Skip to main content
NCBI home page
Springer logoLink to Springer
. 2024 Oct 30;55(1):1–11. doi: 10.1007/s10519-024-10204-y

Using Bifactor Twin Modeling to Assess the Genetic and Environmental Dimensionality of Adult ADHD Symptoms

Jacob Knyspel 1,2,3, Geneviève Morneau-Vaillancourt 1,2,3, Thalia C Eley 1,2,3,✉
PMCID: PMC11790749  PMID: 39476281

Abstract

Attention Deficit Hyperactivity Disorder (ADHD) is a common and heritable neurodevelopmental condition that has been the subject of a wealth of genetics research. Because ADHD has an early age of onset, most of this work has focused on children, meaning that less is known about the genetics of ADHD in adults. Additionally, while much research has assessed the heritability of ADHD as a general dimension, less has assessed the heritability of individual subtypes (inattention, hyperactivity) or symptoms of ADHD. It therefore remains unclear whether the genetic factors underlying ADHD symptoms conform to a unidimensional or multidimensional structure. The aim of this study was to assess the genetic and environmental dimensionality of adult ADHD symptoms. We analyzed data from 10,454 twins of the Twins Early Development Study, who provided self-reports of ADHD symptoms using the Conners scale at age 21 years. The data conformed well to a psychometric bifactor model, providing support for a general dimension of ADHD in addition to secondary dimensions for inattention and hyperactivity. However, a bifactor independent pathway twin model provided support for a general dimension only at the level of non-shared environmental effects and not additive genetic or shared environmental effects. This suggests that symptoms of ADHD cluster together under a general dimension of non-shared environmental effects, although the two subtypes of ADHD (inattention and hyperactivity) are meaningfully genetically distinct. We found the overall heritability of ADHD to be 40%, comparable with previous estimates for adult ADHD symptoms. Our results provide useful insights into the genetic and environmental architecture of specific ADHD symptoms.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10519-024-10204-y.

Keywords: ADHD, Twin study, Adults, Heritability, Behavioral genetics, Bifactor

Attention Deficit Hyperactivity Disorder (ADHD) is a common and heterogeneous neurodevelopmental condition characterized by attentional problems, hyperactivity, and impulsivity (American Psychiatric Association 2013). It has a mean age of diagnosis of approximately 7–9 years (Rocco et al. 2021), although it is believed to be noticeable to caregivers as early as 2 years of age (van Lieshout et al. 2016), making it one of the earliest onset psychiatric conditions. ADHD is especially prevalent in young people, with global prevalence estimates of 7.6% in children aged 3–12 years and 5.6% in adolescents aged 12–18 years (Salari et al. 2023). Unlike other neurodevelopmental conditions, such as autism spectrum disorder and intellectual disability, the symptomology of ADHD often changes considerably with age. Many young people with ADHD no longer meet diagnostic thresholds later in life (Vos and Hartman 2022), with the prevalence of ADHD decreasing to 2.5% in adults (Song et al. 2021). This might reflect a true decrease in ADHD symptoms with age, although making such a conclusion is limited by the fact that studies have often relied on retrospective reporting during adulthood (e.g. Vos and Hartman 2022). Research has also suggested that symptoms of ADHD can emerge for the first time during adulthood (Agnew-Blais et al. 2016; Moffitt et al. 2015), despite childhood symptoms typically being required for an ADHD diagnosis. This has led researchers to consider whether childhood and adulthood ADHD are qualitatively distinct from one another (Castellanos 2015). However, the role that additional factors such as lifestyle changes or assessment methods (diagnosis vs. symptoms) might also explain observed differences between childhood and adulthood ADHD requires further investigation, ideally via prospective longitudinal studies. Despite the wealth of available pharmacological and psychological interventions for ADHD in both children and adults (Caye et al. 2019), the prognosis for ADHD remains poor in many cases. Treatment adherence is notably low among people with ADHD (Biederman et al. 2019), and even among those who adhere to treatment, ADHD remains associated with adverse life outcomes including unemployment, substance abuse, and criminality (Erskine et al. 2016). ADHD is thus a highly concerning condition in need of further research regarding its etiology, pathophysiology, and treatment (Cattoi et al. 2022).

In children, twin and family studies have shown that genetic factors account for a substantial proportion of the variance in ADHD, with consistent heritability estimates of 70–80% (Faraone and Larsson 2019). In adolescents and adults, who have been the focus of fewer studies, heritability estimates are less consistent. Most adolescent and adult twin studies have used self-report ADHD symptom scales, producing heritability estimates of only 30–40% (e.g., Boomsma et al. 2010; Haberstick et al. 2008; Larsson et al. 2013; Li et al. 2020; van den Berg et al. 2006). These findings have led researchers to conclude that a qualitative distinction exists between childhood and adulthood ADHD, which would be congruent with the changing symptomology of ADHD with age. A further possibility is that the heritability of ADHD simply decreases with age and that environmental factors account for a larger proportion of the variance in ADHD over time (Boomsma et al. 2010). However, there is also evidence suggesting that the heritability of clinically diagnosed ADHD in young adulthood, assessed via public health records, is around 70% (Larsson et al. 2014), a much greater heritability estimate than those from studies relying on self-reports (30–40%). These inconsistent findings may be partly attributable to how ADHD is assessed across studies, such as through self-reports, parent-reports, and health records (see Brikell et al. 2015; Faraone and Larsson 2019). Indeed, when combining self-reports and parent-reports into latent factors, Chang et al. (2013) found that the heritability of attention problems remained consistent from ages 8 to 20. Regardless, this difference highlights the need for research focusing on the genetics of ADHD specifically in adults. Better understanding of how the influence of genetics changes with age might help to explain why the presentation of ADHD often differs among children and adults. It might also help to inform interventions for the condition that are better tailored to child and adult populations, improving support for people with ADHD throughout their lifespan.

If researchers wish to use twin data to study the genetics of ADHD symptoms, several statistical models are relevant. The simplest is a univariate model which estimates genetic and environmental effects on a total ADHD symptom score. This univariate model has by far been the most common method for estimating the heritability of adult ADHD symptoms (e.g. Boomsma et al. 2010; Larsson et al. 2013; van den Berg et al. 2006). An alternative is the common pathway model (Kendler et al. 1987), which estimates genetic and environmental effects on a latent factor which itself loads onto numerous observed variables such as ADHD symptoms. Constructing a latent factor out of observed variables can help to minimize measurement error among these variables, potentially providing a more precise and less biased estimate of heritability. A further alternative is the independent pathway model (Kendler et al. 1987). In this model, shared genetic and environmental effects among variables are estimated directly on each variable, rather than via a latent factor. Unlike the common pathway model, the independent pathway model does not assume that the shared genetic and environmental effects on a set of variables are mediated through the same latent factor. It is therefore useful for assessing whether the genetic and environmental effects on variables have unique factor structures (see Franić et al. 2013).

If multiple sets of variables are included in common and independent pathway models, such as symptoms reflecting the inattention and hyperactivity subtypes of ADHD, useful extensions to these models become possible. One extension that has seen little use in research is the inclusion of a general latent factor that represents the unidimensionality of all the observed variables in a model (e.g., Pelt et al. 2022; Zakharin & Bates, 2022; see Fig. 1). Structural equation models which include general latent factors, called “bifactor” models (Holzinger and Swineford 1937), have become increasingly popular in research as they provide an ideal method for modeling the covariances of variables at multiple levels of generality (Reise et al. 2023). This includes covariances at the general level (what a variable shares with all other variables) and group level (what a variable shares with a subset of other variables). They are especially relevant for ADHD, since much psychometric research has established that ADHD symptoms conform well to bifactor models (for a review, see Arias et al. 2018). This means that while ADHD symptoms can load onto latent factors representing different subtypes of the condition, they typically load most strongly onto a shared general factor. It indicates that even after accounting for the presence of subtypes, ADHD symptoms remain unidimensional enough that a general dimension of ADHD is psychometrically meaningful. This general dimension could be said to correspond to a combined-type ADHD presentation that is meaningfully distinct from subtype-specific presentations (Milich et al. 2001). If general latent factors are included in twin models, these models can be used to provide insight into the genetic and environmental dimensionality of symptoms. These terms refer to the extent to which symptoms of a condition share genetic and environmental effects within specific subsets (e.g., inattention, hyperactivity), across all of subsets together (e.g., overall ADHD), or both.

Fig. 1.

Fig. 1

Open in a new tab

Bifactor common pathway model (a) and bifactor independent pathway model (b), both of which include general latent factors. Note X = Observed variable; A = Additive genetic effect; C = Shared environmental effect; E = Non-shared environmental effect; F = Latent factor; G = General latent factor

Our aim with this study was to assess the genetic and environmental dimensionality of ADHD symptoms in a sample of young adult twins as measured using the Conners self-report symptom scale (Conners 2008). First, we performed a psychometric analysis in which an overall bifactor structure was fitted to the ADHD symptoms. Based on previous findings (Arias et al. 2018), we hypothesized that the data would conform to this bifactor structure well. Second, we performed genetic analyses in which a bifactor common pathway model and bifactor independent pathway model were used to the assess the genetic and environmental dimensionality of ADHD symptoms in young adulthood. We had no specific hypothesis as to whether the genetic and environmental dimensionality of ADHD symptoms would be equivalent, as no studies have yet examined the genetic and environmental dimensionality of ADHD symptoms using bifactor twin models. Based on existing literature (e.g., McLoughlin et al. 2007; Nikolas and Burt 2010), we hypothesized that the overall heritability of the two ADHD subtypes, and by extension an ADHD general factor, would be roughly equivalent.

Method

Sample

Existing data were analyzed from the Twins Early Development Study (TEDS; Lockhart et al. 2023), an ongoing twin cohort study investigating the development of 13,759 pairs of twins born in the years 1994–1996 in England and Wales. Specifically, self-reported ADHD symptom data collected at age 21 were used. This included data from 10,454 twins, 35.1% of whom were monozygotic and 64.9% of whom were dizygotic. The mean age of the sample was 22.85 (SD = 0.88). Of the sample, 59.9% and 40.1% were assigned female and male at birth respectively. TEDS has ethical approval from Kings College London Research Ethics Committee (References: PNM/09/10–104 and HR/DP-20/21–22060). Prior to accessing the data, this study was pre-registered at https://osf.io/emx8y.

Measures

ADHD symptoms were measured using the short self-report version of the Conners scale (3rd edition; Conners 2008). This scale includes 20 items assessing ADHD symptoms (e.g. “It is hard for me to sit still”) which are self-rated on an ordinal scale from 0 (“Not true at all”) to 3 (“Definitely true”) in response to the following question: “To what extent do the following statements accurately describe you?”. This question was not timeframe specific. There are two subscales: an 11-item subscale for inattention symptoms and a 9-item subscale for hyperactivity symptoms. In this sample, the internal consistency of the Conners scale was found to be strong (Cronbach’s α = 0.89, McDonald’s ω = 0.89), as was the internal reliability of the inattention subscale (α = 0.88, ω = 0.88) and hyperactivity subscale (α = 0.82, ω = 0.83).

Psychometric Analysis

The psychometric analysis assessed how well the Conners scale conforms to a bifactor structure comprising one general latent factor (overall ADHD) and two specific latent factors (inattention and hyperactivity). Direct Schmid-Leiman orthogonalisation (Waller 2018) was used to estimate this bifactor structure, using the R package “fungible” (Waller et al. 2023). This method was chosen because it performed best in a recent large simulation study (Giordano and Waller 2020). To minimize bias, only the Conners scale item scores from one randomly selected twin per pair were included. Among these twins, 17% had missing data, resulting in a sample of 4338 twins following listwise deletion. The meaningfulness of the general latent factor within the structure was evaluated in terms of standardized factor loadings (λ) in addition to the internal consistency (ωh, cutoff = > 0.5; McDonald 1999), construct reliability (H-index, cutoff = > 0.7; Hancock & Muller, 2001), and explained common variance (ECV; Ten Berge and Sočan 2004) of the general latent factor (for a review of these bifactor-specific indices, see Rodriguez et al. 2016). These indices were selected because model fit statistics do not necessarily provide useful information regarding whether a general latent factor has a meaningful interpretation within a bifactor model (Bonifay and Cai 2017; Reise et al. 2016). Based on general guidelines (e.g., Pituch and Stevens 2016), standardized factor loadings greater than 0.4 were considered to be adequate. Since the bifactor structure was orthogonal in nature, meaning that factors were constrained to be uncorrelated, correlations between factors were not assessed.

Genetic Analyses

We then examined the genetic and environmental dimensionality of the Conners scale items by conducting and comparing bifactor common pathway and independent pathway models. The models included additive genetic, shared environmental, and non-shared environmental effects (i.e., ACE models; see Rijsdijk and Sham 2002; for a description of the twin method). We coded models using the R packages “OpenMx” (Boker et al. 2011) and “umx” (Bates et al. 2019). They were run using the SLSQP optimizer, available in the “OpenMx” package, and 95% confidence intervals were bootstrapped with 1000 replications for all statistics of interest. As with the psychometric analysis, 17% of twins had missing data, which were handled via full information maximum likelihood estimation. The age, sex, and cohort of twins were included as covariates by residualizing them out of the data prior to the fitting of the twin models. As with the psychometric analysis, we assessed the meaningfulness of the general latent factors within each twin model (i.e., the overall general latent factor within the bifactor common pathway model and the separate genetic and environmental general latent factors within the bifactor independent pathway model) in terms of standardized factor loadings, internal consistency, construct reliability, and explained common variance. For each twin model, we tested alternative versions in which additive genetic and/or shared environmental effects were dropped to assess the redundancy of these effects and identify the most parsimonious models. We compared models using five fit statistics: the Root Mean Squared Error of Approximation (RMSEA), Comparative Fit Index (CFI), -2 Log-Likelihood (-2LL), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). Based on previous research (Hu and Bentler 1999), we used cutoffs of RMSEA < 0.06 and CFI > 0.95 to indicate strong model fit. We also performed likelihood ratio tests to compare the different versions of each twin model.

Data Availability

The data used in this study are not openly available. TEDS data can be accessed by researchers by application, see: https://www.teds.ac.uk/researchers/teds-data-access-policy.

Results

Psychometric Analysis

Among the entire sample, the mean Conners scale total score was 13.5 (SD = 8.8) out of a possible 60. The estimated psychometric bifactor structure showed that the general latent factor had adequate internal consistency (ωh = 0.62, cutoff = > 0.5) and construct reliability (H-index = 0.88, cutoff = > 0.7), explaining 52% of the common variance across items. The standardized factor loadings within the bifactor structure are presented in Table 1. All items except for two (“Dislikes complex tasks”, “Prefers being on the go”) had adequate loadings (λ > 0.4) onto the general factor, with most items (60%) loading more strongly onto the general factor than subscale factors. All items correctly loaded onto their intended subscale factor more strongly than the other subscale factor. Together these findings indicate that the Conners scale item data conformed sufficiently well to a bifactor structure to be suitable for subsequent genetic analyses with bifactor twin models.

Table 1.

Standardized factor loadings (λ) of Conners scale items within the psychometric bifactor structure

Item Brief Item Description Subscale General Inattention Hyperactivity
1 Struggles paying attention Inattention 0.53 0.50 0.04
2 Makes mistakes Inattention 0.40 0.38 0.01
3 Struggles focusing on tasks Inattention 0.59 0.57 0.02
4 Struggles focusing on conversations Inattention 0.56 0.50 0.06
5 Struggles following instructions Inattention 0.56 0.52 0.04
6 Struggles finishing tasks Inattention 0.52 0.55 -0.03
7 Struggles with organisation Inattention 0.46 0.53 -0.07
8 Dislikes complex tasks Inattention 0.34 0.38 -0.04
9 Loses things Inattention 0.47 0.38 0.09
10 Gets distracted Inattention 0.59 0.46 0.13
11 Forgets things Inattention 0.50 0.47 0.03
12 Struggles sitting still Hyperactivity 0.53 0.14 0.39
13 Leaves seat inappropriately Hyperactivity 0.57 0.19 0.38
14 Feels restless Hyperactivity 0.56 0.19 0.36
15 Struggles remaining quiet Hyperactivity 0.58 0.10 0.48
16 Prefers being on the go Hyperactivity 0.29 -0.08 0.37
17 Talks too much Hyperactivity 0.41 -0.13 0.54
18 Blurts out answers to questions Hyperactivity 0.47 -0.09 0.56
19 Struggles waiting turns Hyperactivity 0.53 -0.03 0.55
20 Interrupts others Hyperactivity 0.45 -0.02 0.48
Open in a new tab

Note Brief item descriptions are not the full item statements

Genetic Analyses

Model fit statistics for each twin model, as well as the results of the likelihood-ratio tests used to compare nested models, are presented in Table 2. As this table shows, bifactor independent pathway models had consistently better fit statistics than bifactor common pathway models, which indicates that the shared genetic and environmental effects on Conners scale items were best modeled as independent and not mediated by a common latent factor. The best fitting bifactor common pathway model was the AE version, while the best fitting bifactor independent pathway model was the ACE version. Only the best fitting bifactor independent pathway model met the pre-specified cutoffs for strong model fit (RMSEA < 0.06, CFI > 0.95).

Table 2.

Model fit statistics and likelihood ratio tests for the bifactor common pathway and bifactor independent pathway twin models

Model RMSEA CFI -2LL AIC BIC LRT p
Bifactor common pathway model
 ACE 0.027 0.906 336795.3 337060.0 337488.6
 AE 0.027 0.907 336795.3 337011.8 337365.0 > 0.999
 CE 0.030 0.888 338045.8 338262.3 338615.4 < 0.001
 E 0.029 0.888 338045.8 338214.5 338491.8 < 0.001
Bifactor independent pathway model
 ACE 0.018 0.961 333107.8 333523.9 334182.6
 AE 0.020 0.952 333773.3 334061.1 334525.6 < 0.001
 CE 0.023 0.933 335014.4 335302.3 335766.8 < 0.001
 E 0.029 0.888 338045.8 338208.4 338475.7 < 0.001
Open in a new tab

Note A = Additive genetic effects; C = Shared environmental effects; E = Non-shared environmental effects; RMSEA = Root mean squared error of approximation; AIC = Akaike information criterion; BIC = Bayesian information criterion; -2LL = -2 Log-Likelihood; LRT = Likelihood-ratio test. Likelihood-ratio tests compared models to their corresponding ACE versions. Bold indicates the best fitting version

The genetic and environmental effect estimates from the models are presented in Table 3. As this table shows, the bifactor common pathway model produced heritability estimates of 40% for the ADHD general factor, 42% for the inattention subscale factor, and 33% for the hyperactive subscale factor. Although the heritability of inattention items appeared larger than the heritability of hyperactive items, this difference was not statistically significant (see Table 3 for confidence intervals). In the bifactor independent pathway model, the general factor accounted for 50% of the heritability of the Conners scale items on average, while the inattention and hyperactivity factors both accounted for 19%. The remaining 12% was accounted for by genetic effects that were unique to individual items. In contrast to the common pathway model, the general factor in the independent pathway model therefore accounted for a larger proportion of the overall heritability of Conners scale items than both the inattention and hyperactivity factors combined.

Table 3.

Genetic and environmental effect estimates from best fitting bifactor common pathway and bifactor independent pathway twin models

Model A [95% CI] C [95% CI] E [95% CI]
Bifactor common pathway model
 General 0.40 [0.36, 0.45] - 0.60 [0.55, 0.64]
 Inattention 0.42 [0.36, 0.47] - 0.58 [0.53, 0.64]
 Hyperactivity 0.33 [0.26, 0.39] - 0.67 [0.61, 0.74]
Bifactor independent pathway model
 General 0.08 [0.05, 0.12] 0.02 [0.00, 0.04] 0.18 [0.15, 0.22]
 Inattention 0.03 [0.02, 0.04] 0.01 [0.00, 0.01] 0.10 [0.09, 0.11]
 Hyperactivity 0.03 [0.02, 0.04] 0.01 [0.01, 0.01] 0.05 [0.04, 0.05]
 Unique 0.02 [0.02, 0.03] 0.00 [0.00, 0.01] 0.46 [0.46, 0.47]
Open in a new tab

Note A = Additive genetic effect; E = Non-shared environmental effect; CI = Confidence interval (bootstrapped with 1000 iterations). For the bifactor common pathway model, the genetic and environmental effect estimates for each latent factor (general, inattention, hyperactivity) are reported. For the bifactor independent pathway model, genetic and environmental effects are broken down into the effects that were either shared among factors (general, inattention, hyperactivity) or unique to items. These effects were calculated as the means of the squared standardized effects across items. The squared standardized effects for each individual item are reported in Supplementary Table 1

The indices used to evaluate the meaningfulness of the general latent factors within the bifactor independent pathway model are reported in Table 4, along with those from the psychometric analysis as a reference. Only the non-shared environmental general factor was found to have adequate indices (ωh = 0.67, H-index = 0.86), these indices being similar to those from the psychometric analysis. The additive genetic general factor had a marginally adequate ωh value but inadequate H-index (ωh = 0.51, H-index = 0.68), and the shared environmental general factor had wholly inadequate indices (ωh < 0.01, H-index = 0.16). This indicates that the psychometric bifactor structure of Conners scale items was sufficiently reflected only in their non-shared environmental effects. By extension, it suggests that the concept of a general dimension for either additive genetic or shared environmental effects was not supported by the data. The standardized factor loadings within each bifactor structure from the twin models are reported in Supplementary Tables 2–5.

Table 4.

General factor indices for the ACE bifactor independent pathway twin model, in addition to those from the prior psychometric analysis

Model ωh H-Index ECV
Bifactor independent pathway model
 Additive genetic effects 0.51 0.68 0.59
 Shared environmental effects < 0.01 0.16 0.31
 Non-shared environmental effects 0.67 0.86 0.59
Psychometric analysis (reference)
 Psychometric bifactor structure 0.62 0.88 0.52
Open in a new tab

Note ωh = Omega hierarchical (cutoff = > 0.5); H-index = Construct reliability (cutoff = > 0.7); ECV = Explained common variance

Discussion

The aim of this study was to assess the genetic and environmental dimensionality of adult ADHD symptoms, or the extent to which ADHD symptoms share genetic and environmental effects across subtypes as opposed to within subtypes. We applied bifactor twin models to ADHD symptom data measured using the Conners scale from adult twins aged 21–26. First, we showed that the data conformed well to a psychometric bifactor structure with three latent factors (general, inattention, hyperactivity), in line with previous findings about ADHD symptoms (Arias et al. 2018). This suggests that at a phenotypic level, a general dimension of ADHD that can be measured beyond the individual domains of inattention and hyperactivity is highly replicable. This general dimension is theoretically important because it provides a useful reference point for understanding why some people with ADHD exhibit both symptoms of inattention and hyperactivity (i.e., a combined-type ADHD presentation) while others do not.

We then used a bifactor common pathway model and a bifactor independent pathway model. The bifactor independent pathway model fit the data better than the bifactor common pathway model, suggesting that genetic and environmental effects on adult ADHD symptoms were best modeled as independent and not mediated by a common latent factor. Within the bifactor independent pathway model, only non-shared environmental effects, and not additive genetic or shared environmental effects, conformed well to a bifactor structure. This is because the general latent factors estimated for additive genetic and shared environmental effects did not reach the cutoffs required to be considered meaningful. It suggests that the non-shared environmental effects on ADHD symptoms are unidimensional enough that a general dimension of ADHD (i.e., a combined-type presentation) be considered meaningful, even after accounting for subtypes, although the same is not true of genetic and shared environmental effects. This has important implications for understanding the etiology of ADHD. Unidimensionality at the level of non-shared environmental effects indicates that environmental risk factors that can differ among twins, such as childhood head trauma (Adeyemo et al. 2014) or eczema (van der Schans et al. 2017), have a roughly equivalent association with inattention and hyperactivity, thereby contributing to a combined-type presentation. Conversely, a lack of unidimensionality at the level of shared environmental effects indicates that environmental risk factors shared by twins, such as maternal pre-pregnancy obesity (Jenabi et al. 2019), might be more uniquely associated with either inattention or hyperactivity (for a review of environmental risk factors for ADHD, see Kim et al. 2020). In this way, the use of twin modeling revealed important differences between certain types of environmental risk factors (shared vs. non-shared) in their relationship to ADHD symptoms.

As for genetics, the findings from the bifactor independent pathway model suggest that ADHD symptoms in young adulthood are genetically multidimensional enough that a general dimension of ADHD is not as meaningful genetically as phenotypically. In other words, the findings suggest that the two subtypes of ADHD can be considered meaningfully genetically distinct. Findings from the broader ADHD literature support this conclusion. For example, although the genetic correlation between ADHD subtypes is large (Kuntsi et al. 2014; McLoughlin et al. 2007), the genetic correlations among other psychiatric phenotypes, such as between depression and anxiety, are often larger (e.g., Morneau-Vaillancourt et al. 2020). Similarly, some estimates of the genetic correlation between ADHD subtypes (e.g., McLoughlin et al. 2007) have been roughly equivalent to estimates of the genetic correlations between ADHD and other neurodevelopmental conditions (Andersson et al. 2020). If the subtypes of ADHD are genetically multidimensional and have a genetic correlation that is equivalent to or lower than the genetic correlations among other psychiatric phenotypes, this has important implications. Genetic separation between symptoms of inattention and hyperactivity implies the existence of distinct biological pathways that could be targeted by more symptom-specific medications or behavioral interventions. Future research might therefore aim to explore how genetic differences between inattention and hyperactivity can be translated to improved treatments. A more extreme implication might be that inattention and hyperactivity should be considered distinct conditions altogether (see Grizenko et al., 2009). If the subtypes of ADHD truly reflected a single condition, it might be expected that ADHD symptoms would be both genetically and environmentally unidimensional, which was not observed. Considering the results of this and previous studies, findings that the genetic overlap between inattention and hyperactivity is lower than might be expected have now been made in both children (Kuntsi et al. 2014; McLoughlin et al. 2007) and young adults. This suggests that the genetic overlap between inattention and hyperactivity might be one aspect of the genetics of ADHD that remains fairly consistent with age.

The overall heritability of self-reported adult ADHD symptoms was estimated to be 40% in this study. This estimate is consistent with those from previous adult twin studies of self-reported ADHD symptoms (Faraone and Larsson 2019). As previously mentioned, it is notably lower than the estimated heritability of ADHD in children (Faraone and Larsson 2019). This study therefore provided additional evidence that heritability might be an aspect of the genetics of ADHD that changes with age, unlike the genetic overlap between inattention and hyperactivity. However, it is again important to note that this change in heritability might instead reflect methodological issues such as a distinction between diagnoses and self-report scales. Through the use of a bifactor common pathway model, this study operationalized ADHD symptomology as a latent variable, rather than an observed total symptom score. The use of latent variables allows statistical models to better account for the measurement error which frequently occurs among self-report items (Palmer et al. 2002). Since measurement error can inflate estimates of non-shared environmental effects in twin models (van den Berg et al. 2007), accounting for measurement error had the potential to increase the estimated heritability of self-reported adult ADHD symptoms. This would have made the heritability estimate closer to that which has been observed for clinically diagnosed adult ADHD (Larsson et al. 2014). Given that the heritability estimate from this study is consistent with those from previous studies, it is likely that measurement error has had a minimal impact on previous estimates of the heritability of self-reported adult ADHD symptoms.

There are a several limitations to this study. First, twin data are sometimes considered ungeneralizable to the broader population. Several researchers have argued that social and cultural factors might cause families with twins to be systematically different to other families (Mönkediek et al. 2020; Schwabe et al. 2017). Evidence for the existence of systematic differences between twins and non-twins specifically in the presentation of ADHD is mixed (Gjone et al., 1995; Levy et al. 1996), and so the severity of this limitation is unclear. It is therefore important to explore these issues using other genetically informative designs. Second, our analyses assumed no gene-environment interplay in the etiology of ADHD. A wealth of research has found that gene-environment interplay influences the emergence and persistence of ADHD symptoms (Nigg et al. 2010). Importantly, gene-environment interplay can bias the results of twin models (Purcell 2002; Verhulst and Hatemi 2013), including the bifactor twin models used in this study. Future research might therefore aim to combine bifactor twin models with components that allow for the modeling of gene-environment interplay, such as environmental moderators (Purcell 2002). Third, we used only self-reported measures of ADHD symptoms. Future research might aim to expand upon these analyses to include additional raters (e.g. family members) to ensure clarity over presentation of symptoms across different environments (e.g. home, school, workplace).

Despite these limitations, this study provides new insight into the genetic and environmental dimensionality of adult ADHD symptoms. Importantly, we found that although adult ADHD symptoms have substantial psychometric unidimensionality, they have limited genetic unidimensionality. One possible conclusion from this study is that adult ADHD symptoms are genetically multidimensional enough that the subtypes of ADHD be considered meaningfully genetically distinct. In genome-wide association studies, power might be improved by undertaking analyses at the subtype level rather than at the level of ADHD diagnosis per se. ADHD is just one example among many psychiatric conditions for which the use of bifactor twin models might be useful to assess genetic and environmental unidimensionality. Symptom scales used to measure many psychiatric conditions are known to be fitted well by bifactor models, including psychotic disorders (Anderson et al. 2018) and mood disorders (Simms et al. 2008). Bifactor twin models might also provide insight into genetic unidimensionality across conditions, especially those for which their phenotypic unidimensionality has previously been assessed using bifactor modeling, such as ADHD and oppositional defiant disorder (Burns et al. 2020; Lee et al. 2016). This would in turn provide invaluable insight into the contribution of genetics to the risk of common ADHD comorbidities. In this way, the use of bifactor twin models is a desirable method for assessing the dimensionality of conditions such as ADHD in greater detail.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Supplementary Material 1 (25.2KB, xlsx)
Supplementary Material 2 (850.8KB, docx)

Acknowledgements

We gratefully acknowledge the ongoing contribution of the participants in the Twins Early Development Study (TEDS) and their families.

Author Contributions

J.K. wrote the main manuscript text and prepared the supplementary materials. All authors reviewed and edited the manuscript.

Funding

TEDS is funded primarily by the UK Medical Research Council (MR/V012878/1 and previously MR/M021475/1), with additional funding by the US National Institute of Health (AG046938). J.K. is supported by a doctoral studentship from the King’s College London Medical Research Council Doctoral Training Partnership in Biomedical Sciences. G.M.-V. is supported by postdoctoral fellowships from the Social Sciences and Humanities Research Council of Canada (756-2021-0516) and Fonds de recherche du Quebec Société et Culture (2022-B3Z-297753).

Data Availability

The data used in this study are not openly available. Twins Early Development Study data can be accessed by researchers by application, see: https://www.teds.ac.uk/researchers/teds-data-access-policy.

Declarations

Competing Interests

The authors declare no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Adeyemo BO, Biederman J, Zafonte R, Kagan E, Spencer TJ, Uchida M, Kenworthy T, Spencer AE, Faraone SV (2014) Mild traumatic brain injury and ADHD: a systematic review of the literature and meta-analysis. J Atten Disord 18(7):576–584. 10.1177/1087054714543371 [DOI] [PubMed] [Google Scholar]
  2. Agnew-Blais JC, Polanczyk GV, Danese A, Wertz J, Moffitt TE, Arseneault L (2016) Evaluation of the persistence, remission, and emergence of attention-deficit/hyperactivity disorder in young adulthood. JAMA Psychiatry 73(7):713–720. 10.1001/jamapsychiatry.2016.0465 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders (5th ed.). 10.1176/appi.books.9780890425596
  4. Anderson AE, Marder S, Reise SP, Savitz A, Salvadore G, Fu DJ, Li Q, Turkoz I, Han C, Bilder RM (2018) Bifactor modeling of the positive and negative syndrome scale: generalized psychosis spans schizoaffective, bipolar, and schizophrenia diagnoses. Schizophr Bull 44(6):1204–1216. 10.1093/schbul/sbx163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Andersson A, Tuvblad C, Chen Q, Du Rietz E, Cortese S, Kuja-Halkola R, Larsson H (2020) Research review: the strength of the genetic overlap between ADHD and other psychiatric symptoms - a systematic review and meta-analysis. J Child Psychol Psychiatry Allied Discip 61(11):1173–1183. 10.1111/jcpp.13233 [DOI] [PubMed] [Google Scholar]
  6. Arias VB, Ponce FP, Núñez DE (2018) Bifactor models of attention-deficit/hyperactivity disorder (ADHD): an evaluation of three necessary but underused psychometric indexes. Assessment 25(7):885–897. 10.1177/1073191116679260 [DOI] [PubMed] [Google Scholar]
  7. Bates TC, Maes H, Neale MC (2019) Umx: twin and path-based structural equation modeling in R. Twin Res Hum Genet 22(1):27–41. 10.1017/thg.2019.2 [DOI] [PubMed] [Google Scholar]
  8. Biederman J, Fried R, DiSalvo M, Storch B, Pulli A, Woodworth KY, Biederman I, Faraone SV, Perlis RH (2019) Evidence of low adherence to stimulant medication among children and youths with ADHD: an electronic health records study. Psychiatric Serv 70(10):874–880. 10.1176/appi.ps.201800515 [DOI] [PubMed] [Google Scholar]
  9. Boker S, Neale M, Maes H, Wilde M, Spiegel M, Brick T, Spies J, Estabrook R, Kenny S, Bates T, Mehta P, Fox J (2011) OpenMx: an open source extended structural equation modeling framework. Psychometrika 76(2):306–317. 10.1007/s11336-010-9200-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bonifay W, Cai L (2017) On the complexity of item response theory models. Multivar Behav Res 52(4):465–484. 10.1080/00273171.2017.1309262 [DOI] [PubMed] [Google Scholar]
  11. Boomsma DI, Saviouk V, Hottenga JJ, Distel MA, de Moor MH, Vink JM, Geels LM, van Beek JH, Bartels M, de Geus EJ, Willemsen G (2010) Genetic epidemiology of attention deficit hyperactivity disorder (ADHD index) in adults. PLoS ONE 5(5):e10621. 10.1371/journal.pone.0010621 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Brikell I, Kuja-Halkola R, Larsson H (2015) Heritability of attention-deficit hyperactivity disorder in adults. Am J Med Genet 168(6):406–413. 10.1002/ajmg.b.32335 [DOI] [PubMed] [Google Scholar]
  13. Burns GL, Geiser C, Servera M, Becker SP, Beauchaine TP (2020) Application of the bifactor S – 1 model to multisource ratings of ADHD/ODD symptoms: an appropriate bifactor model for symptom ratings. J Abnorm Child Psychol 48(7):881–894. 10.1007/s10802-019-00608-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Castellanos FX (2015) Is adult-onset ADHD a distinct entity? Am J Psychiatry 172(10):929–931. 10.1176/appi.ajp.2015.15070988 [DOI] [PubMed] [Google Scholar]
  15. Cattoi B, Alpern I, Katz JS, Keepnews D, Solanto MV (2022) The adverse health outcomes, economic burden, and public health implications of unmanaged attention deficit hyperactivity disorder (ADHD): a call to action resulting from CHADD Summit, Washington, DC, October 17, 2019. J Atten Disord 26(6):807–808. 10.1177/10870547211036754 [DOI] [PubMed] [Google Scholar]
  16. Caye A, Swanson JM, Coghill D, Rohde LA (2019) Treatment strategies for ADHD: an evidence-based guide to select optimal treatment. Mol Psychiatry 24(3):390–408. 10.1038/s41380-018-0116-3 [DOI] [PubMed] [Google Scholar]
  17. Chang Z, Lichtenstein P, Asherson PJ, Larsson H (2013) Developmental twin study of attention problems: high heritabilities throughout development. JAMA Psychiatry 70(3):311–318. 10.1001/jamapsychiatry.2013.287 [DOI] [PubMed] [Google Scholar]
  18. Conners CK (2008) Conners 3rd Edition (Conners 3) Manual. MHS Assessments
  19. Erskine HE, Norman RE, Ferrari AJ, Chan GC, Copeland WE, Whiteford HA, Scott JG (2016) Long-term outcomes of attention-deficit/hyperactivity disorder and conduct disorder: a systematic review and meta-analysis. J Am Acad Child Adolesc Psychiatry 55(10):841–850. 10.1016/j.jaac.2016.06.016 [DOI] [PubMed] [Google Scholar]
  20. Faraone SV, Larsson H (2019) Genetics of attention deficit hyperactivity disorder. Mol Psychiatry 24(4):562–575. 10.1038/s41380-018-0070-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Franić S, Dolan CV, Borsboom D, Hudziak JJ, van Beijsterveldt CEM, Boomsma DI (2013) Can genetics help psychometrics? Improving dimensionality assessment through genetic factor modeling. Psychol Methods 18(3):406–433. 10.1037/a0032755 [DOI] [PubMed] [Google Scholar]
  22. Giordano C, Waller NG (2020) Recovering bifactor models: a comparison of seven methods. Psychol Methods 25(2):143–156. 10.1037/met0000227 [DOI] [PubMed] [Google Scholar]
  23. Gjone H, Nøvik TS (1995) Parental ratings of behaviour problems: a twin and general population comparison. J Child Psychol Psychiatry Allied Discip 36(7):1213–1224. 10.1111/j.1469-7610.1995.tb01366.x [DOI] [PubMed] [Google Scholar]
  24. Grizenko N, Paci M, Joober R (2010) Is the inattentive subtype of ADHD different from the combined/hyperactive subtype? J Atten Disord 13(6):649–657. 10.1177/1087054709347200 [DOI] [PubMed] [Google Scholar]
  25. Haberstick BC, Timberlake D, Hopfer CJ, Lessem JM, Ehringer MA, Hewitt JK (2008) Genetic and environmental contributions to retrospectively reported DSM-IV childhood attention deficit hyperactivity disorder. Psychol Med 38(7):1057–1066. 10.1017/S0033291707001584 [DOI] [PubMed] [Google Scholar]
  26. Hancock GR, Mueller RO (2001) Rethinking construct reliability within latent variable systems. In: Toit SD, Cudeck R, Sorbom D, Jöreskog KG (eds) Structural equation modeling: Present and future. Scientific Software International, pp 195–216
  27. Holzinger KJ, Swineford F (1937) The bi-factor method. Psychometrika 2:41–54. 10.1007/BF02287965 [Google Scholar]
  28. Hu L-t, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Struct Equ Model 6(1):1–55. 10.1080/10705519909540118 [Google Scholar]
  29. Jenabi E, Bashirian S, Khazaei S, Basiri Z (2019) The maternal prepregnancy body mass index and the risk of attention deficit hyperactivity disorder among children and adolescents: a systematic review and meta-analysis. Korean J Pediatr 62(10):374–379. 10.3345/kjp.2019.00185 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kendler KS, Heath AC, Martin NG, Eaves LJ (1987) Symptoms of anxiety and symptoms of depression: same genes, different environments? Arch Gen Psychiatry 44(5):451–457. 10.1001/archpsyc.1987.01800170073010 [DOI] [PubMed] [Google Scholar]
  31. Kim JH, Kim JY, Lee J, Jeong GH, Lee E, Lee S, Lee KH, Kronbichler A, Stubbs B, Solmi M, Koyanagi A, Hong SH, Dragioti E, Jacob L, Brunoni AR, Carvalho AF, Radua J, Thompson T, Smith L, Oh H, Fusar-Poli P (2020) Environmental risk factors, protective factors, and peripheral biomarkers for ADHD: an umbrella review. Lancet Psychiatry 7(11):955–970. 10.1016/S2215-0366(20)30312-6 [DOI] [PubMed] [Google Scholar]
  32. Kuntsi J, Pinto R, Price TS, van der Meere JJ, Frazier-Wood AC, Asherson P (2014) The separation of ADHD inattention and hyperactivity-impulsivity symptoms: pathways from genetic effects to cognitive impairments and symptoms. J Abnorm Child Psychol 42(1):127–136. 10.1007/s10802-013-9771-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Larsson H, Asherson P, Chang Z, Ljung T, Friedrichs B, Larsson JO, Lichtenstein P (2013) Genetic and environmental influences on adult attention deficit hyperactivity disorder symptoms: a large Swedish population-based study of twins. Psychol Med 43(1):197–207. 10.1017/S0033291712001067 [DOI] [PubMed] [Google Scholar]
  34. Larsson H, Chang Z, D’Onofrio BM, Lichtenstein P (2014) The heritability of clinically diagnosed attention deficit hyperactivity disorder across the lifespan. Psychol Med 44(10):2223–2229. 10.1017/S0033291713002493 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lee S, Burns GL, Beauchaine TP, Becker SP (2016) Bifactor latent structure of attention-deficit/hyperactivity disorder (ADHD)/oppositional defiant disorder (ODD) symptoms and first-order latent structure of sluggish cognitive tempo symptoms. Psychol Assess 28(8):917–928. 10.1037/pas0000232 [DOI] [PubMed] [Google Scholar]
  36. Levy F, Hay D, McLaughlin M, Wood C, Waldman I (1996) Twin sibling differences in parental reports of ADHD, speech, reading and behaviour problems. J Child Psychol Psychiatry Allied Discip 37(5):569–578. 10.1111/j.1469-7610.1996.tb01443.x [DOI] [PubMed] [Google Scholar]
  37. Li L, Taylor MJ, Bälter K, Kuja-Halkola R, Chen Q, Hegvik TA, Tate AE, Chang Z, Arias-Vásquez A, Hartman CA, Larsson H (2020) Attention-deficit/hyperactivity disorder symptoms and dietary habits in adulthood: a large population-based twin study in Sweden. Am J Med Genet Part B Neuropsychiatric Genet 183(8):475–485. 10.1002/ajmg.b.32825 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lockhart C, Bright J, Ahmadzadeh Y, Breen G, Bristow S, Boyd A, Downs J, Hotopf M, Palaiologou E, Rimfeld K, Maxwell J, Malanchini M, McAdams TA, McMillan A, Plomin R, Eley TC (2023) Twins early Development Study (TEDS): a genetically sensitive investigation of mental health outcomes in the mid-twenties. JCPP Adv 3(2):e12154. 10.1002/jcv2.12154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. McDonald RP (1999) Test theory: a unified approach. Erlbaum
  40. McLoughlin G, Ronald A, Kuntsi J, Asherson P, Plomin R (2007) Genetic support for the dual nature of attention deficit hyperactivity disorder: substantial genetic overlap between the inattentive and hyperactive-impulsive components. J Abnorm Child Psychol 35(6):999–1008. 10.1007/s10802-007-9149-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Milich R, Balentine AC, Lynam DR (2001) ADHD combined type and ADHD predominantly inattentive type are distinct and unrelated disorders. Clin Psychol Sci Pract 8(4):463–488. 10.1093/clipsy.8.4.463 [Google Scholar]
  42. Moffitt TE, Houts R, Asherson P, Belsky DW, Corcoran DL, Hammerle M, Harrington H, Hogan S, Meier MH, Polanczyk GV, Poulton R, Ramrakha S, Sugden K, Williams B, Rohde LA, Caspi A (2015) Is adult ADHD a childhood-onset neurodevelopmental disorder? Evidence from a four-decade longitudinal cohort study. Am J Psychiatry 172(10):967–977. 10.1176/appi.ajp.2015.14101266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Mönkediek B, Schulz W, Eichhorn H, Diewald M (2020) Is there something special about twin families? A comparison of parenting styles in twin and non-twin families. Soc Sci Res 90:102441. 10.1016/j.ssresearch.2020.102441 [DOI] [PubMed] [Google Scholar]
  44. Morneau-Vaillancourt G, Coleman JRI, Purves KL, Cheesman R, Rayner C, Breen G, Eley TC (2020) The genetic and environmental hierarchical structure of anxiety and depression in the UK Biobank. Depress Anxiety 37(6):512–520. 10.1002/da.22991 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Nigg J, Nikolas M, Burt SA (2010) Measured gene-by-environment interaction in relation to attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry 49(9):863–873. 10.1016/j.jaac.2010.01.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Nikolas MA, Burt SA (2010) Genetic and environmental influences on ADHD symptom dimensions of inattention and hyperactivity: a meta-analysis. J Abnorm Psychol 119(1):1–17. 10.1037/a0018010 [DOI] [PubMed] [Google Scholar]
  47. Palmer RF, Graham JW, Taylor B, Tatterson J (2002) Construct validity in health behavior research: interpreting latent variable models involving self-report and objective measures. J Behav Med 25(6):525–550. 10.1023/a:1020689316518 [DOI] [PubMed] [Google Scholar]
  48. Pelt DH, de Vries LP, Bartels M (2022) Unraveling the relation between personality and well-being in a genetically informative design. Eur J Pers 08902070221134878. 10.1177/08902070221134878
  49. Pituch KA, Stevens JP (2016) Applied multivariate statistics for the social sciences, 6th edn. Routledge
  50. Purcell S (2002) Variance components models for gene-environment interaction in twin analysis. Twin Res 5(6):554–571. 10.1375/136905202762342026 [DOI] [PubMed] [Google Scholar]
  51. Reise SP, Kim DS, Mansolf M, Widaman KF (2016) Is the bifactor model a better model or is it just better at modeling implausible responses? Application of iteratively reweighted least squares to the Rosenberg Self-Esteem Scale. Multivar Behav Res 51(6):818–838. 10.1080/00273171.2016.1243461 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Reise SP, Mansolf M, Haviland MG (2023) Bifactor measurement models. In: Hoyle RH (ed) Handbook of structural equation modeling, 2nd edn. Guildford, pp 329–348
  53. Rijsdijk FV, Sham PC (2002) Analytic approaches to twin data using structural equation models. Brief Bioinform 3(2):119–133. 10.1093/bib/3.2.119 [DOI] [PubMed] [Google Scholar]
  54. Rocco I, Corso B, Bonati M, Minicuci N (2021) Time of onset and/or diagnosis of ADHD in European children: a systematic review. BMC Psychiatry 21(1):575. 10.1186/s12888-021-03547-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Rodriguez A, Reise SP, Haviland MG (2016) Evaluating bifactor models: calculating and interpreting statistical indices. Psychol Methods 21(2):137–150. 10.1037/met0000045 [DOI] [PubMed] [Google Scholar]
  56. Salari N, Ghasemi H, Abdoli N, Rahmani A, Shiri MH, Hashemian AH, Akbari H, Mohammadi M (2023) The global prevalence of ADHD in children and adolescents: a systematic review and meta-analysis. Ital J Pediatr 49(1):48. 10.1186/s13052-023-01456-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Schans JV, Çiçek R, de Vries TW, Hak E, Hoekstra PJ (2017) Association of atopic diseases and attention-deficit/hyperactivity disorder: a systematic review and meta-analyses. Neurosci Biobehav Rev 74(Pt A):139–148. 10.1016/j.neubiorev.2017.01.011 [DOI] [PubMed] [Google Scholar]
  58. Schwabe I, Janss L, van den Berg SM (2017) Can we validate the results of twin studies? A census-based study on the heritability of educational achievement. Front Genet 8:160. 10.3389/fgene.2017.00160 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Simms LJ, Grös DF, Watson D, O’Hara MW (2008) Parsing the general and specific components of depression and anxiety with bifactor modeling. Depress Anxiety 25(7):E34–E46. 10.1002/da.20432 [DOI] [PubMed] [Google Scholar]
  60. Song P, Zha M, Yang Q, Zhang Y, Li X, Rudan I (2021) The prevalence of adult attention-deficit hyperactivity disorder: a global systematic review and meta-analysis. J Global Health 11:04009. 10.7189/jogh.11.04009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Ten Berge JM, Sočan G (2004) The greatest lower bound to the reliability of a test and the hypothesis of unidimensionality. Psychometrika 69:613–625 [Google Scholar]
  62. van den Berg SM, Willemsen G, de Geus EJ, Boomsma DI (2006) Genetic etiology of stability of attention problems in young adulthood. Am J Med Genet 141B(1):55–60. 10.1002/ajmg.b.30251 [DOI] [PubMed] [Google Scholar]
  63. van den Berg SM, Glas CA, Boomsma DI (2007) Variance decomposition using an IRT measurement model. Behav Genet 37(4):604–616. 10.1007/s10519-007-9156-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. van Lieshout M, Luman M, Twisk JW, van Ewijk H, Groenman AP, Thissen AJ, Faraone SV, Heslenfeld DJ, Hartman CA, Hoekstra PJ, Franke B, Buitelaar JK, Rommelse NN, Oosterlaan J (2016) A 6-year follow-up of a large European cohort of children with attention-deficit/hyperactivity disorder-combined subtype: outcomes in late adolescence and young adulthood. Eur Child Adolesc Psychiatry 25(9):1007–1017. 10.1007/s00787-016-0820-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Verhulst B, Hatemi PK (2013) Gene-environment interplay in twin models. Political Anal 21(3):368–389. 10.1093/pan/mpt005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Vos M, Hartman CA (2022) The decreasing prevalence of ADHD across the adult lifespan confirmed. J Global Health 12:03024. 10.7189/jogh.12.03024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Waller NG (2018) Direct Schmid-Leiman transformations and rank-deficient loadings matrices. Psychometrika 83(4):858–870. 10.1007/s11336-017-9599-0 [DOI] [PubMed] [Google Scholar]
  68. Waller NG, Jones J, Giordano C, Nguyen HV (2023) fungible: Psychometric functions from the Waller Lab (ver. 2.3). https://cran.r-project.org/web/packages/fungible/index.html
  69. Zakharin M, Bates TC (2023) Testing heritability of moral foundations: common pathway models support strong heritability for the five moral foundations. Eur J Pers 37(4):485–497. 10.1177/08902070221103957 [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (25.2KB, xlsx)
Supplementary Material 2 (850.8KB, docx)

Data Availability Statement

The data used in this study are not openly available. TEDS data can be accessed by researchers by application, see: https://www.teds.ac.uk/researchers/teds-data-access-policy.

The data used in this study are not openly available. Twins Early Development Study data can be accessed by researchers by application, see: https://www.teds.ac.uk/researchers/teds-data-access-policy.

Articles from Behavior Genetics are provided here courtesy of Springer

RESOURCES