Relative age as a risk factor for psychiatric diagnoses in children born preterm and to term: a cohort study

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Objective: To assess relative age effects (how the youngest children in a school class are at increased risk compared with their older classmates) on healthcare use with psychiatric diagnoses in childhood and adolescence within preterm and term-born boys and girls. Design/setting/patients: Registry-based cohort study including individuals born in Norway from 1991 to 2012 with follow-up data from the National Patient Registry from 2008 to 2017 when they were aged between 4 and 18 years. Exposures: Relative age was defined according to birth month and grouped in four 3-month periods. Gestational age at birth (GA week+days) was categorised into preterm (GA 23+0–36+6) and term (GA 37+0–41+6). Main outcome measures: The presence of psychiatric diagnoses (any diagnosis and specific diagnosis groups according to ICD-10) in data from specialist healthcare contacts at different ages during follow-up was compared between relative age groups in preterm and term born using generalised estimating equation logistic regression analyses. Results: Of 1 109 411 individuals, 7% were born preterm. Relative age effects for psychiatric diagnosis and specific diagnosis groups were seen for both preterm and term-born boys and girls, with the strongest estimates for any psychiatric diagnosis in the relatively younger preterm girls born in October–December, compared with the relatively older preterm girls born in January–March (OR 1.43 (95% CI 1.25 to 1.63) at ages 4–10 years). Conclusions: Relative age effects were widely demonstrated for healthcare use with psychiatric diagnoses within term and preterm born, for both girls and boys. The excess risk for those born late in the year added to already existing adversity in children born preterm, emphasising the need for additional consideration related to school and societal structures.
Funding Information
  • Fakultet for medisin og helsevitenskap, Norges Teknisk-Naturvitenskapelige Universitet (2015/18873)

This publication has 38 references indexed in Scilit: