To the Editor: Small for gestational age (SGA) is generally defined as birth weight at least 2 standard deviations below the mean or below the 10th percentile for gestational age. SGA children face an increased risk of later-life disease, particularly neurological complications, and have higher rates of cognitive, socioemotional, and behavioral difficulties than those born appropriate for gestational age (AGA). The long-term cognitive implications of SGA, such as lower intelligence quotient (IQ) and academic underachievement, are well documented, but mental disorders related to SGA have been less systematically described. Attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) are prevalent neurodevelopmental conditions with substantial long-term impact on quality of life.1 Moreover, children with ASD and ADHD are at increased risk for depression and anxiety, which can have long-term consequences for mental and physical health.2 Whether SGA is associated with ASD, ADHD, anxiety, and depression remains controversial. Some studies suggest increased risks of ASD, depression, and behavioral problems in SGA, whereas others report no clear association. Earlier meta-analyses of SGA in relation to ASD or depression were based on relatively small samples and rarely distinguished preterm from full-term SGA. Preterm birth (<37 weeks), especially extremely preterm (≤28 weeks), is itself a strong risk factor for mental health disorders. To address these gaps, we conducted a systematic review and meta-analysis to evaluate the association between SGA and ASD, ADHD, depression, and anxiety, with subgroup analyses by gestational age. We initially considered a broader range of psychopathology, but due to very small numbers of other psychiatric diagnoses among SGA individuals, only anxiety and depression were retained as additional psychiatric outcomes. This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was registered in PROSPERO (CRD42023484809) Supplementary Information 1–2, https://links.lww.com/CM9/C791. We searched MEDLINE, PubMed, Web of Science, Embase, and APA PsycINFO from inception to March 1, 2025, using Medical Subject Headings and free-text terms related to “small for gestational age”, “fetal growth retardation”, ASD, ADHD, depression, and anxiety Supplementary Information 3, https://links.lww.com/CM9/C791. Two authors independently screened titles, abstracts, and full texts Supplementary Information 4, https://links.lww.com/CM9/C791. Inclusion criteria were peer-reviewed observational studies with population-based data on intrauterine growth restriction (IUGR) or SGA, AGA controls, and odds ratios (ORs) with 95% confidence intervals (CIs). When multiple reports used the same cohort, the study with the largest sample size was retained. Two authors independently extracted data using a standardized form and assessed quality using the Newcastle–Ottawa Scale (NOS), with scores <4 indicating low quality, 5–7 moderate, and ≥8 high quality. Most studies diagnosed ASD and related conditions according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) or the International Classification of Diseases (ICD)-8/9/10; other tools are detailed in Supplementary Tables 1–4, https://links.lww.com/CM9/C791. When ORs were not reported, they were calculated from crude data. Random-effects models were used to pool effect estimates, accounting for between-study heterogeneity and providing more conservative and robust estimates. Heterogeneity was assessed using the I2 statistic, and publication bias was evaluated with Egger’s test and the Duval and Tweedie trim-and-fill method, where ≥10 studies were available. Sensitivity analyses were performed by excluding individual studies. All analyses used Stata/SE (Version 16.0; StataCorp., College Station, TX, USA). We identified 4672 records and excluded 1709 duplicates. After screening, 131 full texts were assessed; 107 were excluded for irrelevance, overlapping cohorts, or missing full text Supplementary Table 5, https://links.lww.com/CM9/C791. Three additional studies were identified from reference lists. Overall, 27 studies were included Supplementary Figure 1, https://links.lww.com/CM9/C791. For ASD, 12 cohort studies including 668,597 SGA and 5,532,653 AGA controls were analyzed. Individuals born SGA had a higher risk of ASD (OR: 1.15, 95% CI: 1.09–1.22, P <0.001) Supplementary Figure 2, https://links.lww.com/CM9/C791, with moderate overall heterogeneity. In gestational-age-stratified analyses, the OR for ASD was 1.22 (95% CI: 1.08–1.38, P = 0.001) for preterm and 1.07 (95% CI: 1.03–1.13, P = 0.002) for full-term SGA, with low heterogeneity in both subgroups. Studies with unknown gestational age showed more variability, suggesting that overall heterogeneity was largely driven by this group. Four studies with eight samples, including 47,581 SGA and 1,995,288 controls, examined ADHD. The pooled OR for ADHD in SGA births was 1.84 (95% CI: 1.44–2.34, P <0.001) Supplementary Figure 3, https://links.lww.com/CM9/C791. When stratified by gestational age, both preterm and full-term SGA showed significantly increased ADHD risk, with ORs of 2.15 (95% CI: 1.54–3.02, P <0.001) and 1.55 (95% CI: 1.40–1.70, P <0.001), respectively. Substantial heterogeneity was present overall and in preterm and unknown-gestational-age SGA, whereas full-term SGA showed little heterogeneity, suggesting that differences in gestational age, diagnostic criteria, and other confounders contributed to variability in effect estimates. Given limited childhood data on depression after SGA birth, we included outcomes from adolescence to adulthood. Eight studies comprising 30,862 SGA and 120,129 controls evaluated depression Supplementary Figure 4, https://links.lww.com/CM9/C791. Overall, there was no significant association between SGA and depression (OR: 1.07, 95% CI: 0.97–1.18, P = 0.156), with low heterogeneity. In subgroup analyses, preterm SGA was not associated with increased depression risk (OR: 1.03, 95% CI: 0.91–1.17, P = 0.624), whereas full-term SGA was associated with a modestly elevated risk (OR: 1.06, 95% CI: 1.03–1.10, P = 0.001). For anxiety, we analyzed five studies including 39,003 SGA and 1,098,161 controls spanning adolescence to adulthood Supplementary Figure 5, https://links.lww.com/CM9/C791. SGA birth was significantly associated with anxiety (OR: 1.56, 95% CI: 1.37–1.77, P <0.001), again with low overall heterogeneity. Both preterm and full-term SGA were associated with increased anxiety risk, with ORs of 1.60 (95% CI: 1.19–2.15, P = 0.002) and 1.55 (95% CI: 1.26–1.91, P <0.001), respectively. In the SGA–ASD analyses, Egger’s test indicated funnel-plot asymmetry (t = 3.75, P = 0.001). The Duval and Tweedie trim-and-fill method suggested nine missing studies on the left side Supplementary Figure 6, https://links.lww.com/CM9/C791. According to the approach proposed by Sterne et al3, the location of these missing studies suggests that asymmetry was unlikely due to publication bias. Asymmetry was confined to studies with unknown gestational age (t = 2.85, P = 0.019), whereas preterm and full-term subgroups showed no evidence of asymmetry (P = 0.157 and 0.505), suggesting heterogeneity in this subgroup as the likely source. No publication bias was detected for SGA–depression (t = 0.46, P = 0.654) Supplementary Figure 7, https://links.lww.com/CM9/C791. For SGA–ADHD and SGA–anxiety, Egger’s test was not performed because fewer than 10 studies were available.3 Sensitivity analyses showed stable results for SGA–ASD, SGA–ADHD, and SGA–anxiety Supplementary Figure 8A–C, https://links.lww.com/CM9/C791. However, removing the full-term SGA data reported by Upadhyaya et al4 substantially changed the SGA–depression estimate, indicating sensitivity in that subgroup Supplementary Figure 8D, https://links.lww.com/CM9/C791. This finding suggests that the pooled estimate for depression in the full-term SGA subgroup is influenced by a single large study and should be interpreted cautiously. ASD and ADHD frequently cooccur with depression and anxiety, yet the potential link between SGA birth and these neurodevelopmental and psychiatric outcomes has not been fully clarified. By separating preterm and full-term SGA and drawing on large population-based cohorts, our study strengthens the evidence for an association between SGA and mental health outcomes. We confirmed a significant association between SGA and ASD risk, consistent with earlier work such as that by Jenabi et al5, while providing more precise pooled estimates from an expanded sample. A similarly elevated risk of ADHD was observed, supporting the hypothesis that growth restriction in utero can broadly disrupt neurodevelopmental trajectories. The strong comorbidity between ASD and ADHD further supports the plausibility of shared mechanisms. In contrast to earlier meta-analyses, such as that by Loret de Mola et al6, which did not find a significant link between SGA and depression, our results suggest a modest increase in depression risk among full-term SGA individuals; however, this association appears sensitive to individual studies, largely influenced by a single large cohort, and therefore should be interpreted with caution. Notably, our findings also indicate a stronger association with anxiety, a relationship that has been largely overlooked in earlier literature. The inclusion of adolescent cohorts underscores the early psychological burden of SGA status and highlights the need for early identification and support. Stratified analyses by gestational age showed higher risks of ASD and ADHD among preterm than full-term SGA children, consistent with the known neurodevelopmental vulnerability of prematurity. In contrast, no clear increase in depression risk was observed in preterm SGA, likely reflecting limited statistical power. Egger’s test suggested funnel-plot asymmetry for the SGA–ASD association, but this was confined to studies with unknown gestational age and probably reflected heterogeneity in design and sample characteristics rather than publication bias. Preterm and full-term subgroups showed no asymmetry. Sensitivity analyses supported the robustness of the pooled estimates for ASD, ADHD, and anxiety, whereas the SGA–depression association was driven largely by a single large study (Upadhyaya et al7), warranting cautious interpretation. The mechanisms linking SGA to adverse neurodevelopmental and psychiatric outcomes remain multifactorial. IUGR, commonly underlying SGA, may disrupt late-gestation brain development through impaired placental function, leading to reductions in gray and white matter volume and impaired maturation of neurons and glial cells. Neuroinflammation has also been implicated, with IUGR models showing elevated microglial activation and reactive astrocytosis, which release proinflammatory cytokines, disrupting synaptogenesis, and neuronal repair. Excitatory/inhibitory imbalance is another proposed mechanism. In guinea pig models, IUGR altered the ratio of excitatory neurons expressing COUP-TF–interacting protein 2 (Ctip2) and inhibitory somatostatin (SST)-immunoreactive neurons, contributing to functional impairments observed in conditions, such as ASD and schizophrenia. Some limitations should be acknowledged. Although most included studies provided adjusted estimates, a number relied on unadjusted or partially adjusted outcomes, introducing potential residual confounding. Definitions of SGA were not entirely consistent: While most adhered to the <10th percentile standard, others used <3rd, <5th, or <15th percentile threshold, which may have introduced exposure misclassification. Additionally, some studies assessed psychiatric outcomes using nondiagnostic questionnaires, limiting the clinical validity of their findings. Given the overlapping nature of ASD, ADHD, depression, and anxiety, as well as the limited follow-up duration in many included cohorts, we were unable to assess longitudinal progression or clustering of mental disorders. Future large-scale prospective studies are needed to elucidate the developmental trajectories and long-term psychiatric outcomes of individuals born SGA. In conclusion, this study suggested that SGA is adversely associated with mental health. Our analysis demonstrated that SGA birth is associated with an increased risk of neurodevelopmental disorders, such as ASD and ADHD, as well as depression and anxiety in later life. By subgrouping our meta-analysis according to gestational age, we sought to minimize prematurity and offer a more nuanced understanding of the mental risks. Our findings highlighted the fetal origins of mental disorders and emphasized the need for early identification and intervention in children born SGA. Funding This study was supported by the National Key R&D Program of China (No. 2023YFC2706300). Conflicts of interest None.
Tian et al. (Tue,) studied this question.