Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Childhood Symptoms of ADHD Overrule Comorbidity in Relation to Psychosocial Outcome at Age 15: A Longitudinal Study

  • Eva Norén Selinus ,

    Affiliation Department of Clinical Neuroscience, Centre for Psychiatry Research & Education, Karolinska Institutet & Stockholm County Council, Stockholm, Sweden

  • Yasmina Molero,

    Affiliation Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden

  • Paul Lichtenstein,

    Affiliation Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden

  • Tomas Larson,

    Affiliation Center for Ethics, Law and Mental Health (CELAM), University of Gothenburg, Göteborg, Sweden

  • Sebastian Lundström,

    Affiliations Center for Ethics, Law and Mental Health (CELAM), University of Gothenburg, Göteborg, Sweden, Gillberg Neuropsychiatry Centre, University of Gothenburg, Göteborg, Sweden

  • Henrik Anckarsäter,

    Affiliation Center for Ethics, Law and Mental Health (CELAM), University of Gothenburg, Göteborg, Sweden

  • Clara Hellner Gumpert

    Affiliation Department of Clinical Neuroscience, Centre for Psychiatry Research & Education, Karolinska Institutet & Stockholm County Council, Stockholm, Sweden



Neurodevelopmental problems (NDPs) may influence the transition from childhood to adolescence. Our aim was to study long-term psychosocial outcomes of NDPs, focusing on ADHD.


Data was collected through a telephone interview with parents of twins at ages 9 or 12 years. NDP screen-positive children were clinically assessed at age 15; N = 450. Psychosocial outcome concerning peers, school, internalizing problems, antisocial behavior, alcohol misuse, drug misuse, and impaired daily functioning was examined.


Even after controlling for other NDP comorbidity, screen-positivity for ADHD doubled or tripled the odds of later psychosocial problems. When controlling for parental education level, the significant effect of ADHD remained only for antisocial behavior and impaired daily functioning.


Signs of NDPs as well as other psychiatric diagnoses at ages 9 or 12 years are associated with a more problematic adolescence. However, despite the presence of comorbidity, early ADHD symptoms stand out as the most important risk factor for later antisocial development and impaired daily functioning.


Childhood neurodevelopmental problems (NDPs; i.e., attention deficit and hyperactivity disorder [ADHD], autism spectrum disorder [ASD], tic disorder [TD], developmental coordination disorder [DCD], and learning disabilities [LD]), affect around 10% of all children [1]. ADHD is the most common disorder, the prevalence being between 5 and 10% [24], while the prevalence of ASD ranges between 1% and 2.6% [57]. About 1.5% of children meet criteria for LD [810], and 1% to 6.6% of children meet criteria for TD [1114]. DCD has an estimated prevalence of about 5% [15, 16].

Children with NDPs have a higher risk of functional and psychosocial difficulties, increased mortality, and mental health problems, and often require a wide selection of life-long interventions from medical and social services [17]. Several studies have shown that children with ADHD are at risk for school failure, emotional difficulties, substance misuse, antisocial behavior and poor peer relationships in adolescence [1823], and are more impaired in psychosocial, educational and neuropsychological functioning as adults [17]. Studies that have investigated psychosocial problems in NDPs other than ADHD are, however, scarce [24, 25]. The few studies that have examined associated problems of ASD, TD, LD, and DCD, have focused on psychiatric comorbidities [26, 27], academic difficulties [28, 29], and peer victimization [30, 31], and less attention has been paid to substance misuse and antisocial behavior [3234]. Furthermore, research has been restricted by small samples, cross-sectional data or short-term follow-up [35, 36].

Children with one NDP diagnosis frequently demonstrate comorbidity with other NDP diagnoses [3739]; for example, many children with ADHD also exhibit ASD symptoms [9, 4043]. Studies have also reported large co-existence between TD and ADHD, LD and ASD [4447]. Furthermore, up to 50% of children with LD or ADHD or ASD present DCD symptoms [4850]. It has further been shown that individuals drift between NDP diagnoses [1, 51], leading to the suggestion that these diagnoses are not discrete disorders or syndromes. This notion was supported by findings from a study by Larson and colleagues [52]. In a Swedish sample screened for NDP diagnoses at ages 9 or 12 years (n = 198), about 50% received a clinical NDP diagnosis at 15, but only 40% of these received the same diagnosis as at the earlier screening. The same relationship was found in a Swedish clinical study [53].

NDP comorbidity is an important clinical feature and may have implications for diagnosis and treatment, as individuals with comorbid NPDs could present a more severe form of the disorders [32], and subsequently have a poorer long-term prognosis [51, 54]. Thus far, few studies have taken NDP comorbidity into account when examining long-term outcomes as most studies have examined outcomes of ADHD, ASD, DCD, LD and TD separately.

Given that NDPs are not entirely discrete disorders, comorbidity of NDPs needs to be systematically addressed both at baseline and follow-up in longitudinal studies, in order to examine if NDPs should be considered as a group of problem domains with shared outcomes. Likewise, it is equally important to consider a broad spectrum of psychosocial outcomes, since different symptom profiles may be linked to different outcomes.


The aim of the present study was to examine the association between signs of ADHD and/or other NDPs (at ages 9 or 12 years) and negative psychosocial outcomes at age 15. Negative psychosocial outcome was defined as the presence of any of the following: peer problems, school/educational problems, internalizing problems, antisocial behavior, alcohol misuse, substance misuse, and/or impaired level of daily functioning/global impairment.


Subjects at Baseline

The Child and Adolescent Twin Study in Sweden (CATSS) [32] is an ongoing study that tracks all twins born in Sweden from July 1992 and onwards. A telephone interview including the Autism–Tics, ADHD, and other Comorbidities inventory (A-TAC) is conducted with the parents of all participating twins in connection with the children’s 9thor 12th birthday (parents to twins born before July-95 were interviewed at age 12; those born July -95 and after were interviewed at age 9). As of January 2014, 12,500 A-TAC interviews had been performed, covering 25,000 twins. The overall response rate was 75%.

In the present study, 860 twins were eligible for inclusion. The response rate was 52%. 450 participated fully: 247 screen-positive for NDP and/or NDP related behavioral disorder (144 boys, 103 girls), 157 screen-negative (88 boys, 69 girls), and 46 randomly chosen sex-matched controls (30 boys, 16 girls). There were 38 twin pairs where both twins were screen-positive for a NDP (23 pairs of boys and 15 pairs of girls). An analysis of the attrition group in comparison with the study group is presented in Table 1.

Table 1. Descriptive statistics.

Comparison of frequencies in the DOGSS cohort and the attrition cohort.

The A-TAC interview.

The A-TAC was designed as an easy-to-administer, dimensional, and comprehensive interview to assess a broad scope of NDPs, and yields dimensional ratings of symptoms, suffering and dysfunction in a range of neurodevelopmental and psychiatric disorders, including ASD, ADHD, LD, TD and DCD. The investigators from a market research company (”Intervjubolaget”) who have all been trained in using the computerized version of the A-TAC inventory administer the telephone interview. All interviews are performed by laypersons. The average interview time for the A-TAC in the CATSS is about 30 minutes.

In previous validation studies, the A-TAC inventory has shown good test-retest measures, excellent inter-rater reliability and construct validity [52, 5557], and convergent validity with the Child Behaviour Check List [58]. The Hansson et al study from 2005 showed excellent screening properties for ASDs and ADHD in a clinical sample [56]. Hansson et al found that Areas under Receiver Operating Characteristics curves between interview scores and clinical diagnoses were approx. 0.90 for ADHD and ASD and above 0.70 for TD, LD, and DCD. Using optimal cut-off scores for ASD and ADHD, good to excellent kappa levels for interviews and clinical diagnoses were noted. Larson and colleagues replicated these results in 2010 [57], and cut-off scores to identify proxies to clinical diagnoses were established for ASDs, ADHD, DCD, LD and TD. These cut-off scores (concurrent validity) had sensitivities above 0.90 (0.95 for ASD and ADHD). The predictive value for the A-TAC concerning the DOGSS cohort has also been addressed in Larson et al.’s study from 2013. The results in that study showed that the sensitivity and specificity of A-TAC scores for predicting later (3 or 6 years) clinical diagnoses were good to excellent, with values of the area under the receiver operating characteristics curves ranging from 0.77 (ADHD) to 0.91 (ASD).

Screening at baseline (age 9/12)

In the A-TAC interview, cut-off scores for the different NDPs were determined and validated by two earlier studies [56, 57]. ADHD traits were calculated as the sum of scores of both the inattention and hyperactivity-impulsivity domains, yielding between 0 and 19 points. A dichotomous variable was created with a cut-off value of ≥8 points. The ASD cut-off level was set at ≥4.5 (of a maximum of 17 points), LD traits at ≥3.5 (out of 9), TD at ≥2 (out of 4) and DCD at ≥1.5 (out of 5) [57].

Subjects at Follow-up

Twins born 1993–1995 (n = around 4100) were eligible for the clinical follow-up at age 15, referred to as the DOGSS study (Developmental Outcomes in a Genetic twin Study in Sweden). Same-sex twin pairs, in which at least one in the pair screened positive for one or several of the targeted disorders at ages 9/12 (i.e. ASD, ADHD, TD, LD, DCD, and/or behavioral disorders with known NDP comorbidities, such as Obsessive Compulsive Disorder [OCD] and/or Oppositional Defiant Disorder [ODD] and/or Conduct Disorder [CD] and/or Eating Disorder [ED]), were included. Randomly selected controls were included to create a population-based study group enriched for NDPs. From the 1995 cohort, the inclusion criteria were narrowed to include ASDs and ADHD only. The cohort (N = 860) was contacted by mail and invited to participate in a comprehensive clinical assessment, and a total of 452 adolescents consented to participate in the study (total response rate 52%). Two of the co-twins to screen-positive children had to be excluded from the present study due to missing A-TAC-data. In total, 450 participated fully in both CATSS 9/12 and DOGSS, including 247 screen-positive children (144 boys and 103 girls), 157 screen-negative co-twins (88 boys and 69 girls), and 46 randomly chosen, sex-matched controls (30 boys and 16 girls from CATSS-9/12).

Procedure at follow-up

The follow-up assessment procedure included various self-report rating scales from a care-giver (i.e. the parent or parents that accompanied the twin to the clinical assessment participated) and the teenager, as well as a thorough clinical assessment. For the purpose of this paper, we used: a) data from the parent and adolescent self-reports regarding psychosocial functioning at age 15, and b) the clinical psychologists’ assessment of the teenager’s overall functional level (results from the diagnostic validation of the A-TAC interview have been reported elsewhere [52].

Psychosocial outcome at age 15

Psychosocial outcome was defined as indicators of impairment in seven different domains related to everyday-life: peer problems, school problems, internalizing problems, antisocial behavior, alcohol misuse and/or drug misuse. In addition, a global measure of functioning, CGAS [59] was also used as an outcome variable.

Peer problems.

Peer problems were assessed via the SDQ (Strengths and Difficulties Questionnaires) [60] using >5 points for self report (SR) and >3 points for parent report (PR) to indicate peer problems. The scale ranges between 0 and 10 points, where 0 is “no problems” and 10 is “serious problems”. The cutoff was chosen according to the standard for SDQ for abnormality. In addition, the inquiry of problems related to bullying was included. Being bullied was described using Olweus Bully Victim Questionnaire [61], a self-rating scale from 1 (not been bullied) to 5 (several times a week during the last months), and was dichotomized with the recommended cut-off value of ≥3. Bullying others was defined using the same questionnaire, with a scale from 1 (not bullied others) to 5 (several times a week during the last months), and was dichotomized with a cut-off value of ≥3. Any Peer problem was defined as a dichotomous variable, indicating that at least one of the scales described above reached above cut-off level.

School problems.

School problems were defined as truancy, repetition of a school year and/or failure in grades. Truancy was assessed using Self-reported Delinquency [6265], a self-rating scale, which was dichotomized with a cut-off value of ≥4 points (indicating more than 3–5 times of illegal absence from school). Repetition of a school-year was dichotomized as yes/no and Failure in grades was dichotomized as yes/no (Yes—failed in majority of subjects; No—passed in majority/all of subjects), as assessed in the K-SADS interview. Any School problem was defined as dichotomous (yes/no) variable indicating that at least one of the scales was scored above cut-off level.

Internalizing problems.

Internalizing problems were assessed through self-rating and parent-rating scales from SDQ, including the sub-scores for Emotional symptoms. Each score´s range is between 0 (no problems) and 10 (serious problems). Emotional symptoms included psychosomatic symptoms, worry, unhappiness, nervousness, and fearfulness, and was dichotomized with a cut-off value of >6 for SR and >4 for PR, according to the standard for SDQ for abnormality. Any internalizing problem was defined as a dichotomous (yes/no) variable indicating that at least one of the scales was scored above cut-off level.

Antisocial behavior.

Antisocial behavior was defined through a combination of three different self-reported measures of criminality, norm-breaking behavior and violent events using Self-reported Delinquency and SDQ. Criminality included frequency of self-reported: a) non-violent criminal acts and b) violent criminal acts. Non-violent criminal acts ranging from 0 to 10, was dichotomized with a cut-off value of ≥1 (at least one non-violent criminal act). Non-violent criminal acts were defined as vandalism, graffiti, fire, driving without a license, forgery of ID, car theft, theft, breaking into private property, and supplying illicit substances. Violent criminal acts was measured using self-reported violent behavior, including robbery, hurting animals, hurting a human, having sex with someone against her/his will with/without physical violence. It was expressed as frequency of self-reported violent acts, ranging from 0 to 6 and was dichotomized with a cut-off value of ≥1 (once or more of one or several acts). Conduct problems was assessed using: a) the K-SADS diagnose Conduct Disorder (yes/no) (which includes a Definite Diagnose [scored as 4 according to the K-SADS algorithm], a Probable Diagnose [3] and Diagnose in partial remission [5]) and b) SDQ Conduct Problems Score (self-rated and parent-rated). The SDQ questionnaire addressed conduct problems, and ranged between 0 (no problems) and 10 points (conduct problem several times a week during the last months). A dichotomous variable was created with a cut-off value of >4 points for SR and >3 points for PR, according to the standard for SDQ for abnormality. Any Antisocial behavior was defined as at least one of the antisocial problems above.

Alcohol misuse.

Alcohol misuse was assessed using the Substance abuse self-reported alcohol and drug use [6265] and included measures of drinking. Drinking was measured through two questions: a) “How often do you feel drunk when you drink alcohol?” ranging from 1 (“I don’t drink”) to 6 (“every time”) which was dichotomized with a cut-off value of ≥3 (rarely/sometimes/once in a while), b) “Have you been drinking beer, wine or liquor last month?” which was dichotomized into yes/no. Any alcohol misuse was defined as at least one of the alcohol problems above.

Drug misuse.

Drug misuse was assessed using the Substance abuse self-reported alcohol and drug use [6265], and included measures of drug use and substance abuse diagnosis (substance abuse diagnosis was assessed through the K-SADS interview). Drug misuse was defined as the reported use of any illicit drug and was dichotomized with a cut-off value of ≥1 (used at least one illicit drug). The K-SADS classifies the substance abuse diagnosis into a Definite Diagnose (4), a Probable Diagnose (3) and Diagnose in partial remission (5), and we included all of these. Any drug misuse was defined as at least one of the drug problems above.


The Children’s Global Assessment Scale (CGAS) [59] is a clinician-rating tool used for both research purposes or in clinical settings. The rater is asked to assess the lowest overall level of psychosocial functioning during the preceding month. The scores range from 1 (the most impaired level) to 100 (the best level of functioning). CGAS is currently routinely used in many countries, and has also been used in several longitudinal and epidemiological studies [6673]. Recent Swedish studies have been made concerning inter-rater reliability of CGAS [74] and comparing seminar and computer based training programs for CGAS on the accuracy and reliability of raters [75]. A third study showed that having a CGAS ≤ 60 was an independent predictor of several adversities in young adulthood, such as criminal conviction, bipolar disorder and borderline personality disorder [76]. CGAS was therefore dichotomized with a cut-off value of ≤ 60, which was then expected to identify adolescents at higher risk of developing different types of psychosocial problems. The participating psychologists were all trained in evaluation according to CGAS.

Parental education level.

Data on the parents’ educational level was collected from the parent interview at age 9/12. It was divided into three levels: 1) Primary school (≤ 9 years of school) 2) Secondary school (10–12 years) 3) University studies or equivalent (> 12 years). This was recorded for both of the parents separately, but summarized in the analysis to a continuous variable ranging from 2 (both parents had level 1) to 6 (both parents had level 3).

Statistical Analyses

First, frequencies for all seven negative psychosocial outcome domains, including sub-categories, are presented for the whole cohort (n = 450). In a second step, frequencies of all seven psychosocial outcomes are presented for each screen-positive diagnose at baseline (i.e. screen-positive according to the A-TAC interview; NDPs and other diagnoses), for screen-negative siblings of cases and for screen-negative random controls, respectively. Furthermore, a category was added to indicate the frequency of individuals with no reports in any of the seven negative psychosocial outcome domains, labeled as ‘No Problem’.

Since our definition of antisocial behaviors included a wide variety of antisocial acts, we also wanted to determine if the severity of antisocial behavior was associated differently to different NDP groups. We therefore distinguished between non-violent and violent antisocial behavior, as well as between the other sub groups included in the variable “Any antisocial behavior”, i.e. the SDQ conduct scale self- and parent-report, and CD diagnose. All of these behaviors are now displayed in a separate descriptive analysis.

Finally, we analyzed all cases (n = 293; i.e. all those individuals who were screen-positive for diagnoses at ages 9/12). Screen-negative co-twins were not included in this analysis since we wanted to compare screen-positives with pure controls and not with siblings (co-twins). For each individual we calculated the odds of each negative psychosocial outcome. The predictor of main interest in this analysis was ADHD, with or without comorbidity with other NDPs at ages 9/12. The association between ADHD and each of the outcomes was modeled with logistic regression, controlling for the following NDPs: co-existing autism spectrum disorder (ASD), learning disability (LD), tics disorder (TD), and developmental coordination disorder (DCD). The following non-NDPs were also controlled for: oppositional defiant disorder (ODD), conduct disorder (CD), obsessive compulsive disorder (OCD), and eating disorder (ED). In addition, we controlled for parental education level as described above. Since several correlated twin-pairs could be included in the analysis, a robust sandwich estimator was used to adjust for the correlated data when calculating the confidence intervals. The estimated associations are reported as odds ratios with 95% confidence intervals. All p-values less than 0.05 were considered statistically significant. The software package Stata IC12 was used.

Ethical Considerations

For the CATSS study there was a written consent from the parent/legal guardian, and for the DOGSS study there was a written consent from both teenager and parent/legal guardian after written and oral information about the study. Analyses were performed on anonymized data files. The study protocol accorded with the Helsinki declaration and was approved by the ethical review board of Karolinska Institutet, Solna, Sweden (registration numbers CATSS Dnr: 02–289 and 2010/507-31/1 DOGSS Dnr: 03–672 and 2010/1356/31/1). Participants who were considered to be in need of child psychiatric care or social welfare, were taken care of in an adequate way.


The frequency rates of psychosocial outcomes are presented in Table 2. The most frequently reported negative outcome was ‘Any antisocial behavior’ (51.8%), followed by ‘Any alcohol misuse ‘(28.9%), ‘Any peer problem’ (23.6%), ‘Any school problem’ (23.6%), and ‘Any drug misuse’ (14.9%). ‘Impaired daily functioning’ was present in 15.3% of the participants (CGAS rate of ≤ 60; assessed by psychologist and child psychiatrist). Approximately 20% did not report any of the problems in focus.

Table 2. Distribution of age 15 psychosocial outcomes in the whole cohort (n = 450).

Among the NDPs, antisocial behavior was the most common outcome across all proxy diagnoses (Table 3). Peer problems were also present across all the NDP proxy diagnoses (all above 30%, the ASD group reaching the highest level), but were also reported among the other groups. Those who had been screen-positive for ADHD displayed the highest rates of problems in the antisocial domain (66.3%), which was also the most frequent problem area for those who had been screen-positive for LD, TD and DCD. In addition to reporting the highest level of problems concerning peers, the ASD-group was assessed as having lower levels of functioning with 51.9% under the cut-off. School problems varied between 36.9% (ADHD group) and 17.1% (TD group). Internalizing problem levels were all in the 20%-range, with the highest for the ASD group (29.6%). Alcohol misuse was most common in the ADHD group, and the DCD group reported the highest frequency of drug misuse. Those who had been screen-positive for other diagnoses (ODD, CD, OCD or ED) also reported higher frequencies of all negative outcomes, as compared with screen-negative siblings and controls. Antisocial behavior and other problems were the most common outcome in CD/ODD (over 70%), but the cases in this sample were few which limits the generalizability. Screen-positive individuals all displayed more impaired levels of functioning as compared with screen-negative siblings or controls. Screen-negative siblings of cases showed lower rates of problems than the NDP cases, but higher than the screen-negative controls.

Table 3. Screen-positive cases according to A-TAC, screen-negative siblings, and random, screen-negative controls compared to the psychosocial outcomes at age 15.

Baseline: CATSS-9/12 study (n = screen-positive in A-TAC for “proxy” diagnosis). Follow-up: Psychosocial outcomes in the clinical follow-up study, CATSS-15/DOGSS.

An analysis of sub categories of ‘Any antisocial behavior’ was done to determine the severity of antisocial behavior in the different groups (Table 4). In the ADHD group, 41.1% reported non-violent criminality, 31.6% reported violent acts, and 26.3% of their parents reported conduct problems in SDQ. Also, the rate of CD diagnosis was highest in the ADHD group: 7.4%. The ASD, LD, and DCD groups were lower in all sub categories, whereas the TD group reported both higher non-violent criminality (45.7%) and violent acts (40.0%). Of the Non-NDP groups, the ODD group surpassed the ADHD group in all sub categories except CD diagnosis (6.4%). Screen-negative siblings reported higher frequencies in both non-violent criminality (36.9%) and violent acts (34.4%) than did controls (15.2% and 23.9%). In general, non-violent criminality was more common than violent acts.

Table 4. Frequencies of subcategories of the outcome antisocial behavior (see Table 3, column 4) for screen-positive cases, screen-negative siblings, screen-negative controls, and the whole cohort at age 15.

Table 5 shows an in-depth analysis through logistic regression of the NDP subsample with signs of ADHD at ages 9 or 12 years in relation to the other groups. Having been screen-positive for ADHD at 9/12 doubled the odds for school and antisocial problems at age 15 (‘Any school problem’ OR 2.28, 95% CI 1.30–4.01; ‘Any antisocial behavior’ OR 2.67, 95% CI 1.58–4.52), and increased the odds of risky alcohol use (OR 1.97, 95% CI 1.10–3.53). Adjusting for ASD, LD, TD, DCD, and also for ODD, CD, OCD and ED, did not change the odds materially. However, when adjusting for parental education level, the significant association disappeared both for school problems and risky alcohol use, but remained for antisocial behavior (OR 2.16, 95% CI 1.14–4.09) at a slightly lower level. The significant association between ADHD and functional level remained (and even increased) in the adjusted model.

Table 5. Logistic Regression Analyses of the Associations of ADHD at Age 9/12 and Seven Psychosocial Outcomes at Age 15.


The aim of the present study was to examine to what extent signs of NDPs at ages 9 or 12 years of age is associated with negative psychosocial outcome in a range of areas at age 15, while at the same time controlling for comorbidity. Our findings indicate that negative psychosocial outcomes were common in adolescence for all childhood proxy diagnoses; both for NDPs and other conditions. A further analysis of outcomes in relation to childhood ADHD symptoms showed an increased risk of negative psychosocial outcomes during adolescence, regardless of other co-existing symptoms. However, it was common that ADHD co-existed with other symptoms, and that risk levels for different outcomes changed when comorbidity was taken into account. When considering comorbidity as well as parental education level, early ADHD-symptoms still stood out as a contributing factor to both antisocial development and lowered psychosocial functioning.

Outcome in relation to same age peers

On a group level, risk behaviors were more common in our study sample than among general Swedish 15-year olds. According to the School survey, an annual survey on criminality and drug use among Swedish 15-year olds based on self-report data, it is quite common that youth engage in illegal behavior, but the reported levels of such problems were higher in the studied cohort. For example, over one third (33.6%) of our respondents had committed at least one criminal act, compared to 20% in the national sample [77], and one third (29.5%) had committed a violent act compared to the 10% national level. It should also be noted that even though self-rate levels were higher in our study cohort than in population-based samples of the same age, parents generally reported even higher problem levels than did the participating adolescents.

Overall, there was a trend towards a higher problem load among all those who had been screen-positive for a diagnosis at ages 9 or 12 years (NDP or other) compared with those who had been screen-negative co-twins or controls, in that those who had been screen-positive reported more psychosocial problems (including antisocial behaviors) and were more often judged by assessing psychologist and child psychiatrist to have impaired functioning. Thus, not only early ADHD signs but also symptoms of other NDPs and/or other mental health and behavior problems were indicative of a more problematic adolescence. Not surprisingly, those who had been screen positive for ASD reported higher levels of peer problems, and over half were judged to have functional impairment. Given that the A-TAC, which was used for screening, has shown good to excellent psychometric properties for predicting later clinical diagnoses [52], this finding is of significant clinical relevance and stresses the importance of comprehensive assessments including symptoms other than those related to ADHD.

Outcomes of ADHD in relation to other NDPs

Those who had been screen positive for ADHD reported the highest levels of externalizing and school problems at age 15 compared to the other groups, except for those who had been screen-positive for ODD/CD, which is to be expected. Many studies have reported similar findings of negative long-term consequences of ADHD [20, 7885], but fewer studies have explored outcomes in relation to the entire spectrum of neurodevelopmental problems. When adjusting for all other diagnoses, including diagnoses other than NDPs, the doubled or tripled odds for ADHD in relation to several of the psychosocial outcomes remained. Previous studies have focused on the connection between ADHD and externalizing disorders such as oppositional defiant or conduct disorder [8688] and it is well known that ADHD frequently co-occurs with externalizing disorders, and that there is a considerable overlap between ADHD and oppositional defiant or conduct disorder. However, the interplay between these conditions is complicated, and the view that ADHD is a stable condition that precedes the onset of externalizing behaviors has been challenged [88]. However, since our study sample was young, only a few were screen positive for diagnoses such as conduct disorder, which could have underestimated the impact of early externalizing behaviors.

Socioeconomic factors and ADHD

Previous studies have also focused on the impact of socioeconomic factors on both the prevalence of ADHD and of its outcomes [80, 89, 90]. When we added a contextual factor (parental education) to our analysis, the only negative outcomes of ADHD that remained significant were antisocial behavior and impaired social functioning. Thus, the finding that parental education impacts outcome of ADHD is in line with other studies indicating that ADHD and socioeconomic factors are interconnected. In a population based study of over 800 000 Swedish individuals, Larsson et al. found a dose-dependent association with lower family income and the risk of having a child diagnosed with ADHD [91]. Currie and Stabile [80] applied the concept of human capital accumulation to study the effects of ADHD, and concluded that even though a given level of ADHD symptoms had similar effects on the academic test scores of rich and poor children, the richer children were less likely to repeat a grade. These findings were corroborated in a study by Fletcher and Wolfe [92]. In a follow-up of ADHD youth, Fried concluded that even though ADHD was the strongest predictor for academic underachievement, social class and IQ were also significant predictors of high school dropout or repeated grade [89].

Comorbidity between NDPs

It has been suggested that single NDP diagnoses are rare and should be regarded as exceptions rather than the rule [1, 42, 93]. In the DOGSS cohort, comorbidity was common, as were negative psychosocial outcomes during adolescence. The findings clearly point out ADHD as the single most important condition for antisocial behavior during adolescence. However, the results also underscore the need for comprehensive assessment of the entire neurodevelopmental spectrum and the need to take social and family factors into consideration when planning treatment. It is likely that an individual with a combination of ADHD and ASD will need a different treatment approach and support than someone with ADHD combined with e.g. TD or LD. Therefore, assessment in early childhood focusing on a wider neurodevelopmental spectrum seems highly relevant. On the same note, a wider risk assessment perspective during adolescence is important, given the higher levels of various negative outcomes reported among the 15 year olds in this study.


Signs of NDPs as well as other psychiatric diagnoses at ages 9 or 12 years are associated with a more problematic adolescence. However, despite the presence of comorbidity, early ADHD symptoms stand out as the most important risk factor for later antisocial development and impaired daily functioning.


The strengths of the present study lie in the whole-population-based cohort of screen-positive children, their screen-negative siblings, and controls, including the wide assessments of psychosocial outcomes. We have assessed a broad spectrum of NDPs simultaneously at baseline, and assessed a wide range of psychosocial outcomes at follow-up. These psychosocial outcomes were chosen to reflect key aspects of adolescent development and considered to be crucial for the further development into adulthood.


The outcome measures were based mainly on self and parent reports. The time period between the A-TAC parental assessment and the follow-up measuring the psychosocial outcomes was between three and six years. This time period in itself and the complex psychosocial problems associated with puberty may interfere with interpreting the results. There were 38 twin pairs where both twins were NDP screen-positive (23 pairs of boys and 15 pairs of girls). These twin pairs share environmental factors, which could affect our results. The effects of this have been taken care of through a robust sandwich estimator which was used to adjust for the correlated data when calculating the confidence intervals. As mentioned earlier, an analysis of the attrition group showed that there were some differences between the groups regarding sex distribution, ADHD symptoms and parental education levels. The attrition group had fewer girls, more boys, more ADHD screen-positive cases and parental education levels were more concentrated to secondary school level (Table 1). These differences suggest that the attrition group might have had a heavier problem load than the study cohort, indicating that the findings regarding psychosocial problems associated with ADHD symptoms in the study group may still be valid.

Clinical implications

Our results show that early detection and assessment is of paramount importance in children with NDPs, and that treatment and support plans need to be individualized. This includes also assessing and treating co-existing problems, such as other NDPs, and psychosocial problems such as antisocial behavior, substance abuse and peer problems. Our results underscore that children with ADHD have a high risk of becoming marginalized in society and that they may suffer serious consequences for their education, psychosocial health and future occupational life. A wider risk assessment perspective during adolescence is important, given the higher levels of various negative outcomes for the 15-year olds in this study.

Future research

Further studies are needed to explore how gender may modify the associations between ADHD and the psychosocial outcomes included in the present study. It would also be of interest to further explore if ADHD symptoms on a subthreshold level contribute to impairments and psychosocial problems in daily life.

Author Contributions

Conceived and designed the experiments: PL HA CHG. Performed the experiments: ENS TL SL. Analyzed the data: ENS. Contributed reagents/materials/analysis tools: TL SL ENS. Wrote the paper: ENS YM PL TL SL HA CHG.


  1. 1. Gillberg C., The ESSENCE in child psychiatry: Early Symptomatic Syndromes Eliciting Neurodevelopmental Clinical Examinations. Res Dev Disabil, 2010. 31(6): p. 1543–51. pmid:20634041
  2. 2. Faraone S.V., et al., The worldwide prevalence of ADHD: is it an American condition? World Psychiatry, 2003. 2(2): p. 104–13.
  3. 3. Kessler R.C., et al., The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication. Am J Psychiatry, 2006. 163(4): p. 716–23. pmid:16585449
  4. 4. Polanczyk G., et al., The worldwide prevalence of ADHD: a systematic review and metaregression analysis. Am J Psychiatry, 2007. 164(6): p. 942–8. pmid:17541055
  5. 5. Baird G., et al., Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: the Special Needs and Autism Project (SNAP). Lancet, 2006. 368(9531): p. 210–5. pmid:16844490
  6. 6. Idring S., et al., Autism spectrum disorders in the Stockholm Youth Cohort: design, prevalence and validity. PLoS One, 2012. 7(7): p. e41280. pmid:22911770
  7. 7. Kim Y.S., et al., Prevalence of autism spectrum disorders in a total population sample. Am J Psychiatry, 2011. 168(9): p. 904–12. pmid:21558103
  8. 8. Gillberg C. and Soderstrom H., Learning disability. Lancet, 2003. 362(9386): p. 811–21. pmid:13678878
  9. 9. Kotte A., et al., Autistic traits in children with and without ADHD. Pediatrics, 2013. 132(3): p. e612–22. pmid:23979086
  10. 10. Landerl K. and Moll K., Comorbidity of learning disorders: prevalence and familial transmission. J Child Psychol Psychiatry, 2010. 51(3): p. 287–94. pmid:19788550
  11. 11. Bitsko R.H., et al., Health care needs of children with Tourette syndrome. J Child Neurol, 2013. 28(12): p. 1626–36. pmid:23143730
  12. 12. Cubo E., Review of prevalence studies of tic disorders: methodological caveats. Tremor Other Hyperkinet Mov (N Y), 2012. 2.
  13. 13. Khalifa N. and von Knorring A.L., Prevalence of tic disorders and Tourette syndrome in a Swedish school population. Dev Med Child Neurol, 2003. 45(5): p. 315–9. pmid:12729145
  14. 14. Walkup J.T., et al., Tic disorders: some key issues for DSM-V. Depress Anxiety, 2010. 27(6): p. 600–10. pmid:20533370
  15. 15. Kadesjo B. and Gillberg C., Developmental coordination disorder in Swedish 7-year-old children. J Am Acad Child Adolesc Psychiatry, 1999. 38(7): p. 820–8. pmid:10405499
  16. 16. Kirby A., Sugden D., and Purcell C., Diagnosing developmental coordination disorders. Arch Dis Child, 2014. 99(3): p. 292–6. pmid:24255567
  17. 17. Biederman J., et al., Functional impairments in adults with self-reports of diagnosed ADHD: A controlled study of 1001 adults in the community. J Clin Psychiatry, 2006. 67(4): p. 524–40. pmid:16669717
  18. 18. Elkins I.J., McGue M., and Iacono W.G., Prospective effects of attention-deficit/hyperactivity disorder, conduct disorder, and sex on adolescent substance use and abuse. Arch Gen Psychiatry, 2007. 64(10): p. 1145–52. pmid:17909126
  19. 19. Faraone S.V., Attention-deficit hyperactivity disorder and the shifting sands of psychiatric nosology. Br J Psychiatry, 2013. 203(2): p. 81–3. pmid:23908337
  20. 20. Loe I.M. and Feldman H.M., Academic and educational outcomes of children with ADHD. J Pediatr Psychol, 2007. 32(6): p. 643–54. pmid:17569716
  21. 21. Molina B.S.G. and Pelham W.E., Childhood predictors of adolescent substance use in a longitudinal study of children with ADHD. Journal of Abnormal Psychology, 2003. 112(3): p. 497–507. pmid:12943028
  22. 22. Zulauf C.A., et al., The complicated relationship between attention deficit/hyperactivity disorder and substance use disorders. Curr Psychiatry Rep, 2014. 16(3): p. 436. pmid:24526271
  23. 23. Erskine H.E., et al., The global burden of conduct disorder and attention-deficit/hyperactivity disorder in 2010. J Child Psychol Psychiatry, 2014. 55(4): p. 328–36. pmid:24447211
  24. 24. Gilmour J., et al., Social communication deficits in conduct disorder: a clinical and community survey. J Child Psychol Psychiatry, 2004. 45(5): p. 967–78. pmid:15225339
  25. 25. Kanne S.M., Christ S.E., and Reiersen A.M., Psychiatric symptoms and psychosocial difficulties in young adults with autistic traits. J Autism Dev Disord, 2009. 39(6): p. 827–33. pmid:19132522
  26. 26. Piek J.P., et al., Do motor skills in infancy and early childhood predict anxious and depressive symptomatology at school age? Human Movement Science, 2010. 29(5): p. 777–86. pmid:20650535
  27. 27. Simonoff E., et al., The persistence and stability of psychiatric problems in adolescents with autism spectrum disorders. J Child Psychol Psychiatry, 2013. 54(2): p. 186–94. pmid:22934711
  28. 28. St Clair M.C., et al., A longitudinal study of behavioral, emotional and social difficulties in individuals with a history of specific language impairment (SLI). J Commun Disord, 2011. 44(2): p. 186–99. pmid:20970811
  29. 29. Westendorp M., et al., The relationship between gross motor skills and academic achievement in children with learning disabilities. Res Dev Disabil, 2011. 32(6): p. 2773–9. pmid:21700421
  30. 30. Tseng M.H., et al., Cooccurrence of problems in activity level, attention, psychosocial adjustment, reading and writing in children with developmental coordination disorder. International Journal of Rehabilitation Research, 2007. 30(4): p. 327–332. pmid:17975453
  31. 31. Twyman K.A., et al., Bullying and ostracism experiences in children with special health care needs. Journal of Developmental & Behavioral Pediatrics, 2010. 31(1): p. 1–8.
  32. 32. Anckarsater H., et al., The Child and Adolescent Twin Study in Sweden (CATSS). Twin Res Hum Genet, 2011. 14(6): p. 495–508. pmid:22506305
  33. 33. Geluk C.A., et al., Autistic symptoms in childhood arrestees: longitudinal association with delinquent behavior. J Child Psychol Psychiatry, 2012. 53(2): p. 160–7. pmid:21884523
  34. 34. Lundstrom S., et al., Childhood neurodevelopmental disorders and violent criminality: a sibling control study. J Autism Dev Disord, 2014. 44(11): p. 2707–16. pmid:23807203
  35. 35. Einarsson E., et al., Screening for attention-deficit hyperactivity disorder and co-morbid mental disorders among prison inmates. Nord J Psychiatry, 2009. 63(5): p. 361–7. pmid:19247871
  36. 36. Siponmaa L., et al., Juvenile and young adult mentally disordered offenders: the role of child neuropsychiatric disorders. J Am Acad Psychiatry Law, 2001. 29(4): p. 420–6. pmid:11785613
  37. 37. Hurtig T., et al., ADHD and comorbid disorders in relation to family environment and symptom severity. Eur Child Adolesc Psychiatry, 2007. 16(6): p. 362–9. pmid:17401612
  38. 38. Reiersen A.M., Constantino J.N., and Todd R.D., Co-occurrence of motor problems and autistic symptoms in attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry, 2008. 47(6): p. 662–72. pmid:18434922
  39. 39. Simonoff E., et al., Psychiatric disorders in children with autism spectrum disorders: prevalence, comorbidity, and associated factors in a population-derived sample. J Am Acad Child Adolesc Psychiatry, 2008. 47(8): p. 921–9. pmid:18645422
  40. 40. Antshel K.M., Zhang-James Y., and Faraone S.V., The comorbidity of ADHD and autism spectrum disorder. Expert Rev Neurother, 2013. 13(10): p. 1117–28. pmid:24117274
  41. 41. Matson J.L. and Cervantes P.E., Commonly studied comorbid psychopathologies among persons with autism spectrum disorder. Res Dev Disabil, 2014. 35(5): p. 952–62. pmid:24629541
  42. 42. Reiersen A.M., et al., Autistic traits in a population-based ADHD twin sample. J Child Psychol Psychiatry, 2007. 48(5): p. 464–72. pmid:17501727
  43. 43. Ronald A., et al., Exploring the relationship between autistic-like traits and ADHD behaviors in early childhood: findings from a community twin study of 2-year-olds. J Abnorm Child Psychol, 2010. 38(2): p. 185–96. pmid:19908138
  44. 44. Canitano R. and Vivanti G., Tics and Tourette syndrome in autism spectrum disorders. Autism, 2007. 11(1): p. 19–28. pmid:17175571
  45. 45. Ringman J.M. and Jankovic J., Occurrence of tics in Asperger's syndrome and autistic disorder. J Child Neurol, 2000. 15(6): p. 394–400. pmid:10868783
  46. 46. Rizzo R., et al., Tourette syndrome and comorbid conditions: a spectrum of different severities and complexities. J Child Neurol, 2014. 29(10): p. 1383–9. pmid:24832397
  47. 47. Sturm H., Fernell E., and Gillberg C., Autism spectrum disorders in children with normal intellectual levels: associated impairments and subgroups. Dev Med Child Neurol, 2004. 46(7): p. 444–7. pmid:15230456
  48. 48. Missiuna C., et al., Psychological distress in children with developmental coordination disorder and attention-deficit hyperactivity disorder. Res Dev Disabil, 2014. 35(5): p. 1198–207. pmid:24559609
  49. 49. Moruzzi S., et al., The nature of covariation between autistic traits and clumsiness: a twin study in a general population sample. J Autism Dev Disord, 2011. 41(12): p. 1665–74. pmid:21347613
  50. 50. Polatajko H.J. and Cantin N., Developmental coordination disorder (dyspraxia): an overview of the state of the art. Semin Pediatr Neurol, 2005. 12(4): p. 250–8. pmid:16780296
  51. 51. Reiersen A.M., Links between autism spectrum disorder and ADHD symptom trajectories: important findings and unanswered questions. J Am Acad Child Adolesc Psychiatry, 2011. 50(9): p. 857–9. pmid:21871367
  52. 52. Larson T., et al., Predictive properties of the A-TAC inventory when screening for childhood-onset neurodevelopmental problems in a population-based sample. BMC Psychiatry, 2013. 13: p. 233. pmid:24066834
  53. 53. Helles A., et al., Asperger syndrome in males over two decades: stability and predictors of diagnosis. J Child Psychol Psychiatry, 2014.
  54. 54. De Alwis D., et al., ADHD symptoms, autistic traits, and substance use and misuse in adult Australian twins. J Stud Alcohol Drugs, 2014. 75(2): p. 211–21. pmid:24650814
  55. 55. Cubo E., et al., Validation of screening instruments for neuroepidemiological surveys of tic disorders. Mov Disord, 2011. 26(3): p. 520–6. pmid:21259342
  56. 56. Hansson S.L., et al., Psychiatric telephone interview with parents for screening of childhood autism—tics, attention-deficit hyperactivity disorder and other comorbidities (A-TAC): preliminary reliability and validity. Br J Psychiatry, 2005. 187: p. 262–7. pmid:16135864
  57. 57. Larson T., et al., The autism—tics, AD/HD and other comorbidities inventory (A-TAC): further validation of a telephone interview for epidemiological research. BMC Psychiatry, 2010. 10: p. 1. pmid:20055988
  58. 58. Hallerod S.L., et al., The Autism—Tics, AD/HD and other Comorbidities (A-TAC) telephone interview: convergence with the Child Behavior Checklist (CBCL). Nord J Psychiatry, 2010. 64(3): p. 218–24. pmid:20192892
  59. 59. Shaffer D., et al., A children's global assessment scale (CGAS). Arch Gen Psychiatry, 1983. 40(11): p. 1228–31. pmid:6639293
  60. 60. Goodman R., Psychometric properties of the strengths and difficulties questionnaire. J Am Acad Child Adolesc Psychiatry, 2001. 40(11): p. 1337–45. pmid:11699809
  61. 61. Solberg M.E. and Olweus D., Prevalence estimation of school bullying with the Olweus Bully Victim Questionnaire. Aggressive Behavior, 2003. 29(3): p. 239–268.
  62. 62. Elliot D.S., Huizinga D., and Ageton S.S.E., Explaining delinquency and drug use. 1985: Beverly Hills: Sage.
  63. 63. Junger-Tas J., Terlouw G.-J., and K.M. W. , Delinquent behavior among young people in the western world: first results of the international self-report delinquency study. Studies on crime and justice, ed. Justice R.-M.o.. 1994: Kugler Publications—Amsterdam / New York.
  64. 64. Junger-Tas J. and Marshall I.H., The Self-Report Methodology in Crime Research. The University of Chicago, 1999.
  65. 65. Ring, Stöld, våld och droger bland pojkar och flickor i årskurs nio—Resultat från tre självdeklarationsundersökningar, B.r.C. prevention), Editor. 2000.
  66. 66. Ayton A., Keen C., and Lask B., Pros and cons of using the Mental Health Act for severe eating disorders in adolescents. Eur Eat Disord Rev, 2009. 17(1): p. 14–23. pmid:18666096
  67. 67. Bella T., et al., Psychosocial functioning in offspring of parents with bipolar disorder. J Affect Disord, 2011. 133(1–2): p. 204–11. pmid:21463899
  68. 68. Bird H.R., et al., Impairment in the epidemiological measurement of childhood psychopathology in the community. J Am Acad Child Adolesc Psychiatry, 1990. 29(5): p. 796–803. pmid:2228936
  69. 69. Canino G., et al., The DSM-IV rates of child and adolescent disorders in Puerto Rico: prevalence, correlates, service use, and the effects of impairment. Arch Gen Psychiatry, 2004. 61(1): p. 85–93. pmid:14706947
  70. 70. Leaf P.J., et al., Mental health service use in the community and schools: results from the four-community MECA Study. Methods for the Epidemiology of Child and Adolescent Mental Disorders Study. J Am Acad Child Adolesc Psychiatry, 1996. 35(7): p. 889–97. pmid:8768348
  71. 71. Milne J.M., et al., Frequency of phobic disorder in a community sample of young adolescents. J Am Acad Child Adolesc Psychiatry, 1995. 34(9): p. 1202–11. pmid:7559315
  72. 72. Petersen D.J., et al., The population prevalence of child psychiatric disorders in Danish 8- to 9-year-old children. Eur Child Adolesc Psychiatry, 2006. 15(2): p. 71–8. pmid:16477477
  73. 73. Steinhausen H.C. and Metzke C.W., Global measures of impairment in children and adolescents: results from a Swiss community survey. Aust N Z J Psychiatry, 2001. 35(3): p. 282–6. pmid:11437800
  74. 74. Lundh A., et al., Children's Global Assessment Scale (CGAS) in a naturalistic clinical setting: Inter-rater reliability and comparison with expert ratings. Psychiatry Res, 2010. 177(1–2): p. 206–10. pmid:20334931
  75. 75. Lundh A., et al., A comparison of seminar and computer based training on the accuracy and reliability of raters using the Children's Global Assessment Scale (CGAS). Adm Policy Ment Health, 2012. 39(6): p. 458–65. pmid:21785889
  76. 76. Lundh A., On the Children’s Global Assessment Scale (CGAS). 2012: Karolinska Institutet, Stockholm.
  77. 77. Brå, Kriminalstatistik 2011. Brå Rapport 2012. Vol. Rapport 2012:11. 2012, Stockholm, Sweden: Brottsförebyggande rådet.
  78. 78. Barbaresi W.J., et al., Long-term school outcomes for children with attention-deficit/hyperactivity disorder: a population-based perspective. J Dev Behav Pediatr, 2007. 28(4): p. 265–73. pmid:17700078
  79. 79. Biederman J., et al., Impact of executive function deficits and attention-deficit/hyperactivity disorder (ADHD) on academic outcomes in children. J Consult Clin Psychol, 2004. 72(5): p. 757–66. pmid:15482034
  80. 80. Currie J. and Stabile M., Child mental health and human capital accumulation: the case of ADHD. J Health Econ, 2006. 25(6): p. 1094–118. pmid:16730082
  81. 81. Frazier T.W., et al., ADHD and achievement: meta-analysis of the child, adolescent, and adult literatures and a concomitant study with college students. J Learn Disabil, 2007. 40(1): p. 49–65. pmid:17274547
  82. 82. Fredriksen M., et al., Childhood and persistent ADHD symptoms associated with educational failure and long-term occupational disability in adult ADHD. Atten Defic Hyperact Disord, 2014. 6(2): p. 87–99. pmid:24497125
  83. 83. Sayal K., et al., Impact of early school-based screening and intervention programs for ADHD on children's outcomes and access to services: follow-up of a school-based trial at age 10 years. Arch Pediatr Adolesc Med, 2010. 164(5): p. 462–9. pmid:20439798
  84. 84. Pingault J.B., et al., Childhood hyperactivity, physical aggression and criminality: a 19-year prospective population-based study. PLoS One, 2013. 8(5): p. e62594. pmid:23658752
  85. 85. Strine T.W., et al., Emotional and behavioral difficulties and impairments in everyday functioning among children with a history of attention-deficit/hyperactivity disorder. Prev Chronic Dis, 2006. 3(2): p. A52. pmid:16539793
  86. 86. Biederman J., et al., The longitudinal course of comorbid oppositional defiant disorder in girls with attention-deficit/hyperactivity disorder: findings from a controlled 5-year prospective longitudinal follow-up study. J Dev Behav Pediatr, 2008. 29(6): p. 501–7. pmid:19077845
  87. 87. Galera C., et al., Childhood and adolescent hyperactivity-inattention symptoms and academic achievement 8 years later: the GAZEL Youth study. Psychol Med, 2009. 39(11): p. 1895–906. pmid:19335935
  88. 88. Kuja-Halkola R., et al., Codevelopment of ADHD and externalizing behavior from childhood to adulthood. J Child Psychol Psychiatry, 2014.
  89. 89. Fried R., et al., Is ADHD a Risk Factor for High School Dropout? A Controlled Study. J Atten Disord, 2013.
  90. 90. Galera C., et al., Attention problems in childhood and adult substance use. J Pediatr, 2013. 163(6): p. 1677–1683 e1. pmid:23972646
  91. 91. Larsson H., et al., Family income in early childhood and subsequent attention deficit/hyperactivity disorder: a quasi-experimental study. J Child Psychol Psychiatry, 2014. 55(5): p. 428–35. pmid:24111650
  92. 92. Fletcher J. and Wolfe B., Child mental health and human capital accumulation: the case of ADHD revisited. J Health Econ, 2008. 27(3): p. 794–800. pmid:18221807
  93. 93. Anckarsater H., Beyond categorical diagnostics in psychiatry: Scientific and medicolegal implications. Int J Law Psychiatry, 2010. 33(2): p. 59–65. pmid:20080303