Predicting Uncertain Multi-Dimensional Adulthood Outcomes From Childhood and Adolescent Data in People Referred to Autism Services

Abstract

IntroductionAutism spectrum disorder is a highly heterogeneous diagnosis. When a child is referred to autism services or receives a diagnosis of autism spectrum disorder it is not known what their potential adult outcomes could be. We consider the challenge of making predictions of an individual child’s long-term multi-facetted adult outcome, focussing on which aspects are predictable and which are not.MethodsWe used data from 123 adults participating in the Autism Early Diagnosis Cohort. Participants were recruited from age 2 and followed up repeatedly through childhood and adolescence to adulthood. We predicted 14 adult outcome measures including cognitive, behavioral and well-being measures. Continuous outcomes were modeled using lasso regression and ordinal outcomes were modeled using proportional odds regression. Optimism corrected predictive performance was calculated using cross-validation or bootstrap. We also illustrated the prediction of an overall composite formed by weighting outcome measures by priorities elicited from parents.ResultsWe found good predictive performance from age 9 for verbal and non-verbal IQ, and daily living skills. Predictions for symptom severity, hyperactivity and irritability improved with inclusion of behavioral data collected in adolescence but remained modest. For other outcomes covering well-being, depression, and positive and negative affect we found no ability to predict adult outcomes at any age. Predictions of composites based on parental priorities differed in magnitude and precision depending on which parts of the adult outcome were given more weight.ConclusionVerbal and non-verbal IQ, and daily living skills can be predicted well from assessments made in childhood. For other adult outcomes, it is challenging to make meaningful predictions from assessments made in childhood and adolescence using the measures employed in this study. Future work should replicate and validate the present findings in different samples, investigate whether the availability of different measures in childhood and adolescence can improve predictions, and consider systematic differences in priorities.

How do autistic people fare in adult life and can we predict it from childhood?

“…Overall prediction of an individual's mental health in adulthood was very poor, consistent with findings relating to prediction from at least age 12 in the US Early Diagnosis Cohort (Forbes et al, 2021 ). The findings of weak correlations between childhood emotional problems and adult depression and anxiety (Pearson's correlation 0.21 and 0.17, respectively) and poor predictive performance for these outcomes contrast with the moderate correlation (correlation = 0.43) found between childhood and adult emotional problems domain of the SDQ in a previous analysis of this cohort (Stringer et al, 2020 ).…”

“…Very good predictions of a young adults living situation and modest predictions of employment and education from late childhood have been reported previously in a cohort conducted in US Early Diagnosis Cohort (Forbes et al, 2021 ). While in this study we only consider predictions form late childhood, findings from the US early diagnosis cohort indicate that predicting adult outcomes from earlier than this is much more challenging.…”

Ethical dimensions of translational developmental neuroscience research in autism

“…Lloyd-Fox et al, 2018), suggesting that from its earliest emergence, autism is associated with subtle differences across multiple domains of brain development. Initial machine-learning approaches to predicting autism from infant data highlight widespread alterations in cortical thickness (Hazlett et al, 2017), functional brain activity across a range of scalp regions and frequency bands (Gabard-Durnam et al, 2019), and alterations to both social and non-social stimuluslocked processing (Tye et al, 2020). This evidence makes it implausible that a single brain 'deficit' solely responsible for the development of autism will emerge.…”

“…It is important to note that uncertainty is not confined to our capacity to predict whether an infant will later develop autism or not; it also extends to our capacity to predict autistic people’s future outcomes from childhood data. While we have stronger confidence in early predictors such as language development, prediction of global outcomes increases through to age 9 years ( Pickles, McCauley, Pepa, Huerta, & Lord, 2020 ); and for outcomes such as mental health, prediction is only possible with data collected in adolescence, or it is not possible at all ( Forbes, Lord, Elias, & Pickles, 2021 ). Acknowledging that very early development influences , rather than determines , autistic individuals’ future outcomes is fundamental to oppose certain assumptions that may (and do) harm children and their parents.…”

Looking through rainbow-rimmed glasses: Taking neurodiversity perspective is related to subjective well-being of autistic adults

“…At the same time, typical objective indicators of well-being like having a job, living independently or graduating high school, do not predict the mental health and well-being of autistic adults (Forbes et al, 2021). As being on the autism spectrum brings particular challenges and joys, it is possible that subjective factors, such as the perception of one’s own life’s challenges or attitude towards one’s own diagnosis, may be an area for a positive change.…”