Developmental, cognitive and behavioural assessment tools — branching viva

Branching viva — developmental, cognitive and behavioural assessment tools

The examiner releases the stem and then branches into five probes. A strong candidate names the three acts first, defends the schedule and the choice of tool, interprets the M-CHAT-R/F correctly, and escalates regression without prompting.

Opening (examiner)

"A 20-month-old is brought in because he has no words, does not point, and his mother believes he has lost the sounds he used to make. Before you reach for a tool, tell me how developmental assessment is structured." [1]

Branch 1 — The three acts (expected answer)

Developmental assessment is built on three different acts. Surveillance is the longitudinal, opinion-based monitoring performed at every health visit, drawing on history, examination, observation and parental concern. Screening is a brief, standardised tool applied to the whole population at scheduled ages to find the child who needs a closer look. Diagnostic assessment is the lengthy, trained-administered evaluation that confirms and quantifies a disorder. A screen can never do the diagnostic tool's job, and a positive screen is a reason to refer, never a diagnosis. [1]

Probe. "What is the screening schedule?" — Developmental surveillance at every visit, plus a standardised developmental screen at nine, 18 and 30 months and an autism screen (M-CHAT-R/F) at 18 and 24 months. [1]

Branch 2 — The right tool (expected answer)

For a broadband developmental concern, use a parent-completed screen such as the Ages and Stages Questionnaires (one to 66 months, five domains) or PEDS. For an autism concern, use the M-CHAT-R/F (16 to 30 months). For a diagnostic cognitive assessment, refer for the Bayley Scales up to 42 months, or the Griffiths or Wechsler scales for the older child. The parent-completed screens are fast enough for a population schedule; the diagnostic tools trade reach for the precision a diagnosis needs. [3] [9]

Probe. "Why is Bayley the reference standard and not the screen?" — Because a diagnostic tool minimises both false positives and false negatives through standardised administration, basal and ceiling rules, and age-stratified norms; a 2024 meta-analysis used Bayley as the reference standard against which the parent-completed screens were compared. [9]

Branch 3 — The M-CHAT-R/F result (expected answer)

The M-CHAT-R is a 20-item parent report. A child screens positive if they fail three or more of the twenty items, or two or more critical items; the Follow-Up Interview then confirms the risk and lowers the false-positive rate. This child's social-communication concerns meet that threshold, but the regression means the result should drive an urgent referral, not a wait for the next routine visit. [5]

Probe. "The candidate says a positive screen means the child has autism — why is that wrong?" — Because in a low-prevalence population the positive predictive value is modest; a positive screen is a reason to refer for diagnostic assessment, not a diagnosis. [10]

Branch 4 — Regression (expected answer)

The history of lost sounds, no pointing and no eye contact is regression, and regression is never a normal variant. A child who loses any acquired language, social or motor skill needs urgent referral the same week for diagnostic developmental and autism assessment plus audiology. Watchful waiting or a repeat screen at the next visit is the avoidable harm. [5]

Probe. "What accompanies the referral?" — Hearing and vision must be checked, because unrecognised sensory impairment is a reversible mimic of language and social delay; cognitive assessment comes after audiology, not before. Early intervention begins on need, before the diagnosis is confirmed. [9]

Branch 5 — The standard-score model (expected answer)

A standard cognitive score has a mean of 100 and a standard deviation of 15, so a score of 70 or below sits two standard deviations below the mean and marks significant concern. The developmental quotient equals developmental age divided by chronological age times 100, and a value below 70 flags development below two standard deviations. The diagnostic tool reports the standard score, the confidence interval and the clinical interpretation. [9]

Probe. "Why does real-world screening accuracy fall below the validation figures?" — A 2023 meta-analysis of the real-world accuracy of developmental screening tests found pooled sensitivity and specificity varied widely by tool and setting and were lower than developers' figures, which is exactly why a screen is layered onto surveillance and followed by diagnostic confirmation. [10]

Examiner's wrap

Hold the three-acts principle: surveillance at every visit, scheduled screening, and diagnostic confirmation of every positive screen. Reserve the first M-CHAT-R/F threshold, treat a positive screen as a referral and never a label, and escalate regression the same week with audiology in the work-up. The broadband screens (ASQ, PEDS) reach the population; the gold-standard tools (Bayley, Griffiths, Wechsler) confirm and quantify; the behavioural scales (SDQ, Vanderbilt) localise the concern. [7]