Objective: This study aimed to examine profiles of co-occurrence of developmental disorders and their association with birth before full term.
Design: Latent class analysis of cohort data with linked health data.
Setting: Bradford, England.
Patients: 13 172 children were included in the analysis.
Outcome measures: Developmental disorder in medical records.
Methods: Data were censored at each child's 12th birthday. The latent class analysis identified patterns of developmental disorders. Multinomial logistic regression explored the association with gestational age while adjusting for clinical and socio-factors.
Results: The majority (12,536) had a low risk of developmental disorders; this group was named 'typical development'. The remaining children were classified into three groups: 'educational difficulties' (347 children); 'social, emotional, behavioural and communication difficulties' (189 children) and 'early developmental impairment, with physical and intellectual disabilities' (100 children).Compared with 'typical development', very preterm birth was associated with an increased likelihood of being in the 'early developmental impairment, with physical and intellectual disabilities' group, adjusted relative risk ratio (aRRR): 9.22 (95% CI 4.58 to 18.55). Birth before full term was associated with increased likelihood of being in the 'educational difficulties' group; risk was highest <34 weeks (aRRR: 2.64 (95% CI 1.44 to 4.83)) but persisted up to 37-38 weeks: aRRR: 1.41 (95% CI 1.10 to 1.81). There was no association between gestational age and the 'social, emotional, behavioural and communication difficulties'' group.
Conclusion: Four distinct profiles of developmental disorders were identified; gestational age was associated with two of these. Understanding which disorders children are most at risk of and how these co-occur can help provide accurate information to families and contribute to prompt diagnosis.