Summary of findings 2. Parent‐ and clinician‐mediated intervention versus clinician‐mediated intervention alone for language development in young children with Down syndrome.

Parent‐ and clinician‐mediated intervention versus clinician‐mediated intervention alone for communication and language development in young children with Down syndrome
Patient or population: children with Down syndrome aged between birth and six years Setting: home, clinic, or both; interventions delivered through group or one‐to‐one sessions Intervention: parent‐ and clinician‐mediated intervention Comparison: clinician‐only‐mediated intervention
Outcomes	Impact	№ of participants (studies)	Certainty of the evidence (GRADE)
Expressive language (number of words; standard scores; MLUw; NDW; IPSyn; frequency of unique targets; % target talk) Assessed with: parent reports; direct assessment; language sample analysis; experimental task Follow‐up: range 1 day to 12 months	1 study found no child language differences between the groups based on parent‐report or norm‐referenced measures immediately, 6 months or 12 months postintervention. Similarly, there were no differences in child language measures based on trained experimental tasks, apart from the number of utterances with child language targets, which was ranked more than twice as high for the intervention group (P = 0.006) immediately postintervention and almost twice as high for the intervention group (P = 0.043) at 6 months postintervention,compared to the control group. This difference was not maintained 12 months postintervention. No differences were noted in the untrained activities at any time point postintervention.	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b,c
Receptive language (total language; standard scores) Assessed with: direct assessment Follow‐up: range 1 day to 12 months	1 study found no child language differences between the groups for any norm‐referenced measures for any time point postintervention	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b,c
Parental stress (total stress scores) Assessed with: Parenting Stress Index Follow‐up: range 1 day to 12 months	1 study did not find any differences between total parental stress scores at any time point postintervention	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b,c
Changes in parental behaviour/responsivity Assessed with: Milieu Teaching Project KidTalk Code Rating Scale (scored on a scale of 0‐100%; higher scores indicate increased use of target strategies as a % of potential episodes) Follow‐up: range 1 day to 12 months	1 study found that parents in the intervention group were ranked, on average, twice as high on a measure of 'responsive interaction' immediately (P = 0.006, P = 0.005), 6 months (P = 0.006, P = 0.002) and 12 months (P = 0.001, P = 0.030) postintervention in trained and untrained activities, respectively. They were also ranked, on average, twice as high on the number of 'expansions' used at all time points on trained and untrained activities postintervention compared to the control group. With the exception of 12 months postintervention in trained activities, parents had a higher ranking on 'percentage of language modelling' at all time points in trained and untrained activities. The intervention group were ranked almost twice as high on their use of 'milieu teaching prompts' immediately postintervention in untrained activities (P = 0.021) and 6 months postintervention in trained (P = 0.020) and untrained (P = 0.005) activities compared to the control group. This was not maintained 12 months postintervention in trained or untrained activities.	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b,c
Socialisation	Not measured
Language attrition Assessed with: parent reports; direct assessment; language sample; experimental task Follow‐up: range 1 day to 12 months	1 study did not find evidence of language attrition in the intervention or control group postintervention	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b,c
Adherence to treatment	Not measured
CI: confidence interval; IPSyn: Index of Productivity Syntax; MLUw: mean length of utterance in words; NDW: number of different words; RCT: randomised controlled trial; TNW: total number of words
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded by one level as we judged the risk of bias to be high or unclear for some factors.
 ^bDowngraded by one level as it is not possible to measure inconsistency from a single study.
 ^cDowngraded by two levels as the sample size was small and the analysis involved a narrative synthesis and so estimates were not available.