A 20-month-old who does not point or respond to his name
A 20-month-old boy is brought for the second time because he still does not talk. He uses two single words inconsistently, does not point to show something, and does not follow a one-step command unless it is accompanied by a gesture. He does bring toys to his mother to share, waves bye-bye, and walked at 15 months. At 12 months he had glue ear treated with a 6-week course of antibiotics; tympanometry at 14 months was type B bilaterally and no repeat was arranged. His mother reports that he covers his ears when the blender is on and has stopped responding to his name over the past month. There has been no loss of motor or language skills, no seizures, and no dysmorphism. The family recently moved, and the previous doctor wrote 'late talker, review in 6 months'. The parents ask whether stopping tablet use will fix it, and whether the MMR given at 15 months caused it. They show a 12-minute phone video in which he lines up cars and does not look up when called.
1. Give your differential diagnosis in order of what must be excluded first, and justify the order using the findings in the scenario.
Read the worked answer
1. Hearing impairment (conductive, possibly persistent) - this comes first because it is common, treatable, and confounds every other diagnosis. He had bilateral type B tympanometry at 14 months, which means effusion at that time, and no repeat was done; six months of unrecognised conductive loss is more than enough to explain absent words and inconsistent response to name. The ear-covering behaviour could be recruitment or sensory sensitivity. Action: repeat otoscopy and tympanometry now, and request an auditory brainstem response if there is any doubt or if effusion persists.
2. Autism spectrum disorder - the red flags are the absence of pointing to show at 20 months, failure to follow a one-step command without gesture, loss of response to name, and lining up objects without looking up on the video. These are social-communication deficits, not just speech delay. The presence of shared enjoyment (bringing toys to mother, waving) does not exclude ASD, because joint attention (pointing, showing, following a point) is the more discriminating domain.
3. Global developmental delay or receptive-expressive language disorder - less likely with normal gross motor milestones, but needs a developmental quotient assessment rather than a guess.
4. Regression or an epileptic language disorder (for example Landau-Kleffner) - only if there is documented loss of previously acquired words; at present the history is of slow acquisition, not regression, and there are no seizures.
5. Environmental factors - bilingual home, limited stimulation, or excessive screen time can slow vocabulary but do not explain absent pointing or name response; they should not be used as the primary explanation.
Priority is therefore: hearing test this week, formal developmental assessment, and simultaneous referral rather than sequential waiting.
2. How do you interpret the tympanometry at 14 months, and what does a normal screening questionnaire at 18 months add or not add?
Read the worked answer
Tympanometry at 14 months: type B is a flat trace, indicating no measurable middle-ear compliance - in practice effusion, a perforation with discharge, or a blocked probe. It is not a hearing test; it says nothing about the sensorineural pathway and nothing about the severity of any conductive loss. Because it was abnormal bilaterally and never repeated, the correct interpretation is 'untreated possible otitis media with effusion for at least six months', which is a plausible contributor but does not exclude a co-existing sensorineural loss or ASD. A normal otoacoustic emission screen at birth does not help either, because it does not exclude later conductive loss or auditory neuropathy.
Screening questionnaire: a tool such as the M-CHAT-R completed at 18-24 months is a screen, not a diagnosis. A negative screen has limited value in a child with clear clinical red flags, because parents with low awareness of social milestones may under-report, and because the instrument is designed for population use, not for ruling out autism in a referred child. A positive screen does not diagnose autism either; it triggers a diagnostic assessment (structured observation plus developmental testing).
Therefore: in this child the video evidence and the absent pointing outweigh any questionnaire result. Use the questionnaire to structure the history and to prepare the family for assessment, but do not use a normal score to justify waiting another 6 months, and do not use it to avoid the hearing re-test.
3. Counsel the parents about the MMR question and the tablet question. What is evidence-based, and what would you say about the 'wait 6 months' plan?
Read the worked answer
MMR: there is no credible evidence that MMR causes autism. The 1998 paper that started the scare was retracted by the journal, its lead author lost his licence, and subsequent large cohort and case-control studies in several countries found no association, including no association in children who already have developmental concerns. Measles itself can cause pneumonia, encephalitis and death; a 20-month-old who has had one dose is due a second dose according to the national schedule. Say this plainly and document the conversation, and check the vaccination record rather than accepting 'he had his injections'.
Tablet use: the evidence is observational. Screen time is associated with shorter vocabulary and less parent-child interaction, especially if it displaces conversation and play, but it is not a demonstrated cause of autism and removing it is not a treatment. Give practical advice: no screens during meals and the hour before bed, co-viewing rather than solo use, and replace screen time with shared book-reading and play. Set a review date to see whether language changes - but do not let this be the only intervention.
'Review in 6 months': decline this plan, with a reason the parents can understand - a 20-month-old should have at least 50 words and two-word combinations by 24 months, and absent pointing at 20 months is a red flag regardless of word count. Waiting consumes the period of maximal brain plasticity, and the assessment process itself takes weeks. The referral for audiology and developmental assessment should go out today, with the general practitioner copied in, and the family should be given written information on what happens next so the move between services does not lose them.