Abstract
Mok, P. L. H., Pickles, A., Durkin, K., & Conti-Ramsden, G. (2014). Longitudinal trajectories of peer relations in children with specific language impairment. Journal of Child Psychology and Psychiatry, 55, 516–527
Specific language impairment (SLI) is not so specific. SLI significantly affects academic attainments and vocational options (Conti-Ramsden, Durkin, Simkin, & Knox, 2009). SLI puts students at risk not only for academic difficulties but also for social-emotional difficulties. Difficulties in expression and/or comprehension of language make participation in conversation challenging. As a consequence, children with SLI engage less in active interactions than do those with typical language, enter less frequently into positive social interactions, are less sensitive to the initiations offered by others, and manifest situationally inappropriate verbal responses (e.g., Fujiki, Brinton, Isaacson, & Summers, 2001; McCormack, Harrison, McLeod, & McAllister, 2011). Experiencing difficulties with peer relations during childhood means that many children with SLI enter adolescence less equipped in the skills needed for this area of life.
Numerous studies have shown the link between early SLI and literacy skills. Consequently, speech-language pathologists (SLP) incorporate reading and writing activities in their language intervention. Less attention has been given to the social-emotional implications of SLI and the roles that SLPs might play in this area. In a recent study, Conti-Ramsden and colleagues investigated the developmental trajectories of behavioral, emotional, and social difficulties in individuals with a history of SLI from childhood to adolescence (St. Clair, Pickles, Durkin, & Conti-Ramsden, 2011). They observed an increase in peer problems and in the proportion of individuals functioning in the impaired range from childhood to adolescence. By 16 years of age, against national norms, nearly 40% of adolescents with SLI appeared impaired in their interactions with peers.
SLI is a very heterogeneous disorder. In general, children with SLI are sociable (Wadman, Durkin, & Conti-Ramsden, 2008), but there is evidence that they are more likely to be withdrawn (Brinton & Fujiki, 2002) and less likely to exhibit skilled social behaviors (Fujiki, Brinton, Morgan, & Hart, 1999). The authors of this study sought to identify the developmental patterns that lead to specific outcomes. They examined factors that have the potential to influence developmental trajectories of children with SLI, including expressive, receptive, and expressive language skills, as well as emotional symptoms, conduct problems, hyperactivity, and prosocial behavior.
Participants
Participants had a history of SLI and were originally part of a wider study: the Manchester Language Study (Conti-Ramsden & Botting, 1999; Conti-Ramsden, Crutchley, & Botting, 1997).
The initial cohort of 242 children (6;6 to 7;9 years) was a random sample of 50% of all 7-year-olds attending 118 language units across England. Children were excluded from the study if they were reported by their teachers as having frank neurological difficulties, hearing impairment, a diagnosis of autism, or a general learning disability. Participants were contacted again at ages 8, 11, 14, and 16. The average standard scores for receptive language at both ages and for expressive language at age 7 years were around 1 SD below the population mean, while average expressive language score at age 11 years was close to 2 SD below. Participants were classified into four groups based on their concurrent language and Performance Intelligence Quotient (PIQ) scores:
SLI-PIQ standard score 85 or above (i.e., in the normal range) and concurrent receptive or expressive language standard scores below 85.
Non-specific language impaired–PIQ standard scores below 85 and receptive or expressive language standard scores below 85.
Low cognition, resolved language–PIQ standard scores below 85 and receptive and expressive language standard scores 85 or above.
Resolved language–PIQ, receptive and expressive language standard scores all 85 or above.
The majority, 70.5%, of the children had impaired language ability at age 7. There was little change in the language status at age 8. At ages 11 and 16, the percentages of children with SLI fell to 39.6% and 36.1%, respectively, while the percentages with non-specific language impairment rose to 43.8% and 48.9%, respectively. The changing SLI profiles of some of the participants were thus due to their PIQ scores having fallen since they were recruited to the study. There is evidence suggesting that children with low PIQ and language skills perform in important ways much like children with a history of SLI who have PIQ within the normal range (Leonard, 2003). At ages 11 and 16, around 85% of the children showed language ability in the impaired range. None of the children in the original cohort had a diagnosis of autism.
Measures
Peer Relations
Two measures were used to assess peer relations: the peer problem subscale of the teacher-report version of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997), and teachers’ responses to three items in the Rutter Children’s Behaviour Questionnaire (Rutter, 1967). The peer problem subscale of the SDQ was administered at ages 11 and 16. Items of the peer problem subscale include Rather solitary, tends to play alone, Has at least one good friend, Generally liked by other children, Picked on or bullied by other children, and Gets on better with adults than with other children. The Rutter Children’s Behaviour Questionnaire was completed by the participants’ teachers at ages 7, 8, and 11 years. This consists of 26 statements, and the teacher is asked to score each item as doesn’t apply, applies somewhat, or certainly applies. Scores derived from the Rutter questionnaire and from the SDQ have been found to be highly correlated and have equivalent predictive validity (Goodman, 1997).
Prosocial Behaviour and Behavioral Difficulties
The SDQ behavioral measures (teacher-reported version; Goodman, 1997): prosocial behavior, emotional symptoms, conduct problems, and hyperactivity were obtained at age 11 years. The prosocial subscale consists of five positive items: considerate of other people’s feelings, shares readily with other children, helpful if someone is hurt, upset or feeling ill, kind to younger children, and often volunteers to help others. Examples of items constituting three other behavioral difficulty subscales include: Many worries, often seems worried (emotional symptoms); Often has temper tantrums or hot tempers (conduct problems); Constantly fidgeting or squirming (hyperactivity). Thresholds for identifying normal, borderline, and abnormal behavior are available for the subscales.
Performance IQ (PIQ)
Raven’s Coloured Progressive Matrices was used to assess participants’ PIQ at age 7 (Raven, 1986). At age 11, Block Design and Picture Completion of the U.K. version of the Wechsler Intelligence Scale for Children–Third Edition (Wechsler, 1992) was administered.
Receptive and Expressive Language
At ages 7and 11, receptive language was assessed using the Test for Reception of Grammar (Bishop, 1982). Expressive language at age 7 was assessed using the Bus Story Test (Renfrew, 1991), and at age 11, it was measured by the Recalling Sentences subtest of the Clinical Evaluation of Language Fundamentals–Revised (Semel, Wiig, & Secord, 1987).
Reading Accuracy and Comprehension
The Word Reading subtest of the British Abilities Scale (Elliot, 1983) was used to assess reading accuracy at age 7. At age 11, the Basic Reading and the Reading Comprehension subtests of the Wechsler Objective Reading Dimensions (Wechsler, 1993) were used to measure reading accuracy and reading comprehension, respectively.
Pragmatic Language
Of the 171 children in this study, 54 (31.6%) were considered to have semantic/pragmatic difficulties at age 7. Pragmatic language skills were formally assessed at age 11 using the Children’s Communication Checklist (CCC; Bishop, 1998). The checklist has been shown to differentiate between children with pragmatic language impairment and those with more typical SLI.
Results
Peer problems increased significantly between 7 and 11, and continued to increase beyond 11, but not significantly.
A four-group model best fit the data: low-level/no problems in peer relations (22.2%), childhood-limited problems (12.3%), childhood-onset persistent problems (39.2%), and adolescent-onset problems (26.3%).
The group with low/no problems exhibited few problems from childhood to adolescence. For the group with childhood-limited problems, by age 16, the SDQ mean peer problem score had fallen, and none of these children were classified as having borderline or abnormal levels of difficulties. The childhood-onset persistent problems group showed an increase in the Rutter mean scores between ages 7 and 11. Overall, 74.1% of children in this group at age 11 and 62.2% at age 16 were classified as having borderline or abnormal levels of peer problems. For the group with adolescent-onset problems, the percentages scoring above the SDQ threshold for borderline or abnormal levels of peer problems increased from 10% to 40.6% during this period.
Five predictors for group membership were measures of pragmatic language skills at age 11, prosocial behavior, hyperactivity levels, and levels of emotional and conduct disorders.
For children with persistent peer problems, the odds of being rated by their teachers as having pragmatic language difficulties was 2.5 times higher than for children with low/no problems.
Discussion
Somewhat surprising was the finding that there was not a significant relationship between expressive and receptive language abilities and the different trajectories. The authors suggest that this was because the language assessments tapped only some structural language skills and not higher order skills such as making inferences and organizing discourse or texts. This finding could be viewed as a caution for SLPs. The majority of popular assessments for SLI focus on structural aspects of language development. As students get older, they may master these skills, but continue to have difficulty with the language demands in social and academic settings. Scores on these formal tests may be used, inappropriately, to justify discharging them from services. The findings offer positive messages regarding peer relations without problems for some children and of amelioration of difficulties across time in others. The groups with low/no problems or with childhood-limited difficulties comprised approximately one third of the sample. In contrast, children who had persistent, or adolescent-onset, difficulties in peer relations from childhood to adolescence represented two thirds of children with a history of SLI.
Differences in peer relations were observed from early childhood (7 years) and were evident in adolescence, with trajectories diverging more widely as children entered middle childhood.
Based on these findings, the authors suggest that the identification of pragmatic difficulties and emotional problems could be critical to the amelioration of potential difficulties in peer relations. The use of teacher report of pragmatic difficulties was informative at age 7 as was the CCC and the SDQ emotional scale in middle childhood, at age 11. Results of randomized controlled trials show that pragmatic difficulties (Adams et al., 2012) and emotional problems (Sanders, Baker, & Turner, 2012) can be addressed effectively, in at least some children. The authors conclude that clinically, it is advisable not only to identify deficits but to evaluate the extent to which potential protective and/or positive factors are present. The SDQ prosocial scale in middle childhood was found to be a particularly strong predictor of differences in the trajectories of peer relations of children with a history of SLI.
Just as SLPs now typically consider literacy skills when working with students with SLI, they should also consider the early signs of social difficulties and incorporate goals in these areas on students’ individualized educational plans.