study reported an overall increased risk of SMM during hospital admission for delivery among women who had excess gestational weight gain; however, similarly to studies of perinatal outcomes, these findings pertained only to term deliveries [16]. None of the prior studies included all serious adverse birth outcomes, including death and severe maternal and fetal/infant morbidity.

The prevalence of all types of maternal obesity continues to rise in developed countries [6,17-21]. The IOM/ACOG guidelines provide the same optimal weight gain recommendation for women with pre-pregnancy obesity regardless of the obesity class [1,3]. However, suboptimal weight gain likely differs with increasing BMI in these women [6,16], and this lack of specific recommendations by obesity class signifies an important knowledge gap.

This large population-based study investigated the associations between suboptimal gestational weight gain and maternal death and/or severe maternal morbidity, and perinatal death and/or neonatal morbidity, among women with various pre-pregnancy BMIs. In addition, we examined these severe adverse birth outcomes in women with pre-pregnancy class I, II, and III obesity.

Methods

This study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 STROBE Checklist). The study methodology and analysis plan (S1 Text) were developed prior to data manipulation and the actual analysis.

Ethics

This study was exempted from ethics approval by the Washington State Department of Social and Health Services, US. According to the Washington State Institutional Review Board, a research study can be exempt from Washington State Institutional Review Board review if “the research does not involve obtaining data about subjects through interaction or intervention with individuals, and does not involve obtaining identifiable private information about subjects from confidential records.” Our study was granted this exemption (DOH Exempt Request E-050515-H).

Study population and data sources

Information on all singleton births between 20 and 45 weeks’ gestation that occurred in Washington State hospitals from 1 January 2004 to 31 December 2013 was obtained from the Washington State Department of Health. Data sources included 2 linked population databases: BERD (Birth Event Record Database), including information from live birth and fetal death certificates, and CHARS (Comprehensive Hospital Abstract Reporting System), including birth hospitalization data. CHARS data provided information on delivery and newborn hospitalizations (up to 9 diagnostic and 9 procedure codes using the International Classification of Diseases–9th Revision, Clinical Modification [ICD-9-CM]), the type of health insurance coverage, intensive care unit (ICU) admission, and death during hospitalization. BERD included information on maternal characteristics such as pre-pregnancy BMI, age, race/ethnicity, education, marital status, parity, and obstetric history (previous infant death, preterm birth, or SGA birth in parous women); pregnancy characteristics such weight gain during pregnancy, smoking during pregnancy, assisted conception, gestational hypertension, and gestational diabetes; and birth characteristics such as year of birth, infant sex, gestational age at delivery, labour characteristics, mode of delivery, stillbirth, neonatal death, neonatal seizures, and congenital anomalies. Race/ethnicity was recorded in the birth certificate (BERD) as mother’s self-identified race/ethnicity with the categories non-Hispanic white, black or African American, Native American or Alaska Native, and other (e.g., Asian: Indian, Chinese, Filipino, Japanese; also including “other” as an open-ended category); Hispanic origin was also self-reported and recorded as a separate category. All data were abstracted by trained abstractors using standardized forms (S1 Form). Pre-pregnancy diabetes mellitus and pre-pregnancy hypertension were identified from both BERD and CHARS; the condition was classified as present if indicated in at least 1 dataset. The accuracy and completeness of the BERD (birth certificate) and CHARS (hospitalization) data was monitored by Washington State Department of Health through annual assessments and consistency checks [22]. Multiple births were excluded because twin or triplet sets could not be identified in the data source.

Self-reported pre-pregnancy weight and height was used to calculate BMI, classified according to the World Health Organization [23] into the following categories: normal BMI (18.5–24.9 kg/m2), underweight (<18.5 kg/m2), overweight (25.0–29.9 kg/m2), and obesity class I (30.0–34.9 kg/m2), class II (35.0–39.9 kg/m2), and class III (≥40 kg/m2). BMI values were checked for consistency by the Department of Health: Maternal BMI values outside the expected range were flagged as potentially erroneous and rectified before the data were released. Gestational age at delivery was based on ultrasound dating; date of last menstrual period was used for women with missing ultrasound data.

Weight gain during pregnancy

Pregnancy weight gain was defined as the woman’s weight measured at childbirth minus her self-reported pre-pregnancy weight. Optimal weight gain was defined by IOM/ACOG guidelines [1]. For term pregnancy (≥37 weeks), optimal weight gain was defined as the recommended range of total weight gain during pregnancy according to the pre-pregnancy BMI category: 28–40 lbs for underweight (<18.5 kg/m2), 25–35 lbs for normal BMI (18.5–24.9 kg/m2), 15–25 lbs for overweight (25.0–29.9 kg/m2), and 11–20 lbs for obese (≥30.0 kg/m2). Women who delivered at term and had gestational weight gain lower or higher than the recommended weight gain were classified as having low or excess weight gain, respectively, i.e., suboptimal gestational weight gain. To account for gestational age at delivery among women who delivered preterm, optimal weight gain per week was calculated using the recommended weekly weight gain range according to the pre-pregnancy BMI category: 1.0–1.3 lbs for underweight, 0.8–1.0 lbs for normal BMI, 0.5–0.7 lbs for overweight, and 0.4–0.6 lbs for obese [1]. These recommendations pertain to the second trimester, assuming a weight gain range of 1.1–4.4 lbs in the first trimester; all calculations were performed consistently for each BMI category [1]. Comparisons were made between women with low and excess weight gain during pregnancy and those who achieved a recommended (optimal) weight gain.

Outcomes

The primary outcomes were composite measures including severe adverse outcomes as follows: (1) maternal death (death during delivery hospitalization) and/or SMM and (2) perinatal death (including stillbirth or neonatal death within 28 days) and/or severe neonatal morbidity. SMM included maternal conditions that have a high case fatality rate or that lead to vital organ damage or serious long-term sequelae. A list of such conditions was developed by the Canadian Perinatal Surveillance System [24]; in addition, we included conditions recognized as SMM by the US Centers for Disease Control and Prevention [25-27], for instance, conditions requiring life-saving procedures such as mechanical ventilation or conversion of cardiac rhythm (see full list and definitions in S1 Table). Information on SMM was obtained from the CHARS dataset.

Severe neonatal morbidity was also identified from the CHARS database using ICD-9-CM diagnostic codes and included the following conditions: bronchopulmonary dysplasia, respiratory distress syndrome, retinopathy of prematurity, intraventricular hemorrhage grade 3 or more, periventricular leukomalacia, neonatal sepsis, necrotizing enterocolitis, and severe birth trauma. The occurrence of any of these conditions or neonatal seizures (information obtained from birth certificates) composed severe neonatal morbidity.

Statistical analyses

Logistic regression was used to obtain adjusted odds ratios (AORs) and 95% CIs expressing the association between pregnancy weight gain and the composite outcomes and their components stratified by pre-pregnancy BMI. AORs were adjusted for demographic and pre-pregnancy characteristics that are known to be associated with adverse maternal and perinatal outcomes. These included maternal age (<25 years, 25–34 years, or ≥35 years), maternal education (high school graduation or higher versus less than high school graduation), maternal marital status (single, widowed, or separated versus married or cohabitating), maternal race/ethnicity (Hispanic, African American, Native American, or other versus non-Hispanic white), parity (nulliparous or parity ≥ 4 versus parity 1–3), assisted conception (no versus yes), maternal smoking during pregnancy (no versus yes), type of health insurance (Medicaid versus private, self-pay, or other), year of birth, and fetal sex (female versus male). For neonatal outcomes, AORs were also adjusted for congenital anomalies.

Supplementary analyses

In the supplementary analyses, we first examined associations between pregnancy weight gain and SMM components. We then assessed the associations between suboptimal weight gain and composite outcomes among obese women stratified by obesity class I, II, and III. We also assessed associations between weight gain and SGA and LGA, as secondary perinatal outcomes, to allow for comparisons with previous studies.

Sensitivity analyses

We examined the associations between pregnancy weight gain and composite outcomes in a subset restricted to term pregnancies. The aim of this analysis was to facilitate comparisons with previous studies and to assess whether the results differ markedly from those for the total study population.

All analyses were performed using R statistical software (R version 3.5.1). Complete case analyses were performed (the number of records with any missing value was <3% of the study population). All p-values are reported as recommended by the American Statistical Association [28].

Results

Study population

Overall, 952,212 mothers gave birth (live birth or stillbirth) in Washington State between 1 January 2004 and 31 December 2013. We excluded births that occurred out of state, multiple births, births before 20 weeks’ gestation, births to women aged <15 or >60 years (35,598 mothers, 3.7%), births that occurred out of hospital (24,716 mothers, 2.6%), and births that could not be matched with hospital records (64,609 mothers, 6.8%). From the remaining 827,289 births, 104,450 were excluded due to missing pre-pregnancy BMI or gestational weight gain (12.6%). The study population consisted of 722,839 women with a singleton pregnancy for whom information on pre-pregnancy weight and gestational weight gain was available. Of these, 3.1% were underweight, 48.1% had normal BMI, 25.8% were overweight BMI, and 23% were obese prior to pregnancy. Among the obese women, 56.6% had class I obesity, 26.7% had class II obesity, and 16.7% had class III obesity.

Overall, there were 143,509 (19.9%) women with low weight gain, 227,715 (31.5%) women with optimal weight gain, and 351,615 (48.6%) women with excess weight gain during pregnancy. Maternal demographic and clinical characteristics of women who had low, optimal, and excess weight gain differed by pre-pregnancy BMI.

Low weight gain

Low weight gain was observed in 28.7% of women who were underweight, 21.1% of women with normal BMI, 14.1% of women who were overweight, and 22.6% of obese women (Fig 1). In general, women with low weight gain were more likely to be African American, more likely to have Medicaid health insurance, and less likely to be married than women with optimal weight gain in each pre-pregnancy BMI category. Except for underweight women, they were also more likely to be young, to have pre-pregnancy diabetes, and to smoke during pregnancy (Table 1). Overall, women with low weight gain had higher rates of spontaneous vaginal delivery and lower rates of prolonged labour. The rates of gestational diabetes were higher in these women, especially in those who were underweight and those with normal BMI prior to pregnancy (Table 2).

Fig 1

Pre-pregnancy BMI category and gestational weight gain for singleton births, Washington State, 2004–2013.

Table 1

Demographic and pregnancy characteristics by pre-pregnancy BMI and gestational weight gain—Singleton births, Washington State, 2004–2013.

Characteristic	Pre-pregnancy underweightN = 22,714 (3.1%)			Pre-pregnancy normal BMIN = 347,555 (48.1%)			Pre-pregnancy overweightN = 186,662 (25.8%)			Pre-pregnancy obeseN = 165,908 (23.0%)
	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain
Total population (percent of BMI category)	10,237(45.1)	6,510(28.7)	5,967(26.3)	132,572(38.1)	73,290(21.1)	141,693(40.8)	45,746(24.5)	26,284(14.1)	114,632(61.4)	39,160(23.6)	37,425(22.6)	89,323(53.8)
Maternal age, years
15–19	1,280(12.5)	810(12.4)	1,051(17.6)	9,428(7.1)	6,795(9.3)	13,949(9.8)	2,701(5.9)	1,929(7.3)	8,205(7.2)	1,551(4.0)	1,604(4.3)	5,392(6.0)
20–24	2,844(27.8)	1,786(27.4)	1,786(33.7)	25,838(19.5)	16,542(22.6)	33,370(23.5)	9,551(20.9)	5,907(22.5)	26,105(22.8)	8,490(21.7)	8,470(22.6)	22,052(24.7)
25–29	2,824(27.6)	1,719(26.4)	1,511(25.3)	37,758(28.5)	19,567(26.7)	40,760(28.8)	13,147(28.74)	7,369(28.0)	34,215(29.8)	12,214(31.2)	11,411(30.5)	27,442(30.7)
30–34	2,193(21.4)	1,386(21.3)	964(16.2)	37,232(28.1)	18,220(24.9)	34,510(24.4)	11,905(26.0)	6,383(24.3)	29,155(25.4)	10,133(25.9)	9,530(25.5)	21,625(24.21)
35–39	923(9.0)	642(9.9)	375(6.3)	18,285(13.8)	9,761(13.3)	15,851(11.2)	6,707(14.7)	3,650(13.9)	13,830(12.1)	5,384(13.7)	5,056(13.51)	10,374(11.6)
≥40	173(1.7)	167(2.57)	58(1.0)	4,031(3.0)	2,405(3.3)	3,253(2.3)	1,735(3.8)	1,046(4.0)	3,122(2.7)	1,388(3.5)	1,354(3.6)	2,438(2.7)
Race/ethnicity
Non-Hispanic white	6,657(65.03)	3,774(58.0)	4,351(72.9)	97,334(73.4)	46,579(63.5)	111,825(78.9)	30,095(65.8)	15,653(59.5)	88,435(77.1)	27,968(71.4)	26,212(70.0)	68,472(76.7)
African American	305(3.0)	291(4.5)	207(3.5)	3,303(2.5)	3,021(4.1)	3,778(2.7)	1,959(4.3)	1,647(6.3)	3,849(3.4)	1,969(5.0)	2,427(6.5)	3,755(4.2)
Native American	106(1.0)	74(1.1)	106(1.8)	1,460(1.1)	1,040(1.4)	2,270(1.6)	922(2.0)	604(2.3)	2,474(2.2)	1,206(3.1)	1,289(3.4)	2,748(3.1)
Hispanic	733(7.2)	661(10.1)	406(6.8)	11,759(8.9)	10,543(14.4)	9,860(7.0)	8,299(18.1)	5,288(20.1)	12,176(10.6)	5,857(15.0)	5,377(14.4)	9,712(10.9)
Other	2,405(23.5)	1,693(26.0)	864(14.5)	18,423(13.9)	11,937(16.3)	13,639(9.6)	4,368(9.5)	2,990(11.4)	7,438(6.5)	2,077(5.3)	2,018(5.4)	4,471(5.0)
Maternal education less than high school	220(2.1)	215(3.3)	140(2.3)	3,256(2.46)	3,543(4.8)	2,571(1.8)	2,924(6.4)	1,899(7.2)	3,534(3.1)	1,968(5.0)	1,583(4.2)	2,744(3.1)
Smoking	1,293(12.6)	759(11.7)	1,164(19.5)	9,187(6.9)	6,486(8.8)	14,780(10.4)	4,046(8.8)	2,941(11.2)	11,132(9.7)	4,364(11.1)	5,011(13.4)	10,794(12.1)
Married or cohabitating	6,776(66.2)	4,271(65.6)	3,119(52.3)	98,379(74.2)	48,806(66.6)	95,464(67.4)	30,618(66.9)	16,327(62.1)	76,480(66.7)	25,238(64.45)	23,331(62.3)	55,209(61.8)
Insurance
Medicaid	4,148(40.5)	2,928(45.0)	2,983(50.0)	41,078(31.0)	32,015(43.7)	47,936(33.8)	21,224(46.4)	14,078(53.6)	42,874(37.4)	19,074(48.7)	18,947(50.6)	39,406(44.1)
Private	5,269(51.5)	3,061(47.0)	2,445(41.0)	82,752(62.4)	36,124(49.3)	83,198(58.7)	21,420(46.8)	10,363(39.4)	63,299(55.2)	17,170(43.8)	15,490(41.4)	43,123(48.3)
Self-pay	114(1.1)	88(1.3)	75(1.3)	1,348(1.0)	863(1.2)	1,410(1.0)	387(0.8)	261(1.0)	980(0.8)	284(0.7)	303(0.8)	597(0.7)
Othera	647(6.3)	402(6.2)	445(7.5)	6,750(5.1)	3,957(5.4)	8,447(6.0)	2,532(5.5)	1,400(5.3)	6,966(6.1)	2,497(6.4)	2,440(6.52)	5,867(6.6)
Parity
Nulliparous	5,324(52.0)	3,060(47.0)	3,457(57.9)	57,911(43.7)	28,477(38.9)	73,305(51.7)	13,604(29.7)	8,165(31.1)	50,022(43.6)	10,680(27.3)	10,032(26.8)	35,644(39.9)
1–3 prior births	4,687(45.8)	3,247(49.9)	2,308(38.7)	70,039(52.8)	41,276(56.3)	63,809(45.0)	28,911(63.2)	15,900(60.5)	59,369(51.8)	25,116(64.1)	23,745(63.5)	48,357(54.1)
≥4 prior births	174(1.7)	170(2.6)	142(2.4)	3,884(2.9)	3,106(4.2)	3,701(2.6)	3,005(6.6)	2,025(7.7)	4,571(4.0)	3,148(8.0)	3,376(9.0)	4,783(5.3)
Assisted conception	57(0.6)	49(0.7)	36(0.6)	1,451(1.1)	621(0.9)	1,412(1.0)	390(0.8)	169(0.64)	1,110(1.0)	371(0.9)	318(0.9)	838(0.9)
Pre-pregnancy diabetes mellitus	15(0.2)	10(0.1)	15(0.3)	364(0.3)	277(0.4)	587(0.4)	355(0.8)	238(0.9)	885(0.8)	765(1.9)	816(2.2)	1,750(2.0)
Pre-pregnancy hypertension	19(0.2)	23(0.3)	21(0.3)	529(0.4)	323(0.4)	761(0.5)	419(0.9)	214(0.8)	1,342(1.2)	1,276(3.3)	1,241(3.3)	2,876(3.2)
Male fetal sex	5,231(51.1)	3,150(48.4)	3,142(52.7)	67,192(50.7)	35,923(49.0)	75,215(53.1)	22,846(49.9)	13,031(49.6)	59,853(52.2)	19,810(50.6)	18,453(49.3)	46,758(52.4)
Congenital anomaliesb	634(6.2)	422(6.5)	327(5.5)	7,411(5.6)	4,433(6.0)	7,930(5.6)	2,723(5.9)	1,715(6.5)	6,541(5.7)	2,444(6.2)	2,391(6.4)	5,466(6.1)

Data given as N (%). Underweight defined as BMI < 18.5 kg/m2, normal weight as BMI 18.5–24.9 kg/m2, overweight as BMI 25.0–29.9 kg/m2, and obese class I–III as BMI ≥ 30.0 kg/m2.

aIncludes other government insurance, student insurance, Indian Health Service, and other programs.

bIncludes the following conditions observed within first 24 hours after birth: anencephaly, meningomyelocele or spina bifida, cyanotic congenital heart disease, congenital diaphragmatic hernia, omphalocele, gastroschisis, limb reduction, cleft lip, cleft palate, Down syndrome, chromosomal disorders, and hypospadias.

Table 2

Labour and delivery characteristics by pre-pregnancy BMI and gestational weight gain—Singleton births, Washington State, 2004–2013.

Characteristic	Pre-pregnancy underweightN = 22,714 (3.1%)			Pre-pregnancy normal BMIN = 347,555 (48.1%)			Pre-pregnancy overweightN = 186,662 (25.8%)			Pre-pregnancy obeseN = 165,908 (23.0%)
	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain	Optimal weight gain	Low weight gain	Excess weight gain
Total population (percent of BMI category)	10,237(45.1)	6,510(28.7)	5,967(26.3)	132,572(38.1)	73,290(21.1)	141,693(40.8)	45,746(24.5)	26,284(14.1)	114,632(61.4)	39,160(23.6)	37,425(22.6)	89,323(53.8)
GA at delivery
20–33 weeks	154(1.5)	184(2.8)	162(2.7)	1,265(0.9)	1,416(1.9)	2,761(2.0)	579(1.3)	734(2.8)	1,786(1.6)	663(1.7)	1,094(2.9)	1,634(1.8)
34–36 weeks	559(5.4)	519(8.0)	437(7.3)	4,595(3.5)	3,612(4.9)	9,359(6.6)	1,701(3.7)	1,619(6.2)	6,077(5.3)	1,901(4.9)	2,475(6.6)	5,412(6.1)
≥37 weeks	9,524(93.0)	5,807(89.2)	5,368(90.0)	126,712(95.6)	68,262(93.1)	129,573(91.5)	43,466(95.0)	23,931(91.1)	106,769(93.1)	36,596(93.5)	33,856(90.5)	82,277(92.1)
Vaginal delivery
Spontaneous	7,379(72.1)	4,794(73.6)	4,083(68.4)	92,888(70.1)	52,508(71.6)	93,472(66.0)	30,477(66.62)	18,115(68.9)	71,393(62.3)	22,888(58.4)	22,496(60.1)	48,399(54.2)
VBAC	106(1.0)	74(1.1)	56(0.9)	1,933(1.5)	1,175(1.6)	1,681(1.2)	905(1.98)	491(1.9)	1,649(1.4)	663(1.7)	636(1.7)	1,185(1.3)
Forceps	112(1.1)	68(1.0)	58(1.0)	1,147(0.9)	509(0.7)	1,330(0.9)	228(0.5)	136(0.5)	806(0.7)	163(0.4)	142(0.4)	455(0.5)
Forceps VBAC	2(0.0)	2(0.0)	0(0.0)	32(0.0)	21(0.0)	35(0.0)	19(0.04)	5(0.02)	28(0.02)	10(0.0)	15(0.0)	15(0.0)
Vacuum	719(7.0)	417(7.0)	468(7.8)	7,415(5.6)	3,715(5.6)	8,226(5.8)	1,766(3.9)	1,063(3.9)	5,070(4.4)	1,133(2.8)	1,144(2.9)	2,910(3.3)
Vacuum VBAC	19(0.2)	10(0.1)	12(0.2)	233(0.2)	133(0.2)	167(0.1)	100(0.2)	50(0.2)	171(0.2)	61(0.2)	51(0.1)	102(0.1)
C-section
Primary with labour	633(6.2)	345(5.3)	504(8.4)	9,287(7.0)	4,028(5.5)	14,271(10.1)	3,325(7.3)	1,539(5.9)	12,692(11.1)	3,657(9.3)	3,165(8.5)	12,149(13.6)
Primary with no labour	722(7.0)	440(6.8)	465(7.8)	9,608(7.2)	5,422(7.4)	12,368(8.7)	3,556(7.8)	2,219(8.4)	10,938(9.5)	4,119(10.5)	4,004(10.7)	10,978(12.3)
Repeat with labour	39(0.4)	27(0.4)	20(0.3)	767(0.6)	442(0.6)	749(0.5)	385(0.8)	201(0.8)	865(0.7)	428(1.1)	365(1.0)	821(0.9)
Repeat with no labour	504(4.9)	331(5.1)	300(5.0)	9,241(6.97)	5,304(7.2)	9,364(6.6)	4,980(10.9)	2,445(9.3)	10,999(9.6)	6,034(15.4)	5,390(14.4)	12,291(13.8)
One previous C-section	558(5.4)	357(5.5)	310(5.2)	9,734(7.34)	5,475(7.5)	9,484(6.7)	4,666(10.2)	2,209(8.4)	10,355(9.0)	5,018(12.8)	4,356(11.6)	10,397(11.6)
≥1 previous C-section	106(1.0)	85(1.3)	77(1.3)	2,411(1.8)	1,570(2.1)	2,464(1.7)	1,699(3.7)	961(3.7)	3,312(2.9)	2,160(5.5)	2,069(5.5)	3,979(4.4)
Induced labour	1,853(18.1)	1,046(16.1)	1,241(20.8)	26,284(19.8)	12,733(17.4)	33,785(23.8)	9,797(21.4)	4,902(18.6)	29,672(25.9)	10,105(25.8)	9,268(24.8)	25,867(29.0)
PROM (≥12 hours)	542(5.3)	322(4.9)	327(5.5)	6,759(5.1)	3,125(4.3)	8,821(6.2)	2,066(4.5)	969(3.7)	6,511(5.7)	1,604(4.1)	1,599(4.3)	4,684(5.2)
Precipitous labour	338(3.3)	300(4.6)	146(2.4)	4,758(3.6)	3,065(4.2)	3,693(2.6)	1,556(3.4)	995(3.8)	2,947(2.6)	1,058(2.7)	1,116(3.0)	1,895(2.1)
Prolonged labour (≥20 hours)	210(2.1)	110(1.7)	138(2.3)	2,825(2.1)	1,234(1.7)	3,648(2.6)	876(1.9)	404(1.5)	2,954(2.6)	827(2.1)	641(1.7)	2,287(2.6)
Gestational hypertension	228(2.2)	107(1.6)	226(3.8)	3,320(2.5)	1,547(2.1)	7,333(5.2)	1,738(3.8)	808(3.1)	7,862(6.9)	3,004(7.7)	2,379(6.4)	10,062(11.3)
Gestational diabetes	285(2.8)	265(4.1)	130(2.2)	4,713(3.6)	4,437(6.1)	3,652(2.6)	4,049(8.8)	2,622(10.0)	5,820(5.1)	5,507(14.1)	5,920(15.8)	8,670(9.7)
Chorioamnionitis	205(2.0)	113(1.7)	138(2.3)	2,541(1.9)	1,107(1.5)	3,558(2.5)	763(1.7)	342(1.3)	2,508(2.2)	536(1.4)	476(1.3)	1,700(1.9)
Previous infant death or preterm or SGA infant	148(1.4)	102(1.6)	84(1.4)	1,756(1.3)	1,133(1.5)	1,858(1.3)	789(1.7)	477(1.8)	1,702(1.5)	767(2.0)	788(2.1)	1,713(1.9)

Data given as N (%). Underweight defined as BMI < 18.5 kg/m2, normal weight as BMI 18.5–24.9 kg/m2, overweight as BMI 25.0–29.9 kg/m2, and obese class I–III as BMI ≥ 30.0 kg/m2.

C-section, cesarean section; GA, gestational age; PROM, premature rupture of membranes; SGA, small-for-gestational age; VBAC, vaginal birth after cesarean section.

Excess weight gain

Excess weight gain was observed in 26.3% of women who were underweight, 40.8% of women with normal BMI, 61.4% of women who were overweight, and 53.8% of obese women (Fig 1). In general, women with excess weight gain were more likely to be young, non-Hispanic white, and nulliparous than women who had optimal weight gain in each pre-pregnancy BMI category (Table 1). Except for underweight women, women with excess weight gain were more likely to have private health insurance. Among those with low and normal pre-pregnancy BMI, women who had excess weight gain were more likely to smoke during pregnancy and less likely to be married than women with optimal weight gain (Table 1). Women with excess weight gain were also more likely to have gestational hypertension, chorioamnionitis, induced labour, premature rupture of membranes, and cesarean delivery (Table 2).

Maternal outcomes in women with low gestational weight gain

Overall, women with low gestational weight gain had a significantly higher rate of maternal death than women with optimal weight gain (7.97 versus 2.63 per 100,000, p = 0.027). The rates of maternal death/SMM were 1.57%, 1.55%, 1.90%, and 1.75% among underweight, normal weight, overweight, and obese women, respectively (S2 Table). Among normal weight and overweight women, those with low gestational weight gain had higher adjusted odds of maternal death/SMM compared with those who had optimal gestational weight gain (AOR 1.12, 95% CI 1.04–1.21, p = 0.004, and AOR 1.17; 95% CI 1.04–1.32, p = 0.009, respectively; Table 3). Women who had normal pre-pregnancy BMI and low weight gain had elevated odds of cerebrovascular/central nervous system morbidity, while overweight women had increased odds of ICU admission (S4 Table).

Table 3

Adjusted odds ratios (AORs) for severe adverse birth outcomes by gestational weight and pre-pregnancy BMI—Singleton births, Washington State, 2004–2013 (AOR relative to optimal weight gain in each pre-pregnancy BMI category).

Outcome	Pre-pregnancy underweight		Pre-pregnancy normal BMI		Pre-pregnancy overweight		Pre-pregnancy obese
	Low weight gain	Excess weight gain	Low weight gain	Excess weight gain	Low weight gain	Excess weight gain	Low weight gain	Excess weight gain
Maternal death/severe maternal morbiditya	0.96 (0.74–1.24)0.737	1.28 (1.00–1.63)0.049	1.12 (1.04–1.21)0.004	1.20 (1.12–1.28)<0.001	1.17 (1.04–1.32)0.009	1.07 (0.98–1.18)0.209	1.07 (0.95–1.02)0.244	1.12 (1.01–1.23)0.019
Perinatal deathb	2.03 (1.26–3.29)0.003	1.28 (0.75–2.20)0.369	2.14 (1.86–2.46)<0.001	1.41 (1.24–1.61)<0.001	2.53 (2.09–3.07)<0.001	0.87 (0.73–1.04)0.099	2.04 (1.72–2.42)<0.001	0.82 (0.69–0.97)0.032
Perinatal death/severe neonatal morbidityb	1.92 (1.50–2.46)<0.001	1.49 (1.15–1.95)0.003	1.41 (1.31–1.53)<0.001	1.38 (1.29–1.85)<0.001	1.65 (1.47–1.84)<0.001	0.99 (0.90–1.08)0.808	1.55 (1.40–1.72)<0.001	1.05 (0.96–1.15)0.203
Combined maternal and/or perinatal adverse outcomeb	1.35 (1.13–1.62)<0.001	1.38 (1.15–1.66)<0.001	1.26 (1.19–1.33<0.001	1.28 (1.22–1.34)<0.001	1.39 (1.27–1.51)<0.001	1.02 (0.95–1.09)0.575	1.32 (1.22–1.42)<0.001	1.09 (1.01–1.16)0.017

Values given as AOR (95% CI) and p-value. Two-sided p-values were calculated using multivariable logistic regression Wald chi-squared test; statistically significant AORs given in bold.

aAdjusted for maternal age (<25 years, 25–34 years, or ≥35 years), maternal education (high school graduation or higher versus less than high school graduation), marital status (single, widowed, or separated versus married or cohabitating), race/ethnicity (Hispanic, African American, Native American, or other versus non-Hispanic white), parity (nulliparous or parity ≥ 4 versus parity 1–3), assisted conception (no versus yes), smoking during pregnancy (no versus yes), type of health insurance (Medicaid versus private, self-pay, or other), year of birth, and fetal sex (female versus male).

bAdjusted for congenital anomalies in addition to covariates above.

Maternal outcomes in women with excess gestational weight gain

The maternal death rate among women with excess weight gain was 3.98 per 100,000 (versus 2.63 per 100,000 in the optimal weight gain group, p = 0.394). The rates of maternal death/SMM were 1.94%, 1.54%, 1.56%, and 1.81% among underweight, normal weight, overweight, and obese women, respectively (S2 Table). Adjusted for other covariates, among underweight, normal weight, and obese women, excess gestational weight gain was associated with elevated odds of maternal death/SMM compared with those with optimal weight gain (AOR 1.28, 95% CI 1.00–1.63, p = 0.049; AOR 1.20, 95% CI 1.12–1.28, p < 0.001; and AOR 1.12, 95% CI 1.01–1.23, p = 0.019; respectively; Table 3). All women with excess weight gain had elevated odds of sepsis (mainly puerperal sepsis). In women who were underweight and obese, excess weight gain was also associated with higher odds of respiratory morbidity; those with normal weight had increased odds of postpartum hemorrhage with transfusion and potentially life-saving medical interventions; overweight women had higher odds of ICU admission (S3 Table).

Perinatal outcomes

In each pre-pregnancy BMI category, women with low weight gain had higher odds of the composite adverse perinatal outcome—perinatal death and severe neonatal morbidity—than women with optimal weight gain (Table 3). Among women with low weight gain, the odds of stillbirth was significantly elevated in all pre-pregnancy BMI categories, except for a non-statistically-significant increase in underweight women (S4 Table).

In underweight and normal weight women, excess weight gain was associated with higher odds of perinatal death/severe neonatal morbidity compared with those with optimal weight gain (Table 3). Among women with normal pre-pregnancy BMI, those who had excess gestational weight gain had increased odds of perinatal death/severe neonatal morbidity, while in overweight women, excess weight gain did not increase the odds of any severe adverse perinatal outcomes. Obese women with excess weight gain had lower odds of perinatal death than obese women with optimal weight gain (Table 3 and S4 Table).

Secondary analyses including only obese women

Regardless of the pre-pregnancy obesity class, obese women with low or excess gestational weight gain did not have a significantly increased odds of the composite outcome maternal death/SMM (Table 4). An increased odds of ICU admission was observed among women who had pre-pregnancy class I obesity and low gestational weight gain (S6 Table). Excess gestational weight gain was significantly associated with increased odds of sepsis among women with class II obesity (S6 Table). Perinatal death/severe neonatal morbidity was elevated among women with low weight gain in each pre-pregnancy obesity class group (Table 4).

Table 4

Adjusted odds ratios (AORs) for severe adverse outcomes by gestational weight gain and obesity class—Singleton births, Washington State, 2004–2013 (AOR relative to optimal weight gain in each obesity class category).

Outcome	Pre-pregnancy BMI
	Obesity class I		Obesity class II		Obesity class III
	Low weight gain	Excess weight gain	Low weight gain	Excess weight gain	Low weight gain	Excess weight gain
Maternal death/severe maternal morbiditya	1.05 (0.90–1.25)0.493	1.07 (0.94–1.22)0.298	1.10 (0.88–1.36)0.397	1.20 (1.00–1.44)0.05	1.03 (0.81–1.30)0.816	1.23 (0.99–1.54)0.059
Perinatal death/severe neonatal morbidityb	1.52 (1.30–1.76)<0.001	1.04 (0.91–1.18)0.441	1.68 (1.39–2.02)<0.001	1.06 (0.89–1.27)0.375	1.42 (1.15–1.76)0.002	1.16 (0.94–1.44)0.180
Combined maternal and perinatal severe adverse outcomesb	1.29 (1.15–1.44)<0.001	1.05 (0.96–1.16)0.244	1.40 (1.21–1.61)<0.001	1.12 (0.98–1.28)0.081	1.24 (1.06–1.46)<0.001	1.21 (1.03–1.42)0.016

Values given as AOR (95% CI). Two-sided p-values were calculated using multivariable logistic regression Wald chi-squared test; statistically significant AORs given in bold.

aAdjusted for maternal age (<25 years, 25–34 years, or ≥35 years), maternal education (high school graduation or higher versus less than high school graduation), marital status (single, widowed, or separated versus married or cohabitating), race/ethnicity (Hispanic, African American, Native American, or other versus non-Hispanic white), parity (nulliparous or parity ≥ 4 versus parity 1–3), assisted conception (no versus yes), smoking during pregnancy (no versus yes), type of health insurance (Medicaid versus private, self-pay, or other), year of birth, and fetal sex (female versus male).

bAdjusted for congenital anomalies in addition to covariates above.

Results restricted to women with term pregnancy included 672,141 women (93% of all women; S7 Table). Among women with normal pre-pregnancy BMI, both low and excess weight gain were associated with increased odds of maternal death/SMM compared with optimal weight gain. The increased risk of perinatal death/severe neonatal morbidity was no longer observed, except for obese women with low and excess weight gain (S7 Table).

Discussion

In this study, we found that most women do not achieve optimal weight gain during pregnancy. Low weight gain was consistently associated with increased risk of severe adverse birth outcomes, and in particular with maternal death and perinatal death. On the other hand, excess weight gain was associated with adverse maternal and perinatal outcomes only in women who had pre-pregnancy BMI in the normal range and below. Women who were overweight prior to pregnancy and gained excess weight during pregnancy did not have elevated risks of serious adverse birth outcomes compared with overweight women with optimal weight gain. Pre-pregnancy obesity and excess weight gain was associated with elevated risk of SMM.

Our results show that only about one-third of women achieved optimal weight gain, while almost half gained excess weight during pregnancy. The excess weight gain was more common in overweight and obese women (61% and 53%, respectively), which is consistent with the literature [29,30]. Previous studies underscored increases in SMM only among women with excess weight gain [14,16]; however, our results show an increase in maternal death/SMM also in women with low gestational weight gain. Prior studies were smaller and were often restricted to women with term pregnancy [14,16]. Our sensitivity analyses show that such restriction provides different results for perinatal outcomes, as women with low and excess weight gain are more likely to deliver preterm. Two large meta-analyses focusing on less severe adverse birth outcomes showed elevated rates of LGA and macrosomia in women with excess weight gain in all pre-pregnancy BMI categories, and elevated rates of SGA in women with low weight gain [2,31], which is consistent with our results. Low weight gain is known to be associated with preterm birth [4,5] and with infant death during the first year after birth [13,32]. The new findings in our study include the association between low weight gain and adverse perinatal outcomes, including stillbirth and neonatal death. Our results suggest that low weight gain during the second and third trimester can potentially serve as a marker for an increased risk of stillbirth. As such, weight gain monitoring may help to prevent stillbirth by increased prenatal surveillance, ultrasound checks, and early delivery in cases of worsened intrauterine growth restriction.

We did not observe any association between excess weight gain in overweight women and severe adverse maternal and perinatal outcomes, confirming the findings of Platner et al. [16] from a large study of pregnant women in New York City. This finding, together with the very high rate of excess weight gain in this group, suggests that the IOM/ACOG guidelines for optimal weight gain in these women may need a re-evaluation.

With respect to individual maternal morbidities, the association between excess weight gain and pregnancy complications requiring blood transfusion and ventilation is consistent with the literature [14,16]. In addition, we observed an increased risk of sepsis, especially puerperal sepsis, among women with excess weight gain regardless of pre-pregnancy BMI. This may indicate a potential link between excess weight gain and maternal immune response or placental function, which is consistent with a previously reported association between excess weight gain and chorioamnionitis [14]. Of note is also an elevated rate of gestational diabetes among women with low weight gain, consistently observed in all pre-pregnancy BMI groups. Such an association has been reported previously [14], suggesting that gestational diabetes may affect optimal nutritional intake in the second and third semesters, and potentially lead to adverse perinatal outcomes [33]. It is not entirely clear whether gestational diabetes control leads to low weight gain or whether another common cause leads to both gestational diabetes and low weight gain, and subsequently negatively impacts perinatal outcomes. While low and excess weight gain are potentially modifiable risk factors and may be causally related to the adverse outcomes observed, suboptimal weight gain can also result from underlying medical conditions and therefore serve as a “marker” for adverse pregnancy and fetal/infant health outcomes.

Our study has several strengths. We used a large population database with consistent data collection of important maternal and birth characteristics. The linkage between birth certificate and hospitalization data improved data accuracy (positive and negative predictive values for most birth outcomes were greater than 80% and 98%, respectively) compared with similar non-linked databases [22].

This study has some limitations. First, information on pre-pregnancy BMI was self-reported. Previous studies have shown that self-reported weight measurements are usually within close range of the actual weight and are reliable [34,35]. However, if women intentionally reported lower pre-pregnancy weight than their actual weight, for instance due to social desirability, the proportion of those with excess weight gain would have been overestimated. This overestimation may have led to bias towards the null in our results pertaining to excess weight gain, especially among women with normal and higher pre-pregnancy BMI. Second, the IOM/ACOG recommendations for per-week weight gain are uniform for each gestational week preterm. We calculated optimal weight gain in women who delivered preterm by multiplying weekly recommended minimum (and maximum) weight gain by the number of weeks. However, gestational weight gain does not naturally occur in a linear fashion and tends to accelerate as pregnancy progresses during the second and third trimester. Thus, some women with early preterm deliveries may appear to have had lower than recommended weight gain, while their gain was in fact optimal [36]. Nevertheless, we aimed to follow the IOM/ACOG per-week recommendations in a pragmatic manner, the same way as they are used by maternity care providers. Third, we relied on ICD-9-CM codes to identify severe maternal conditions, which may be under-reported [37,38], resulting in potential underestimation of the associations with suboptimal weight gain. Similarly, potential omissions and coding errors in other covariates may have led to non-differential misclassification and may have biased the results towards the null [38]. Fourth, we did not have information on some behavioural factors, such as alcohol/drug use, that are associated with weight gain and adverse pregnancy outcomes. However, we did not aim to examine potential causes or contributors to suboptimal weight gain and their direct or indirect effects on adverse outcomes. We did adjust, however, for common demographic factors including marital status, race, education, and type of medical insurance. Last, our finding that more than half of women did not achieve optimal weight gain during pregnancy may be influenced by the ethnic/racial composition and other characteristics of the population of pregnant women in Washington State. And while similar proportions with suboptimal weight gain have been observed in the US and Europe, in Asian countries, for instance, the proportion of women with excess weight gain is much lower when standard BMI categories are applied [31].

This study adds to the body of literature that underscores the importance of pre-pregnancy BMI and weight gain during pregnancy with respect to severe adverse pregnancy outcomes. And while these 2 risk factors are modifiable, we still need effective strategies for women to achieve optimal BMI and weight gain during pregnancy. Pregnancy is an important opportunity to advocate for and support women in behavioural changes towards a healthy lifestyle.

To our knowledge, this is the first large study to assess the associations between serious adverse outcomes and suboptimal weight gain during pregnancy in both term and preterm gestations, making our results more generalizable and applicable to all women in the second and third trimester of pregnancy and their healthcare providers. Our findings are comprehensive with respect to all serious adverse outcomes (death and severe morbidity) that are relevant to mother, fetus, and infant. This facilitates counselling for pregnant women about severe adverse effects associated with weight gain during pregnancy, according to their pre-pregnancy BMI. Future research is needed to create algorithms that would identify optimal weight gain trajectories tailored to individual women, based on their pre-pregnancy BMI and other prognostic factors. In addition, future research is required to investigate the potential utility of low weight gain as an early sign of increased risk of stillbirth.

Washington State birth filing form sample.

(DOCX)

Click here for additional data file.

STROBE checklist of items that should be included in reports of observational studies.

(DOCX)

Click here for additional data file.

Severe maternal morbidity definitions.

(DOCX)

Click here for additional data file.

Rates of maternal death and severe maternal morbidity (rates per 10,000 births) by pre-pregnancy BMI and gestational weight gain.

(DOCX)

Click here for additional data file.

Adjusted odds ratios (AORs) for SMM components by gestational weight gain and pre-pregnancy BMI.

(DOCX)

Click here for additional data file.

Adjusted odds ratios (AORs) for stillbirth, neonatal death, and severe neonatal morbidity by gestational weight gain and pre-pregnancy BMI.

(DOCX)

Click here for additional data file.

Rates of maternal death and severe maternal morbidity by gestational weight gain and pre-pregnancy obesity class.

(DOCX)

Click here for additional data file.

Adjusted odds ratios (AORs) for the components of SMM by gestational weight gain and pre-pregnancy obesity class.

(DOCX)

Click here for additional data file.

Adjusted odds ratios (AORs) for severe adverse maternal and perinatal outcomes by gestational weight gain and pre-pregnancy BMI—Singleton births at term gestation (≥37 weeks).

(DOCX)

Click here for additional data file.

Study analysis plan.

(DOCX)

Click here for additional data file.

30 Sep 2019

Dear Dr. Ukah,

Thank you very much for submitting your manuscript "Association between gestational weight gain and severe adverse birth outcomes: Does pre-pregnancy body mass index matter?" (PMEDICINE-D-19-02499) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also discussed with an academic editor with relevant expertise, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Oct 21 2019 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Clare Stone, PhD

Acting Chief Editor

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

Title – please add a study design in title, and remove rhetorical question

Please comment on the limitation of the data and how old it is.

Please provide a summary of demographic details and limitations in abstract (limitations as the final sentence of the Methods and Findings section.)

Please provide p values where 95%Cis are given.

data – please be explicit about why access to data is prohibited (IRB or ethical restrictions, for example).Also we need the email or URL for (eligible) people to access data. Noting this cannot be an author.

At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

Ethics – please say why it was exempt.

Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis—including those made in response to peer review comments—should be identified as such in the Methods section of the paper, with rationale.

at lines 55 and 345, "In this study ... we found that ... was ..." or similar

Please ensure that the study is reported according to the [STROBE] guideline, and include the completed [STROBE or other] checklist as Supporting Information. When completing the checklist, please use section and paragraph numbers, rather than page numbers. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

References should be in square brackets instead of round throughout text.

Line 397 – I wonder if “We did not aim to examine causal associations” should be removed. As this is an observational study, causality couldn’t be assessed anyway. Otherwise rephrase to outline that this is not possible in this study.

Please use the "Vancouver" style for reference formatting, and see our website for other reference guidelines https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references

Please ensure that the study is reported according to the STROBE guideline, and include the completed STROBE checklist as Supporting Information. Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

The STROBE guideline can be found here: http://www.equator-network.org/reporting-guidelines/strobe/

When completing the checklist, please use section and paragraph numbers, rather than page numbers.

Comments from the reviewers:

Reviewer #1: This is a well-conducted study on the association between gestational weight gain and severe adverse birth outcomes, analysed in different pre-pregnancy BMI categories. The study design, datasets, statistical methods and analyses, and presentation (tables and figures) and interpretations of results are mostly adequate and of a good standard. Only a couple of minor points needing attention:

1) Table 2 and 3 are massive tables with a lot of details which are not as important as other tables in terms of primary analyses. To be concise, suggest to remove both or one of the tables to the supplementary information.

2) Table 1 is a huge descriptive table but without any statistical test. I know it's already huge, but it would be helpful to add a column of p-values in each of the pre-pregnancy BMI categories to test the statistical significance of the baseline characteristics across the levels of weight gain, which will make it easy for the readers to read and follow the main message.

Reviewer #2: This large observational study shows adverse health impacts of inadequate or excess gestational weight gain, with a focus on severe morbidity and mortality. Linked data sets and sheer size of the analysis are strengths of this work.

However, the limitations need also to be addressed. The findings are largely consistent with prior literature. The clear gaps proposed in the introduction are not strongly justified. The studies referenced in the introduction have previously addressed many of the key gaps in this field and the contribution of prior literature and the very specific nuances aims and hypotheses should be more clearly outlined. It is not clear what the specific initial hypotheses were and why?

It is not clear why ethics review was waivered and this should be further clarified, especially given the use of linked data bases.

A clear rationale for the composite outcomes should be provided.

The observation on line 221 that those with GDM were more likely to have low GWG is expected as treatment of GDM reduces maternal weight gain. This presents an interesting subgroup to study relationships between GWG and outcomes and could perhaps be explored as a subgroup here.

The data presented here reflects the nuances of the specific ethnicities in the US and the unique health system in the US, both of which are limitations for generalisability. This is especially regarding the health system which requires around 20% GDP and has significant access and disparity issues, limiting generalisability and this should be acknowledged. Likewise the ethnic variation shown to vary in Asian subgroups specifically (eg Goldstein et al BMC Medicine), limits the broader generalisability for example in the majority of the international population who are of Asian ethnicity.

Table 4 is difficult to interpret, as there seems to be a separate column of data superimposed on the last column.

The focus on the inconsistent findings of rare outcomes across multiple subgroups as noted in the first paragraph on the discussion, are difficult to interpret and appear inconsistent. For example findings in normal and obese but not overweight women. These subgroup results appear inconsistent and not driven by plausible hypotheses, an area that needs to be more thoroughly explored.

Broader existing literature, evidence synthesis and meta-analyses need to be considered in the discussion. For example the discussion in lines 373-377 should be revised to extend to evidence that does not support these findings including meta analyses and the potential reasons why. Specifically the differences from JAMA. 2019;321(17):1702-1715. doi:10.1001/jama.2019.3820 and JAMA. 2017;317(21):2207-2225. doi:10.1001/jama.2017.3635 as large scale intergrated evidence synthesis studies and the findings here should be discussed in this context.

Reviewer #3: The purpose of this study was to investigate the association between suboptimal and excessive weight gain and the effects on adverse birth outcomes in women with a range of pre-pregnancy BMI. This was a population-based study of a singleton hospital births in the state of Washington between 2004 and 2013. The primary composite outcome included: 1) maternal death or severe maternal morbidity, and 2) perinatal death and/or severe neonatal morbidity. Overall 722,839 women with information on pre-pregnancy BMI were included in the analysis. The distribution of BMI weight classes, according to WHO criteria, was consistent with what has been previously reported. Low gestational weight gain was associated with higher rates of maternal death and severe maternal morbidity in women with normal pre-pregnancy BMI and overweight women. Similarly, excessive gestational weight gain was associated with an increased rate of death and severe maternal morbidity among women with normal pre-pregnancy BMI and obese women. Low gestational weight gain was associated with perinatal death and severe neonatal morbidity regardless of pre-pregnancy BMI, including obese classes 1, 2, and 3. The authors conclude that most women do not achieve optimal weight during pregnancy. Both low and excessive gestational weight gain were associated with severe adverse birth outcomes regardless of pre-pregnancy BMI except for overweight women with excessive gestational weight gain.

Specific Comments:

1. This is a well-written manuscript with detailed tables, which because of the large number of cases could easily serve as a reference manuscript. The inclusion of both pre-term and term pregnancies, along with detailed information about class I, II, and III obesity, are important components of this manuscript.

2. As noted by the authors in the introduction, there was a lack of specific recommendations relating to gestational weight gain in the IOM report in class I, II and III obese women because of the lack of data in 2008. The information in this manuscript helps to improve that knowledge gap.

3. Gestational age at delivery was based on ultrasound dating. One assumes that the ultrasound dating in the vast majority of subjects was done in early pregnancy. It would be helpful if the authors would provide information on gestational age at the time of the ultrasound.

4. Pregnancy weight gain was defined as a woman's weight measured at childbirth subtracted from her pre-pregnancy self-reported weight. Many investigators use weight at the last ante-natal visit as the end point of weight gain during pregnancy. In the state of Washington, was maternal weight actually measured in the labor and delivery unit?

5. To account for gestational age at delivery among women who delivered pre-term, weight pre-term was calculated using the IOM criteria. These criteria provide a range of weight gain per week. Which part of the range was used in the analysis and was it used consistently among the various BMI groups?

6. Very detailed information is provided in both the tables in the manuscript and in the supplementary tables. Is all of this this information available in the BERD and CHARS data bases or were additional means used to attain the granularity of detail presented in these tables?

7. The percentage of women in each pre-pregnancy BMI weight category is pretty much consistent as what has been reported in the literature, including data from the CDC.

8. On the bottom of page 11, women with low weight gain had higher rates of spontaneous vaginal delivery and lower rates of prolonged labor. The rates of gestational diabetes were higher in these women, especially for those who were underweight and those with a normal BMI prior to pregnancy. These data are very interesting since many previous studies have related the increased risk of gestational diabetes on excessive gestational weight gain and not low gestational weight gain.

9. The data presented in table 3 at the bottom of page 18 are important as noting that low gestational weight gain had a significantly higher rate of maternal death as compared with women with optimal weight gain (7.97 vs. 2.63 per 100,000 p=0.027). Among normal weight and overweight women, those with low gestational weight gain had higher adjusted odds of maternal death and severe morbidity compared with optimal gestational weight gain. These data are important to note given that many individuals recommended gestational weight gain less than IOM guidelines to improve some perinatal outcomes.

10. Similarly in Table 4, it is important to note that the adjusted odds ratio for severe adverse birth outcomes is significantly higher in obese women with low weight gain for perinatal death and severe neonatal morbidity, as well as the combined maternal and perinatal adverse outcomes.

11. On page 26, lines 373-377, the authors state that they did not observe any association between excessive weight gain in overweight women in severe adverse maternal and perinatal outcomes. Although these women had a very high rate of excessive weight gain, the authors suggest that the IOM guidelines for optimal weight gain in overweight women may need to be re-evaluated. How would the authors recommend changing the IOM guidelines for overweight women based on their findings?

12. In multiple places in the manuscript, for example page 3 line 84 the authors refer to the IOM/AJOG guidelines. AJOG should be ABOG.

Any attachments provided with reviews can be seen via the following link:

[LINK]

1 Nov 2019

Submitted filename: Response to editors and reviewers_final.docx

Click here for additional data file.

19 Nov 2019

Dear Dr. Ukah,

Thank you very much for re-submitting your manuscript "Association between gestational weight gain and severe adverse birth outcomes: Does pre-pregnancy body mass index matter?" (PMEDICINE-D-19-02499R1) for review by PLOS Medicine.

I have discussed the paper with my colleagues and the academic editor and it was also seen again by xxx reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.

We look forward to receiving the revised manuscript by Nov 26 2019 11:59PM.

Sincerely,

Louise Gaynor-Brook, MBBS PhD

Associate Editor

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

General comments

Please add reference brackets at the end of each sentence, before the full stop and and after a space following the last word.

Please remove ‘Appendix’ in references to Supplementary Information files

Data Availability: The data link provided takes us to a paid for data access site, costing $50 per file, per year. Access to data via a paid service does not comply with our data policy. Please provide a non-payment website where data can be accessed / where authors can apply for access. Please note that a study author cannot be the contact person for the data.

Title – Please include setting and dates of study. We suggest revising the title to “Association between gestational weight gain and severe adverse birth outcomes in Washington, USA: A population-based, retrospective cohort study, 2004-2013”

Abstract

Please begin the Methods and Findings with ‘We conducted’ or similar.

Lines 42-45: Please consider breaking this sentence up into 3 shorter sentences.

Lines 50-52: Please clarify what is meant by ‘a composite maternal death’ in sentence beginning ‘In addition, low weight gain...’.

Author Summary

Line 71 - please add (IOM) after Institute of Medicine

Introduction

Line 122 - please clarify what is meant by ‘delivery hospitalization’

Methods

Line 142 - Please revise to ‘...guideline (S1 Checklist).’ Please rename the prospective analysis plan file as ‘S1 Text’ and include a reference to this your Methods.

Results

Table 2 - please provide definitions for all abbreviated terms used, in the table legend.

Tables 3 & 4 - When a p value is given, please specify the statistical test used to determine it.

Discussion

Line 407 - please reformat reference in superscript to PLOS house style (please see below).

Please add a short section after discussing the strengths/limitations to discuss the implications of the study and next steps for research, clinical practice, and/or public policy

Please remove the subtitle of ‘Conclusion’

Line 482 - please correct to ‘trimester’

Figure 1 – Please revise figure to use solid blocks of colour (rather than patterned), and ideally avoid using green and red together.

References

Please make sure references are appropriately capitalised and formatted e.g. ref 11 ‘BMC pregnancy and childbirth’ to ‘BMC Pregnancy and childbirth’

Please use the "Vancouver" style for reference formatting, and see our website for other reference guidelines https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references

Supplementary Files

S1 STROBE Checklist - When completing the checklist, please use section and paragraph numbers, rather than line numbers.

Tables S3, S4, S6, S7 - please provide definitions for all abbreviated terms used, in the table legend

Tables S4, S5, S8 - When a p value is given, please specify the statistical test used to determine it.

Comments from Reviewers:

Reviewer #1: I am satisfied with the response and the revision. No further issues needing attention.

Reviewer #3: The authors have responded to all my queries and I have no further comments of questions.

Any attachments provided with reviews can be seen via the following link:

[LINK]

27 Nov 2019

Submitted filename: Response to requests from Editors and production_v2.docx

Click here for additional data file.

3 Dec 2019

Dear Dr Ukah,

On behalf of my colleagues and the academic editor, Dr. Louise Gaynor-Brook, I am delighted to inform you that your manuscript entitled "Association between gestational weight gain and severe adverse birth outcomes in Washington State, USA: A population-based retrospective cohort study, 2004-2013" (PMEDICINE-D-19-02499R2) has been accepted for publication in PLOS Medicine.

PRODUCTION PROCESS

Before publication you will see the copyedited word document (in around 1-2 weeks from now) and a PDF galley proof shortly after that. The copyeditor will be in touch shortly before sending you the copyedited Word document. We will make some revisions at the copyediting stage to conform to our general style, and for clarification. When you receive this version you should check and revise it very carefully, including figures, tables, references, and supporting information, because corrections at the next stage (proofs) will be strictly limited to (1) errors in author names or affiliations, (2) errors of scientific fact that would cause misunderstandings to readers, and (3) printer's (introduced) errors.

If you are likely to be away when either this document or the proof is sent, please ensure we have contact information of a second person, as we will need you to respond quickly at each point.

PRESS

A selection of our articles each week are press released by the journal. You will be contacted nearer the time if we are press releasing your article in order to approve the content and check the contact information for journalists is correct. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact.

PROFILE INFORMATION

Now that your manuscript has been accepted, please log into EM and update your profile. Go to https://www.editorialmanager.com/pmedicine, log in, and click on the "Update My Information" link at the top of the page. Please update your user information to ensure an efficient production and billing process.

Thank you again for submitting the manuscript to PLOS Medicine. We look forward to publishing it.

Best wishes,

Louise Gaynor-Brook, MBBS PhD

Associate Editor

PLOS Medicine

plosmedicine.org