5.1 STUDY I OBJECTIVE
In Study I, the aim was to validate the diagnostic predictive properties of the A-TAC interview.
METHODS
The study included NDP screen-positive twins from three birth year cohorts (1993-1995) in CATSS, along with their co-twins and randomly selected, healthy controls, who all were invited to a comprehensive clinical follow-up at age 15 (N = 452, participation rate 52%).
RESULTS
Sensitivity and specificity of the A-TAC scores for predicting later 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 (ASDs). Among children who were screen-positive for an ASD at age 9/12, 48% received a clinical diagnosis of ASDs at 15. For ADHD, the
corresponding figure was also 48%, for LDs 16%, and for TDs 60%. Between 4% and 35%
of screen-positive children did not receive any diagnosis at the clinical follow-up three years later. Among screen-negative controls, prevalence of ASDs, ADHD, LDs, and TDs was 0%, 7%, 4%, and 2% respectively.
5.2 STUDY II OBJECTIVE
The aim of Study II was to assess long-term psychosocial outcomes in adolescence associated with childhood NDPs, focusing on ADHD but considering other NDPs as well.
METHODS
We used the same cohort as in Study I: This included NDP screen-positive twins from three birth year cohorts (1993-1995) in CATSS, along with their co-twins and randomly selected, healthy controls, who all were invited to a comprehensive clinical follow-up at age 15 (N = 450, participation rate 52%, with two cases excluded due to incomplete data). The included psychosocial outcomes were peer problems, school problems, internalizing problems, antisocial behavior, alcohol misuse, drug misuse, and daily functioning.
RESULTS
Among the NDPs, antisocial behavior was the most common outcome across all NDP diagnoses. We performed a logistic regression which demonstrated that having been screen-positive for ADHD at 9/12 doubled the odds for school related and antisocial problems at age 15, and also increased the odds of risky alcohol use. Adjusting for ASD, LD, TD, DCD, and
also for ODD, CD, OCD and ED, did not change the odds substantially. When adjusting for parental education level, however, the significant association disappeared both for school problems and risky alcohol use, but remained for antisocial behavior. The significant association between ADHD and functional level remained in the adjusted model.
5.3 STUDY III OBJECTIVES
The aim of Study III was to examine possible gender differences in psychosocial outcome at age 15. Moreover, we explored outcomes in relation to different levels of childhood ADHD symptoms.
METHODS
The CATSS-15 cohort from 1993-1997 was used for the follow-up. A sample of Swedish twins 9-12 years of age (N=4635) was screened for ADHD symptoms through a telephone interview with their parents. Participants were grouped into the following groups based on the degree of ADHD symptoms: screen-negative, screen-intermediate, and screen-positive. At follow-up (age 15), parents and teenagers completed questionnaires regarding: 1) ongoing hyperactivity/inattention, 2) peer problems, 3) school problems, 4) internalizing problems, 5) antisocial behavior, 6) alcohol misuse, and 7) drug misuse.
RESULTS
High levels of ADHD symptoms at age 9/12 were associated with high levels of hyperactivity and impulsivity at follow-up. After adjusting for baseline comorbidity, high levels of ADHD symptoms were also associated with higher probability for all outcomes in both genders, except for alcohol misuse in girls and drug misuse in boys. The results demonstrated a higher probability of the following outcomes among girls (but not boys) in the screen-positive group: peer problems, school problems (truancy), internalizing problems, and drug misuse.
The probability of drug use was significantly higher in girls than in boys.
Girls reported more internalizing problems than boys across all ADHD symptom levels (OR:
1.50, 95% CI 1.20-1.87). Girls who had been screen-positive for ADHD displayed higher problem levels during adolescence than their male counterparts on several of the psychosocial outcomes, particularly for drug misuse (OR: 1.64, 95% CI 1.25-2.14).
In both genders, the screen-intermediate groups reported higher levels of psychosocial problems than the screen-negatives.
5.4 STUDY IV OBJECTIVES
In Study IV we aimed to explore the association between ADHD and internalizing problems.
Furthermore, we aimed to investigate the relative contribution of genetic and environmental influences to symptoms of ADHD and internalizing problems in childhood on symptoms of internalizing problems in adolescence, including gender differences.
METHODS
The CATSS-15 cohort from 1993-1997 was used for the follow-up. A sample of Swedish twins 9-12 years of age (N=4635) was screened for internalizing and ADHD symptoms through a telephone interview with their parents. At follow-up (age 15), parents completed questionnaires regarding internalizing problems in the adolescent. In the analyses, the internalizing problems at 15 were allowed to be influenced by ADHD and internalizing problems at the earlier age 9/12. The variance in internalizing problems at 15 was divided into effects of ADHD and internalizing at 9/12 and new effects at age 15. Each source was divided into additive genetic effects, shared environmental effects and unique environmental effects.
RESULTS
We found that the data was appropriate for quantitative genetic modeling. There was a positive correlation between ADHD and internalizing problems in both boys and girls. The strength of the associations between ADHD and internalizing problems at baseline (9/12) were similar to the corresponding associations between ADHD at baseline and internalizing at 15.
Higher correlations in MZ compared to DZ twins indicated a genetic influence in the phenotype (intra-class correlations), or in the covariation between phenotypes (cross-twin-cross-trait-correlations). The greatest difference was observed within the phenotypes (i.e.
adhd-9, internalizing-9, or internalizing-15). Also, MZ twins consistently displayed higher cross-twin-cross-trait correlations than DZ. Boys and girls had a slightly different pattern of intra-class and cross-twin-cross-trait correlations.
We fitted the model including A, C and E sources of covariance, as well as both quantitative and qualitative gender differences. We could exclude the C source of covariance without the model explaining the data significantly worse. But nothing else could be excluded from the model, so the model with A and E-sources of covariance and quantitative and qualitative gender differences was the model that best fitted the data.
Quantitative genetic results – explained variance in internalizing-15
The fraction of the variance explained by genetic effects for internalizing-15 (A) was 30% for girls and 35% for boys. The contribution of genetic effects from adhd-9 and internalizing-9 differed between the genders. In girls, the genetic effect from adhd-9 explained 11% of the variance in internalizing-15, whereas in boys, the genetic effect from adhd-9 only 1% of the variance in internalizing-15. In both girls and boys, internalizing-9-adhd-9 (A) explained 3%
of the variance. Also, at age 15 new non-shared environmental effects accounted for a little more than half of the variation in internalizing problems at age 15 in both genders.