Psychiatric disorders in term-born children with marginally low birth weight: a population-based study.

Background: Marginally low birth weight (MLBW) is defined as a birth weight of 2000 ~ 2499 g. Inconsistent findings have been reported on whether children with low birth weight had higher rates of neurological, attention, or cognitive symptoms. No studies have explored the occurrence of clinically diagnosed psychiatric disorders in term- born MLBW infants. We aimed to investigate the risk of subsequent psychiatric disorders in term-born children with MLBW.
Methods: This is a nationwide retrospective cohort study, by analysing the data from Taiwan's National Health Insurance Research Database from 2008 to 2018. The study population includes propensity-score-matched term-born infants with MLBW and those without MLBW (birth weight ≥ 2500 g). Cox proportional hazard analysis was used after adjustment for potential demographic and perinatal comorbidity confounders. Incidence rates and hazard ratios (HR) of 11 psychiatric clinical diagnoses were evaluated.
Results: A total of 53,276 term-born MLBW infants and 1,323,930 term-born infants without MLBW were included in the study. After propensity score matching for demographic variables and perinatal comorbidities, we determined that the term-born MLBW infants (n = 50,060) were more likely to have attention deficit and hyperactivity disorder (HR = 1.26, 95% confidence interval (CI) [1.20, 1.33]), autism spectrum disorder (HR = 1.26, 95% CI [1.14, 1.40]), conduct disorder (HR = 1.25, 95% CI [1.03, 1.51]), emotional disturbance (HR: = 1.13, 95% CI [1.02, 1.26]), or specific developmental delays (HR = 1.38, 95% CI [1.33, 1.43]) than term-born infants without MLBW (n = 50,060).
Conclusion: MLBW was significantly associated with the risk of subsequent psychiatric disorder development among term-born infants. The study findings demonstrate that further attention to mental health and neurodevelopment issues may be necessary in term-born children with MLBW. However, possibilities of misclassification in exposures or outcomes, and risks of residual and unmeasured confounding should be concerned when interpreting our data.
(© 2024. The Author(s).)