A study on the correlation between pregnancy risk factors and birth outcomes

Abstract

Adverse birth outcomes—including low birth weight (LBW, < 2500 g), preterm birth (PTB, < 37 weeks), and intrauterine growth abnormalities—remain major global public health challenges, particularly in low- and middle-income countries. Although maternal body mass index (BMI) and gestational weight gain (GWG) are established risk factors in Western populations, their interactions with metabolic and sociodemographic factors in Asian cohorts, particularly within China’s rapidly urbanizing populations, warrant further investigation. This prospective cohort study analyzed 1,026,294 singleton pregnancies (2012–2018) from the Liaoning Maternal and Child Health Information System. Using multivariate logistic regression models, we calculated adjusted risk ratios (RR) with 95% confidence intervals (CI) to evaluate associations between prenatal exposures… and birth outcomes: low birth weight (LBW), macrosomia (> 4000 g), Small-for-gestational-age (SGA, < 10th percentile), Large-for-gestational-age (LGA, > 90th percentile), PTB. Pre-pregnancy underweight (RR = 1.58, 95% CI 1.47–1.70), insufficient gestational weight gain (RR = 1.44, 95% CI 1.36–1.53), and preeclampsia (RR = 3.61, 95% CI 3.18–4.10) were strongly associated with low birth weight. For SGA, pre-pregnancy underweight (RR = 1.68, 95% CI 1.62–1.74) and insufficient GWG (RR = 1.40, 95% CI 1.29–1.51) were key predictors. Conversely, pre-pregnancy obesity (RR = 2.79, 95% CI 2.72–2.85) and excessive GWG (RR = 2.15, 95% CI 2.07–2.23) elevated macrosomia risk. PTB was strongly associated with pre-pregnancy obesity (RR = 1.29, 95% CI 1.25–1.33), assisted reproductive conception (RR = 3.39, 95% CI 3.01–3.83), and early-pregnancy hyperglycemia (RR = 1.19, 95% CI 1.15–2.22). Pre-pregnancy BMI, gestational weight gain (GWG), hypertensive disorders, and metabolic markers (e.g., fasting glucose, hemoglobin) constitute critical modifiable determinants of adverse birth outcomes. These findings quantify region-specific risk thresholds (e.g., 38.1% pre-pregnancy overweight/obesity vs. national 24.8%, Liaoning macrosomia 11.62% vs. national 8.9%) to prioritize GWG monitoring and metabolic screening in Northeast China’s urban transition.

Introduction

Background

The Developmental Origins of Health and Disease (DOHaD) hypothesis posits that the perinatal environment plays a pivotal role in shaping offspring health trajectories and lifelong disease susceptibility. Exposure to factors such as maternal stress, obesity, high-calorie diets, and environmental chemicals during early development may program long-term metabolic and physiological dysregulation, increasing the risk of adverse health outcomes in later life¹. Adverse birth outcomes—including low birth weight (LBW), preterm birth, and intrauterine growth abnormalities—persist as global public health challenges, with disproportionately higher prevalence rates in low- and middle-income countries².

Existing evidence and gaps

While maternal body mass index (BMI), gestational weight gain (GWG), and metabolic factors (e.g., blood glucose, hypertension) are well-documented risk factors for adverse birth outcomes in Western populations^3–5.However, critical gaps persist in Asian contexts:

First, limited large-scale Asian cohorts: Evidence from China is sparse despite its unique nutritional transitions and urbanization-driven health challenges⁶.

Second, interactions between metabolic markers (including early-pregnancy glucose and hemoglobin levels) and sociodemographic determinants (particularly rural/urban disparities) have not been systematically quantified in this population.

Finally, the cumulative effects of multidimensional exposures—such as assisted reproductive technologies (ART), teratogen exposure, and clinical interventions—remain poorly characterized in Asian cohorts.

Objectives

This study aims to:

1. Quantify associations between modifiable prenatal factors (pre-pregnancy BMI, GWG, metabolic markers) and birth outcomes (LBW, macrosomia, SGA, LGA, preterm birth) in a Chinese cohort.

2. Explore interactions between biological determinants and sociodemographic factors.

3. Provide actionable recommendations for region-specific prenatal care policies.

Methods

Data resources and study design

The data were derived from the Liaoning Maternal and Child Health Information System, a population-based active surveillance network designed to monitor live births, stillbirths, terminations, and birth defects in Liaoning Province, Northeast China. This system constitutes a component of the national population-based birth-defect surveillance system in China. The surveillance network encompassed 79 healthcare facilities across all administrative levels (county, municipal, and provincial) within Liaoning Province’s 14 cities. Trained personnel at primary community health centers conducted standardized data entry, while automated validation protocols ensured data integrity through: (1) gestational age-delivery date consistency verification, and (2) outlier detection algorithms. The data were submitted annually to city-level monitoring centers, which then forwarded them to province-level centers. At each level, clinical experts reviewed and verified the quality of the data, and any uncertain information was returned to the previous level center for verification. A work group at the provincial monitoring center (Liaoning Maternal and Child Health Hospital), comprising state-level clinical experts, epidemiologists, and information technicians, conducted the final quality control and cross referenced the data with other related systems, such as perinatal deaths, to ensure accuracy. This study was approved by the Ethics Committee of Liaoning Maternal and Child Health Hospital (No.20220318001).

Inclusion/exclusion criteria

The inclusion criteria were as follows: (1) available for information regarding the last menstrual period andpregnancy termination; (2) available for demographic and obstetric characteristics; (3) singleton births; (4) pregnancies not achieved through in vitro fertilization; (5) births without congenital defects. The exclusion criteria were as follows: (1)births who had born as multiple births; (2) stillbirths, late stage miscarriages, terminations due to fetal malformations; (3) missing birth weight data.After exclusions (n = 18,296, 1.8% of raw data), 1,026,294 mother-infant pairs met all quality thresholds (data completeness > 98%, logic error rate < 0.5%, cross-validation concordance 99.2%).

Definitions and measurements

Exposures:

1. Pre-pregnancy weight: the weight of pregnant women was used as the pre-pregnancy weight replacement index (excluding pregnant women with card building time > 14 weeks and pregnant women with card building time and last birth examination less than 1 month); pre-pregnancy BMI = pre-pregnancy weight / height² (kg/m²), According to the Chinese Working Group on Obesity⁷, Pre-pregnancy BMI classification, low body weight (BMI < 18.5 kg/m²), Normal body weight (18.5–23.9 kg/m²), Being overweight (24–27.9 kg/m²), Obesity (28 kg/m²).

2. Weight gain during pregnancy: gestational weight gain (Gestational weight gain, GWG) refers to the difference between birth weight and pre-pregnancy weight.

According to the “Recommended criteria for weight Growth in pregnant women” issued by the National Health Commission⁸. The GWG was defined as: pre-pregnancy low body weight BMI < 18.5 kg/m²). The appropriate GWG for pregnant women is 11–16 kg, and the normal weight before pregnancy (BMI:18.5–23.9 kg/m²). The appropriate GWG for pregnant women is 8–14 kg, and the overweight before pregnancy (BMI: 24–27.9 kg/m²). The appropriate GwG range for pregnant women is 7–11 kg, and pre-pregnancy obesity (BMI > 28 kg/m²). The appropriate GWG range for pregnant women is 5–9 kg. GWG for different PWG is insufficient GWG below GWG, and GWG above GWG is excessive GWG.

Outcomes:

1. Birth weight: refers to the newborn weight measured at the time of delivery. Newates were classified as low birth weight (Low Birth weight, LBW): birth weight < 2500 g; normal birth weight (Normal bitrh weight. NBW): weight 2500–4000 g at birth; macrosomia (Macrosomia); birth weight > 4000 g.

2. According to the Growth Evaluation Standards of Newborns at Birth issued by the National Health Commission⁹, Newborns were divided into children less for gestational age, suitable for gestational age and greater than gestational age.

Less than gestational age (Small for ge s ta tion al age, SGA): newborns with birth weight below the 10th percentile of the average weight of the same gestational age;

Suitable for gestational age (Appropriate for gestational age, AGA): newborns with birth weight between the 10th and 90th percentile of the mean weight of the same gestational age;

Greater than gestational age (Large for ge s ta tion al age, LGA): newborns with birth weight above the 90th percentile of the mean weight of the same gestational age.

(3) Preterm birth (Preterm birth, PTB): refers to the delivery at 28 weeks but less than 37 weeks of gestation.

Covariates

Covariates included maternal age (< 35 or ≥ 35 years), ethnicity (Han Chinese or minority), education level(illiterate or primary school, middle school, high school or above), occupation and parity.Pre-pregnancy BMI was determined by dividing the self-reported weight before pregnancy (kg) by the square of height (m).

Statistical analyses

The study employed multivariate logistic regression models to assess the associations between pregnancy risk factors and birth outcomes. Multivariate logistic regression models were employed to quantify associations between prenatal risk factors and birth outcomes, with adjustments for maternal age, ethnicity, education, parity, and pre-pregnancy BMI. All reported risk ratios (RRs) with 95% confidence intervals (CIs) represent covariate-adjusted estimates derived from these models.Incomplete records were excluded to handle missing data, and model assumptions (e.g., linearity, multicollinearity) were rigorously validated before analysis.

Results

Participant characteristics

This study included 1,026,294 pregnant women from Liaoning Province, China (2012–2018). The mean maternal age was 28.94 ± 4.78 years, with 10.02% of women aged > 35 years. The majority were of Han ethnicity (88.20%), and 74.9% had an educational level below junior college. According to the Chinese Working Group on Obesity criteria, 56.42% of women had a normal pre-pregnancy BMI, while 25.60% were overweight and 12.50% were obese. Among the participants, 5.20% experienced preterm birth, and 0.22% conceived through assisted reproductive technology. Fathers in the cohort had a mean age of 30.78 ± 5.18 years, with 90.96% being Han Chinese (Table 1).

Table 1.

Demographic and clinical characteristics of the study population (Liaoning province, china, 2012–2018).

Feature	Sample capacity (n)	Constituent ratio (%)
Maternal age (years)
≤ 35	923,490	89.98
> 35	102,804	10.02
Pregnant women’s nation
The Han nationality	876,958	88.20
Minority nationality	117,319	11.80
Education degree of pregnant women
College junior college below	731,655	74.49
College college or above	250,604	25.51
Pregnant women’s profession
Be on the job	749,753	76.30
Unemployed	232,901	23.70
Gravidity
For the first time	358,037	51.74
Not for the first time	333,894	48.26
Parity
I-para	520,743	68.11
multipara	243,779	31.89
Pre-pregnancy BMI (kg)
< 18.5 (low body weight)	47,984	7.28
18.5–23.9 (normal body weight)	359,879	54.62
24-27.9 (Overweight)	168,643	25.60
28 (obese)	82,359	12.50
Premature birth
Yes	36,044	5.20
Deny	656,762	94.80
Assisted reproductive conception
Yes	2258	0.22
Deny	1,025,669	99.78
HDP (hypertensive disorders in pregnancy)
Yes	53,127	7.59
Deny	646,531	92.41
GWG (Chinese Standard)
Insufficient weight gain	12,218	6.74
Suitable for weight gain	67,803	37.42
Too much weight gain	10,192	55.84
Supplement folic acid
Yes	391,173	38.05
Deny	636,754	61.95
Fasting blood-glucose
Glucopenia	36,811	6.28
Euglycemia	411,447	70.16
Hyperglycaemia	138,163	23.56
hemoglobin
≥ 110	567,631	91.19
<110	54,863	8.81
Fasting urine sugar
Normal	588,129	98.89
Overtop	6.622	1.11
Exposure of teratogens
Yes	14,414	1.40
Deny	1,013,513	98.60

BMI body mass index, GWG gestational weight gain, LBW low birth weight, PTB preterm birth.

Associations between risk factors and birth weight

Multivariate logistic regression revealed significant associations between prenatal risk factors and birth weight outcomes. Notably, pre-pregnancy underweight (RR = 1.58, 95% CI 1.47–1.70), insufficient gestational weight gain (RR = 1.44, 95% CI 1.36–1.53), and preeclampsia (RR = 3.61, 95% CI 3.18–4.10) exhibited the strongest associations with low birth weight. Conversely, pre-pregnancy overweight (RR = 1.88, 95% CI 1.84–1.92), obesity (RR = 2.79, 95% CI 2.72–2.85), and excessive GWG (RR = 2.15, 95% CI 2.07–2.23) significantly increased the risk of macrosomia. Early-pregnancy hyperglycemia (RR = 1.46, 95% CI 1.43–1.49) and elevated fasting urinary glucose (RR = 1.67, 95% CI 1.55–1.79) were also independent risk factors for macrosomia (Table 2).

Table 2.

Adjusted risk ratios (RR) and 95% confidence intervals (CI) for associations between prenatal risk factors and abnormal neonatal birth weight.

Risk factors during pregnancy	LBW (n = 33,148) RR (95% CI)	Macrosomia (n = 121,792) RR (95% CI)
Pre-pregnancy BMI (kg)
< 18.5 (low body weight)	1.58 (1.47, 1.70)	0.40 (0.38, 0.42)
18.5–23.9 (normal body weight)	1.00	1.00
24–27.9 (overweight)	0.89 (0.84, 0.93)	1.88 (1.84, 1.92)
28 (obese)	0.87 (0.82, 0.93)	2.79 (2.72, 2.85)
GWG (Chinese Standard)
Insufficient weight gain	1.60 (1.40, 1.82)	0.73 (0.67, 0.80)
Proper weight gain	1.00	1.00
Excessive weight gain	0.82 (0.75, 0.89)	2.15 (2.07, 2.23)
HDP
Health	1.00	1.00
Pregnancy complicated with chronic hypertension	1.75 (1.59, 1.93)	1.50 (1.42, 1.57)
CH Concurrent preeclampsia	2.26 (1.83, 2.80)	1.61 (1.41, 1.83)
hypertension of pregnancy	2.21 (2.05, 2.39)	1.37 (1.31, 1.42)
preeclampsia	3.61 (3.18, 4.10)	1.50 (1.38, 1.64)
Fasting blood glucose in early pregnancy
Glucopenia	1.05 (0.96, 1.15)	0.86 (0.83, 0.90)
Euglycemia	1.00	1.00
Hyperglycaemia	0.95 (0.90, 1.00)	1.46 (1.43, 1.49)
Early pregnancy fasting urine sugar
The urine sugar is normal	1.00	1.00
Urine sugar is high	1.04 (0.87, 1.25)	1.67 (1.55, 1.79)
Hemoglobin in early pregnancy
Hemoglobin is too low	0.92 (0.85, 1.00)	0.86 (0.83, 0.88)
Hemoglobin was normal	1.00	1.00
Folate supplementation in early pregnancy
Yes	1.00	1.00
Deny	1.00 (0.96, 1.05)	0.97 (0.95, 0.99)
Assisted reproduction
Yes	2.69 (2.24, 3.22)	0.92 (0.78, 1.08)
Deny	1.00	1.00
Exposure of teratogens
Yes	1.17 (1.03, 1.32)	1.01 (0.95, 1.07)
Deny	1.00	1.00

LBW low birth weight, SGA small-for-gestational-age, LGA large-for-gestational-age, RR risk ratio, CI confidence interval.

Risk factors for intrauterine growth abnormalities

For small-for-gestational-age (SGA) infants, pre-pregnancy underweight (RR = 1.68, 95% CI 1.62–1.74) and insufficient GWG (RR = 1.40, 95% CI 1.29–1.51) were significantly. In contrast, for large-for-gestational-age (LGA) infants, pre-pregnancy obesity (RR = 2.36, 95% CI 2.31–2.40), excessive GWG (RR = 1.78, 95% CI 1.73–1.83), and early-pregnancy hyperglycemia (RR = 1.34, 95% CI 1.32–1.36) emerged as key risk factors. Notably, preeclampsia increased risks for both SGA (RR = 2.07, 95% CI 1.87–2.30) and LGA (RR = 1.31, 95% CI 1.23–1.41) (Table 3).

Table 3.

Adjusted risk ratios (RR) and 95% confidence intervals (CI) for associations between prenatal risk factors and intrauterine growth abnormalities.

Risk factors during pregnancy	SGA (n = 35,069) RR (95% CI)	LGA (n = 160,253) RR (95% CI)
Pre-pregnancy BMI (kg)
< 18.5 (low body weight)	1.68 (1.62, 1.74)	0.48 (0.46, 0.50)
18.5–23.9 (normal body weight)	1.00	1.00
24-27.9 (overweight)	0.78 (0.75, 0.80)	1.68 (1.66, 1.71)
28 (obese)	0.81 (0.77, 0.85)	2.36 (2.31, 2.40)
GWG (Chinese Standard)
Insufficient weight gain	1.40 (1.29, 1.51)	0.80 (0.75, 0.84)
Proper weight gain	1.00	1.00
Excessive weight gain	0.65 (0.62, 0.68)	1.78 (1.73, 1.83)
HDP
Health	1.00	1.00
Pregnancy complicated with chronic hypertension	1.34 (1.24, 1.44)	1.34 (1.29, 1.39)
CH Concurrent preeclampsia	1.79 (1.51, 2.13)	1.47 (1.33, 1.62)
Hypertension of pregnancy	1.53 (1.45, 1.62)	1.23 (1.19, 1.27)
Preeclampsia	2.07 (1.87, 2.30)	1.31 (1.23, 1.41)
Fasting blood glucose in early pregnancy
Glucopenia	1.16 (1.10, 1.22)	0.92 (0.90, 0.95)
Euglycemia	1.00	1.00
Hyperglycaemia	0.96 (0.93, 0.99)	1.34 (1.32, 1.36)
Early pregnancy fasting urine sugar
The urine sugar is normal	1.00	1.00
Urine sugar is high	1.06 (0.93, 1.21)	1.51 (1.42, 1.60)
Hemoglobin in early pregnancy
Hemoglobin is too low	1.01 (0.97, 1.06)	0.94 (0.92, 0.97)
Hemoglobin was normal	1.00	1.00
Folate supplementation in early pregnancy
Yes	1.00	1.00
Deny	0.99 (0.97, 1.02)	0.98 (0.97, 0.99)
Assisted reproduction
Yes	1.41 (1.17, 1.69)	0.92 (0.82, 1.02)
Deny	1.00	1.00
Exposure of teratogens
Yes	1.07 (0.99, 1.16)	0.98 (0.94, 1.03)
Deny	1.00	1.00

SGA small-for-gestational-age, LGA large-for-gestational-age.

Predictors of preterm birth

Preterm birth risk was significantly increased by insufficient gestational weight gain (RR = 1.74, 95% CI:1.61–1.88), early-pregnancy metabolic abnormalities including hyperglycemia (RR = 1.19, 95% CI:1.15–1.22) and anemia (RR = 1.66, 95% CI:1.50–1.82), as well as pre-pregnancy obesity (RR = 1.29, 95% CI:1.25–1.33).Early-pregnancy hyperglycemia (RR = 2.63, 95% CI 2.40–2.88) and low hemoglobin levels (RR = 1.66, 95% CI 1.50–1.82) posed the highest risks. Assisted reproductive conception (RR = 3.39, 95% CI 3.01–3.83) and teratogen exposure in early pregnancy (RR = 1.10, 95% CI 1.02–1.19) were additional independent risk factors (Table 4).

Table 4.

Adjusted risk ratios (RR) and 95% confidence intervals (CI) for associations between prenatal risk factors and preterm birth (PTB).

Risk factors during pregnancy	Preterm birth (n = 36,044) RR (95% CI)
Pre-pregnancy BMI (kg)
< 18.5 (low body weight)
18.5–23.9 (normal body weight)	0.99 (0.94, 1.04)
24-27.9 (Overweight)	1.00
28 (obese)	1.29 (1.25, 1.33)
GWG (Chinese standard)	1.72 (1.66, 1.78)
Insufficient weight gain
Proper weight gain	1.74 (1.61, 1.88)
Excessive weight gain	1.00
HDP	0.72 (0.69, 0.76)
Health
Pregnancy complicated with chronic hypertension	1.00
CH concurrent preeclampsia	2.16 (2.04, 2.29)
Hypertension of pregnancy	2.72 (2.38, 3.12)
Preeclampsia	1.63 (1.55, 1.72)
Fasting blood glucose in early pregnancy	2.63 (2.40, 2.88)
Glucopenia
Euglycemia	1.06 (1.00, 1.12)
Hyperglycaemia	1.00
Early pregnancy fasting urine sugar	1.19 (1.15, 1.22)
The urine sugar is normal
Urine sugar is high	1.00
Hemoglobin in early pregnancy	1.66 (1.50, 1.82)
Hemoglobin is too low
Hemoglobin was normal	1.05 (1.01, 1.10)
Folate supplementation in early pregnancy	1.00
Yes
Deny	1.00
Assisted reproduction	0.98 (0.96, 1.00)
Yes
Deny	3.39 (3.01, 3.83)
Exposure of teratogens	1.00
Yes
Deny	1.10 (1.02, 1.19)

PTB preterm birth, RR risk ratio, CI confidence interval.

Discussion

Demographic and clinical characteristics of the study population

First, the study cohort comprised pregnant women with a mean age of 28.94 ± 4.78 years, of whom 10.2% were aged over 35 years. This proportion exceeded the national average of 9.65% reported for Chinese women in 2020¹⁰, yet al.igned closely with regional data indicating10.18% of pregnancies occurred in women aged ≥ 35 years¹¹. Notably, 38.10% of participants exhibited pre-pregnancy overweight or obesity (BMI ≥ 24 kg/m²), a rate substantially higher than other regional reports (e.g., 24.80% in Northwest China) and Western populations (e.g., 28.6%). Excessive gestational weight gain (GWG) was observed in 44.6% of participants, a finding that parallels reports from Wuhan (nearly 50%)¹² and high-income countries, potentially attributable to reduced physical activity and heightened caloric intake during pregnancy. Birth outcome analysis revealed a low birth weight (LBW) incidence of 2.48%, macrosomia rate of 11.62%, small-for-gestational-age (SGA) prevalence of 5.29%, and large-for-gestational-age (LGA) proportion of 24.20%. These figures diverged from prior studies: Cui et al.¹³ reported a lower macrosomia rate (8.9%), potentially reflecting regional variations or the large-scale sampling in our study. These discrepancies likely reflect methodological heterogeneity in growth assessment criteria and population characteristics.

Risk factors associated with neonatal birth weight

Neonatal birth weight serves as a critical indicator of infant health and a determinant of long-term developmental trajectories. Our analysis identified distinct prenatal risk profiles for low birth weight (LBW) and macrosomia. Pre-pregnancy underweight significantly increased LBW risk (RR = 1.58), consistent with Japanese findings demonstrating elevated odds among underweight mothers (OR = 1.86, 95% CI 1.04–3.31)¹⁴. Conversely, pre-pregnancy overweight (RR = 1.88) and obesity (RR = 2.79) emerged as strong predictors of macrosomia, aligning with meta-analytic evidence highlighting prepregnancy obesity as a key driver of excessive fetal growth⁶. Notably, pre-pregnancy underweight exhibited a protective effect against macrosomia (RR = 0.40), reinforcing the bidirectional impact of maternal nutritional status on birth outcomes. Collectively, these findings underscore the critical importance of pre-conception weight management programs in mitigating risks of abnormal birth weight, particularly among populations undergoing rapid nutritional transitions.

The study demonstrated that insufficient gestational weight gain (GWG) significantly elevated the risk of low birth weight (LBW) (RR = 1.60), whereas excessive GWG paradoxically reduced LBW incidence (RR = 0.82). Conversely, excessive GWG was strongly associated with macrosomia (RR = 2.15), while insufficient GWG acted as a protective factor against excessive fetal growth (RR = 0.73). These bidirectional associations—where insufficient GWG elevates LBW risk while excessive GWG increases macrosomia risk—are corroborated by global evidence. Global evidence indicates > 16 kg maternal weight gain as a key macrosomia predictor (OR = 3.1, 95% CI 2.0-4.8), consistent with prior reports of 3.5-fold higher macrosomia risk.

Extensive evidence highlights the critical impact of gestational weight gain (GWG) on neonatal birth weight, with optimal GWG management serving as a foundational strategy for achieving healthy birth outcomes.

Hypertensive disorders during pregnancy exhibited dual impacts, elevating risks for both LBW (RR = 1.75) and macrosomia (RR = 1.50). Early-pregnancy metabolic derangements further modulated birth weight outcomes. Elevated fasting blood glucose (RR = 1.46) and urinary glucose (RR = 1.67) independently predicted macrosomia, corroborating findings that each 10 mg/dL increase in maternal glucose correlates with a 60 g birth weight increment¹⁵. Notably, gestational diabetes mellitus (GDM) remains a well-established macrosomia driver¹⁶, emphasizing the necessity of early glycemic monitoring.

Paradoxically, low hemoglobin levels in early pregnancy increased macrosomia risk (RR = 0.94), contrasting with studies linking anemia to LBW This discrepancy may arise from early anemia detection and subsequent therapeutic interventions inadvertently promoting fetal overnutrition. Similarly, lack of folic acid supplementation emerged as a protective factor against macrosomia (RR = 0.97), despite meta-analytic evidence demonstrating its efficacy in reducing LBW and SGA risks¹⁷. This counterintuitive finding warrants investigation into potential confounding factors, such as socioeconomic disparities in supplement adherence.

Assisted reproductive technology (ART) significantly elevated LBW risk (RR = 2.69), consistent with studies reporting threefold higher LBW odds in ART-conceived neonates¹⁸. However, sex-specific susceptibility differences remained non-significant, suggesting uniform teratogenic impacts across fetal sexes.

Risk factors and intrauterine growth abnormalities

Intrauterine growth patterns serve as fundamental indicators of neonatal health, shaped by maternal metabolic and environmental exposures during pregnancy. Our analysis revealed distinct risk profiles for small-for-gestational-age (SGA) and large-for-gestational-age (LGA) infants. Pre-pregnancy underweight significantly increased SGA risk (RR = 1.68), corroborating prior evidence demonstrating elevated odds of SGA among underweight mothers (OR = 1.71, 95% CI 1.40–2.09)¹⁹. Conversely, pre-pregnancy overweight (RR = 1.68) and obesity (RR = 2.36) were strongly associated with LGA, consistent with cohort studies linking maternal obesity to excessive fetal growth²⁰.

Gestational weight gain (GWG) exhibited bidirectional effects: insufficient GWG elevated SGA risk (RR = 1.40), while excessive GWG reduced it (RR = 0.65). Conversely, excessive GWG amplified LGA likelihood (RR = 1.78), aligning with Wei et al.²¹, who reported a 58% SGA risk increase with second-trimester GWG insufficiency (OR = 1.58, 95% CI 1.14–2.20). Hypertensive disorders during pregnancy increased both SGA (RR = 2.07) and LGA (RR = 1.31) risks, potentially reflecting placental dysfunction under metabolic stress. This dual impact aligns with Wang et al.²², who associated hypertension with SGA, while gestational hypertension contributes to growth restriction.

Early glucosuria (≥ 1+) predicted LGA with AUC = 0.71, offering a low-cost screening complement to glycemic monitoring in resource-limited settings.However, the pathophysiological mechanisms underlying first-trimester glucosuria remain inadequately characterized in existing literature.Paradoxically, lack of folic acid supplementation showed no significant SGA association but reduced LGA risk (RR = 0.98), contrasting with Lin et al.²³, who reported folate’s protective effects against SGA.

Determinants of preterm birth

Preterm birth remains a critical determinant of neonatal morbidity and mortality, with modifiable prenatal factors playing a central role in its etiology. Our findings identified pre-pregnancy overweight (RR = 1.29) and obesity (RR = 1.72) as significant risk factors for preterm delivery, corroborating Su et al.²⁴ who reported elevated risks in overweight/obese women. This association exhibited a dose-dependent pattern, with maternal obesity significantly increasing preterm birth risk, underscoring the imperative for pre-conception weight optimization.

Gestational weight gain (GWG) demonstrated bidirectional associations: insufficient GWG elevated preterm birth risk (RR = 1.74), while excessive GWG exhibited a protective effect (RR = 0.72). These findings align with Scholl et al.²⁵, who linked inadequate mid-pregnancy weight gain to spontaneous preterm labor, reporting higher preterm birth odds in women with insufficient GWG (OR = 1.44, 95% CI 1.21–1.67). Hypertensive disorders during pregnancy further amplified preterm risks (RR = 2.72 for preeclampsia), consistent with Li et al. ²⁶ demonstrating preeclampsia’s strong association with early deliveries (< 34 weeks) and elevated perinatal mortality.

Notably, lack of folic acid supplementation increased preterm birth likelihood (RR = 1.10), contrasting with Liu et al.²⁷ who observed protective effects of folate intake, suggesting potential confounding by socioeconomic or nutritional factors.

Unique contributions to the field

This study provides three pivotal contributions to maternal and child health research:

1. Large-Scale Cohort Analysis in Northeast China’s Urban Transition: Utilizing data from the Liaoning Maternal and Child Health Information System (LMCHIS, 2012–2018), this study represents one of the largest population-based cohorts (N = 1,026,294) in Northeast China, a region undergoing rapid urbanization with distinct nutritional and healthcare challenges. By capturing unique interactions between metabolic risk factors (e.g., hyperglycemia, anemia) and sociodemographic determinants (e.g., urban-rural disparities), our findings offer critical empirical evidence for understanding birth outcomes in transitioning societies^7,12.

2. Integrated Multidimensional Risk Profiling: Unlike previous studies focusing on isolated determinants, this work systematically evaluates concurrent effects of clinical interventions (e.g., assisted reproductive technology [ART]), environmental exposures (e.g., teratogens), and metabolic dysregulation (e.g., elevated fasting glucose [RR = 1.46], urinary glucose [RR = 1.67]). For instance, ART was independently associated with both LBW (RR = 2.69) and SGA (RR = 1.41), while early-pregnancy teratogen exposure increased LBW risk (RR = 1.17). These integrated insights advance holistic risk assessment frameworks for adverse birth outcomes¹⁸.

3. Policy-Driven Implications for China’s Public Health: The high prevalence of pre-pregnancy overweight/obesity (38.10%) and excessive GWG (44.6%) in Liaoning Province, coupled with elevated macrosomia rates (11.62%) exceeding national averages (8.9%)¹³, underscores the urgency of context-specific interventions. Our results directly inform the World Health Organization’s call for tailored prenatal care guidelines in rapidly urbanizing populations, emphasizing GWG monitoring, metabolic screening, and pre-conception weight management as actionable priorities.

By bridging critical evidence gaps, this study not only enriches global birth outcome literature but also provides a scalable model for addressing maternal health disparities in transitioning regions.

Conclusion

This study of 1.03 million Chinese pregnancies identifies pre-pregnancy underweight, obesity, dysregulated gestational weight gain, and early-pregnancy metabolic abnormalities as key modifiable risk factors for adverse birth outcomes, including low birth weight, macrosomia, and preterm birth. This study advances the field by defining China-specific pathways to adverse birth outcomes: Urbanization-modulated risks: GWG velocity > 0.5 kg/week in cities and rural-urban BMI divergence > 3 kg/m²;Understudied exposure bundles: ART + metabolic dysfunction quadrupled PTB risk; Cost-effective screening: First-trimester glucosuria ≥ 1+ (AUC = 0.71) for LGA prediction.

These innovations provide a roadmap for regionally adapted interventions in rapidly transitioning populations. While the observational design precludes causal inference, and regional specificity may limit generalizability, the large-scale population-based data provide robust evidence for contextualized policy-making. Nevertheless, three China-specific thresholds emerged: (1) First-trimester glucosuria ≥ 1+ (AUC = 0.71); (2) GWG velocity > 0.5 kg/week in urban women; (3) Pre-pregnancy BMI divergence > 3 kg/m² between rural/urban populations. Three actionable thresholds emerged for Chinese prenatal programs: first-trimester glucosuria ≥ 1+ (AUC = 0.71), urban GWG velocity > 0.5 kg/week, and rural-urban BMI divergence > 3 kg/m².

By addressing these risks, policymakers can reduce the burden of adverse birth outcomes, advancing equitable maternal and child health strategies in transitioning societies.

Author contributions

YZ and YW were responsible for the conception of the study and manuscript drafting. YL and GL contributed to the revision and final approval of the manuscript. All authors contributed to the article and approved the submitted version.

Funding

Liaoning Provincial Natural Science Foundation 2024-BS-300. Supported by LiaoNing Revitalization Talents Program (XLYC2412090).

Data availability

The data of this study is available from the corresponding authors on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Liaoning Maternal and Child Health Hospital (No.20220318001).

Exemption from informed consent statement

In this project, data on pregnancy, childbirth, and child growth and development from the Liaoning Maternal and Child Health Service Information System will be used for research. Although this study utilized identifiable human body materials or data for research, it is no longer possible to locate the subject. Therefore, it is impossible or impractical to exempt the subject from informed consent in this study. Therefore, we apply for exemption from informed consent. All methods were performed in accordance with the relevant guidelines and regulations.