Autism and Early Development

We know that the first signs of autism appear during early childhood. Yet autistic people often have to wait for years for a diagnosis. One reason is that previous studies often focused on single likelihood factors for autism, such as family history or pregnancy complications, without integrating or translating their findings into clinical practice. This research programme will use new methods, such as machine learning, to combine clinical cohort data, such as family history, pregnancy complications and early neurodevelopment. The aim will be to develop a new clinical pathway, so that children can be prioritised for early assessment, enabling more timely access to effective developmental support.

This approach will be informed by recent discoveries and extensive community engagement with both autistic people and their families, as these discoveries are directly relevant to them. For example, new findings across the world have shown that if a mother has been diagnosed with polyendocrine metabolic ovarian syndrome (PMOS, previously known as PCOS), her children are substantially more likely to be autistic. These findings also suggest that autistic women themselves are at a higher likelihood for hormonal imbalance and may benefit from screening and early diagnosis to improve their quality of life.

In addition, while an autism diagnosis remains more common in boys, especially during childhood, we now know that a lot of the early signs and difficulties are missed in girls, who may be internalising their difficulties and anxiety. This is why the programme will combine data from studies that follow children as they grow (prospective cohorts) and studies that look back at people’s early lives (retrospective cohorts), so that the earliest signs can be identified even when a diagnosis has been given after childhood. To overcome reliance on external behaviours, the study will also include detailed measures of brain development and brain function, which have been challenging to measure in infants and young children, until now.

Finally, the programme will bridge early neurodevelopment with studies on pregnancy and the earliest environmental exposures, while the brain is developing in the womb. This will not be done with prenatal diagnosis in mind, as this remains scientifically impossible, ethically inappropriate and contrary to our research values. Instead, the study of pregnancy in the context of autism, will aim to inform early screening after birth, as well as understand the healthcare needs of autistic women themselves during and after their pregnancy.

This programme will have clinical benefits for autistic people and their families, as well as value for basic science and our understanding of neurodevelopment. It can provide new answers to key questions about how biology before and after birth contributes to different developmental pathways in autism, to early brain development in general and to the rich neurodiversity of human society.