Highlights
-
•
Maternal age is not an independent risk factor for pediatric ophthalmic morbidity.
-
•
Perinatal factors, not maternal age, drive ophthalmic disease and treatment needs.
-
•
Maternal age-related diagnoses do not increase invasive treatments or costs.
-
•
Term infants of older mothers are not at increased risk for ophthalmic disorders.
-
•
Counseling should prioritize perinatal health status over maternal demographics.
Purpose
To evaluate the independent effect of advanced maternal age (AMA) on pediatric ophthalmic morbidity, invasive interventions, and healthcare costs.
Design
Nationwide, population-based cohort study
Participants
A total of 2,500,044 mother–infant pairs from a national health insurance database, followed from birth to 5 years of age. A health-screening subgroup of 365,494 pairs with detailed maternal metabolic and lifestyle data was analyzed to control for potential confounding.
Methods
Mother–infant pairs were categorized by maternal age (<35 vs ≥ 35 years) and birth status (term/normal birth weight [NBW] vs preterm or low birth weight [LBW]). Cox proportional hazards models and Poisson or negative binomial regression models were used to estimate adjusted hazard ratios (HRs) and incidence rate ratios (IRRs) for ophthalmic outcomes, invasive procedures, and healthcare utilization. Models were sequentially adjusted for neonatal factors, socioeconomic status, maternal comorbidities, and maternal metabolic and lifestyle variables.
Main Outcomes Measures
Composite ophthalmic morbidity (amblyopia, strabismus, congenital cataract), retinopathy of prematurity (ROP), invasive ophthalmic procedures, ophthalmic outpatient visits, hospitalizations, and ophthalmology-related healthcare costs.
Results
Among Term/NBW infants, older maternal age was associated with a lower risk of composite ophthalmic morbidity (adjusted HR, 0.93; 95% CI, 0.92–0.94), primarily driven by a reduced hazard of strabismus. However, this did not translate into reductions in invasive procedures or hospitalization rates. In contrast, Preterm/LBW status was the primary driver of all ophthalmic risks (HRs 1.52–1.81) and procedures (IRRs 1.84–2.00) regardless of maternal age. Among term/NBW infants, higher ROP diagnosis rates observed in the AMA group (adjusted HR 1.331; 95% CI, 1.211–1.450) were not accompanied by increased ROP-related invasive treatment. Apparent differences in ophthalmology-related healthcare costs across maternal age groups were attenuated and no longer significant after adjustment for maternal metabolic health.
Conclusions
Advanced maternal age is not an independent biological determinant of early pediatric ophthalmic disease or treatment-requiring morbidity. In the absence of prematurity or low birth weight, children born to older mothers are not at increased risk of early ophthalmic disorders. These findings provide evidence-based reassurance for older mothers and support counseling that emphasizes perinatal health status rather than maternal age itself.
INTRODUCTION
M aternal age has been steadily increasing worldwide, driven by shifting demographic patterns and socioeconomic changes. , Traditionally, advanced maternal age (AMA) has been labeled a “high-risk” obstetric factor, primarily due to its association with adverse perinatal outcomes such as prematurity and low birth weight (LBW). ,, Consequently, older mothers often face significant psychological distress and societal pressure, fueled by the perceived risk that their age alone may compromise the long-term health of their offspring. , However, emerging evidence in general pediatrics suggests that the impact of AMA may be neutral or even protective when socioeconomic and environmental factors are optimized, challenging the conventional “biological decline” narrative. ,
Pediatric ophthalmic disorders—including strabismus, amblyopia, and retinopathy of prematurity (ROP)—are major causes of long-term visual impairment and socioeconomic burden. ,,, While some studies have reported higher rates of these conditions in children of older mothers, these associations remain controversial as they often fail to disentangle the independent effect of maternal age from the confounding influence of prematurity and maternal metabolic health. ,, Most existing literature is limited by small sample sizes, reliance on maternal recall, or incomplete adjustment for the complex interplay between maternal lifestyle and neonatal status. ,,, Furthermore, the economic implications and healthcare utilization patterns associated with AMA in pediatric ophthalmology have not been rigorously evaluated on a national scale.
In this study, we sought to determine whether maternal age independently increases the risk of pediatric ophthalmic morbidity or if this perceived risk is merely a reflection of coexisting perinatal and maternal health factors. Using a massive nationwide cohort of 2.5 million mother–infant pairs, we evaluated the association between AMA and various ophthalmic outcomes, including invasive procedures and healthcare costs. By uniquely integrating detailed maternal health-screening data—including BMI, metabolic parameters, and lifestyle factors—we aimed to provide a definitive answer to whether the “AMA risk” in pediatric ophthalmology is a biological reality or a statistical artifact driven by manageable health variables.
METHODS
Data Source
This study was designed as a retrospective, population-based cohort analysis using the National Health Insurance Service (NHIS) claims database of South Korea. The NHIS provides comprehensive nationwide medical records and covers approximately 97% of the Korean population, including information on demographics, outpatient and inpatient visits, procedures, prescriptions, and diagnostic codes based on the Korean Classification of Diseases, 8th revision (KCD-8), aligned with ICD-10.
Mother–infant pairs were identified through family insurance registration numbers that link dependents to a primary insurance holder. Newborns are typically registered under their mother’s insurance account at birth, enabling deterministic linkage between maternal health information and neonatal healthcare claims. Maternal health screening variables (including body mass index, blood pressure, cholesterol level, fasting glucose, creatinine, smoking status, alcohol intake, and physical activity) were available for a subset of mothers who participated in the National Health Screening Program.
All data were fully de-identified before analysis. The study adhered to the principles of the Declaration of Helsinki and applicable national regulations, and was approved by the Institutional Review Board of Yonsei University Severance Hospital (IRB No. 9-2024-0161). Informed consent was waived owing to the retrospective nature of the study and the use of de-identified, routinely collected data.
Participants
A nationwide maternal–newborn cohort was constructed from infants born between 2010 and 2023 using the NHIS claims database. Of 5,360,085 neonates initially identified, those whose maternal insurance numbers were unavailable, those with leukomalacia or myopathy-related diagnostic codes (Supplementary Table 1), those with missing maternal delivery claim codes, and those whose birth year could not be clearly defined between 2010 and 2018 were excluded. After additional exclusion of neonates with congenital diseases (Q codes, except ophthalmic Q codes) and those who died within 5 years of birth, 2,500,044 mother–infant pairs remained for analysis ( Figure 1 ). Participants were categorized into 4 groups according to maternal age and neonatal birth status according to diagnostic codes (Supplementary Table 2): (1) Term birth and normal birth weight (NBW) infants born to younger mother (<35 years), (2) Term birth and NBW infants born to older mother (≥35 years), (3) Preterm or LBW infants born to younger mother (<35 years), and (4) Preterm or LBW infants born to older mother (≥35 years). A subgroup analysis was additionally conducted among 365,494 mother–infant pairs for whom maternal health screening data were available. These subgroup participants were classified using the aforementioned 4 maternal age–birth status categories ( Figure 1 ).
Flowchart of the Study Population Selection. From 5,360,085 neonates identified in the National Health Insurance Service (NHIS) database (2010–2023), we excluded those without maternal insurance linkage, those with leukomalacia or myopathy-related diagnostic codes, those lacking maternal delivery claim codes, and those with unclear birth year (2010–2018). Neonates with congenital diseases (Q codes, except ophthalmic Q codes) and those who died within 5 years of birth were further excluded, yielding 2,500,044 mother–infant pairs for analysis. Participants were categorized into 4 maternal age–birth status groups: (1) Term/NBW–younger mothers (<35 years), (2) Term/NBW–older mothers (≥35 years), (3) Preterm or LBW–younger mothers, and (4) Preterm or LBW–older mothers. A subgroup of 365,494 pairs with available maternal health screening data was classified using the same 4 categories.
Outcomes
Pediatric ophthalmic outcomes were assessed from birth through 5 years of age. Ophthalmic outcomes included amblyopia, congenital cataract, and strabismus (diagnostic codes listed in Supplementary Table 3). Nasolacrimal duct (NLD) disorders were evaluated separately because of their limited impact on long-term visual function. Retinopathy of prematurity (ROP) was also analyzed independently, as it occurs almost exclusively in preterm or low birth weight (LBW) infants and therefore differs fundamentally from term infant outcomes.
Invasive ophthalmic procedures included strabismus surgery, ROP-related procedures (intravitreal anti–VEGF injection or laser photocoagulation), cataract surgery, retinal detachment surgery, and NLD-related procedures (codes listed in Supplementary Table 4).
Ophthalmic outpatient visits were defined as encounters in which the primary diagnostic code was an ophthalmic condition. Ophthalmic hospitalizations were defined as inpatient admissions with a primary ophthalmic diagnostic code. Total ophthalmology-related cost was calculated as the sum of all claim-based expenditures incurred during ophthalmic outpatient visits and ophthalmic hospitalizations.
Covariates included neonatal sex, plurality (singleton vs multiple), socioeconomic status, residential area (urban vs rural), and maternal comorbidities (hypertension, diabetes mellitus, dyslipidemia, hyperthyroidism, hypothyroidism, and autoimmune diseases). A prespecified subgroup analysis was conducted among mother–infant pairs with available maternal health screening data. For this subgroup, additional covariates included body mass index, systolic and diastolic blood pressure, total cholesterol, fasting glucose, creatinine levels, smoking status, alcohol intake, and regular physical activity.
Statistical Analysis
Descriptive statistics were used to compare baseline characteristics across maternal age–birth status groups. Continuous variables were summarized as means with standard deviations (SDs) and compared using the Student’s t-test, whereas categorical variables were presented as counts and percentages and compared using the chi-square test.
For time-to-event outcomes (all ophthalmic outcomes except retinopathy of prematurity [ROP]), incidence rates were calculated per 1000 person-years. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Because ROP occurs almost exclusively in preterm or low–birth weight infants and is extremely rare among term infants, logistic regression was used to estimate odds ratios (ORs) and 95% CIs for ROP.
For event-count outcomes (invasive procedures, ophthalmic outpatient visits, and ophthalmic hospitalizations), incidence rate ratios (IRRs) and 95% CIs were estimated using Poisson regression, or negative binomial regression when overdispersion was detected. All count models included an offset term of log(person-time) to account for differential follow-up duration.
Total ophthalmology-related costs were analyzed using a generalized linear model with a Tweedie distribution and log link, which appropriately accommodates semicontinuous cost data with a large proportion of zeros and right-skewed positive values. Results were expressed as cost ratios with 95% CIs for each maternal age–birth status group.
All multivariable models were adjusted for the following covariates: neonatal sex, birth plurality (singleton or multiple birth), socioeconomic status, residential area (urban or rural), and comorbidities, including hypertension, diabetes mellitus, dyslipidemia, hyperthyroidism, hypothyroidism, and autoimmune disorders. Sensitivity analyses additionally adjusted for maternal health screening variables, including body mass index (BMI), systolic and diastolic blood pressure, total cholesterol, fasting glucose, creatinine, smoking status, alcohol consumption frequency, and regular exercise.
All analyses were performed using R (version 4.0.3; R Foundation for Statistical Computing) and SAS software (version 9.4; SAS Institute Inc.), and two-sided P values <.05 were considered statistically significant.
RESULTS
Baseline Characteristics
A total of 2,500,044 mother-infant pairs were classified into 4 groups according to maternal age and birth status: (1) Term/NBW–Younger mothers (n = 1,964,420), (2) Term/NBW–Older mothers (n = 449,661), (3) Preterm or LBW–Younger mothers (n = 64,254), and (4) Preterm or LBW–Older mothers (n = 21,709) ( Table 1 ). Older mothers (≥35 years) were more likely to have higher socioeconomic status (SES Q4: 31.9% vs. 23.3%), higher rates of Cesarean delivery, and a greater burden of pre-existing comorbidities, including hypertension and diabetes, compared to younger mothers.
Table 1
Baseline Characteristics of the Study Population By Maternal Age and Birth Status.
| Characteristics |
Term/NBW-Younger Mother
(n = 1,964,420) |
Term/NBW-Older
Mother (n = 449,661) |
Preterm or LBW-Younger Mother
(n = 64,254) |
Preterm or LBW-Older Mother
(n = 21,709) |
p -Value |
|---|---|---|---|---|---|
| Maternal factors | |||||
|
Maternal age, years
(mean ± SD) |
30.51 ± 3.31 | 37.86 ± 1.95 | 30.83 ± 3.34 | 38.02 ± 2.06 | <.001 |
| Parity, n (%) | <.001 | ||||
| Primiparity | 1,841,348 (93.73%) | 423,272 (94.13%) | 43,188 (67.21%) | 14,131 (65.09%) | |
| Multiparous | 123,072 (6.27%) | 26,389 (5.87%) | 21,066 (32.79%) | 7,578 (34.91%) | |
| Delivery mode, n (%) | <.001 | ||||
| Vaginal | 1,249,469 (63.60%) | 223,533 (49.71%) | 25,265 (39.32%) | 6,016 (27.71%) | |
| Cesarean section | 714,951 (36.40%) | 226,128 (50.29%) | 38,989 (60.68%) | 15,693 (72.29%) | |
| Socioeconomic status, n (%) | <.001 | ||||
| Null | 79,814 (4.06%) | 15,397(3.42%) | 2,557 (3.98%) | 729 (3.36%) | <.001 |
| Q1 (lowest) | 309,702 (15.77%) | 65,701 (14.61%) | 10,266 (15.98%) | 3,138 (14.45%) | <.001 |
| Q2 | 499,720 (25.44%) | 85,964 (19.12%) | 15,789 (24.57%) | 4,085 (18.82%) | <.001 |
| Q3 | 695,463 (35.40%) | 150,040 (33.37%) | 22,589 (35.16%) | 7,257 (33.43%) | <.001 |
| Q4 (highest) | 379,721 (19.33%) | 132,559 (29.48%) | 13,053 (20.31%) | 6,500 (29.94%) | <.001 |
| Residence, n (%) | <.001 | ||||
| Urban | 1,121,316 (57.08%) | 243,137 (54.07%) | 36,219 (56.37%) | 11,815 (54.42%) | |
| Rural | 843,104 (42.92%) | 206,524 (45.93%) | 28,035 (43.63%) | 9,894 (45.58%) | |
| Comorbidities, n (%) | |||||
| Hypertension | 12,338 (0.63%) | 6,708 (1.49%) | 1,838 (2.86%) | 1,175 (5.41%) | <.001 |
| Diabetes | 25,924 (1.32%) | 12,718 (2.83%) | 1,804 (2.81%) | 1,058 (4.87%) | <.001 |
| Dyslipidemia | 69,336 (3.53%) | 25,685 (5.71%) | 3,715 (5.78%) | 1,948 (8.97%) | <.001 |
| Thyroid disease | 152,358 (7.75%) | 46,633 (10.37%) | 7,014 (10.91%) | 2,994 (13.79%) | <.001 |
| Autoimmune disease | 37,021 (1.88%) | 11,699 (2.60%) | 1,878 (2.92%) | 792 (3.65%) | <.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
