Article Text

Download PDFPDF
Parent training improves language development in children with, or at risk of, language impairment
  1. Kate Christina Harvey
  1. Department of Paediatrics, Royal Wolverhampton Hospitals NHS Trust, Wolverhampton, UK
  1. Correspondence to Dr Kate Christina Harvey, Department of Paediatrics, Royal Wolverhampton Hospitals NHS Trust, Wolverhampton WV10 0QP, UK; k.richardson{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Review of: Roberts MY, Curtis PR, Sone BJ, Hampton LH. Association of Parent Training With Child Language Development: A Systematic Review and Meta-analysis. JAMA Pediatr 2019;173(7):671–680

Study design: Systematic review and meta analysis. Four databases searched (ERIC, Academic Search Complete, PsycINFO and PsycARTICLES) for studies in which parents were taught strategies to support their child’s language and communication.

Number of studies included: 76

Setting: 57% North American studies, 26% European studies. 33% of studies were performed within low-income populations.

Patients: 5848 included children had a mean age of 3.5 years. 27 trials studied patients with autistic spectrum disorder (ASD), 10 patients with a developmental language disorder and 34 patients at risk of language impairment (eg, premature birth, siblings with ASD and low-income families).

Intervention: 57 (77%) studies were randomised and 17 (23%) were non-randomised. 63 (83%) studies taught naturalistic language strategies (ie, responding to child communication); 16 (21%) studies taught a dialogue-reading approach (ie, discussion during book reading). 49 (64%) studies taught parents through coaching; 17 (22%) studies taught through workshops; and 21 (28%) studies taught through therapist modelling. Interventions lasted a mean of 23 weeks (median 12 weeks, range 4–120 weeks); frequency of intervention ranged from 1 hour/month to 14 hours/week.

Outcomes: Data were extracted from studies using mean and SD. The Hedges effect size examined the difference in communication and language outcomes between control and test groups. An effect size of 0.2 is generally considered small, 0.5 medium and 0.8 large. Consideration of the context is necessary, as a small effect size may be clinically important in some settings.1

Main results

The results are summarised in table 1.

Table 1

Effect size (95% CI) across measure types and populations

Parent training was associated with improved communication and language in children with a moderate effect size; the most benefit was seen by children with developmental language disorders. Intervention characteristics (length, frequency and type) were not significantly associated with the effects seen.


Parental training in communication and language intervention techniques were associated with improved outcomes for children.


The traditional model of speech and language therapy was to deliver a ‘block’ of treatment to a child; this was done by a therapist with the parents as onlookers. Now, services are increasingly structured around training parents (and others who care for the child) to deliver therapy. This allows the family to integrate therapy into everyday life while also reducing the burden on speech and language therapists.

This paper seeks to amalgamate the evidence supporting or refuting the efficacy of parent training programmes. It is unconventional to perform a meta-analysis on such a heterogenous group of studies, with so many different outcomes. The precision of the reported effect sizes is therefore questionable. That aside, it is otherwise a well-designed systematic review, including mainly randomised trials published within the past 9 years. The most impressive effect sizes are seen for children with developmental language disorder, although there were fewer included studies looking at this patient group, and not all outcomes were reported on.

This paper supports the use of a parent-delivered model by giving evidence of its efficacy. Much of the work done by children’s therapy teams is lacking robust evidence; therefore, this paper is refreshing and encouraging. As paediatricians, this evidence may help us to reassure parents who are sceptical about the changes from a therapist-delivered to a parent-delivered model (with the often associated reduction in session numbers), although of course this paper did not directly address the question of new vs old.

The paper also does not give any guidance about the optimal duration and frequency of parent training sessions. In the included studies, there was a large variation in the intervention length and dose (hours/week), with no significant difference in outcomes based on these variables. Further studies in this area would guide practice and rationalise the commissioning of services.

Abstract and commentary by: Dr Kate Harvey, Royal Wolverhampton Hospitals NHS Trust, Wolverhampton, UK;


The author thanks Marty Richardson for her statistical support and Sarah Drayton for her expertise in Speech and Language Therapy service provision.



  • Contributors KH wrote the abstract and the commentary.

  • Provenance and peer review Not commissioned; internally peer reviewed.