Do Preschoolers’ Neuropsychological Functioning and Hyperactivity/Inattention Predict Social Functioning Trajectories Through Childhood?

Abstract

Objective

This longitudinal study examined whether preschool attention deficit hyperactivity disorder (ADHD) symptoms and neuropsychological functioning predicted trajectories of children’s social functioning from age 3 to 4 through 12 years.

Methods

Three- and four-year-old children (N = 208; 72.6% boys) were evaluated annually through age 12. Parent and teacher Attention Deficit/Hyperactivity Disorder Rating Scale, Fourth Edition responses during the initial evaluation were used to categorize preschoolers as “High” or “Low” risk for ADHD. Children’s neuropsychological functioning was assessed using the NEPSY. Teachers’ reports of children’s social functioning were obtained annually from preschool through age 12 years using the Adaptability, Functional Communication, and Social Skills subscales of the Behavioral Assessment System for Children, Second Edition. Hierarchical linear modeling was used to assess the trajectories of social functioning and determine whether preschoolers’ neuropsychological functioning and ADHD risk status predicted social functioning at age 12 years, and/or change in social functioning throughout childhood. All models controlled for baseline socioeconomic status.

Results

High Risk children had significantly lower teacher-rated Adaptability and Social Skills at age 12 years. High Risk children and those with lower Verbal neuropsychological functioning in preschool had lower teacher-rated Functional Communication at 12 years old. Lower preschool Verbal neuropsychological functioning predicted greater positive change in teacher-rated Functional Communication across childhood.

Conclusions

Early identification of and intervention for children exhibiting ADHD behaviors is critical given the enduring negative impact of these behaviors on social functioning. Screening preschoolers for verbal difficulties is encouraged given their long-term impact on children’s ability to clearly express thoughts and feelings and obtain and provide information.

Introduction

Social functioning is an area of marked impairment for many individuals with attention deficit hyperactivity disorder (ADHD; Hoza, 2007; Solanto et al., 2009; Wehmeier et al., 2010). Social difficulties can be manifested in many ways. During interactions with peers, children with ADHD may engage in less cooperation, sharing, and turn taking, act impulsively, be intrusive and/or overbearing, be more disruptive, show off, and express anger or frustration more than their non-ADHD peers (Wehmeier et al., 2010). Children with ADHD also have difficulties within the family, typically experiencing more conflict and stress in the parent–child relationship than their non-ADHD peers (Deault, 2010).

As children with ADHD develop into adolescents and young adults, their social difficulties tend to persist (Bagwell et al., 2001; Barkley et al., 1991), especially if their ADHD symptoms also show a chronic course (Bagwell et al., 2001). When compared with peers with no childhood ADHD history, adolescents/young adults who had ADHD in childhood tend to have fewer close friendships (Bagwell et al., 2001), experience more frequent and intense family conflicts (Barkley et al., 1991), struggle to overcome their negative social reputation (Hoza, 2007), are seen as rigid and inflexible leaders, and have trouble expressing and understanding feelings and ideas (Wehmeier et al., 2010).

Social functioning is a multidimensional concept that requires comprehensive evaluation. Rose-Krasnor (1997) proposed a prism model of social competence that incorporates theoretical levels (e.g., understanding the transactional nature of a social interaction) through to skills (e.g., perspective taking). Many of the domains in her model have been highlighted as areas of weakness for individuals with ADHD. Furthermore, Dirks et al. (2007) highlighted the importance of the match between an individual’s social context and behavior, and how an individual must read his/her environment and adapt interactions to fit the social context. Using Rose-Krasnor’s (1997) model and Dirks et al.’s (2007) conceptualization as guides, a comprehensive evaluation of social functioning should include assessment of an individual’s social skills, communication, and flexibility/adaptability.

Given the persisting nature of social problems associated with ADHD, elucidating risk factors for these difficulties is particularly important. The traditional view of ADHD has been that cognitive deficits underpin symptoms, which in turn drive functional impairment (Coghill et al., 2014). However, this view has been challenged given that impairment may persist despite remitting ADHD symptoms. For many individuals whose ADHD symptom severity has decreased, improvements in adaptive functioning (if any) are not of the same magnitude (O’Connor et al., 2015). This is one reason that Coghill et al. (2014) suggested that cognition and symptoms may make independent contributions to impairment; this revised model may better explain the social difficulties observed among children with ADHD.

Consistent with Coghill et al.’s (2014) revised model, Hawkins et al. (2016)^,1 found that impairments in executive function were related to behavioral and social communication problems. They postulated that “features of social interaction such as monitoring and maintaining appropriate topics of conversation, and planning coherent speech” (Hawkins et al., 2016, p. 12) rely on working memory and focused attention.

To date, the relation between neuropsychological functioning and social impairment has largely been explored in elementary school-aged children and adolescents. Yet, inattention and/or hyperactivity/impulsivity often emerge during early childhood. Therefore, understanding how social functioning changes over time and relates to early cognitive functioning and symptom severity could provide key insights into how to more effectively support children with ADHD and tailor interventions to better meet their needs.

This study investigated relations among children’s neuropsychological functioning and hyperactivity/inattention at preschool age, and their social functioning trajectories from ages 3 to 4 through 12 years. Three- and four-year-old children were classified as “High Risk” or “Low Risk” for ADHD based on parent and teacher ratings of ADHD behaviors. At baseline, preschoolers were administered a neuropsychological assessment, and teachers rated children’s social functioning annually through 12 years of age using three subscales from the Behavioral Assessment System for Children, Second Edition (BASC-2; Reynolds & Kamphaus, 2004): Social Skills, Functional Communication, and Adaptability. Growth curves of children’s social functioning over this 9-year period were examined to determine whether preschoolers’ ADHD risk status and/or their neuropsychological functioning predicted social functioning at age 12 years and/or change in social functioning from preschool through middle school. Consistent with Coghill et al.’s (2014) model, we hypothesized that ADHD symptoms and neuropsychological functioning would independently predict change in social functioning across early childhood and social functioning at age 12 years. Specifically, High Risk preschoolers would have weaker growth in the three domains of social functioning across childhood and would show poorer social functioning at 12 years of age than Low Risk children. Furthermore, preschoolers with stronger neuropsychological functioning would have greater growth in social functioning over time and demonstrate better social functioning at age 12 years.

Materials and Methods

Participants

Racially and ethnically diverse preschoolers (N = 208) were recruited into a longitudinal study via local preschools and direct clinical referrals (Table I). Entry was determined based on parent and teacher reports on the Attention Deficit/Hyperactivity Disorder Rating Scale, Fourth Edition (ADHD-RS-IV; DuPaul et al., 1998). Children whose parents and teachers endorsed fewer than three symptoms rated as Often or Very Often on the Hyperactivity/Impulsivity and Inattention subscale(s) were considered “Low Risk” for ADHD (N = 73, 35.1%). “High Risk” children (N = 135, 64.9%) received six or more symptoms rated as Often or Very Often on the Hyperactivity/Impulsivity and/or Inattention subscale(s) by either their parent or teacher. High Risk preschoolers were oversampled at a ratio of 2:1.

Children were excluded if they had a pervasive developmental disorder; posttraumatic stress disorder; a neurological disorder; Full-Scale IQ (FSIQ) < 80 on the Wechsler Preschool and Primary Scale of Intelligence, 3rd Edition (Wechsler, 2002); were taking systemic medication for a chronic medical or psychiatric disorder, including ADHD; did not attend a preschool/daycare facility; or their parents were not fluent in English. Preschoolers who met inclusion criteria were administered the NEPSY (Korkman et al., 1998) in the laboratory. In addition, teachers completed the BASC-2 questionnaire to assess children’s social functioning annually from preschool through age 12 years. Children’s mean (SD) age at recruitment (T0) and follow-up T1, T2, T3, T4, T5, T6, T7, T8, were: 4.31 (0.47); 5.33 (0.49); 6.37 (0.53); 7.42 (0.53); 8.34 (0.52); 9.35 (0.50); 10.36 (0.52); 11.38 (0.51), and 12.45 (0.48) years.

The study was approved by the university’s Institutional Review Board. Parents provided written consent.

Measures

Preschool Hyperactivity/Inattention

Preschool hyperactivity/inattention was measured using the parent and teacher versions of the ADHD-RS-IV (DuPaul et al., 1998), an 18-item, 4-point scale comprising the DSM–IV symptoms of ADHD. As indicated above, preschoolers were classified as High or Low Risk for ADHD using this scale. For this sample, Cronbach’s alpha for parent ratings was .95 and for teachers it was .97.

Neuropsychological Functioning

Neuropsychological functioning was assessed using the NEPSY (Korkman et al., 1998), a standardized neuropsychological test battery for 3- to 12-year-old children. Subtests cover five cognitive domains including Language, Memory, Attention-Executive Functioning, Visuospatial, and Sensorimotor. In the normative sample, test–retest reliability for the five domains ranged from 0.70 to 0.91 for children 3–4 years old (Korkman et al., 1998).

Social Functioning

Social functioning was measured using the BASC-2 (Reynolds & Kamphaus, 2004), which was completed by teachers at baseline (T0), and annually thereafter until 12 years of age. Teacher ratings were used to reduce method bias as children were generally rated by different teachers each year. Teachers rated how often a behavior occurs using a 4-point rating scale (0 = Never; 3 = Almost Always). The Adaptive Functioning subscales; Adaptability, Functional Communication, and Social Skills were used to reflect children’s social functioning, with higher scores indicating better functioning. These three scales capture several aspects of social functioning that are often challenging for children with ADHD symptoms, such as rigidity and lack of reciprocity (Wehmeier et al., 2010).

The Adaptability subscale measures a child’s ability to adjust to changes in routine or plans, recover after a setback and to be flexible when things do not go according to plan. Functional Communication assesses an individual’s ability to express ideas and feelings, to obtain and convey information and generally communicate clearly with others. The Social Skills subscale measures the individual’s ability to employ Social Skills and engage in a socially acceptable manner in their interactions with peers and adults. In this study, Cronbach’s alpha values ranged from .83 to .89 on the preschool form (ages 2–5 years), .79 to .95 on the school-age form (ages 6–11 years), and from .71 to .91 on the adolescent form (age 12 and older). These values are consistent with those of previous reliability studies (Reynolds & Kamphaus, 2010).

For each child, mean raw scores for each of the BASC-2 teacher-rated Adaptability, Functional Communication, and Social Skills subscales were calculated for each time. If a child had ratings from more than one teacher at a given time point, the scores were aggregated to generate an overall mean for the scale at that assessment period. The mean score was used, rather than highest or lowest score, to reflect that children behave differently in different classrooms.

Data Analysis

The NEPSY generates five domains of neuropsychological functioning, but the authors have advised against using the scales individually (Korkman et al., 2007). Thus, a factor analysis of the domain standard scores was carried out using Maximum Likelihood estimation and Oblimin with Kaiser Normalization rotation (Goodness-of-Fit test χ² [1] =.24, p =.62). Factors with eigenvalues >1 were retained, resulting in two factors being extracted—a “Nonverbal” factor accounting for 38.28% of the variance and a “Verbal” factor accounting for 17.58% of the variance. Sensorimotor and Visuospatial domains loaded onto the Nonverbal factor, whereas Language and Memory domains loaded on the Verbal factor. The Attention-Executive Functioning domain split its variance across both factors (see Supplementary 1 for factor loadings). Given that Attention-Executive Functioning performance loaded onto both Verbal and Nonverbal domains, we saved factor scores based on their unique loadings on each factor and used these in subsequent analyses.

Bivariate correlations were calculated among the social functioning variables, ADHD risk status and neuropsychological factors. Hierarchical linear modeling (HLM; Raudenbush & Bryk, 2002) was used to examine intercept (age 12 years) and slope (change over time) of children’s social functioning, while taking into account the lack of independence between repeated observations of each child. HLM enables direct-likelihood estimation of missing data (Raudenbush & Bryk, 2002), allowing individuals who did not come at a particular time point, but who returned a year or two later to be included in the analysis (see Supplementary 2 for details of HLM methodology and Supplementary 3 for preliminary multiple linear regression analyses testing the interaction between Preschoolers’ ADHD Risk Status and their Verbal neuropsychological functioning on Functional Communication at each time point before running HLMs stratified on preschoolers’ ADHD Risk Status for this aspect of social functioning).

The first model investigated trajectories of change in each domain of social functioning across 9 time points, with age centered at 12 years. Social functioning was reflected in the mean raw scores on BASC-2 Adaptability, Functional Communication, or Social Skills subscales, with separate models run for each subscale.

Preschool ADHD risk status (coded as 0 = Low Risk and 1 = High Risk), and Nonverbal and Verbal neuropsychological functioning (standardized regression factors) were then entered into the Level 2 model to examine their associations with social functioning at age 12 years (intercept) and change in social functioning from preschool through age 12 years (slope). Family socioeconomic status (SES; grand mean centered) was included as a covariate due to significant associations between SES and social functioning at several time points (rs = .20–.55, all p ≤ .01).

Missing Data

Total Ns for teacher-reported social functioning at each time period were: T0, N = 200; T1, N = 159; T2, N = 143; T3, N = 133; T4, N = 136; T5, N = 128; T6, N = 119; T7, N = 105; and T8, N = 98. Participants with NEPSY and teacher-reported social functioning data from at least one time point did not differ from those who were missing NEPSY data or teacher ratings from all points of time (N = 8) on gender, race, ethnicity, FSIQ, age, ADHD Risk status, SES, parent-rated ADHD-RS-IV severity, or teacher-rated ADHD-RS-IV severity at T0 (all p ≥ .26).

Results

The means for teacher reports of Adaptability, Functional Communication, and Social Skills at each time point can be found in Figure 1. Independent samples t-tests revealed that High Risk children had poorer Social Skills, Functional Communication, and Adaptability at all years with the exception of Adaptability at T8 (p = .08). After Bonferroni correction was applied to correct for multiple testing (p = .002), group differences were no longer seen for Social Skills at T2, T3, and T8; Functional Communication at T1, T2, T3, and T8; and Adaptability at T3 and T8.

Preschool Nonverbal and Verbal Neuropsychological functioning were significantly correlated, r = .44, p < .0001. Verbal Neuropsychological functioning was significantly associated with teacher-rated Adaptability for 7 of 9 time points, rs = .18–.33, p ≤ .046; Functional Communication for all 9 time points, rs = .31–.58, p ≤ .001, and Social Skills for 8 of 9 time points, rs = .22–.37, p ≤ .026. Nonverbal Neuropsychological functioning was significantly associated with Adaptability for 5 of 9 time points, rs = .18–.34, p ≤ .04; Functional Communication for 8 of 9 time points, rs = .22–.40, p ≤ .029, and Social Skills at 5 of 9 time points, rs = .17–.24, p ≤ .045 (see Supplementary 4 [Adaptability], Supplementary 5 [Functional Communication], and Supplementary 6 [Social Skills]).

Unconditional Models Testing Change in Social Functioning Over Childhood

This model provided average social functioning for each domain at 12 years and the variability in the intercept, as well as rate of change across time. Mean Adaptability at age 12 was 1.92 (T ratio = 43.45; SE = 0.04; p < .001). The average slope or rate of change in Adaptability was 0.03 units (T ratio = 3.66; SE = 0.01; p = .001). There was significant variation around the intercept π² (187) = 443.71; p < .001 and the slope π² (187) = 257.78; p = .001.

The estimated mean Functional Communication score at age 12 was 2.15 (T ratio = 53.10; SE = 0.04; p < .001). The average slope for Functional Communication across time was 0.06 units (T ratio = 8.91; SE = 0.01; p < .001). There was significant variation around the intercept π² (187) = 446.05; p < .001 and the slope π² (187) = 267.63; p < .001.

The estimate of the mean Social Skills at age 12 was 1.57 (T ratio = 31.01; SE = 0.05; p < .001). The average slope for Social Skills at age 12 was 0.02 units (T ratio = 2.66; SE = 0.01; p = .009). There was significant variation around the intercept π² (187) = 440.63; p < .001 and the slope π² (187) = 245.10; p = .003.

Taken together, positive slopes for all three social functioning domains assessed suggests that, on average, children’s social functioning improved as they got older. The variation around all intercepts and slopes for the three outcome variables indicated that the children varied in their social functioning capability at age 12 and also in the rate of growth of social functioning from preschool to middle school. Therefore, these parameters were modeled using the predictors ADHD risk status and preschool neuropsychological functioning.

Conditional Models Evaluating Preschool ADHD Risk Status and Neuropsychological Functioning as Predictors of Social Functioning

ADHD risk status was significantly associated with teacher-rated Adaptability at age 12 years, such that children at High Risk for ADHD as preschoolers had lower Adaptability at age 12 than those who were at Low Risk. Neither preschool ADHD risk status nor neuropsychological functioning was associated with change in Adaptability from preschool through school age (see Table II).

Preschoolers at High Risk for ADHD had weaker Functional Communication skills at age 12 than preschoolers who were Low Risk for ADHD. In addition, lower Verbal neuropsychological functioning at preschool was independently associated with poorer Functional Communication at age 12. Further, lower preschool Verbal functioning predicted greater improvement in Functional Communication (see Table II). Given the significant association of both early Verbal Skills and Risk for ADHD with changes in Functional Communication, we sought to understand how Verbal Skills differently predicted Functional Communication between High and Low Risk ADHD children. Figure 2 shows the association between trajectories of Functional Communication skills and early ADHD risk status with changes in Verbal neuropsychological functioning. Children with lower Verbal neuropsychological functioning at preschool (at or below the 25th percentile of scores) showed greater improvement in Functional Communication skills over time compared with those children with higher Verbal neuropsychological functioning (at or above the 75th percentile of scores). Low ADHD risk children who were in the bottom quartile of Verbal scores had greater positive change in Functional Communication relative to children who were in the top quartile of their Verbal scores but had a High Risk of ADHD.

High ADHD risk, but not neuropsychological functioning in preschool, was significantly associated with poorer Social Skills at age 12. Neither ADHD risk status nor neuropsychological functioning was associated with growth in Social Skills subscale (see Table II).

Discussion

This prospective longitudinal study investigated whether ADHD symptom severity and neuropsychological functioning predict social functioning. The aim was to elucidate risk factors for ADHD-related impairment, as well as to test Coghill et al.’s (2014) model of a dissociation between symptoms and neuropsychological functioning in predicting impairment. To our knowledge, this is the first study to examine how preschoolers’ neuropsychological functioning and hyperactivity/inattention affect trajectories of social functioning through school-age.

Irrespective of social domain assessed, elevated levels of inattention/hyperactivity during preschool was associated with poorer social functioning at 12 years, but not rate of change in social functioning across childhood. This finding is robust across many studies and myriad ways of measuring social functioning; ADHD behaviors are a potent risk factor for diminished social competence, and the risk ADHD poses persists over time (Bagwell et al., 2001; Barkley et al., 1991; Rinsky & Hinshaw, 2011). Of note, preschoolers in this study did not necessarily meet diagnostic criteria for ADHD, but were identified by parents and/or teachers as having elevated ADHD symptoms. Our findings are similar to those of Diamantopoulou et al. (2007), who showed that ADHD symptom severity in 8 years old was related to social functioning 1 year later. Findings underscore the impairing nature of the symptoms, even if subthreshold for a diagnosis, which is consistent with a continuous view of the ADHD construct (Frick & Nigg, 2012). They also highlight the oftentimes chronic course of ADHD over time. Although not every preschooler with elevated symptoms goes on to receive an ADHD diagnosis by the time they enter school, many do, especially when high symptom levels are seen at home and in at least one other setting (O’Neill et al., 2014).

Overall, we found a more limited role for preschoolers’ neuropsychological functioning affecting later social functioning. Verbal, but not Nonverbal neuropsychological functioning was associated with Functional Communication skills at age 12 years, such that weaker functioning in preschool was associated with poorer Functional Communication at middle school. Those with poorer Verbal neuropsychological functioning skills in preschool also showed greater growth in Functional Communication skills over time, although they never managed to catch up to their peers who had stronger baseline skills. Children with Low ADHD risk and low Verbal skills started out with poorer functional communication but improved this skill more than High ADHD risk children with High Verbal skills pointing to the importance of early ADHD Risk in shaping changes in functional communication. Functional Communication, as operationalized by the BASC-2, focuses on children’s ability to provide relatively well-rehearsed information upon being asked for it, thus it can be considered to activate semantic memory networks. The Functional Communication scale also assesses children’ ability to express ideas clearly so others can understand them. Given this, it is not surprising that early Verbal neuropsychological functioning, which comprised emergent language skills, memory, and to a lesser extent, executive functioning, was associated children’s functional communication. Variability in both age 12 outcomes and rate of growth of skills over time remained unaccounted for. For an individual to know whether someone else can understand what they are saying requires perspective taking, of which executive functioning is an important predictor (Nilsen & Graham, 2009). Although executive functioning was assessed by the NEPSY, its variance was split over the Verbal and Nonverbal neuropsychological factors. It is possible that a more robust measure of executive functioning would have emerged as a significant predictor.

Neuropsychological functioning during preschool was not associated with Social Skills or Adaptability at age 12 years, or rate of change in domains from preschool to age 12. The BASC-2 Social Skills domain captures a child’s ability to draw upon routinized norms of social interaction (e.g., use of manners); this narrow focus may have limited the influence of neuropsychological functioning. The Social Skills domain also assesses how frequently children engage in prosocial behaviors. Utilizing a novel chat-room task to assess children’s social functioning, Huang-Pollock et al. (2009) found no relation between executive functioning and the number of times children made prosocial comments.

The Adaptability scale assesses emotional self-regulation, especially when things do not go according to plan, and how easily children can transition from one thing to another. Surprisingly, we did not see a role for neuropsychological functioning. Individuals with stronger verbal skills are likely better able to communicate and symbolize their experiences, making them less easily frustrated and more able to adapt when conditions change. Cognitive flexibility, planning, attention and processing speed are all functions that may be involved in Adaptability. Our neuropsychological functioning measure may not have adequately captured these processes.

Greater focus on executive functions rather than global neuropsychological functioning may have yielded a stronger relation between cognitive functioning and social functioning (Huang-Pollock et al., 2009; Rinsky & Hinshaw, 2011, but see Diamantopoulou et al., 2007). Our participants were very young; given that executive functions show protracted development through early adulthood, we focused on a broad assessment of neuropsychological functioning. However, development of executive functions is nonlinear and many domains undergo a developmental spurt during preschool (Best & Miller, 2010), thus a more comprehensive assessment of executive functions may have been useful. In line with this, Gewirtz et al. (2009) found that weaker inhibitory control in 4.5 years old was associated with poorer parent- and teacher-rated social skills in third grade. ADHD severity was a separate outcome and so it is unclear if these findings generalize to children at high risk for ADHD.

Our findings provide partial support for Coghill et al.’s (2014) model that posits dissociation between ADHD symptomatology and neuropsychological functioning in driving ADHD-related functional impairment. Our findings build on those of Rinsky & Hinshaw (2011) who found that in a large sample of girls, ADHD group status and children’s planning and response inhibition independently predicted social skills 5 years later. Marton et al. (2009) found ADHD severity was an independent predictor of 8- to 12-year-old children's social perspective taking after accounting for language and IQ, explaining a further 8% of the variance.

This study had several strengths including annual independent observations of children’s social functioning over 9 years, beginning when ADHD symptoms are often first emerging and critical development of neuropsychological functions is occurring. The longitudinal design enables clearer elucidation of temporal relations among variables. Finally, neuropsychological functioning was assessed using a well-established standardized and age-normed battery.

Limitations must also be noted. A fuller assessment of executive functions may have been helpful. Also, there may be a disconnect between teacher ratings and how children actually act in a given social context. Observation of children’s behavior and/or sociometric ratings may have yielded different findings (Huang-Pollock et al., 2009). In addition, myriad pre/perinatal factors increase risk for ADHD and neuropsychological deficits, such as weak language skills, and social difficulties (e.g., Huang et al., 2018; D’Onofrio et al., 2013; Tomblin et al., 1997). Including confounding variables or examining moderators of the model (e.g., prematurity, prenatal tobacco exposure) may strengthen the findings. Finally, the impact of multicollinearity was not assessed in these HLMs, which may have been an issue given the expected significant correlations among predictor variables at both levels of the model. However, it is likely that the results are valid given that irrespective of severity of multicollinearity, HLM produces largely unbiased estimates of both fixed and random effects (Yu et al., 2015).

Future studies should determine whether ADHD presentation affects trajectories of social functioning. Combined presentation has been associated with aggression and struggles with self-control (Diamantopoulou et al., 2007; Solanto et al., 2009), whereas inattentive presentation has been linked to lower assertiveness (Solanto et al., 2009). This is an avenue for future research.

Early identification and treatment of ADHD risk status and screening of children’s verbal neuropsychological functioning may help ensure more positive trajectories of social functioning. When compared with Low Risk children, those at High Risk for ADHD in preschool did not recover to the same degree; their Adaptability, Functional Communication, and Social Skills started out worse and although improved over time, never reached the same level as preschoolers who were Low Risk. This means that when children are experiencing ADHD symptoms in preschool, they are likely already on a trajectory of social functioning that is impaired relative to their typically developing peers. The amount of time a child is struggling without intervention may be critical to his/her social development. Feil et al. (2016) showed that treatments that address behavior problems in preschoolers at risk for ADHD led to positive improvement in Social Skills and fewer behavior problems 4 months later. Early intervention for verbal difficulties may promote better social functioning, although this needs to be empirically tested. Preliminary evidence suggests such an approach could be fruitful. For example, using the same sample as this study, Mlodnicka et al. (2016) found that High Risk preschoolers who received speech and language therapy at school had greater improvement in both parent- and teacher-rated Social Skills from baseline to 2-year follow-up compared with their peers who did not receive treatment.

Overall, screening and early identification of children at risk for ADHD as well as those with weak verbal neuropsychological ability may help to achieve a more positive trajectory of social functioning as it allows early access to intervention. Given the enduring nature of social impairment, early detection and action is essential.

Supplementary Data

Supplementary data can be found at: https://dbpia.nl.go.kr/jpepsy.

Funding

This work was supported by the National Institute of Mental Health (grant number R01 MH068286 to J.M.H.); and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant number SC2 HD086868 to S.O.) of the National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institutes of Health.

Conflicts of interest: None declared.

Footnotes

References