Background
Conduct problems refer to antisocial behaviors displayed in childhood and/or adolescence that are symptomatic of conduct disorder and oppositional defiant disorder [
1]. They account for a substantial proportion of personal [
2], familial [
3], and societal burden [
4‐
6], and are associated with negative outcomes across multiple domains, including mental and physical health problems [
7,
8]. It is therefore crucial to thoroughly understand the etiology of such difficulties and to develop effective prevention and intervention programs.
According to Moffitt’s developmental taxonomic theory [
9,
10], individuals with elevated conduct problems can be classified into two subtypes:
early-onset persistent (also called
‘life-course persistent’) and
adolescence-limited. Early-onset persistent conduct problems are proposed to emerge in childhood, originating from genetic, congenital, or acquired neuropsychological deficits. Accumulating interactions with high-risk environments culminate in antisocial and aggressive behavior that persists throughout the lifespan. Thus, neurocognitive impairments, difficult child temperament, and adverse family environments have been proposed as the main risk factors for early-onset persistent conduct problems [
10]. By contrast, adolescence-limited conduct problems are considered to be caused by an extended period of dependence, labeled the
maturity gap, in which the individual is treated as a child despite being biologically mature [
9,
10]. This leads them to imitate the behavior of their peers showing early-onset persistent conduct problems. Thus, delinquent peer relationships are proposed to be the main determinant of adolescence-limited conduct problems [
10]. Accordingly, while early-onset persistent conduct problems are considered a neurodevelopmental disorder, adolescence-limited conduct problems are viewed as normative and transient – an exaggerated form of normal teenage rebellion [
9,
10].
The developmental taxonomic theory has been crucial in shifting focus from considering adolescent conduct problems as a unitary phenomenon to understanding different trajectories of conduct problems that may result from distinct risk factors. Nonetheless, accumulating empirical evidence from a range of disciplines suggests three potential revisions to this model: (i) the addition of a second
adolescence-onset subtype that emerges in adolescence but persists into adulthood; (ii) the inclusion of a second childhood-onset subtype, in which antisocial behavior remits in the transition from childhood to adolescence (
childhood-limited); and (iii) the reformulation of the assumption of distinct etiological causes to a model of common individual and environmental risks across subtypes, albeit with different timings and magnitudes of exposure [
11]. In sum, whereas the original developmental taxonomic theory proposes a
qualitative distinction between early-onset persistent and adolescence-limited conduct problems in terms of etiology and developmental course, there is accumulating evidence for additional conduct problem trajectories, including adolescence-onset and childhood-limited, and
quantitative differences across all subtypes – with children with early-onset persistent conduct problems being exposed to the highest levels of individual and environmental risk and those with adolescence-limited conduct problems exposed to the lowest. More precisely, the magnitude, number, and range of risk factors may be more influential in differentiating between early-onset persistent and adolescence-limited conduct problems than any individual risk factor [
12‐
14].
A key environmental risk factor implicated in the development of conduct problems is child abuse (i.e., physical, psychological, or sexual) and neglect (i.e., physical or psychological), with evidence from prospective longitudinal studies showing that those exposed to abuse and neglect in childhood and/or adolescence are at increased risk of developing conduct problems compared to those who are not exposed [
15,
16]. With respect to conduct problem trajectories, several studies have reported that child abuse and/or neglect are associated with childhood-onset conduct problems (i.e., early-onset persistent or childhood-limited), but not those that develop in adolescence (i.e., adolescence-limited or adolescence-onset), which is broadly consistent with the hypothesis of distinct risk factors across these groups [
17‐
19]. By contrast, Odgers et al. [
20] reported higher rates of child abuse and neglect in individuals with early-onset persistent, childhood-limited,
and adolescence-onset conduct problems. Collectively, these studies all provide evidence that child abuse and neglect is associated with the early-onset persistent conduct problems trajectory, but the findings for conduct problems emerging in adolescence are less conclusive. Furthermore, child abuse and neglect was exclusively measured in childhood in these studies, rather than in adolescence or both developmental periods; consequently, existing evidence is limited in terms of understanding the relationship between developmental timing of abuse and different trajectories of conduct problems, especially considering that exposure to abuse may be more common in adolescence than in childhood [
21].
Determining how the developmental timing of abuse or its persistence maps onto trajectories of conduct problems may provide new insights into the mechanisms underlying risk for conduct problems. For example, while some have proposed that childhood may be a period of particular sensitivity to adverse rearing environments, due to their potential impacts on neural, cognitive, and social development [
22], others have argued that adolescence may be a sensitive period, as it is a key stage of maturation of specific brain regions, such as the medial prefrontal cortex [
23,
24]. Alternatively, the
accumulation of negative experiences may be most relevant in determining outcomes, irrespective of their timing [
25]. However, evidence relating to timing or persistence of exposure to abuse in relation to conduct problems is limited to studies examining outcomes in adolescence and adulthood only, and these have yielded mixed findings. Thornberry and colleagues found adolescence-only and persistent abuse to be consistently predictive of adolescent and adult antisocial behavior, whereas childhood-only abuse showed weaker or null associations [
26‐
28]. By contrast, Mersky et al. [
29] found childhood-only, adolescence-only, and persistent abuse to be equally predictive of antisocial behavior in adolescents. However, these studies did not investigate conduct problem
trajectories, meaning that our understanding of the impact of timing of abuse on the longitudinal development and course of conduct problems remains limited.
To address these gaps in the literature, we examined developmental trajectories of conduct problems in a large population-based sample and differentiated between childhood-only and adolescence-only abuse, in addition to abuse occurring in both developmental periods (hereafter referred to as ‘persistent’ abuse). The main objectives of the current study were: (i) to estimate developmental trajectories of conduct problems from ages 4–17 years in a longitudinal population-based sample, extending existing trajectories from the same sample which only covered the period from 4 to 13 years [
30]; and (ii) to examine associations between exposure to abuse across childhood and/or adolescence and our derived conduct problem trajectories. We expected
temporal ordering effects; while exposure to abuse in childhood may predict the subsequent development of conduct problems in adolescence, the converse relationship would not apply. According to this logic, exposure to adolescence-only abuse would be associated with adolescence-onset, but not childhood-limited, conduct problems. Consistent with a
dose-response or
accumulative effect, we further hypothesized that ‘persistent’, as opposed to time-limited, abuse would yield the strongest effects for all elevated conduct problem trajectories, especially for the early-onset persistent trajectory, as it may cause the emergence of conduct problems in childhood and contribute to their maintenance in adolescence. For our primary analyses, we used an aggregate measure of abuse, encompassing physical, psychological, and sexual abuse, as the base rates of individual abuse subtypes for some conduct problem trajectories were low in our sample. However, in a set of exploratory analyses, we also investigated whether particular abuse subtypes were more strongly associated with the elevated conduct problem trajectories than others.
Discussion
Using data from a prospective longitudinal study with a large, population-based sample, we identified developmental trajectories of conduct problems from ages 4–17 years, and investigated links between abuse experienced at different times during development and the derived conduct problem trajectories. In contrast to previous research using developmental trajectories of conduct problems that focused on abuse experienced during childhood [
17‐
20], we used measures covering both childhood and adolescence, which enabled us to explore the impact of abuse timing and persistence. We found that abuse exposure was associated with substantially greater odds of being in the early-onset persistent and adolescence-onset conduct problem classes, particularly when it was present across both childhood and adolescence. We did not find stronger associations between child abuse and membership of the early-onset persistent compared to the adolescence-onset class, which is in contrast to some previous findings [
17‐
19]. However, it has to be noted that the adolescence-onset class showed slightly raised conduct problems already in childhood, a pattern that has also been observed in prior modeling of conduct problem trajectories [
19,
20,
30]. We also did not replicate previous findings showing an association between abuse exposure and increased odds of being in the childhood-limited class (compared to the low conduct problems class) [
19,
20]. Overall, our findings suggest that conduct problems with an onset in adolescence show similar associations with abuse to conduct problems that emerge in childhood and persist, with any differences between these trajectories being quantitative (i.e., implying common risk factors) rather than qualitative (i.e., distinct risk factors) in nature.
We extended previously published conduct problem trajectories from ages 4–13 years up to age 17 years in a large UK birth cohort [
30]. Using a full information maximum likelihood estimator and the updated 4-band categorization of the SDQ conduct problems subscale, we were able to increase the sample size (
N = 10,648) and capture more variability in conduct problems, compared to the sample size previously used to estimate developmental trajectories (
N = 7218), which also used a dichotomous approach, classifying individuals as either ‘high risk’ or ‘not high risk’ in terms of conduct problems [
30]. This has the potential to enable other researchers to examine associations between other environmental or genetic risk factors and conduct problem trajectories covering both childhood and adolescence. Furthermore, the current study brings together two areas of developmental psychopathology, namely: (i) studies using conduct problem trajectories, which, however, measured child abuse exclusively during childhood, rather than in adolescence or in both developmental periods [
17‐
20]; and (ii) studies examining the impact of timing of child abuse, which have been limited to adolescent and adult antisocial behavior, rather than developmental trajectories [
26‐
29].
In line with official UK government statistics from 2020 on child abuse in England and Wales [
47], we found that one in five participants (19.6%) reported at least one form of child abuse (i.e., ‘any abuse’). Prevalence rates for specific types of abuse were also broadly comparable with official statistics, ranging from 8 to 11%. The current study was limited to child abuse, rather than child neglect. Thus, comparisons with official statistics on the prevalence of neglect are not possible.
The current study builds on previous research by examining timing of exposure to child abuse in relation to developmental trajectories of conduct problems. Importantly, our findings support the hypothesis that persistent abuse has a more detrimental effect than time-limited abuse [
25]. Thus, in line with the cumulative risk hypothesis, abuse exposure in both childhood and adolescence was associated with greater odds of being in the early-onset persistent and adolescence-onset classes, with effect sizes twice the size of those observed for childhood-only abuse. In addition, different patterns were observed for childhood-only versus adolescence-only exposure when using the aggregate measure of abuse (‘any abuse’). Specifically, whereas childhood-only abuse was associated with increased odds of being in the early-onset persistent and adolescence-onset conduct problem classes, adolescence-only abuse was not associated with membership of any of the elevated conduct problem trajectories. The latter observation runs counter to previous research suggesting that adolescence-only abuse has more detrimental effects than childhood-only abuse [
26‐
28]. On the contrary, the current results indicate that abuse occurring in childhood may be more influential than that occurring in adolescence (at least in terms of increasing risk for conduct problems), suggesting there may be a sensitive period in which abuse is particularly likely to lead to persistent conduct problems. Alternatively, it may be that abuse occurring specifically in adolescence, versus in childhood or in both developmental periods, is experienced differently by the individual or arises for different reasons, given that significant conflict in the parent-child relationship is relatively common (and possibly normative) during adolescence [
48].
These findings for childhood-only versus adolescence-only abuse were not replicated in an exploratory analysis that examined the impacts of physical and psychological abuse separately. More precisely, adolescence-only abuse also emerged as predictive of these trajectories, alongside the positive associations already identified for childhood-only abuse. In line with many studies published in this field, small cell sizes mean that caution is essential in interpreting these findings. They also prevented us from investigating the specific impact of sexual abuse, which has been consistently linked to adolescent conduct problems [
15,
16], because of particularly low frequencies in our sample when split across conduct problem classes. As shown in the correlation matrix (see Supplementary Fig.
1), physical and psychological abuse were highly correlated, whereas correlations between these forms of abuse and sexual abuse were much weaker. This may indicate shared risk environments in which both physical and psychological abuse occur, which may explain the similar pattern of effects for these abuse subtypes. Although sexual abuse was most commonly reported in adolescence, it might be less likely to result in conduct problems if experienced within this developmental period. Consequently, the inclusion of sexual abuse in our aggregate measure of abuse may have suppressed associations with adolescence-only abuse. Future studies with larger and/or high-risk samples with a higher prevalence of sexual abuse are needed to further investigate the association between sexual abuse and conduct problem trajectories.
In contrast to the effects observed for the early-onset persistent and adolescence-onset classes, we did not find any evidence of associations between abuse and childhood-limited conduct problems, which contradicts some previous findings in this area [
19,
20]. These studies, however, also included child neglect, a form of child maltreatment not investigated in the current study, which may have influenced associations. Alternatively, individual risk factors, such as neurodevelopmental problems, may be particularly pronounced in these individuals [
11], and, thus, more relevant in the etiology of this trajectory compared to environmental risk factors such as child abuse. For example, Raine et al. (2005) found a range of neurocognitive impairments related to intelligence and memory especially in children with childhood-limited conduct problems compared to those on the low trajectory [
19].
The relationship between child abuse and the early-onset persistent and adolescence-onset conduct problem trajectories may be explained with recourse to social information processing theory [
49]. Children with aggressive behavior show biases in social information processing (e.g., hostile attributional biases) [
50]. These biases have been shown to mediate the relationship between harsh and abusive parenting and conduct problems [
51,
52]. Children may internalize their parents’ aggressive and threatening behaviors, and, as a result, rely on these aggressive schemata in future social interactions. Equipped with this limited repertoire of behaviors, children may struggle to generate non-aggressive responses to situations of conflict and may also evaluate physically and verbally aggressive responses more positively than their non-abused peers [
49]. Furthermore, there are well-established bidirectional effects in the relationship between harsh and abusive parenting and child conduct problems [
53]. Consequently, children showing conduct problems may become ensnared in coercive exchanges with their parents [
54]. By contrast, abuse experienced in adolescence might be less likely to be internalized and viewed as a behavior to emulate, which may explain the null findings for adolescence-only abuse when using the aggregate measure of abuse. Alternatively, adolescence-only abuse may be more relevant for other types of antisocial behavior, which are not assessed by the SDQ conduct problems subscale, and other forms of psychopathology. For example, Mersky et al. (2012) found that adolescence-only abuse was linked to juvenile offending (i.e., arrests, court petitions, and various types of offenses) and particularly drug-related convictions in adulthood [
29]. Finally, the non-significant associations for adolescence-only abuse may partly reflect the relative rarity of abuse only occurring in this developmental period.
Limitations
First, the findings should be interpreted in the context of limitations relating to our measures of abuse in the current study. A highly varied set of experiences could lead to an individual being classified as having experienced child abuse. This problem is inherent in any measure that attempts to capture something as complex as exposure to adversity in a scale score, but is compounded in cohort studies where low prevalence of child abuse necessitates the use of categorical variables. In addition, child abuse was assessed using retrospective self-report at age 22 years, which may have been subject to recall bias. Prospective and retrospective measures of child abuse often show poor agreement, representing two constructs with limited overlap [
55]. However, despite this discrepancy, false positives of retrospective reports of child abuse in adulthood have been shown to be rare [
56]. Furthermore, most instances of child abuse are not reported to authorities [
57], which compromises the representativeness of officially documented child abuse cases – the main alternative to self-report. Therefore, while retrospective self-report measures have limitations, it is difficult to develop feasible and ethically acceptable alternatives, particularly in large prospective cohort studies. We further used a brief measure of child abuse which has not been fully validated, although the included items are extremely similar to those included in well-established measures (e.g., Childhood Trauma Questionnaire [
58]). Future studies need to replicate our findings using a larger number of items from a measure with established psychometric properties. Finally, our measure of ‘persistent’ abuse, defined as exposure to abuse occurring in both childhood and adolescence, may have captured two isolated instances of abuse, rather than a repeated and ongoing pattern of abuse that spans childhood and adolescence. Unfortunately, the available data did not permit a more detailed approach but this issue merits investigation in future research. Second, as already noted, despite this study deriving from a large, representative birth cohort, frequencies of some forms of abuse were low across the different conduct problem trajectories. The findings relating to physical and psychological abuse particularly require replication, and we were not able to examine sexual abuse as a separate category due to very small cell counts for some classes. Third, the present study suffered from high attrition rates, possibly due to the high assessment burden and/or participants’ unwillingness to answer questions about highly intrusive experiences, potentially resulting in systematic differences between the sample included in the analyses estimating conduct problem trajectories (
N = 10,648) and the sample for whom retrospective data on child abuse were available (
n = 3172). More precisely, those with missing data were more likely to be male and in the early-onset persistent and childhood-limited conduct problem classes. This may have led to an underestimation of the effects of abuse and compromised the generalizability of our findings, particularly given that conduct problem trajectories where associated with missingness. However, we employed IPW to minimize the impact of this bias by allocating sample weights to complete cases [
45], and the findings of weighted and unweighted analyses yielded almost identical results. Fourth, the SDQ conduct problems subscale showed modest internal consistency, similar to previous research modeling developmental trajectories of behavior and emotional problems in the ALSPAC sample [
59]. Although, the SDQ is an extremely widely used measure, our findings require replication, using a measure of conduct problems with better psychometric properties. Similarly, the ESYTC, which we used to validate the derived conduct problem trajectories, showed poor reliability. Collectively, these limitations highlight the need for more reliable measures of behavior problems in young people. Fifth, relying on parent-reported conduct problems in adolescence may have underestimated the level of behavioral problems, as parents may be unaware of their child’s antisocial behavior in this developmental phase [
60]. However, the use of different informants for conduct problems versus abuse experiences minimizes potential for inflation of effects by informant bias. Moreover, considering the age range of our sample from ages 4–17 years, neither parent- nor self-report would have perfectly captured conduct problems occurring in both childhood and adolescence. Crucially, we were able to validate our conduct problem trajectories using self-reported measures of antisocial behavior during adolescence, which showed higher rates of antisocial behavior in the elevated conduct problems classes. This information, which is not typically available for studies of this type, supports the validity of our derived trajectories. Nevertheless, an important area of future research will be to compare associations between child abuse and developmental trajectories of conduct problems based on self- versus parent-report. Finally, the temporal overlap between our derived conduct problem trajectories and measures of child abuse precludes causal inferences. Thus, child abuse may be a risk factor for conduct problems or conduct problems may elicit more harsh and abusive parenting, or both factors may interact with each other in a transactional way.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.