Commentary: Increased risk of later emotional and behavioural problems in children with SLI – reflections on Yew and O’Kearney (2013)


Specific language impairment (SLI) is a common disorder. One child in every primary school reception class has SLI. Children with SLI have difficulties with language, i.e. using and producing words to communicate and/or understanding what is said to them, while ‘everything else’ appears to be normal. This ‘everything else’ has traditionally been defined to include adequate input from the senses: normal hearing and normal/corrected vision. It also includes an adequate biological basis to develop language (they have no obvious signs of brain damage) and an adequate basis for learning, i.e. their nonverbal abilities as measured by IQ are similar to those of their peers of the same age. Children with SLI are sociable: they are interested in interacting with others. Because affected children look ‘normal’, SLI is a hidden disability.

Children with SLI find it effortful to learn to talk and these difficulties can be persistent. Given the importance of language to human behaviour, it is not surprising to find that language difficulties are a risk factor for associated difficulties in other aspects of children’s development. For example, research indicates developmental interactions between language impairments and difficulties acquiring literacy skills, making educational progress, the development of memory skills and more general nonverbal abilities throughout middle childhood, adolescence and beyond (Conti-Ramsden, St. Clair, Pickles, & Durkin, 2012). Although children growing up with SLI have the desire to engage socially, they experience greater difficulties in peer relations and social interaction than do typical children and adolescents (Durkin & Conti-Ramsden, 2007). There is also strong evidence that SLI and emotional and behavioural disorders (EBD) commonly co-occur, with estimates of 50–80% comorbidity of children identified with either SLI or an EBD. Thus, children with SLI appear to be more vulnerable to EBD (Toppelberg & Shapiro, 2000). But how strong is the link between childhood SLI and later emotional and behavioural difficulties?

Most of the evidence we have on the relationship between SLI and EBD is cross-sectional in nature. Although informative, such studies are less suited to address developmental relationships. What is needed are prospective studies. However, there are few longitudinal studies of children with SLI. Such investigations are expensive, both in terms of time and cost. Thus, Yew and O’Kearney’s (2013) meta-analyses of 19 controlled prospective studies of SLI and later EBD provide a very useful systematic review of the evidence available: what we can and importantly, what we cannot conclude from the findings to date.

First, it appears that there is not enough evidence to infer a strong link between children with SLI and specific EBDs. Children with SLI do not appear to show a higher risk of specific emotional or behavioural disorders, with one exception, ADHD. Children with SLI are over one and half times more likely to meet criteria for ADHD in later childhood and adolescence than their typically developing peers. In terms of specific disorders associated with SLI, the knowledge available so far indicates that ADHD is one to clinically look out for. However, it needs to be noted that currently there are a limited number of studies examining outcomes with regard to specific mental disorders other than ADHD (e.g. anxiety disorders, depression). This makes it difficult to pool findings together and reach firm conclusions about the relationship between SLI and other later specific EBDs. This is an area were future research is warranted. Second, the evidence does appear to suggest an association between childhood SLI and later more general emotional and behavioural symptoms. In other words, in later childhood and adolescence, individuals with SLI may exhibit some symptoms of mental health difficulties, but it is unlikely that these will be sufficient to meet clinical criteria for specific psychiatric disorders such as depression or anxiety. Thus, individuals with SLI, in general, are not exhibiting difficulties which alert professionals to the need for referral. This may mean that, in practice at least, some individuals with a history of childhood SLI are living with difficulties for which they are not receiving support. This fact needs to be taken into consideration in developing policies in relation to child and adolescent mental health service provision, as well as in clinical practice where assessment of current SLI or a history of language impairment should be included. The magnitude of the risk warrants this suggestion; it is double what would be expected from the general population. Children with SLI are approximately two times more likely to experience emotional difficulties and two times more likely to experience behavioural difficulties in later childhood and adolescence than their typically developing peers.

The third conclusion relates to affected individuals (individuals with SLI who are later exhibiting EBD problems) and applies to both the specific disorder of ADHD as well as to the more general emotional and behavioural symptoms. The evidence suggests mild-to-moderate levels of severity of the aforementioned difficulties for affected individuals with SLI in later childhood and adolescence. Importantly though, findings also show that children with SLI are at a higher risk of having a disorder/impairment level of these general emotional and behavioural problems as well as ADHD. Therefore, on average, the problems experienced by individuals are not likely to be acute, but this does not mean they do not require clinical support. This should be based on informed intervention which takes into consideration the fact that the very instrument of therapeutic input, i.e. language, is likely to be a source of difficulty for these individuals. There is evidence that compared with other neurodevelopmental disorders, knowledge of SLI remains limited. An important message of the meta-analyses is that children with SLI are as deserving of professional attention as others, and this is likely to involve support in later childhood and adolescence (see Raising Awareness of Language Learning Impairments,

Who is likely to be more at risk? The meta-analyses by Yew and O’Kearney make a unique contribution to our understanding of the relationship between SLI and later EBDs. They underlie the fact that there is a dearth of information in this area and that the evidence we have is preliminary in nature and limited to two potential mediators: type of language difficulty and gender. There is preliminary evidence of increased risk of EBDs for children with SLI who have a deficit profile that includes comprehension difficulties. This is therefore an important area for future research. It also informs methodology. It suggests the type of participant information that would be useful to include in future research and clinical studies. That is, providing details of participants’ profiles of difficulties which specify what language modalities are affected. Interestingly, comprehension difficulties in what is known as mixed SLI (whereby both expressive and receptive modalities are affected) has been found to be prognostic of outcomes in other areas of development, for example, quality of friendships in adolescence (Durkin & Conti-Ramsden, 2007). In terms of gender, there appear to be no gender effects in terms of general emotional symptoms in later childhood and adolescence. Both boys and girls with childhood SLI appear to have the increased risk. The rest of the evidence points to potentially interesting areas for future research: for example, are boys with childhood SLI more likely to have later externalizing problems such as conduct difficulties? Part of the difficulty with the study of gender effects in SLI is the higher ratio of males to females that are referred to services (a ratio of 3:1). This reduces the power to identify gender differences given that in most clinical samples of individuals with SLI, there are a smaller number of females. To the above variables, one may add severity of the language impairments as a factor that may influence the level of risk for EBDs. Finally, it is important to note that a number of children with SLI do not experience later EBD symptoms (Snowling, Bishop, Stothard, Chipchase, & Kaplan, 2006). There is heterogeneity of outcomes in SLI. The investigation of pathways to resilience and the role of protective factors are likely to provide important insights into the nature of interactive developmental associations which ameliorate or exacerbate risk of later EBDs in individuals with childhood SLI.


This commentary article, invited by the Editors of JCPP, has not been externally peer reviewed, but has been through a process of editorial review. The author has declared that she has no competing or potential conflict of interests.


Gina Conti-Ramsden, School of Psychological Sciences, The University of Manchester, Ellen Wilkinson Building, Oxford Road, Manchester M13 9PL, UK; Email: