Maternal and Perinatal Risk Factors for Speech Delay in Toddlers: A Systematic Review

Speech delay in toddlers is a sentinel of wider neurodevelopmental risk, yet evidence spanning the perinatal continuum remains fragmented. We conducted a preferred reporting items for systematic reviews and meta-analyses (PRISMA)-compliant systematic review to map maternal, perinatal, neonatal, and environmental determinants of speech and early language delay (12–48 months). Searches of PubMed/MEDLINE, Embase, Scopus, Web of Science, and the Cochrane Library, plus citation chaining, identified 3,300 records (last search August 29, 2025). After deduplication and dual-review screening, 43 studies were included for qualitative synthesis; 20 met high-quality thresholds using the Newcastle–Ottawa Scale (primary studies) and AMSTAR-2/ROBIS (reviews). Data were extracted in duplicate; risk of bias and certainty were appraised, and findings were organized by domain with grading of recommendations assessment, development and evaluation (GRADE) judgments. Across cohorts and case–control studies, low maternal education and socioeconomic disadvantage showed consistent associations with toddler speech/language delay. Maternal mental health problems (depression, stress) and prenatal smoking were repeatedly linked to poorer outcomes. Perinatal pathways were dominated by prematurity and low birth weight, with additive effects of intraventricular hemorrhage, periventricular leukomalacia, bronchopulmonary dysplasia, and neonatal sepsis; Appearance, pulse, grimace, activity, respiration (Apgar) scores, neonatal intensive care unit (NICU) admission, and mode of delivery functioned largely as proxies or were confounded by indication. Evidence for preeclampsia, gestational diabetes, maternal infection, and nutrition was limited or indirect. Mechanistically, convergent models implicated hypoxia–ischemia, white-matter vulnerability of the language network, chronic stress biology, and epigenetic/inflammatory programming. Overall certainty ranged from moderate (prematurity/very low birth weight, severe neonatal morbidities, and maternal education) to low/very low (delivery mode and isolated obstetric exposures). Findings support an integrated perinatal strategy: targeted maternal support (education, mental health, smoking cessation), risk-stratified NICU follow-up for preterm infants, and promotion of language-rich caregiving environments. Future work should harmonize outcomes, leverage causal inference and machine learning, and test multilevel interventions across pregnancy–infancy to narrow developmental inequities.