Longitudinal Stability of Genetic and Environmental Influences on Irritability: From Childhood to Young Adulthood
Publication: American Journal of Psychiatry
Abstract
Objective:
Little is known about genetic influences on juvenile irritability and whether such influences are developmentally stable and/or dynamic. This study examined the temporal pattern of genetic and environmental effects on irritability using data from a prospective, four-wave longitudinal twin study.
Method:
Parents and their twin children (N=2,620 children) from the Swedish Twin Study of Child and Adolescent Development reported on the children’s irritability, defined using a previously identified scale from the Child Behavior Checklist.
Results:
Genetic effects differed across the sexes, with males exhibiting increasing heritability from early childhood through young adulthood and females exhibiting decreasing heritability. Genetic innovation was also more prominent in males than in females, with new genetic risk factors affecting irritability in early and late adolescence for males. Shared environment was not a primary influence on irritability for males or females. Unique, nonshared environmental factors suggested strong effects early for males followed by an attenuating influence, whereas unique environmental factors were relatively stable for females.
Conclusions:
Genetic effects on irritability are developmentally dynamic from middle childhood through young adulthood, with males and females displaying differing patterns. As males age, genetic influences on irritability increase while nonshared environmental influences weaken. Genetic contributions are quite strong in females early in life but decline in importance with age. In girls, nonshared environmental influences are fairly stable throughout development.
Studies in youths suggest that irritability is a common (prevalence, ∼3%) (1, 2), stable (3, 4), and often impairing trait (2, 5, 6). However, research has yet to explore the genetic and environmental contribution to individual differences in irritability. Indeed, few studies have examined the heritability of irritability per se (7), with most work being done in the context of other traits (e.g., aggression) or related psychiatric disorders (8–15). These limited available data suggest that irritability is moderately heritable, with estimates averaging around 0.3 (7, 16) and ranging from 0.22 to 0.51 (14).
Research on pediatric psychopathology, including anxiety, depression, and attention deficit hyperactivity disorder, demonstrates that these clinical phenotypes are developmentally complex, characterized by both continuity and genetic innovation (17, 18). This may also be true of irritability, a stable, heritable trait (3, 4) that has been shown to have both genetic and environmental influences (14). Whereas genetic continuity indicates developmental stability in genetic risk factors over time, genetic innovation suggests that new genes, previously without effect at one developmental age, become influential during subsequent periods (18). Research has yet to examine the relative contribution of genetic influences on irritability across the lifespan. Understanding evolving genetic risk factors for irritability has profound research, clinical, and treatment implications.
In this study, we examined irritability in a population-based cohort of Swedish twins assessed four times between ages 8 and 20 using a developmental model, which permits us to assess both genetic stability and genetic innovation. We measured irritability using the Achenbach System of Empirically Based Assessment (19–22). Our primary goal was to discriminate between two trajectories that might describe the developmental course of genetic risk factors for irritability. A “developmentally stable” pattern would predict a single set of genetic risk factors that emerge early in development and affect irritability throughout development, while a “developmentally dynamic” pattern would suggest that genetic innovation occurs (i.e., new or different genes are activated during development) and that genetic effects on irritability vary over time (see the Method section for additional details). A second goal was to examine environmental effects on irritability to determine whether such influences are stable or dynamic across development.
Our period of observation extends from childhood into young adulthood, thereby including puberty, a developmental transition of special interest given that irritability relates to internalizing disorder development (23) and that this developmental period is associated with an increase in the prevalence of internalizing conditions (24–27). Moreover, while extensive research has shown sex differences in the prevalence and manifestations of internalizing disorders and irritability (2, 24–32), research has yet to examine sex differences in genetic influences on irritability.
Method
Participants
The data analyzed in this study are from the Swedish Twin Study of Child and Adolescent Development. All twin pairs born in Sweden between May 1985 and December 1986 in which both twins were alive and residing in Sweden in 1994 were contacted for study participation (22). Data from 2,719 twin individuals were available; because zygosity was not known for some twin pairs and in some cases only one member of the pair participated, data for 1,310 twin pairs (2,620 individuals) were available for the present analyses. Among these pairs, there were 267 female monozygotic pairs, 199 female-female dizygotic pairs, 254 male monozygotic pairs, 182 male-male dizygotic pairs, and 408 opposite-sex dizygotic pairs.
Twins were assessed at four waves through a mailed questionnaire. Twin ages at the time of the assessments were 8–9 years at wave 1 (parent report only, N=1,109), 13–14 years at wave 2 (parent report, N=1,063; child report, N=2,263), 16–17 years at wave 3 (parent report, N=1,067; child report, N=2,369), and 19–20 years at wave 4 (parent report, N=619; child report, N=1,705). The Ethics Committee of the Karolinska Institute approved the questionnaires. Zygosity was assessed by testing of DNA extracted from saliva samples using OraGene DNA self-collection kits (DNA Genotek, Ontario). For twins for whom DNA samples were not available, zygosity was determined using an algorithm derived from discriminant analyses of twins’ and parents’ responses to validated zygosity questionnaires. In cases of any contradictions between the assignments (N=100, 3.4%), the zygosity was set to unknown and the twins were excluded from the analyses.
Measures
Between ages 8 and 21, the twins’ irritability was assessed by parent report obtained using items from the Child Behavior Checklist (19). At ages 13–14 and 16–17, twins also completed the Youth Self-Report Child Behavior Checklist (20), and at ages 19–20, the Adult Behavior Checklist (21). All items were scored on a 3-point scale (1=not true, 2=somewhat or sometimes true, and 3=very true or often true). Details on the derivation of the irritability score used in all analyses are provided in the data supplement that accompanies the online edition of this article.
Data Analysis
A Cholesky decomposition was used to address questions regarding the magnitude of genetic and environmental influences on longitudinal measurement of irritability. Twin models, such as the Cholesky decomposition, allow for the determination of the degree of similarity or dissimilarity between monozygotic (sharing 100% of their genes) and dizygotic twins (sharing on average 50% of their genes, by descent). Multiple measures of irritability may be correlated because they share common genes and/or common environmental influences. Twin data allow for the partitioning of the covariation between measures into genetic and environmental components. Specifically, Cholesky decomposition allows for the disaggregation of the covariance into additive genetic (A), common (shared) environmental (C), and unique or nonshared environmental (E) components.
The longitudinal Cholesky decomposition is presented in Figure 1 (for simplicity, only additive genetic effects [A] are illustrated). The model contains four major elements. First, it includes four latent irritability scores (T1–T4), which reflect the “true” level of irritability at each assessment wave. These latent variables are indexed by ratings of irritability made by parents (P) and by twin self-report (S). Both informants provided ratings at waves 2–4, but only parent reports are available for wave 1. The paths λP and λS reflect the degree to which the parent- and self-reported irritability scores index the latent level of irritability. Next, the genetic and environmental influences on the latent levels of irritability at waves 1–4 are modeled using a Cholesky decomposition. This approach divides genetic risk into four factors (F1–F4), with the first beginning in childhood (ages 8–9) and potentially remaining active over the entire developmental period. The strength of this factor at each age is reflected in the path coefficients f11, f12, f13, and f14. The second factor begins in early adolescence (ages 13–14) and influences irritability assessed at waves 2–4 via paths f22, f23, and f24. The third factor begins in late adolescence (ages 16–17) and affects waves 3 and 4 via paths f33 and f34. The fourth and final factor begins at wave 4, when twins are in young adulthood (ages 19–20), via path f44. The “developmentally stable” (18) hypothesis predicts that genetic liability to irritability originates solely in the first factor, with no significant subsequent genetic innovation. By contrast, the “developmentally dynamic” (18) hypothesis predicts both genetic innovation (new genetic variation emerging later in development to affect irritability) and genetic attenuation (the impact of genetic factors acting early in development declining over time).
FIGURE 1. Longitudinal Cholesky Decomposition With Reference to Additive Genetic Effectsa
a For simplicity, common (shared) and unique (nonshared) environmental factors are not modeled. The model contains four latent irritability scores, T1–T4, reflecting irritability at waves 1 through 4, respectively. These latent variables are indexed by ratings of irritability by parent report (P) (available for waves 1–4) and by self-report (S) (available for waves 2–4). The degree to which the parent- and self-reported irritability ratings index the latent irritability level is reflected by the paths λP and λS. See text for details. F1–F4 indicate the four genetic risk factors; f, the path from the genetic factors to the latent irritability scores at each of the four waves; R, residual effects; and FP and FS, informant-specific common factors for parent and child reports, respectively.
The model also includes two informant-specific common factors, one for parent (FP) and one for twin self-report (FS), as well as rater- and time-specific residuals. The informant-specific common factors allow the model to estimate genetic and environmental influences on ratings that are unique to the parents or to the child.
Our analyses focused on the latent measures of irritability (T1–T4, in the upper portion of Figure 1) reported by parents and children separately, because these measures are likely to be most valid, reflecting both the subjective and objective manifestations of irritability. Informant-specific factors (FP, FS) are part of the model (in the lower portion of Figure 1) and are briefly reviewed here.
Estimates of heritability and shared environmental effects obtained in this model are not comparable to those obtained from standard twin models, because in standard twin models, errors of measurement contribute to individual-specific environmental effects, thereby reducing estimates of heritability and shared environment. Our use of multiple raters permits us to distinguish true individual-specific environmental effects, which have an impact on the latent irritability scores (T1–T4), from measurement error, which contributes to the rater-specific effects (P1–P4 and S2–S4).
Qualitative and quantitative sex effects on irritability scores were also examined. Qualitative sex effects, which arise when genetic factors influencing a trait are not identical in males and females, are measured by the genetic correlation, rg, which can vary from zero (i.e., entirely distinct sets of genes in the two sexes) to unity (i.e., identical genetic factors affecting males and females). Quantitative sex effects arise when the same genetic factors affect males and females but to different degrees. These were identified by estimating all path coefficients separately by sex.
Irritability scores were treated as a continuous trait. To evaluate the fit of our entire model, we used the Akaike information criterion (33). The lower the Akaike information criterion value, the better the balance of explanatory power and parsimony. All models were conducted in the Mx program (34).
Results
Descriptive Results
Table 1 lists the mean raw irritability scores by wave, informant, sex, and zygosity. Higher irritability scores were observed for females compared with males by both parent and self-report, except at wave 1 (ages 8–9). Moreover, twins generally self-reported higher levels of irritability compared with parent report. After wave 2, irritability symptoms generally decreased with age in females and males by both parent and self-report. Table S2 in the online data supplement contains Pearson correlations between self-report and parent report of irritability scores within and across waves. Parent report was moderately correlated over time (r values, 0.32–0.49), with the correlations declining in a monotonic fashion across waves. This same pattern was observed for twin self-report (r values, 0.31–0.45). Parent and twin reports of twin irritability demonstrated significant modest to moderate levels of association (r values, 0.13–0.36).
| Informant and Wave | Monozygotic Female Twins | Dizygotic Female Twins | Monozygotic Male Twins | Dizygotic Male Twins | Dizygotic Opposite-Sex Twins | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
| Parent report | ||||||||||
| Wave 1 | 3.96 | 1.26 | 4.06 | 1.24 | 3.99 | 1.28 | 4.13 | 1.31 | 4.12 | 1.40 |
| Wave 2 | 4.15 | 1.46 | 4.11 | 1.34 | 3.77 | 1.08 | 3.97 | 1.25 | 3.96 | 1.26 |
| Wave 3 | 4.07 | 1.31 | 4.05 | 1.38 | 3.75 | 1.15 | 3.82 | 1.25 | 3.91 | 1.20 |
| Wave 4 | 3.86 | 1.12 | 3.96 | 1.16 | 3.40 | 0.66 | 3.66 | 0.89 | 3.74 | 1.08 |
| Self-report | ||||||||||
| Wave 2 | 4.98 | 1.55 | 5.27 | 1.61 | 4.51 | 1.47 | 4.80 | 1.56 | 5.12 | 1.56 |
| Wave 3 | 4.88 | 1.50 | 5.18 | 1.63 | 4.34 | 1.33 | 4.47 | 1.50 | 4.88 | 1.57 |
| Wave 4 | 4.34 | 1.33 | 4.67 | 1.49 | 3.66 | 0.98 | 3.81 | 1.21 | 4.40 | 1.45 |
Twin-twin correlations are presented in Table 2. Stronger associations were observed for irritability scores between monozygotic twins compared with dizygotic twins. Parent-rated irritability scores for same- and opposite-sex twins were similar in strength, whereas twin self-report irritability scores were slightly higher in same-sex dizygotic twins relative to opposite-sex dizygotic twins. No evidence of dominance was observed (that is, monozygotic correlations were much greater than twice the dizygotic correlations). Therefore, an ACE model was fitted to the data.
| Informant and Wave | Monozygotic Female Twins | Dizygotic Female Twins | Monozygotic Male Twins | Dizygotic Male Twins | Dizygotic Opposite-Sex Twins |
|---|---|---|---|---|---|
| Parent report | |||||
| Wave 1 | 0.77 | 0.30 | 0.55 | 0.29 | 0.39 |
| Wave 2 | 0.80 | 0.49 | 0.64 | 0.44 | 0.45 |
| Wave 3 | 0.69 | 0.54 | 0.80 | 0.44 | 0.49 |
| Wave 4 | 0.63 | 0.48 | 0.62 | 0.39 | 0.31 |
| Self-report | |||||
| Wave 2 | 0.40 | 0.18 | 0.49 | 0.24 | –0.06 |
| Wave 3 | 0.44 | 0.16 | 0.48 | 0.12 | 0.02 |
| Wave 4 | 0.41 | 0.05 | 0.33 | –0.11 | 0.23 |
Longitudinal Cholesky Decomposition
The longitudinal Cholesky decomposition, where repeated measures of irritability scores were modeled, suggested that the best Akaike information criterion value was obtained for a model with quantitative, but not qualitative, sex effects (Table 3). Thus, the quantitative sex-effects model was selected as the best-fit model, and therefore parameter estimates were examined for males and females separately.
| Variables | Qual / Quan | rg | –2LL | DF | AIC | Δ(–2LL) | ΔDF | ΔAIC |
|---|---|---|---|---|---|---|---|---|
| ACE | + / + | 1.00 | 34488.964 | 13419 | 7650.964 | — | — | — |
| ACEb | – / + | 34488.919 | 13420 | 7648.919 | –0.045 | 1 | –2.045 | |
| ACE | + / – | 0.64 | 34611.159 | 13473 | 7665.159 | 122.195 | 54 | 14.195 |
| ACE | – / – | 34617.768 | 13474 | 7669.768 | 128.804 | 55 | 18.804 | |
| AE | – / + | 34503.228 | 13420 | 7663.228 | 14.264 | 1 | 12.264 | |
| CE | – / + | 34641.908 | 13420 | 7801.908 | 152.944 | 1 | 150.944 |
a
In the ACE model, A denotes additive genetics, C denotes common (shared) environmental effects, and E denotes unique (nonshared) environmental effects. Qual=Qualitative; Quan=Quantitative; rg=genetic correlation; AIC=Akaike information criterion. For qualitative effect, “+” indicates that rg is estimated and equals the genetic correlation between opposite sex twins, and “–” indicates that rg is fixed at 1.0. For quantitative effect, “+” indicates that separate path estimates are computed for males and females, and “–” indicates that the path estimates are held equal for males and females.
b
Best-fitting model.
Males.
In males, genetic factors played a strong role in influencing irritability, as indexed by both self and parent ratings (see Table 4 for parameter estimates). Overall, both genetic innovation and stability were observed, with these two processes together producing a substantial increase in heritability of irritability over the course of development. For males, heritability at each sequential wave was estimated by squaring the wave 1 minus wave 4 “A” parameter estimates. The heritability estimate at wave 1 was 36% (A12=0.602), 68% at wave 2 (A12+A22=0.552+0.612), 76% at wave 3 (A12+A22+A32=0.472+0.622+0.392), and 89% at wave 4 (A12+A22+A32+A42=0.692+[−0.11]2+0.632+[−0.01]2).
| Wave 1 | Wave 2 | Wave 3 | Wave 4 | |||||
|---|---|---|---|---|---|---|---|---|
| Sex and Model Parameter | Estimate | 95% CI | Estimate | 95% CI | Estimate | 95% CI | Estimate | 95% CI |
| Males | ||||||||
| Genetic: total a2 (%) | ||||||||
| 36 | 0.60 | 0.37, 0.96 | ||||||
| 68 | 0.55 | 0.35, 0.72 | 0.61 | 0.38, 0.77 | ||||
| 76 | 0.47 | 0.23, 0.65 | 0.62 | 0.37, 0.80 | 0.39 | –0.59, 0.59 | ||
| 89 | 0.69 | 0.45, 0.92 | –0.11 | –0.26, 0.46 | 0.63 | 0.56, 0.81 | –0.01 | –0.67, 0.67 |
| Shared environment: total c2 (%) | ||||||||
| 5 | –0.23 | –0.66, 0.08 | ||||||
| 3 | 0.01 | –0.31, 0.34 | –0.18 | –0.43, 0.43 | ||||
| 1 | –0.10 | –0.37, 0.22 | 0.02 | –0.31, 0.32 | 0.01 | –0.36, 0.36 | ||
| 2 | 0.15 | –0.31, 0.48 | 0.01 | –0.50, 0.49 | 0.01 | –0.55, 0.55 | –0.01 | –0.51, 0.43 |
| Nonshared environment: total e2 (%) | ||||||||
| 59 | 0.77 | 0.21, 0.85 | ||||||
| 30 | 0.10 | –0.13, 0.66 | 0.54 | 0.37, 0.67 | ||||
| 24 | 0.20 | 0.08, 0.65 | 0.39 | –0.59, 0.59 | 0.21 | –0.46, 0.47 | ||
| 10 | 0.10 | –0.56, 0.44 | 0.28 | –0.38, 0.48 | –0.06 | –0.43, 0.44 | –0.01 | –0.44, 0.43 |
| Females | ||||||||
| Genetic: total a2 (%) | ||||||||
| 66 | 0.81 | 0.72, 1.0 | ||||||
| 64 | 0.62 | 0.47, 0.79 | 0.51 | 0.25, 0.69 | ||||
| 56 | 0.54 | 0.35, 0.70 | 0.52 | 0.21, 0.70 | 0.05 | –0.37, 0.44 | ||
| 46 | 0.49 | 0.27, 0.69 | 0.47 | 0.00, 0.73 | 0.01 | –0.52, 0.60 | 0.01 | –0.57, 0.57 |
| Shared environment: total c2 (%) | ||||||||
| 0 | 0.07 | –0.20, 0.37 | ||||||
| 1 | –0.10 | –0.42, 0.27 | 0.01 | –0.37, 0.37 | ||||
| 9 | 0.27 | –0.13, 0.54 | –0.11 | –0.45, 0.45 | 0.01 | –0.43, 0.43 | ||
| 26 | 0.27 | –0.13, 0.54 | 0.43 | –0.69, 0.69 | 0.01 | –0.65, 0.65 | 0.03 | –0.60, 0.60 |
| Nonshared environment: total e2 (%) | ||||||||
| 35 | 0.59 | –0.41, 0.66 | ||||||
| 34 | 0.17 | –0.66, 0.62 | 0.56 | 0.05, 0.70 | ||||
| 35 | –0.04 | –0.63, 0.54 | 0.36 | 0.16, 0.58 | 0.47 | –0.61, 0.61 | ||
| 28 | 0.13 | –0.52, 0.64 | 0.21 | –0.64, 0.64 | 0.47 | 0.19, 0.65 | –0.01 | –0.11, 0.13 |
a
In the ACE model, A denotes additive genetics, C denotes common (shared) environmental effects, and E denotes unique (nonshared) environmental effects.
Consistent with the “developmentally stable” hypothesis, genetic factors that first appeared before puberty continued to influence irritability significantly throughout adolescence and into young adulthood (see Figure 2A). Indeed, stable genetic influences on irritability constituted a majority of the total genetic effect after ages 8–9. Specifically, the genetic factor measured during childhood (F1) accounted for significant genetic variance in childhood (f11=36%), early adolescence (f12=30%), late adolescence (f13=22%), and young adulthood (f14=48%). Genetic effects emerging during early adolescence also explained significant genetic variation in late adolescence (f23=38%), and there were genetic contributions to irritability in young adulthood associated with late adolescence (f34=40%). However, evidence of genetic innovation was also found, with new genetic influences becoming active during early adolescence (f22=37%) and, to a modest degree, late adolescence (f33=15%). There were no new genetic effects activated after late adolescence.
FIGURE 2. Proportion of Total Variance in Irritability Accounted for by Genetic Factors Through Development in a Longitudinal Twin Studya
a The y-axis represents the total phenotypic variance, so the sum of all genetic factors equals total heritability. The upper panel, illustrating the proportion of total variance for males, shows an increasing trend in heritability, and the lower panel, illustrating the proportion of total variance for females, shows a declining trend in heritability. Factor 1 represents the first genetic factor, starting at ages 8–9, factor 2, the second genetic factor, starting at ages 13–14, and factor 3, the third genetic factor, starting at ages 16–17.
Unlike genetic factors, shared environmental influences on irritability were small, accounting for only 1%–5% of variance in liability across waves. Shared environmental effects were observed in the informant-specific portion of the model, where parent report yielded stronger shared environmental effects than twin self-report. By contrast, nonshared environmental factors demonstrated a robust influence early in development, followed by considerable attenuation, accounting for 59%, 30%, 24%, and ultimately 10% of the variance in irritability. Innovation also was noted, with a new set of unique environmental factors emerging during early adolescence (f22=29%).
Examination of informant-specific estimates (see the lower portion of Figure S1A in the online data supplement ) for the parent and child report indicate that the parent (λP) path was somewhat higher than the child self-report (λS) path, suggesting that parent ratings were a better index of irritability than twin self-report. More consistent informer-specific genetic factors were seen for self-ratings than for parent ratings.
Females.
Although females exhibited robust genetic liability for irritability earlier in development, temporal attenuation occurred such that total heritability estimates declined from childhood into young adult life (66%, 64%, 56%, and 46% from waves 1 to 4, respectively). Genetic stability was found, with genetic effects assessed during childhood (F1) and earlier adolescence (F2) predicting significant genetic variance through adolescence into young adulthood (f12=38%, f13=29%, f14=24%; f23=27%, f24=22%). These genetic processes are illustrated in Figure 2B.
Shared environmental influences on irritability were nearly absent in females, as in males, with the exception of shared environmental influences measured during early adolescence, which explained a significant portion of the shared environmental effect during young adulthood (f24=18%). Informer-specific effects for shared environment also were found, with parents reporting stronger shared environmental effects compared with twin self-report.
By contrast, unique environmental factors accounted for 35%, 34%, 35%, and 28% of the variance in irritability at waves 1 through 4 as reported by both parent and child, demonstrating temporal stability. See Table 4 for parameter estimates; see also Figure S1B in the online data supplement .
Unique environmental influences measured at ages 13–14 contributed modestly to environment effects on irritable mood at ages 16–17 (f23=13%). Similarly, unique environmental factors at ages 16–17 contributed to irritable mood during young adulthood (f34=22%), suggesting stability of unique environmental exposures during late adolescence on irritable mood assessed in young adulthood. Moreover, new unique environmental influences emerged during early and late adolescence (f22=31% and f33=22%, respectively).
The λP path was somewhat higher than the λS path, again suggesting that parent ratings were a better index of irritability than female self-report. Parameter estimates are listed in Table 4. More consistent informer-specific genetic factors were found for self-ratings compared with parent ratings.
Discussion
Our primary aim in this study was to clarify the developmental nature of genetic and environmental risk factors for irritability. We examined data from a general-population twin sample in which both parent and child reported on irritability from childhood through the adolescent years. We found quantitative sex differences for irritability, indicating that the magnitude of genetic and environmental effects differed in males and females. Our results provide clear support for both stable and dynamic genetic effects in males, in whom genetic influences expressed early in childhood were stable into young adulthood and new genetic influences emerged during adolescence. The total of these stable and new genetic influences produced a consistent increase in heritability of irritability for males over the course of development. A different pattern of additive genetic influences was observed for females, who exhibited a robust genetic effect early in development that diminished over time. Although total heritability decreased over time, significant new genetic influences came online during early adolescence, and these emergent genetic effects continued to exert influence through adolescence into young adulthood. Thus, irritability was associated with strong genetic effects in both males and females, although the precise patterns of these effects differed by sex.
We also examined common (shared) and unique (nonshared) environmental contributions to the longitudinal course of irritability. Common environment did not contribute appreciably to the expression of irritability in males or females, suggesting that familial environmental factors that affect twins similarly are not critical in the pathogenesis of irritable mood. Stronger effects were observed for unique environmental influences (i.e., environmental factors that cause twins to be dissimilar) on irritable mood. Males and females both exhibited a strong unique environmental contribution to irritable mood early in childhood, but this effect attenuated substantially from childhood into early adolescence. This suggests that the environmental effects that affect irritability early in life ceased to influence irritability from adolescence into young adulthood. New environmental influences also were detected during early adolescence for both sexes, but this effect also weakened substantially over time. For females, a new environmental contribution to irritability emerged during late adolescence and remained stable into young adulthood. Thus, neither unique environmental exposures occurring during early childhood nor those that emerged during adolescence had an impact on irritability in adulthood; however, unique environmental influences emerging during late adolescence did have a lasting effect on irritability expression during young adulthood.
Irritability is a transdiagnostic dimensional construct that lends itself well to exploration with multiple methods across numerous levels of analysis in human and animal models. For this reason, studies of irritability are consistent with the Research Domain Criteria approach, and a scientific focus on irritable mood is likely to generate tractable and clinically relevant critical questions about the etiology, pathophysiology, and development of mood disorders and other psychiatric conditions. Longitudinal data indicate that chronic irritability predicts increased risk for major depression, dysthymia, and generalized anxiety disorder in young adulthood (23). Since irritability manifests early in life, it would be important to elucidate critical developmental periods during which chronic irritability segues into an internalizing disorder and how genetic and environmental factors contribute to this shift.
We calculated correlations between irritability and anxious/depressed symptoms (see Table S3 in the online data supplement ), which indicated moderate levels of association between these emotional constructs for parent and child self-report. As a next step, the developmental unfolding of this relation would be interesting to examine. Pubertal development likely plays a primary role in the transition from irritability to internalizing disorder, particularly for depression, given its robust association with pubertal stage (27). Another developmentally relevant observation was that no new genetic or environmental effects emerged in young adulthood for males or females. This finding reinforces the importance of early identification of irritability and delivery of empirically supported interventions for children with impairing irritable mood.
The possibility of an evocative gene-by-environment correlation exists for the present study. An evocative gene-by-environment correlation occurs when a child’s inherited characteristics evoke a response from their environment. In our case, irritable mood in a child may elicit certain behaviors from their parents. These parental behaviors are correlated with the child’s irritable mood, which is a genetically influenced trait. Unfortunately, we do not have the type of data needed (e.g., parent and child ratings of parent behaviors) to determine the presence of an evocative gene-by-environment correlation, but the possibility that it is present is certainly plausible and should be considered in future studies.
This study relied on items of the Child Behavior Checklist to generate a previously identified irritability construct measure. Although high factor loadings were observed in this study and a previous one (16), construct validation of this phenotypic measure is lacking. Moreover, our outcomes pertain to the irritability phenotype as defined by the Child Behavior Checklist; additional research into the genetic and environmental contributions to juvenile irritable mood using other measures (e.g., the Affective Reactivity Index [35]) is needed. We should note too that we observed a decrease in the heritability of irritability in females, which is somewhat inconsistent with several other studies examining related phenotypes using comparable, albeit less robust, methods. Also, this study did not include a child report of irritability at ages 8–9, so we relied on parent report only for wave 1. The inclusion of parent and child self-report in one model to examine latent irritability indexed is a strength of the study design, as it reduces rater-specific effects. Moreover, our outcomes suggest that parent ratings were better indices of latent levels of irritable mood than child self-ratings, which may reflect better insight of parents regarding their child’s irritability levels.
Irritability is common and impairing, and it presents in the context of a number of psychiatric conditions in its chronic and episodic form. Given the transdiagnostic nature of irritability, a better understanding of the genetic and environmental contributions to its chronic and episodic expression could inform the literature and promote development of psychological and pharmacological therapeutics. Overall, genes appear to impart robust and dynamic effects on irritability throughout development, while the environment is a strong force primarily during childhood. These effects are worthy foci for future inquiry.
Supplementary Material
File (appi.ajp.2015.14040509.ds001.pdf)
- View/Download
- 315.99 KB
References
1.
Althoff RR, Verhulst FC, Rettew DC, et al: Adult outcomes of childhood dysregulation: a 14-year follow-up study. J Am Acad Child Adolesc Psychiatry 2010; 49:1105–1116
2.
Brotman MA, Schmajuk M, Rich BA, et al: Prevalence, clinical correlates, and longitudinal course of severe mood dysregulation in children. Biol Psychiatry 2006; 60:991–997
3.
Caprara GV, Paciello M, Gerbino M, et al: Individual differences conducive to aggression and violence: trajectories and correlates of irritability and hostile rumination through adolescence. Aggress Behav 2007; 33:359–374
4.
Olweus D: Stability of aggressive reaction patterns in males: a review. Psychol Bull 1979; 86:852–875
5.
Fava M, Hwang I, Rush AJ, et al: The importance of irritability as a symptom of major depressive disorder: results from the National Comorbidity Survey Replication. Mol Psychiatry 2010; 15:856–867
6.
Pickles A, Aglan A, Collishaw S, et al: Predictors of suicidality across the life span: the Isle of Wight study. Psychol Med 2010; 40:1453–1466
7.
Coccaro EF, Bergeman CS, Kavoussi RJ, et al: Heritability of aggression and irritability: a twin study of the Buss-Durkee aggression scales in adult male subjects. Biol Psychiatry 1997; 41:273–284
8.
Derks EM, Dolan CV, Hudziak JJ, et al: Assessment and etiology of attention deficit hyperactivity disorder and oppositional defiant disorder in boys and girls. Behav Genet 2007; 37:559–566
9.
Hudziak JJ, Derks EM, Althoff RR, et al: The genetic and environmental contributions to oppositional defiant behavior: a multi-informant twin study. J Am Acad Child Adolesc Psychiatry 2005; 44:907–914
10.
Hudziak JJ, Althoff RR, Derks EM, et al: Prevalence and genetic architecture of Child Behavior Checklist–juvenile bipolar disorder. Biol Psychiatry 2005; 58:562–568
11.
Derks EM, Hudziak JJ, van Beijsterveldt CEM, et al: A study of genetic and environmental influences on maternal and paternal CBCL syndrome scores in a large sample of 3-year-old Dutch twins. Behav Genet 2004; 34:571–583
12.
Burt SA, Klump KL: Delinquent peer affiliation as an etiological moderator of childhood delinquency. Psychol Med 2013; 43:1269–1278
13.
Eley TC, Lichtenstein P, Stevenson J: Sex differences in the etiology of aggressive and nonaggressive antisocial behavior: results from two twin studies. Child Dev 1999; 70:155–168
14.
Seroczynski AD, Bergeman CS, Coccaro EF: Etiology of the impulsivity/aggression relationship: genes or environment? Psychiatry Res 1999; 86:41–57
15.
Deater-Deckard K, Plomin R: An adoption study of the etiology of teacher and parent reports of externalizing behavior problems in middle childhood. Child Dev 1999; 70:144–154
16.
Stringaris A, Zavos H, Leibenluft E, et al: Adolescent irritability: phenotypic associations and genetic links with depressed mood. Am J Psychiatry 2012; 169:47–54
17.
Kendler KS, Gardner CO, Annas P, et al: A longitudinal twin study of fears from middle childhood to early adulthood: evidence for a developmentally dynamic genome. Arch Gen Psychiatry 2008; 65:421–429
18.
Kendler KS, Gardner CO, Lichtenstein P: A developmental twin study of symptoms of anxiety and depression: evidence for genetic innovation and attenuation. Psychol Med 2008; 38:1567–1575
19.
Achenbach TM: Manual of the Teacher’s Report Form and 1991 profile. Burlington, University of Vermont, Department of Psychiatry, 1991
20.
Achenbach YM: Integrative guide for the 1991 CBCL/4-18, YSR, and TRF profiles. Burlington, University of Vermont, Department of Psychiatry, 1991
21.
Achenbach TM, Rescorla LA: Manual for the ASEBA Adult Forms and Profiles. Burlington, University of Vermont, Research Center for Children, Youth, and Families, 2003
22.
Lichtenstein P, Tuvblad C, Larsson H, et al: The Swedish Twin Study of Child and Adolescent Development: the TCHAD study. Twin Res Hum Genet 2007; 10:67–73
23.
Leibenluft E: Severe mood dysregulation, irritability, and the diagnostic boundaries of bipolar disorder in youths. Am J Psychiatry 2011; 168:129–142
24.
Costello EJ, Mustillo S, Erkanli A, et al: Prevalence and development of psychiatric disorders in childhood and adolescence. Arch Gen Psychiatry 2003; 60:837–844
25.
Costello EJ, Copeland W, Angold A: Trends in psychopathology across the adolescent years: what changes when children become adolescents, and when adolescents become adults? J Child Psychol Psychiatry 2011; 52:1015–1025
26.
Merikangas KR, He J, Burstein M, et al: Lifetime prevalence of mental disorders in US adolescents: results from the National Comorbidity Survey Replication–Adolescent Supplement (NCS-A). J Am Acad Child Adolesc Psychiatry 2010; 49:980–989
27.
Angold A, Costello EJ, Worthman CM: Puberty and depression: the roles of age, pubertal status, and pubertal timing. Psychol Med 1998; 28:51–61
28.
Leibenluft E, Cohen P, Gorrindo T, et al: Chronic versus episodic irritability in youth: a community-based, longitudinal study of clinical and diagnostic associations. J Child Adolesc Psychopharmacol 2006; 16:456–466
29.
Martin LA, Neighbors HW, Griffith DM: The experience of symptoms of depression in men vs women: analysis of the National Comorbidity Survey Replication. JAMA Psychiatry 2013; 70:1100–1106
30.
Crick NR: Relational aggression: the role of intent attributions, feelings of distress, and provocation type. Dev Psychopathol 1995; 7:313–322
31.
Bongers IL, Koot HM, van der Ende J, et al: Developmental trajectories of externalizing behaviors in childhood and adolescence. Child Dev 2004; 75:1523–1537
32.
Aebi M, Müller UC, Asherson P, et al: Predictability of oppositional defiant disorder and symptom dimensions in children and adolescents with ADHD combined type. Psychol Med 2010; 40:2089–2100
33.
Akaike H: Factor analysis and AIC. Psychometrika 1987; 52:317–332
34.
Neale MC, Boker SM, Xie G, et al: Mx: Statistical Modeling, 6th ed. Richmond, Virginia Commonwealth University School of Medicine, 2003
35.
Stringaris A, Goodman R, Ferdinando S, et al: The Affective Reactivity Index: a concise irritability scale for clinical and research settings. J Child Psychol Psychiatry 2012; 53:1109–1117
Information & Authors
Information
Published In
History
Received: 21 April 2014
Revision received: 29 December 2014
Accepted: 26 January 2015
Published online: 23 April 2015
Published in print: July 01, 2015
Revision received: 23 September 2015
Revision received: 26 November 2015
Authors
Funding Information
National Institute of Mental Health10.13039/100000025: MH101518
Swedish Research Council Formas10.13039/501100001862: 2011-2492
Supported by NIMH grants K01-MH080953 and R01-MH101518 (Dr. Roberson-Nay); the Swedish Council for Working Life and Social Research (project 2004-0383) and the Swedish Research Council (project 2004-1415) (Dr. Lichtenstein); and the Division of Intramural Research Programs, NIMH (Drs. Leibenluft and Brotman).Dr. Larsson has served as a speaker for Eli Lilly and has received a research grant from Shire. The other authors report no financial relationships with commercial interests.Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBGet Access
Login options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).