Key findings
This study investigated the combined influence of maternal age and parity on both pregnancy outcomes (BMI, GDM and hypertension), and birth outcomes (birth weight and length z-scores, SGA, LGA and gestational age) in a South African urban cohort. Overall, we found evidence partially in support of our hypothesised U-shaped relationship, whereby both extremes of maternal age and parity were expected to be associated with variations in outcomes. Nulliparous women had lower infant birth weight (2960 g vs. 3185 g, p <.001), higher rates of SGA births (22.9% vs. 15.3%, p <.001), and lower BMI (26.28 vs. 28.57 kg/m², p <.001). In contrast, mothers > 23 years, ≥ 1 child had higher BMI (26.3 vs. 28.6 kg/m², p <.001), and higher prevalence of hypertension (13.7% vs. 6.3%, p <.001) and GDM (13.6% vs. 5.8%, p =.012), as well as a greater proportion of LGA infants (8.8% vs. 3.2%, p =.009). While median BMI values for the nulliparous and mothers > 23 years, ≥ 1 child groups are not classified as obese, associations with birth outcomes were found in our study, particularly SGA among nulliparous women and LGA among mothers > 23 years, ≥ 1 child. Mothers ≤ 23 years, ≥ 1 child delivered infants with higher birth weight and length z-scores compared to nulliparous women (2960 g vs. 3185, p <.001), suggesting a protective effect of reproductive experience when combined with younger maternal age. These findings address a notable gap in the literature by simultaneously examining multiple maternal risk factors, thereby offering insights into reproductive risk factors in LMICs and informing targeted antenatal interventions based on age and parity.
Contextualisation of findings
Group difference analysis revealed significant differences across maternal age-parity groups. In line with global and South African trends [35, 36], nulliparous women were more likely to have completed higher levels of education. Mothers > 23 years, ≥ 1 child had a higher prevalence of HIV, consistent with prior research [37,38,39], which is possibly due to prolonged sexual activity and greater cumulative exposure to HIV risk factors [40]. Significant group differences also occurred for maternal anthropometric measures. Mothers > 23 years, ≥ 1 child had moderately higher weight and BMI compared to nulliparous women, and a greater proportion classified as overweight, which is associated with increased risk of metabolic complications, gestational diabetes, and LGA deliveries. These findings correspond with studies showing factors contributing to age-related weight-gain in women, such as poor-quality diets, decreased physical activity, hormonal changes, cumulative weight, retention and metabolic changes associated with successive pregnancies [41]. Despite having a higher BMI in early pregnancy, multiparous women had moderately lower gestational weight gain per week compared to nulliparous women, consistent with findings from a large retrospective cohort [42]. This greater weight gain in nulliparous women is likely due to a combination of factors, such as their body’s initial physiological adaptation to pregnancy as well as potential behavioural and psychological influences [43, 44]. Gestational weight gain patterns have important implications, with both inadequate and excessive gain being associated with pregnancy outcomes. Insufficient weight gain is associated with increased risks of IUGR, low birth weight, and preterm birth, which may translate to SGA infants [45] who are more prone to complications such as hypoglycaemia [46], temperature instability, and respiratory distress [47], as well as developmental delays [48], and long-term health conditions, including metabolic and cardiovascular diseases [49]. Conversely, excessive weight gain increases the likelihood of LGA infants [50], gestational diabetes, hypertensive disorders [51, 52], labour complications such as dystocia, as well as an increased rate of caesarean section deliveries [50].
Partly in line with our hypothesis, we found that parity at one extreme (i.e., nulliparity) was largely associated with lower birth weight, as well as a higher rate of SGA births, which have important implications given that SGA is associated with neonatal morbidity [45–47, 53], developmental delays [48], and long-term health conditions, including metabolic and cardiovascular diseases [49]. Our findings align with research demonstrating first pregnancy complications associated with uterine inexperience. Without prior gestational experience, the uterus of nulliparous women has yet to undergo adaptive remodelling and vascular changes [54,55,56,57], which may lead to suboptimal maternal–fetal resource allocation [54, 58, 59]. Further evidence in support of uterine priming, whereby prior pregnancies facilitate physiological adaptations that promote fetal growth, is our finding that multiparity, independent of maternal age, was significantly associated with birth outcomes. Infants from mothers ≤ 23 years, ≥ 1 child and mothers > 23 years, ≥ 1 child groups had infants with moderately higher birth weight scores than nulliparous women. However, only infants born to mothers ≤ 23 years, ≥ 1 child had moderately higher birth length z-scores, which have significant implications given that greater birth length is associated with better growth trajectories [60]. These findings suggest that linear growth (i.e., birth length) is more sensitive to maternal age, which is supported by prior research. In particular, studies suggest that linear growth depends more on placental and vascular function, which often declines with maternal age. Vascular ageing, characterised by increased arterial stiffness, reduced blood flow to the uterus, and impaired endothelial function, can limit the placental capacity to support sustained linear growth [61, 62].
Similar to age and parity, maternal BMI was also largely linked to infant birth weight, with normal and overweight mothers delivering infants with higher birth weight z-scores and having increased odds of LGA births compared to underweight women. Moreover, maternal BMI partially explained the association between age-parity, particularly in mothers > 23 years, ≥ 1 child, and birth weight and LGA. These findings are consistent with existing evidence not only showing that older and multiparous women tend to have higher BMI but also that higher maternal BMI promotes greater fetal growth and fat deposition due to increased nutrient transfer [41, 63,64,65]. Despite the moderate effect, the associated risks of LGA births, such as delivery complications and increased likelihood of future metabolic disease, underscore the importance of these findings [50,51,52].
In contrast to birth weight and linear growth outcomes, maternal age-parity was not significantly associated with gestational age at delivery in our study. However, gestational age was independently and moderately associated with maternal GDM and hypertension, both linked to shorter gestations. Shorter gestational age and preterm birth are associated with higher rates of mortality and morbidity, such as cerebral palsy, respiratory conditions, and increased need for intensive neonatal care [66]. These results are possibly attributed to metabolic disturbances linked to elevated maternal blood glucose levels, and vascular dysfunction associated with hypertension, both of which contribute to placental insufficiency, triggering early delivery, spontaneously or through medical intervention [67,68,69].
Together, these findings suggest a dual burden of age and parity, particularly for mothers > 23 years, ≥ 1 child. While prior pregnancies may enhance physiological readiness for childbirth, supporting fetal growth, age-related increases in BMI and metabolic risk seemingly offset these benefits. In particular, these conditions increase the likelihood of LGA births and associated complications such as obstructed labour, perineal trauma, and postpartum haemorrhage [28, 50, 51]. In contrast, young nulliparous women, particularly those with low BMI, may be at higher risk of SGA births.
Implications
Our findings suggest that maternal age and parity together influence pregnancy and birth outcomes. This highlights the limitations of age-based risk frameworks, suggesting that solely relying on maternal age in prenatal risk assessment may be insufficient and potentially overlook important reproductive and metabolic risk factors.
Research implications of our study include highlighting the need to incorporate more detailed reproductive histories, for example age at first birth, intervals between pregnancies and cumulative parity, to better understand how these factors influence maternal and fetal outcomes. Additionally, the findings demonstrate the value of investigating how underlying mechanisms, such as placental function, vascular and metabolic adaptations, and inflammation, influence the relationship between maternal age, parity, and birth outcomes, which may clarify causal pathways.
The findings emphasise the need for antenatal care to go beyond standard maternal age screening but also account for reproductive history (i.e., parity) and emerging chronic disease risks, such as GDM and hypertension. In particular, they underscore the importance of more tailored antenatal interventions. For example, nutritional counselling and fetal growth monitoring can be applied for underweight, nulliparous women, whereas weight management support and proactive metabolic screening may be beneficial for overweight, older multiparous women. These findings can, therefore, guide healthcare providers in optimising resource allocation, facilitating earlier identification and intervention for at-risk women, and improving maternal and infant outcomes.
Finally, this study highlights the need for refined antenatal care guidelines in South Africa and other LMICs. While maternal age, parity, and chronic disease risk factors are recorded in existing frameworks such as the Basic Antenatal Care (BANC) the Road to Health Card (RTHC), this study emphasise the need to improve the application of these data. In particular, our findings underscore the importance of risk stratification models that not only document information but also uses it to improve risk identification and targeted interventions. This goal aligns with the South Africa’s Reproductive, Maternal, Newborn, Child, and Adolescent Health (RMNCAH) policy framework and the WHO 2020 antenatal guidelines [70], which prioritises personalised, context-specific risk-based models of antenatal care. More effective use of data on key factors identified in this study, namely maternal age, parity, and chronic disease risk, within existing approaches can improve risk stratification by better identifying women who are more likely to experience adverse outcomes, thereby enabling the earlier detection and more targeted intervention.
Limitations and future research
Our study has several limitations that need to be considered when interpreting the findings. Age at recruitment and parity were used as a proxy for estimating the age at first childbirth, assuming regular birth intervals, with a cutoff age of 23 applied rather than the widely accepted threshold of 35 years. This approach may lead to misclassification of maternal age-parity groupings, limit our ability to distinguish the effects of early childbearing from those of higher parity on birth outcomes and, therefore, not fully capture the complexity of maternal-age related risks. Due to the inclusion criteria of 18 years and the absence of older mothers in the sample, this study does not capture the extremes of maternal age. This is a notable limitation given that both adolescent and advanced maternal age are associated with adverse maternal and infant outcomes. The recruitment through clinic attendance may introduce selection bias, limiting the representativeness of the sample. While BMI, GDM, and hypertension were included in the analysis of age-parity groups and infant outcomes, there remains the potential for various other residual confounders not accounted for in this study that may have influenced the observed associations (e.g., nutritional status, substance use, environmental exposures). Due to the cross-sectional nature of the study, causality between maternal age, parity, and infant outcomes cannot be inferred. Finally, missing birth record data and the specific urban, South African context of the cohort limit the generalisability of our findings to other populations and settings. Specifically, our findings may not be applicable to rural communities, mothers accessing private healthcare, or populations in high-income countries with different healthcare systems.
Future research should focus on incorporating more detailed reproductive histories (i.e., age at first birth, total number of pregnancies, and the interval between pregnancies), to provide clarity on how timing and frequency of childbirth influence maternal and fetal outcomes. Biological and physiological indicators should be considered in future studies to provide an understanding of the underlying processes that link maternal age, parity, and metabolic risk to birth outcomes. Additionally, future studies should include maternal age extremes to provide a more comprehensive assessment of the influence of maternal age and parity on infant outcomes.
link