A digital parent-completed developmental check was more sensitive than usual GP assessment

Design

Diagnostic accuracy sub-study within a cluster randomised trial

Population

655 children aged 16–35 months in 40 Australian general practices (151 with reference assessment)

Primary outcome

Sensitivity and specificity against Mullen, ADOS and Vineland

Effect

Sensitivity 93.9% vs 75.9% (p = 0.035); specificity 71.4% vs 66.7%

Within a cluster randomised trial in 40 Australian general practices, parents of children aged 16–35 months either completed the Watch Me Grow Integrated (WMG-I) digital screener or received usual GP assessment. WMG-I asks a short primary screen and, if concerns arise, an automated secondary screen; results go to both parent and GP, and reminders are sent at later ages.

All children flagged, plus a 10% sample without concerns, had a reference assessment including the Mullen Scales, ADOS and Vineland. Of 655 children, 151 were assessed. Sensitivity was 93.9% (95% CI 88.1–99.7%) with WMG-I versus 75.9% (57.2–94.0%) with usual care; specificity was similar (71.4% vs 66.7%).

Developmental delay and autism are often missed in short general practice consultations, and parent report is the best early signal. A structured, parent-completed tool linked to an existing contact such as an immunisation visit is a practical way to catch more children. The specific tool is Australian, but parent-completed screeners in local languages can serve the same purpose.

• Use a structured parent-completed developmental screener at toddler visits rather than relying on impression alone.
• Tie screening to immunisation visits, when the child is already in the practice.
• Treat a positive screen as a reason for full assessment, not a diagnosis — specificity was about 70%.
• Ask directly about speech, social communication and play; parent concern is the strongest early signal.
• Refer early — waiting to see if a child 'catches up' delays intervention.

Why it matters

It shows that a quarter of children with developmental needs were missed by usual primary care assessment.

Don't overread it

Only 151 children had the reference assessment and the confidence intervals are wide; the tool itself has not been validated outside Australia.

The statistics, in plain English

Sensitivity of 94% means about 6 in 100 children with a condition were missed, against about 24 in 100 with usual care. Specificity of around 70% means roughly 3 in 10 children without a condition were flagged and referred for assessment.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.