Review highlights systemic gaps after autism screening, urging better pathways to diagnosis

A new review in Pediatric Investigation reveals that after initial autism screening, children often slip through the cracks due to referral and assessment delays, prompting calls for more cohesive diagnostic pathways.

A new review in Pediatric Investigation argues that the hardest part of identifying autism early is often not the screening itself, but what happens next. The study says children aged 0-5 can be missed at multiple points after a concern is raised, including referral, assessment, service capacity and access to support, leaving families stuck between an initial flag and a confirmed diagnosis.

Led by Ruslan Kurmashev of Munster Technological University in Cork, the review examined evidence from major biomedical databases and guidance resources, focusing on research published from 2010 to April 2026. It set out a pathway-based model that links developmental concern, screening, referral, diagnostic assessment and support, with the aim of showing where children and families most often fall through the cracks.

The authors found that screening can speed up diagnosis for some children, but it is not a substitute for ongoing developmental surveillance. In one study of 25,999 toddlers, a screening tool identified fewer than four in 10 children later diagnosed with autism. In another involving 36,233 toddlers, children who screened positive were diagnosed earlier, at about 38.5 months compared with 48.5 months for those who did not. The review says that shows screening can help some families move through the system sooner, even if it does not catch every child who may need assessment.

The bigger problem, the review suggests, is what happens after a result. One study found that only 31% of children who failed an autism screen were referred for specialist assessment. Another found that 40.2% of children who screened positive received at least one recommended referral, while only 3.7% received all of them. A negative result, the authors argue, should not be treated as the end of follow-up, because developmental differences can become clearer over time.

Kurmashev said screening should be seen as “an entry point to a responsive pathway, not as a diagnosis or a finish line”. He added that families need repeat observation, prompt next steps and access to assessment and support whether the first screening result is positive or negative.

The review also points to system pressures that can slow access even when concerns are recognised. In one UK survey, only 17.9% of autism assessment services always met national guidance. Referrals rose by 115% between 2015 and 2019, while 75.8% of services said funding had stayed the same or fallen. One study found families waited an average of 375 days for a diagnosis.

Access can also vary because of language, cost, geography, culture and the difficulty of navigating services. The review cites evidence that, in Australia, adding an autism screening tool to routine developmental checks did not significantly improve identification. It also draws on earlier research, including studies in Brazil and the US, showing that missed opportunities, uneven surveillance and limited provider confidence can all shape who gets identified and when.

The authors argue that the answer is not to abandon screening, but to build better pathways around it. They call for repeat surveillance after unclear or negative results, closed-loop referral tracking, family navigation, coordinated assessment and support based on need before diagnosis is formally completed. In their view, earlier, more coherent systems could help families get help sooner while diagnostic uncertainty is still being resolved.