Emotional problems, especially anxiety, have become increasingly common in recent generations. Few population-based studies have examined trajectories of emotional problems from early childhood to late adolescence or investigated differences in psychiatric and functional outcomes.

Using the Avon Longitudinal Study of Parents and Children (ALSPAC, n = 8286, 50.4% male), we modeled latent class growth trajectories of emotional problems, using the parent-reported Strength and Difficulties Questionnaire emotional scale (SDQ-E) on seven occasions (4–17 years). Psychiatric outcomes in young adulthood (21–25 years) were major depressive disorder (MDD), generalized anxiety disorder (GAD), and self-harm. Functional outcomes were exam attainment, educational/occupational status, and social relationship quality.

We identified four classes of emotional problems: low (67.0%), decreasing (18.4%), increasing (8.9%), and persistent (5.7%) problems. Compared to those in the low class, individuals with decreasing emotional problems were not at elevated risk of any poor adult outcome. Individuals in the increasing and persistent classes had a greater risk of adult MDD (RR: 1.59 95% CI 1.13–2.26 and RR: 2.25 95% CI 1.49–3.41) and self-harm (RR: 2.37 95% CI 1.91–2.94 and RR: 1.87 95% CI 1.41–2.48), and of impairment in functional domains. Childhood sleep difficulties, irritability, conduct and neurodevelopmental problems, and family adversity were associated with a persistent course of emotional problems.

Childhood emotional problems were common, but those whose symptoms improved over time were not at increased risk for adverse adult outcomes. In contrast, individuals with persistent or adolescent-increasing emotional problems had a higher risk of mental ill-health and social impairment in young adulthood which was especially pronounced for those with persistent emotional problems.

Understanding the causes of poor mental health in early childhood and adolescence is important and one potential and relatively unexplored factor is residential mobility in formative years. Previous studies in this area have been relatively small, and potentially limited due to methodological issues.

To investigate relationship between early residential instability and poor mental health among adolescents and young adults in Northern Ireland.

A census-based record linkage study of 28% of the NI population was used. Our sample was children aged 0-8 years at 2001 (n=49,762) with self-reported chronic mental ill-health in the 2011 census as outcome and address changes assessed six-monthly. Logistic regression was used with adjustment for socio-economic status (SES) and household composition and marital dissolution. The relationship between address-change and non-mental health outcomes was also tested.

Overall, 54% had moved house at least once, and 0.5% of the cohort aged 10-18 at 2011 reported chronic mental ill-health. There was a graded relationship between address-change and mental ill-health (ORadj 3.67, 95%CIs 2.11, 6.39 for 5 or more moves). This relationship was not modified by SES or household composition. Marital dissolution was associated with poor mental health but did not modify the relationship between address-change and mental health.

This large study clearly confirms the close relationship between address change in early years and later poor mental health. Life events, including mobility, should be distinguished based on the overall impact which a transition can have over the individual's life.