The critical need to assess pubertal development in studies of child and adolescent psychopathology

Summary

Nearly all developmental studies of youth psychopathology assess the effects of age on risk factor-youth outcomes, yet very few examine the effects of pubertal development on developmental trajectories. Growing evidence underscores the importance of both stages of puberty (adrenarche and gonadarche) in risk for psychopathology and the need to consider these developmental stages as predictors and moderators of mental health outcomes and trajectories. The purpose of this Editorial is to provide examples of this evidence and highlight gaps in our literature base as well as opportunities for future research.

Developmental science provides the foundational roots for the fields of child psychology and psychiatry. The range of developmental processes described in this issue of the Journal of Clinical Psychology and Psychiatry (JCPP) is remarkable and underscores the strong scholarship in the field. Indeed, several studies in this issue (e.g., Antony et al., 2022) examine changes in developmental trajectories of psychopathology using longitudinal and within-person designs that are capable of drawing robust conclusions regarding risk and protective factors.

As noted by Loughnan et al. (2022) in this issue, follow-up studies of children into adolescence are likely to yield even more critical insights, given substantial increases in youth psychopathology during adolescence and shifts in etiologic and developmental processes. It will be vitally important for these follow-up studies to examine key developmental processes during adolescence that are often absent in developmental studies. Pubertal development is one such process. Nearly all studies of child/adolescent psychopathology examine age as a covariate, predictor, or moderator, but few studies routinely examine pubertal development as a key ingredient to changes in psychopathology during pre-adolescent and adolescent development. Growing evidence underscores the importance of both stages of puberty (adrenarche and gonadarche) in risk for psychopathology. In this Editorial, I will provide examples of this evidence and outline the critical need to examine pubertal processes (rather than just age) in studies of youth psychopathology to address critical gaps in our developmental science.

Stages of Pubertal Development: Adrenarche and Gonadarche

Broadly speaking, pubertal development can be broken down into two stages - adrenarche and gonadarche. Adrenarche typically begins between ages 6 and 8 and is characterized by increases in adrenal androgens (e.g., dehydroepiandrosterone) that prepare the body for the transition from childhood into adolescence. Gonadarche is the second stage of puberty that typically begins two years after adrenarche. This stage is characterized by increases in testosterone in boys and estrogen in girls. Progesterone also increases in girls, but these increases occur late in gonadarche (i.e., after first ovulation). Gonadarche is marked by the well-known physical changes of puberty (e.g., growth spurts, skin changes, menses (girls), voice changes (boys) and can take several years to complete.

In general, gonadarche is more well-studied than adrenarche, and it is typically the stage of puberty used in studies of “pubertal status” (i.e., current stage of puberty, including pre-puberty, early puberty, mid-puberty, late-puberty, post-puberty) and “pubertal timing” (i.e., timing of pubertal stage relative to peers, including early, on-time, or late timing). Studies of adrenarche are relatively sparse, perhaps because of the lack of outward physical signs that can make it easy to miss in psychopathology research. Unfortunately, age alone is a poor proxy for adrenarche and gonadarche, as children who are the same age can vary pretty dramatically in their pubertal status and timing across these two pubertal stages. Indeed, the entire concept of pubertal timing is premised on the fact that some youth develop at a much earlier (or later) rate than their peers. The routine examination of age, but not puberty, in child/adolescent studies is therefore insufficient for understanding the developmental processes that may be key to psychopathology trajectories across childhood and adolescence.

Example Findings in the Field

In our lab, we have embarked on a series of studies examining how adrenarche and gonadarche may impact risk for eating disorders and their symptoms differentially in males and females. These studies have shown that adrenarche and gonadarche have sex-specific effects on genetic risk for eating disorder symptoms (e.g., binge eating, body weight and shape concerns) that would have been missed if we had simply focused on age. For example, we observed dramatic differences in genetic influences on eating disorder symptoms during adrenarche in boys, such that there are no genetic influences (heritability ~ 0%) on these symptoms in pre-adrenarche, but substantial genetic influences in late adrenarche (heritability = 44%) (Culbert et al., 2017; Mikhail et al., in press). By contrast, in girls, we found that gonadarche was critical for genetic effects, as there were substantially stronger genetic influences in mid-late gonadarche (heritability ~ 50%) than pre-early gonadarche (heritability ~ 0%) (e.g., see Klump, 2013 for a review; Mikhail et al., 2021). In both sets of studies, developmental differences in genetic influences were not due to differences in age and once activated, the magnitude of genetic effects remained constant across later ages and developmental periods, suggesting that all of the genetic diathesis for eating disorder symptoms became activated during these pubertal stages.

Taken together, these data suggest that pubertal development, rather than age, substantially impacts genetic risk for disordered eating and provides important insights into the developmental processes that contribute to sex-differentiated risk. Indeed, findings from these studies and others (e.g., Culbert et al., 2018; Klump et al., 2018; Klump et al., 2020) suggest that developmental differences in the activation and production of androgens in boys, and ovarian hormones in girls, may contribute to eating disorder risk, potentially through hormone-mediated changes in gene transcription. Although DNA sequence is fixed at birth, gene expression is not. Gene expression changes throughout development, and the adrenal and gonadal hormones activated during puberty are potent regulators of gene transcription and expression in several neural systems (e.g., serotonin, dopamine) that are disrupted in youth psychopathology. Dramatic increases in these hormones during adrenarche and gonadarche may therefore lead to differential transcription of risk and protective genes and drive sex-differentiated increases in genetic influences on eating disorder symptoms across development. Importantly, had we focused on age rather than pubertal development, we would have missed these critical developmental processes.

Gaps to Address

Although findings discussed above were specific to eating disorders, the results illustrate the substantial scientific gains that could be made by including measures of adrenarche and gonadarche in our study designs and conceptualizations of youth psychopathology. Rates of several psychological disorders significantly increase across gonadarche status in girls (e.g., depression; Chan et al., 2022), and timing of gonadarche has important effects on risk for psychopathology in both girls and boys (e.g., Hamlet et al., 2019). Yet few studies examining developmental trajectories of general risk and protective factors incorporate adrenarche (especially) or gonadarche into their assessment protocols. This lack of research substantially limits our etiologic and conceptual models.

The lack of research may also constrain the range of risk factors that we identify as clinically and statistically significant. For example, in this issue, Chen et al. (2022) conducted comprehensive and sophisticated analyses of the impact of polygenic risk scores (PRSs) on the development of psychopathology in pre-adolescent (ages 9 and 12) and adolescent (age 15) girls and boys. Findings suggested significant associations between PRSs and a number of psychological traits and disorders; however, as noted by the authors, observed associations were relatively modest. This is a relatively common finding in genetic research and may be due to the fact that psychopathology is highly polygenic, and the PRSs may not capture all relevant genetic variance. However, another possibility is that significant moderation is present – that associations between PRSs and outcomes vary by some important, additional variable(s). In studies of developmental psychopathology, that important, additional variable may be adrenarche and/or gonadarche, such that the degree of genetic influence and association between a gene or PRS may vary by pubertal status, with stronger associations between genes and psychopathology during particular stages of puberty. If this is the case, then aggregating youth across all adrenarche/gonadarche stages will yield more modest associations between risk genes and outcomes, as genetic effects may be more strongly expressed in youth at later stages of pubertal development. Indeed, changes in gene expression across adrenarche/gonadarche could be lurking in the more modest PRS-psychopathology associations and limiting a full understanding of genetic risk for psychopathology in youth across development.

It is important to note that incorporating measures of puberty will not only increase understanding of biological/genetic processes, but it may also improve our understanding of environmental risk factors that are critical to adolescent development. The Xu et al. (2022) and Lengua et al. (2022) studies in this issue show the profound impacts of minority stress and economic disadvantage, respectively, on youth psychopathology. Xu et al. (2022) described the potential importance of pubertal processes in heightening distress and psychological difficulties in lesbian, gay, and bisexual (LGB) adolescents (see Discussion), and recent findings suggest that economic disadvantage significantly shifts the curve of genetic risk for eating disorder symptoms to much earlier in pubertal development in both boys and girls (Mikhail et al., in press; Mikhail et al., 2021). Examining minority stress and economic disadvantage alone, without considering pubertal development, would miss these critically important developmental processes and points of intervention for these at-risk and often multiply disadvantaged populations.

Conclusions

In summary, we must do more to dive into the key developmental processes that characterize transitions from childhood into adolescence. Data thus far suggest that adrenarche and gonadarche are critical processes to consider as predictors of youth psychopathology as well as moderators of existing risk factor/psychopathology associations. Age alone does not adequately capture these developmental processes, and a focus on age will undoubtedly limit our etiologic and conceptual models. Importantly, in some cases, age might be a better predictor or moderator than pubertal development. However, if we fail to measure both age and pubertal development, we will be unable to identify the specificity of these predictors/processes and the active ingredients underlying developmental trajectories of risk.

Although more complicated than measuring age, there are straightforward questionnaire measures of gonadarche that show acceptable reliability and validity (see Petersen et al., 1988) and, when combined with age information, can allow for approximations of adrenarche as well (see Culbert et al. 2017 for an example). It would be ideal to measure adrenal androgens and gonadal hormones longitudinally to fully characterize pubertal development and impact on risk for psychopathology. However, given limited resources and funding in our field, these proxy measures are good approximations for initial studies of developmental trajectories that can be followed by in-depth neuroendocrine studies that fully elucidate the developmental processes and mechanisms underlying child and adolescent psychopathology.