Symposium: Understanding the mental health and wellbeing of autistic young people: findings from the Millennium Cohort Study
Mental health and social difficulties of early and late-diagnosed autistic children across childhood and adolescence (William Mandy, UCL)
Mental health and social difficulties are common in autistic young people, but their development may vary depending on when autism is diagnosed. Understanding the experiences of young people diagnosed later in childhood is important, as this group may have unmet needs prior to diagnosis. Using data from the Millennium Cohort Study, we examined trajectories of emotional, behavioural, and social difficulties from early childhood to adolescence in autistic youth diagnosed earlier (?7 years) and later (8–14 years) in childhood. Difficulties were measured at ages 3, 5, 7, 11 and 14 using the Strengths and Difficulties Questionnaire, and growth curve models were used to compare levels and change over time between groups. Earlier-diagnosed children showed higher difficulties in early childhood; however, late-diagnosed children also had elevated difficulties and showed steeper increases over time. By adolescence, late-diagnosed youth had higher levels of emotional, behavioural, and social problems. These differences remained after adjusting for sex, IQ, ADHD diagnosis, and maternal characteristics. Findings suggest that many autistic young people experience significant mental health difficulties before receiving a diagnosis, highlighting the importance of earlier recognition and support.
A participatory qualitative approach to co-designing research questions on factors influencing mental health in autistic youth using longitudinal cohort data (Sarah Hampton, UCL)
Understanding mental health in autistic youth using longitudinal cohort data requires research questions that reflect the priorities of autistic people and their families. Qualitative and participatory methods offer a way to achieve this, but there is limited guidance on how to use lived experience to design quantitative cohort research. We describe how co-design can be used to generate research questions suitable for large population cohorts such as the Millennium Cohort Study. Qualitative findings from interviews with autistic young people and parents were shared with autistic stakeholders, who voted on priority topics. In structured workshops, stakeholders and researchers then translated these priorities into testable questions that could be examined using longitudinal cohort data. The resulting research questions differed from those originally proposed by researchers and were more aligned with community priorities. Stakeholders reported that involvement in shaping the research felt meaningful and empowering. This work provides a practical model for incorporating co-production into cohort-based autism research, helping ensure that analyses of large longitudinal datasets address questions that are both scientifically robust and relevant to the autistic community.
Bidirectional Associations of Internalising Problems with Friendship Quantity and Satisfaction: Differences by Autistic Social Communication Traits (Reesha Zahir, UCL)
Friendships play an important role in mental health during childhood and adolescence, but it is unclear whether the quantity or satisfaction with friendships matters most for autistic young people. We examined the relationship between friendship experiences and internalising problems according to autistic trait level, using data from the Avon Longitudinal Study of Parents and Children. Participants were grouped into high and low trait groups based on Social Communication Disorders Checklist scores. Random intercept cross-lagged panel models were used to examine bidirectional associations of self-reported friendship satisfaction and friendship quantity with parent-reported internalising problems. In the low trait group, both friendship measures predicted later reductions in internalising problems, and internalising symptoms predicted declines in both friendship measures. In the high trait group, only friendship satisfaction showed a bidirectional association with internalising problems, while friendship quantity was not related to mental health. These findings suggest that having fewer friends may not be inherently distressing for individuals with an autistic social communication profile, instead, subjective experiences of friendships matter more for mental health.
Autism and Problematic Substance Use in Adolescence: Evidence from the Millennium Cohort Study (Adam James Shea, UCL)
Although autism is associated with elevated rates of mental health conditions, evidence on its relationship with substance use in adolescence is limited, particularly in population-representative samples. This study examined problematic substance use in autistic adolescents at ages 14 and 17. We used data from the Millennium Cohort Study. Problematic substance use outcomes were generated using items from MCS sweeps 6 and 7. Logistic and proportional odds regression models were applied, adjusting for key confounders. Differences in consistency of problematic substance use (between sweep 6 and 7) in autistic vs non-autistic adolescents was determined using Fisher’s Z test. Autistic adolescents showed lower odds of overall problematic substance use at both ages, as well as lower odds of problematic alcohol use. There was no evidence of differences for smoking, cannabis, or other substances, and no differences in consistency of use over time. In contrast to findings in adult samples, our findings suggest that autistic adolescents may be at a reduced risk of problematic substance use, highlighting the importance of longitudinal, life course perspectives.
