Positive friendships significantly reduce the long-term negative association between ADHD symptoms and social problems in children (Becker SP et al, 2013). The aims of this study were to: (1) assess the fit of a two-factor model of positive and negative friendship quality, and (2) use a Multiple Indicators Multiple Causes (MIMIC) model to determine whether comorbid externalising and internalising disorders incrementally predict positive and negative friendship quality in children with ADHD.
Children with ADHD, diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5TM), a friend and both sets of parents were included in the study. Children taking ADHD medication and those with psychiatric comorbidities (e.g. anxiety disorders, conduct disorder, depressive disorders, learning disabilities or oppositional defiant disorder) could be enrolled. However, children with autism spectrum disorder, an IQ <75 or a severe condition (e.g. psychosis, suicidality or trauma) requiring immediate intervention were excluded from participating. Comorbid externalising* and/or internalising† disorders were determined for each child. Children with ADHD, their parents, their friend, and friend’s parents independently completed the Friendship Quality Questionnaire (FQQ) – Short Form to assess the friendship between the child with ADHD and their friend.‡
In total, 165 children with ADHD (mean [standard deviation] age, 8.59 [1.51] years; 33% female) were included in the study. Most children had combined type ADHD (69%), 27% had inattentive type and 4% had hyperactive type. In terms of psychiatric comorbidities, 20.6% had an externalising disorder, 18.8% had an internalising disorder and 10.3% had both.
Generally, children with ADHD tended to agree with their own parents (r = 0.42) rather than with their friend (r = 0.26; p = 0.056) or friend’s parents’ (r = 0.23; p = 0.03) perceptions of negative friendship quality. The friends of children with ADHD also tended to agree more with their own parents (r = 0.57) perception of negative friendship quality, rather than with the children with ADHD (r = 0.26; p < 0.0001) or their parents (r = 0.26; p < 0.0001). These data suggest that the initial two-factor model of friendship quality with four residual correlations (child with ADHD, their parents, a friend, friend’s parents) to model method effects did not acceptably fit the data. This is because children with ADHD, their parents, their friend and their friend’s parents did not provide independent reports of negative friendships quality. Consequently, a modified model with two residual correlations, (1) child with ADHD and their parents and (2) friends and their parents, for negative friendship quality improves the model fit.
Psychiatric comorbid externalising disorders in children with ADHD were associated with more negative friendship quality, and was associated with friends’ parents’ reports of less positive friendship quality. Whereas, psychiatric comorbid internalising disorders in children with ADHD were linked with only the friend’s parents’ reports of less negative friendship quality. The perceptions of positive friendship quality of children with ADHD were not significantly different from their friends’ (p = 0.892). However, children with ADHD perceived more positive friendship quality than their parents (p = 0.000) and their friends’ parents (p = 0.000). Children with ADHD also perceived significantly less negative friendship quality than their friend (p = 0.01), their parents (p = 0.000) and their friends’ parents (p = 0.000). Externalising and internalising psychiatric comorbidities accounted for approximately 15.1% and 14.9% of the variance in positive friendship quality, respectively. The moderate and negative residual correlation between these friendship quality variables suggest that the shared variance in positive and negative friendship quality is not entirely explained by the presence of externalising and internalising psychiatric comorbidities. After controlling for the presence of an internalising disorder, children with ADHD and an externalising psychiatric comorbidity were more likely to have a less positive and a more negative friendship quality. Whereas, when controlling for the presence of an externalising disorder, an internalising psychiatric disorder was not related to friendship quality.
These findings must be considered in view of some study limitations. The measure of friendship quality was limited to only one friendship per child with ADHD, and the family of the friend was invited to participate in the study by the family of the child with ADHD. The raw items or the subscales comprising the positive FQQ composite were not included in the model, as this would have results in a very large model in terms of numbers of variables. The sample size also prevented other covariates being included in the MIMIC modelling, and the cross-sectional study design precluded conclusions regarding the temporal sequence between psychiatric comorbidities and friendship quality.
In summary, these data highlight the utility of parents as informants for children’s friendship quality and suggest combining parental reports with those of children with ADHD and their friends. Children with ADHD and externalising psychiatric comorbidities may also have more negative and less positive friendship quality than children with ADHD but no externalising comorbidity. The authors therefore suggest that further assessment and intervention efforts should be considered for this group of children with ADHD.
Read more about children with ADHD and friendship quality here
*A comorbid externalising disorder was considered if the parent of the child with ADHD endorsed conduct disorder or oppositional defiant disorder on the Kiddie Schedule for Affective Disorders and Schizophrenia (K-SADS), and teacher ratings corresponded to a T score ≥60 on the DSM scales of the Conduct Problems and/or Oppositional Defiant Problems on the Teacher Report Form
†A comorbid internalising disorder was considered if the parent endorsed an anxiety or depressive disorder on the K-SADS, and either: (1) the children endorsed a T score ≥60 on the Multidimensional Anxiety Scale for Children-2, or a T score ≥60 on the Children’s Depression Inventory-2, or (2) the T-score teacher rating on the DSM scales of Anxiety Problems and/or Depressive Problems was ≥60 on the Teacher Report Form
‡The FQQ assesses six features: validation and caring, conflict resolution, help and guidance, companionship and recreation, intimate exchange, and conflict and betrayal. The short version of the questionnaire has 22 items, each scored on a 5-point scale (0=not at all; 4=really true)
Becker SP, Fite PJ, Luebbe AM, et al. Friendship intimacy exchange buffers the relation between ADHD symptoms and late social problems among children attending an after-school care programme. J Psycholpathol Behav Assess 2013; 35: 142-152.
Normand S, Mikami AY, Savalei V, et al. A Multiple Indicators Multiple Causes (MIMIC) model of friendship quality and comorbidities in children with attention-deficit/hyperactivity disorder. Psychol Assess 2020; Epub ahead of print.