Cohort demographics by data sweep.

Objective To investigate associations between clustered adverse neonatal events and later-life multimorbidity. Design Secondary analysis of birth cohort data. Setting Prospective birth cohort study of individuals born in Britain in one week of 1970. Population Respondents provided data at birth (n = 17,196), age 34 (n = 11,261), age 38 (n = 9,665), age 42 (n = 9,840), and age 46 (n = 8,580). Methods Mixed components analysis determined included factors, ‘Birthweight’; ‘Neonatal cyanosis’; ‘Neonatal cerebral signs’; ‘Neonatal illnesses’; ‘Neonatal breathing difficulties’; and ‘Prolonged duration to establishment of respiratory rate at birth’, within the composite adverse neonatal event score. Log-binomial regression quantified the unadjusted and covariate-adjusted (paternal employment status and social class; maternal smoking status; maternal age; parity; cohort member smoking status and Body Mass Index) associations between the adverse neonatal event score and risk of multimorbidity in adulthood. Outcome measures Multimorbidity at each adult data sweep, defined as the presence of two or more Long-Term Conditions (LTCs). Results 13.7% of respondents experienced one or more adverse neonatal event(s) at birth. The percentage reporting multimorbidity increased steadily from 14.6% at age 34 to 25.5% at age 46. A significant association was only observed at the 38 years sweep; those who had experienced two or more adverse neonatal events had a 41.0% (95% CI: 1.05 – 1.88) increased risk of multimorbidity, compared to those who had not suffered any adverse neonatal events at birth. This association was maintained following adjustment for parental confounders and adult smoking status. Conclusions Adverse neonatal events at birth may be independently associated with the development of midlife multimorbidity. Programmes and policies aimed at tackling the growing public health burden of multimorbidity may also need to consider interventions to reduce adverse neonatal events at birth.