Fluctuating course of attention-deficit/hyperactivity disorder across development: Multifactorial influences

Abstract

Attention-deficit/hyperactivity disorder (ADHD) is increasingly conceptualized as a dynamic neurodevelopmental condition, marked by fluctuating symptom trajectories across development rather than the traditional static patterns of persistence or remission. This review synthesizes empirical evidence on the varied trajectories of ADHD symptoms-including late-onset, partial remission, and recurrent fluctuation patterns-and underscores their clinical significance in long-term functioning. We adopt a multifactorial framework to explore how genetic, environmental, and gene–environment interactions contribute to the emergence and evolution of ADHD symptoms over time. In addition, we consider how medication-related variables-particularly tolerance and adherence-may influence symptom fluctuation. Characterizing these developmental dynamics offers critical guidance for designing flexible, personalized interventions that align with individual trajectories and transitional vulnerabilities.

Key Words: Attention-deficit/hyperactivity disorder; Symptom fluctuation; Developmental trajectories; Longitudinal studies; Gene-environment interaction

Core Tip: This review shifts the focus from static views of attention-deficit/hyperactivity disorder (ADHD) persistence to the developmental fluctuation of symptoms over time. Instead of focusing on prevalence or average trajectories, it emphasizes individual-level variability and synthesizes longitudinal evidence on late onset, partial remission, and recurrent fluctuation patterns. Through a multifactorial perspective, the review discusses the independent and interactive roles of genetic and environmental factors in shaping these trajectories. This lens offers a refined understanding of ADHD heterogeneity and informs developmentally tailored clinical approaches.

INTRODUCTION

Attention-deficit/hyperactivity disorder (ADHD) is a multifaceted neurodevelopmental condition characterized by core features of inattention, hyperactivity, and impulsivity[1]. Traditionally, it has been conceptualized as a distinct categorical entity, with diagnostic criteria uniformly applied across age groups using standardized behavioral measures [e.g., Diagnostic and Statistical Manual of Mental Disorders 5^th ed (DSM-5)][2,3]. However, an increasing number of clinical and longitudinal studies[4,5] challenge this static view, indicating that the core symptoms of ADHD exhibit substantial developmental variability. Rather than following a uniform, linear trajectory of remission or persistence, symptoms may fluctuate over time, reflecting diverse individual trajectories[6]. Notably, one of the most common patterns involves diagnostic shifts between syndromic, symptomatic, and asymptomatic states[7].

From a clinical perspective, fluctuating symptom patterns present significant challenges. They complicate diagnostic categorization, obscure prognostic assessments, and disrupt effective treatment planning. These fluctuations may become particularly salient during critical developmental transitions-such as adolescence and early adulthood-when heightened functional demands (e.g., independent learning, goal-directed planning, and social integration) intersect with core ADHD-related vulnerabilities[8,9]. Without timely recognition and support, these challenges may contribute to the persistence or exacerbation of long-term impairments[10]. This highlights the urgent need for clinical frameworks that explicitly incorporate the temporal dynamics of ADHD symptom expression across development.

However, there is a contradiction between the observed symptom trajectory and the rigid diagnostic criteria. This highlights two key limitations of traditional classificatory systems: (1) Their failure to account for the dynamic nature of neurodevelopment; and (2) Their oversight of the functional adaptation requirements of individuals at various developmental stages. For instance, residual attention deficits in adolescence may be misclassified as symptom remission, when they have instead transformed into a more subtle manifestation, such as executive dysfunction. Recent scientific advancements have proposed that ADHD is not a static condition limited to childhood but rather a dynamic and evolving syndrome that covers an individual’s lifespan[6,11].

The fluctuation of ADHD symptoms, as a core manifestation of its dynamic nature, reflects heterogeneity along the temporal dimension and represents a complex outcome of genetic susceptibility, environmental influences, and their interactions across biological and behavioral domains. Compared to traditional categorical models, dimensional approaches provide a more precise and effective framework for characterizing the developmental trajectory of ADHD[12]. Accordingly, building on earlier research based on static models, emphasizing the temporal dependency of symptom fluctuation and systematically investigating the impact of genetic susceptibility and environmental factors on symptom trajectories-as well as their interactions in shaping an individual long-term developmental course-will enable a more effective establishment of dynamic theoretical models. This will further drive the dynamic adaptation of general treatment strategies, ultimately enabling precision management based on individualized symptom trajectories.

In light of the above, this review aims to explore these complexities by addressing several key areas, in that we: (1) Provide an overview of the fluctuations in ADHD symptoms across development; (2) Discuss the characteristics and heterogeneity of ADHD across different developmental stages; (3) Delve into the factors influencing these fluctuations; and (4) Highlight the future research challenges in understanding and addressing the evolving nature of ADHD across the lifespan.

DYNAMIC NATURE OF ADHD: SYMPTOM TRAJECTORIES AND DIAGNOSTIC FLUCTUATIONS

Most studies have adopted a broad framework to classify ADHD symptom trajectories into three main types. The persistent type[9,13-15] refers to symptoms that begin in childhood and persist into adolescence or adulthood. The late-onset type[9,13] describes cases where symptoms first emerge during adolescence or adulthood. The remitting type[9,14] involves symptoms that diminish or even disappear over time, typically by adolescence or adulthood. These classifications are summarized in Table 1, which outlines cohort characteristics and the corresponding ADHD symptom trajectory types identified across studies[5,6,9,13-16].

Table 1 Summary of attention-deficit/hyperactivity disorder symptom trajectory classifications across longitudinal studies.

Ref.	Cohort details	Trajectory types
Agnew-Blais et al[9], 2016	Environmental risk longitudinal twin study (United Kingdom; n = 2232)	Persistent ADHD
		Remitted ADHD
		Late-onset ADHD
Caye et al[13], 2016	1993 Pelotas birth cohort (Brazil; n = 5249)	Persistent ADHD
		Childhood-limited ADHD
		Late-onset ADHD
Grevet et al[14], 2024	Midlife ADHD clinical cohort (Brazil; n = 323)	Stable persistence
		Unstable persistence
		Remission
Moffitt et al[15], 2015	Dunedin multidisciplinary health and development study (New Zealand; n = 1037)	Childhood-only ADHD
		Adult-only ADHD
		Persistent ADHD
Sibley et al[6], 2022	Multimodal Treatment Study of ADHD (United States; n = 558)	Recovered
		Stable partial remission
		Stable persistence
		Fluctuating status
Wang et al[5], 2024	Children’s school functions and brain development project (China; n = 487)	Ascending-high
		Stable-low
		Descending-medium
Carter et al[16], 2024	Millennium cohort study (United Kingdom; n = 10262)	Unaffected
		Mildly affected
		Subclinical remitting
		Adolescent onset
		Stable high

ADHD: Attention-deficit/hyperactivity disorder.

The Multimodal Treatment Study of ADHD (MTA), originally designed to compare 14 months of pharmacological and psychosocial treatments in a cohort of 579 children, has continued prospective follow-up assessments approximately biennially until 16 years post-baseline. Findings from Sibley et al[6] indicate that the percentage of fully remitted cases ranged from 1.4% (2-year assessment) to 18.5% (10-year assessment). The majority of ADHD participants who did not recover experienced either stable partial remission (15.6%) or fluctuating, waxing, and waning ADHD symptoms from childhood into adulthood (63.8%). Only a small proportion of participants exhibited stable ADHD symptoms across the entire follow-up period.

One study utilized a latent class mixed model on data from a 3-year longitudinal cohort of 487 children and adolescents to investigate the developmental trajectories of ADHD symptoms. They identified three distinct trajectory types: (1) Ascending-high; (2) Stable-low; and (3) Descending-medium[5]. This framework specifically highlights the symptom fluctuations occurring during childhood and adolescence.

Given the rapid neurodevelopmental transitions occurring during adolescence, another study extended previous analyses of ADHD symptom development by applying latent class growth analysis to data from the United Kingdom Millennium Cohort Study (n = 10262), tracking individuals from age 3 years to 17 years[16]. The optimal model identified five distinct symptom trajectory classes: (1) Unaffected (37.6%); (2) Mildly affected (34.8%); (3) Subclinical remitting (14.4%); (4) Adolescent onset (7.6%); and (5) Stable high (5.6%).

To achieve a more comprehensive characterization of ADHD, researchers have also investigated the functional impairments associated with the disorder and their relationship to symptom fluctuation. Evidence indicates that individuals with persistent or late-onset ADHD exhibit significantly greater impairments across multiple domains-including mental health, substance use, social adaptation, physical health, and socioeconomic outcomes-compared to other individuals with ADHD. Moreover, symptom remission does not equate to complete functional recovery[9], as individuals whose symptoms have remitted continue to show lagging educational attainment, such as lower levels of academic achievement[16,17].

The functional impairments associated with late-onset ADHD have prompted a reevaluation of its underlying etiology. One explanatory model[9] suggests that individuals in this subgroup may carry a genetic predisposition to ADHD, but symptom expression in childhood remains latent due to protective factors, such as a supportive family environment or well-developed cognitive abilities. However, as these individuals encounter increased cognitive demands during adolescence or early adulthood-such as multitasking and autonomous planning-previous compensatory mechanisms may become insufficient. This shift can lead to the overt manifestation of symptoms and a subsequent decline in functional capacity.

THE EVOLUTION OF ADHD SYMPTOM CLUSTERS ACROSS DEVELOPMENTAL STAGES

The network analysis of one study to examine the structure of ADHD symptoms changes across developmental stages[18] indicates a gradual differentiation of ADHD symptoms across developmental stages. During the preschool period (ages 3–6 years), ADHD symptoms present as a tightly clustered network, with inattentive and hyperactive/impulsive symptoms tightly integrated. Two symptoms are less significant and central to the network: (1) Fails to pay close attention and makes careless mistakes; and (2) Often blurts out an answer.

By childhood (ages 6–12 years), symptoms begin to differentiate into two distinct clusters: (1) Inattention; and (2) Hyperactivity/impulsivity. Within the inattention cluster, “fails to follow through or finish tasks” is the core symptom. The symptom “does not seem to listen” observed in preschool becomes a bridging symptom linking the inattention and hyperactivity/impulsivity clusters. Further, it is associated with “easily distracted”, “difficulty waiting turn”, and “interrupts or intrudes on others”. “Difficulty sustaining attention” and “easily distracted” remain pivotal in the ADHD symptom network at this stage.

In adolescence (ages 13–17 years), ADHD symptoms differentiate even further, with hyperactivity emerging as a distinct cluster, separate from inattention and hyperactivity/impulsivity, and exhibiting more nuanced differences. Among these, inattention and impulsivity symptoms form a tight network, while hyperactivity symptoms, particularly “talks excessively” and “interrupts others” are more peripheral. “Easily distracted”, a core symptom during childhood, remains a bridging symptom in adolescence, linking inattention and impulsivity symptoms. Additionally, inattention symptoms in adolescence show greater differentiation, with “reluctance to engage in tasks requiring sustained mental effort” becoming the core symptom of this stage, while the importance of “does not seem to listen” significantly diminishes.

As individuals transition into adulthood (ages 18–36 years), the ADHD symptom network maintains a structure similar to that of adolescence; however, a distinct verbal impulsivity cluster emerges, characterized by language-related impulsive behaviors, such as interrupting others and excessive talking. Simultaneously, the inattention symptoms in adulthood become more differentiated and complex, with “difficulty sustaining attention”, “reluctance to engage in tasks requiring sustained mental effort”, and “easily distracted” occupying central positions in the inattention symptom cluster. As individuals with ADHD age, the prominence of their symptoms may decrease as they are able to choose environments that align with their interests and needs independently. In some cases, individuals may subjectively perceive themselves as no longer being affected by ADHD.

However, perceived symptom remission may not directly correlate to an individual’s early life experiences or the challenges they faced[9]. This shift in understanding is reflected in the revisions to the definition of ADHD in the DSM-5, which replaced the earlier concept of fixed “subtypes” with the more flexible term “presentations”. This underscores that ADHD symptoms are not static, but instead typically exhibit dynamic changes over an individual’s lifespan[19]. From the initial recognition of the fluctuating nature of ADHD symptoms to the revisions in its definition in the DSM-5, there has been a progressive deepening of the academic understanding of this disorder.

FACTORS INFLUENCING ADHD SYMPTOM FLUCTUATIONS

Although a substantial body of research has identified a range of genetic and environmental factors contributing to the onset and persistence of ADHD, relatively few studies have explicitly examined how these influences shape short-term and long-term symptom fluctuations. In this section, we reexamine well-established genetic and environmental risk factors for ADHD through a developmental lens, with a particular focus on how their duration and interactions may help explain heterogeneity and fluctuation in symptom trajectories over time.

GENETIC FACTORS

Converging evidence indicates that ADHD is a highly heritable condition, with heritability estimates ranging from 60% to 80% during adolescence[20,21] and declining to moderate levels of approximately 37% in adulthood[22]. Notably, different symptom dimensions-such as inattention and hyperactivity/impulsivity-may be governed by distinct genetic mechanisms[22].

Importantly, longitudinal twin studies suggest that genetic influences on ADHD are not static but shift across development. Studies using Cholesky decomposition models-designed to separate age-specific from shared genetic influences-have shown that while a portion of genetic risk remains stable across development, approximately half of the genetic variance in ADHD symptoms emerging during adolescence and early adulthood arises from new, age-specific genetic factors. These findings support the presence of developmentally dynamic genetic effects and align with broader evidence of both genetic stability and innovation underlying ADHD symptom trajectories[20,23-25].

Moreover, gene–environment interactions (G × E) may further modulate the expression of genetic risk, with susceptibility becoming more pronounced under conditions of heightened environmental stress[26]. Together, these findings emphasize that genetic influences on ADHD are temporally dynamic and may play a critical role in driving symptom fluctuations across development.

A cross-sectional study found that certain genes, such as neurotrophic factor-3 and neurotrophic receptor tyrosine kinase 2, were significantly associated with ADHD only in child samples. Other genes, including serotonin receptor 2A gene, synaptosomal-associated protein 25, and nitric oxide synthase, showed associations primarily in adult samples. This finding provides indirect evidence of the association between genetic susceptibility to ADHD and different developmental stages[27].

Longitudinal studies provide more direct evidence for the sustained influence of genetic factors throughout the developmental course of ADHD. For example, one study found that carrying the dopamine D4 receptor (DRD4) 7-repeat allele was significantly associated with a persistent trajectory of ADHD symptoms into adulthood, whereas polymorphisms in dopamine gene transporter (DAT1) and serotonin transporter (5-HTTLPR) did not show similar associations. This finding highlights the potential role of specific genetic variants in the persistence vs remission of ADHD symptoms[28].

Moreover, certain genetic factors exert influence across the entire lifespan of ADHD. For instance, the ciliary neurotrophic factor receptor gene, which regulates hippocampal neuron survival and neurotransmitter systems, has been associated with susceptibility to ADHD in both childhood and adulthood, supporting the continuity of symptoms across developmental stages[27]. Additionally, oxidative stress biomarkers such as malondialdehyde, superoxide dismutase, paraoxonase activity, and total antioxidant status are widely expressed in individuals with ADHD and persist from childhood through adulthood, suggesting that oxidative stress may represent a fundamental biological mechanism shaping the course of ADHD[29]. Furthermore, the interaction between latrophilin 3-rs6551665 and the NTAD gene cluster variant rs2303380 has been shown to exert differential effects at various stages of ADHD development, influencing not only its onset in childhood but also the persistence or remission of symptoms in adulthood[30].

Polygenic risk scores (PRS), which quantify the cumulative burden of common genetic variants, have emerged as a crucial tool for investigating symptom heterogeneity in ADHD[31,32]. They offer insights not only across individuals but also within clinically defined subgroups differing in age of diagnosis, comorbidity patterns, and functional outcomes[33,34]. Studies have demonstrated that ADHD-related PRS is associated with cognitive impairments, attentional difficulties, and lower educational attainment[31] and predicts the persistence of symptoms across developmental stages. Individuals with higher PRS exhibit more severe hyperactivity, impulsivity, and inattention symptoms from early childhood through adolescence[35]. Further analyses reveal that the impact of PRS on symptom trajectories is dimension-specific, showing a particularly strong association with the worsening of inattention symptoms[35]. These advancements provide a more comprehensive genetic foundation for understanding the fluctuating course of ADHD symptoms.

Although the high heritability of ADHD is well established[20,21], existing research has largely focused on explaining static diagnostic status, with comparatively little attention to the genetic underpinnings of symptom fluctuations across development. This static orientation risks obscuring the temporal dynamics inherent in genetic liability itself. Emerging evidence from developmental twin studies demonstrates that new genetic influences during adolescence and early adulthood not only support genetic innovation[20,23,24], but may also constitute key drivers of symptom recurrence, remission, or transformation into other comorbid presentations. Currently, PRS is primarily applied to risk prediction[33,35]; however, it holds greater promise in characterizing heterogeneity in symptom trajectories and identifying subgroups with heightened vulnerability to fluctuation[34]. Integrating PRS with dynamic, developmentally sensitive approaches to symptom monitoring represents a critical yet unmet challenge for future translational research in ADHD genetics.

ENVIRONMENTAL FACTORS

In the developmental course of ADHD, environmental influences can be categorized into shared and non-shared environmental factors. The effects of shared environmental factors (e.g., family and school) on ADHD symptoms vary across developmental stages[36]. Conversely, non-shared environmental influences are typically modest in magnitude and time-specific, primarily arising from individualized life experiences that emerge during adolescence. These factors serve as critical environmental drivers of symptom fluctuations in ADHD[20,36].

The influence of birth weight on ADHD symptoms can persist into adolescence, with lower birth weight associated with more severe symptomatology. However, this effect exhibits heterogeneity across symptom dimensions. Specifically, the impact of birth weight on overall ADHD severity and hyperactivity/impulsivity symptoms tends to decline with age, remaining relatively weak[37]. In contrast, its association with inattentive symptoms appears more robust and enduring.

Focusing on follow-up data ranging from several months to 5 years, short-term longitudinal studies have revealed that the impact of the caregiving environment on short-term ADHD symptom fluctuations is both context specific and time sensitive. For instance, a previous study demonstrated that negative parental responses exacerbate ADHD symptoms by increasing children’s delay aversion tendencies, exclusively observed in waiting tasks[38].

Further studies have shown that mothers with ADHD symptoms are more likely to adopt hostile parenting styles, contributing to a bidirectional reinforcement cycle that exacerbates both maternal and child ADHD symptoms[39]. Another study found that lower levels of observed maternal warmth and paternal sensitivity during the preschool years predicted more severe inattentive and hyperactive symptoms in middle childhood[40]. Notably, the protective effects of positive environmental factors are not stable over time. Although parental warmth can temporarily buffer externalizing symptoms, its effectiveness diminishes as children age, highlighting the need for short-term interventions to dynamically adjust strategies according to developmental stage requirements[41].

Long-term longitudinal studies that follow individuals across developmental stages have revealed the cumulative impact of environmental factors on ADHD symptom trajectories. In early and middle childhood, low socioeconomic status risk and negative parenting behaviors significantly influence ADHD symptoms[42]. Moreover, higher levels of parental psychopathology, particularly emotional disorders like depression, are associated with the long-term persistence of ADHD symptoms, whereas lower depressive symptoms and positive early treatment responses may facilitate full symptom remission[4].

As children enter adolescence, the socio-emotional interactions become more influential. During this period, perceived parental rejection is associated with persistently high ADHD symptom trajectories, whereas greater perceived parental emotional warmth is linked with reduced ADHD symptom severity[43].

Additionally, digital media exposure may influence ADHD symptom fluctuations through stage-dependent effects. Short-term longitudinal studies indicate that frequent use of multiple forms of modern digital media is associated with an increased likelihood of ADHD symptom onset over a 24-month period in mid-adolescence[44]. When extending the research focus to earlier developmental stages, increased screen time in preschool is linked to more severe inattention problems. However, long-term studies using stringent clinical diagnoses (rather than symptom ratings) have found no direct association between early childhood screen time and eventual ADHD diagnosis, suggesting that the effects of media exposure may operate indirectly through comorbid behaviors, such as externalizing symptoms[45].

Most current research has focused on individual-level environmental factors associated with changes in ADHD symptoms, but these findings should not be viewed as fragmented or insufficient[4,38,39,42]. They offer valuable entry points for understanding how specific exposures may trigger or buffer symptom fluctuations over time. Although the field has yet to fully integrate these isolated findings into a cohesive developmental framework, their cumulative value lies in demonstrating that ADHD is dynamically influenced by developmentally specific, context-dependent environmental inputs-underscoring the need to move beyond static models toward a more integrated understanding of symptom fluctuation.

G × E

A central challenge in understanding the dynamic fluctuations of ADHD symptoms lies in elucidating how G × E shape neurodevelopmental trajectories through dynamic molecular mechanisms, thereby mediating symptom variability. G × E effects operate through three pathways: (1) Passive, where genetic and environmental influences are correlated; (2) Evocative, where genetic predispositions influence behaviors that elicit environmental responses; and (3) Active, where genetic factors drive environmental selection[46]. For instance, studies have shown that children with higher genetic risk for ADHD may influence their family environment through behavioral characteristics, such as inattention and impulsivity, contributing to increased household chaos[47]. Another study, employing two genetically sensitive designs-adoption-at-birth and in vitro fertilization-found that biological mothers’ ADHD symptoms predicted impulsive/activating behaviors in their children. This, in turn, influenced hostile parenting practices by adoptive mothers[39]. These findings provide empirical support for the evocative G × E correlation.

Further investigation into specific genes and environmental factors has revealed that individuals carrying at least one COMT 158Met allele (Met carriers) are more susceptible to early adverse environments. When exposed to high-risk socioeconomic status during infancy, these individuals exhibit more severe ADHD symptoms (inattention and hyperactivity/impulsivity) in childhood. In contrast, Val/Val carriers appear relatively resilient to environmental influences, showing minimal changes in ADHD symptoms even under high-risk conditions[42]. These findings provide critical evidence for the role of G × E in shaping ADHD developmental trajectories.

However, two independent longitudinal studies failed to find significant evidence supporting the effects of candidate G × E. The first study, which tracked ADHD symptom trajectories during adolescence, found that DRD4 and 5-HTTLPR genotypes did not significantly moderate the influence of family or school environments on symptom development[48]. Further investigations expanded the scope to include polygenic interactions involving DRD4, MAOA, and 5-HTTLPR; however, no significant interactions were found between genetic risk and parenting styles (rejection, overprotection, or emotional warmth)[43]. These discrepancies may reflect fundamental limitations inherent in traditional candidate gene studies, particularly their weak explanatory power of single genetic loci for complex behavioral phenotypes. In addition, such approaches often fail to account for polygenic accumulation and the dynamic temporal sequencing of environmental exposures[49].

In recent years, epigenetics has garnered increasing attention as a critical bridge linking genetic and environmental influences. From as early as the fetal stage-or even earlier-environmental exposures (e.g., maternal stress, toxins) and genetic susceptibility interact continuously through epigenetic processes such as DNA methylation, histone modifications, and noncoding RNA regulation[46]. These dynamic interactions collectively shape the risk trajectory and symptom progression patterns of ADHD.

Early epigenetic research on ADHD has primarily focused on cross-sectional studies examining the interactions among genetic factors, environmental influences, and epigenetic regulation. Large-scale epigenome-wide association studies (EWAS) have identified significant associations between DNA methylation levels and ADHD. Differentially methylated regions and single CpG sites within the apolipoprotein B and lysophosphatidic acid receptor 5 genes show significant differences between individuals with persistent ADHD and those with intermittent symptoms[50]. Regarding environment–epigenome interactions, methylation of the DAT1 gene’s 5′-UTR region is regulated by both genetic variation (variable number tandem repeat polymorphisms) and family environment. Notably, CpG1 and CpG6 sites exhibit antagonistic methylation patterns: Increased CpG6 methylation is associated with symptom remission, whereas elevated CpG1 methylation correlates with symptom exacerbation[50].

These studies primarily employ independent time-point analysis strategies. However, a key question regarding epigenetic mechanisms is whether DNA methylation levels change over time and whether such changes are associated with fluctuations in ADHD symptoms.

A series of EWAS tracking individuals from birth through late adolescence has demonstrated that DNA methylation undergoes extensive, nonlinear dynamic changes across development. This methylation trajectory, extending from early life through adolescence, provides an epigenetic framework for understanding the dynamic progression of the disorder[51]. To further explore the potential role of DNA methylation in ADHD symptom trajectories, recent longitudinal studies have provided supporting evidence. Walton et al[52] found that at birth, methylation levels at 13 CpG sites-associated with genes such as SKI, ZNF544, ST3GAL3, and PEX2-were significantly related to ADHD symptom trajectories. However, by age 7, these CpG sites were no longer associated with ADHD trajectories. Similarly, Neumann et al[53] reported that DNA methylation at birth (ERC2, CREB5) was linked to ADHD symptom trajectories from ages 4 to 15, but this association was not replicated in school-age methylation data (7–11 years). Furthermore, studies have shown that ADHD-associated DNA methylation sites tend to exhibit higher methylation levels in cord blood. Lower methylation at these sites is associated with more severe ADHD symptoms in childhood[53]. Recent adult EWAS findings also point to age-specific methylation patterns in ADHD[54].

PHARMACOLOGICAL TREATMENT

Pharmacological treatment constitutes a central component of ADHD management. However, clinical evidence indicates that symptom improvement does not follow a linear trajectory. Frequent treatment modifications-such as medication discontinuation, switching agents, dose adjustments, or treatment reinitiation-can disrupt intervention continuity, undermining stable symptom control and eliciting fluctuations in functional status[55-57].

Among treatment modifications, medication discontinuation is a commonly observed behavior. A large retrospective study reported a marked increase in treatment discontinuation rates from late adolescence into early adulthood, with a peak occurring between the ages of 18 years and 19 years. Notably, a subset of individuals resumes pharmacological treatment following initial discontinuation, resulting in an “on–off” treatment pattern[55]. Discontinuation is often attributable not to symptom remission but to a range of non–disease-related factors[55]. In both pediatric and adult populations, common contributors include adverse effects (e.g., insomnia, appetite suppression, anxiety), adherence issues (e.g., interruptions during holidays, subjective resistance), parental choices, comorbidities, and unmet treatment expectations. These factors frequently lead to treatment adjustments, thereby compromising treatment stability[56-58].

Selective dosing and scheduled medication breaks are frequently observed among children and adolescents, while in adults unplanned treatment interruptions are more likely to result from side effects or the perceived risk of misuse[56,57]. Moreover, pharmacological studies have shown that abrupt discontinuation of stimulant medications-such as methylphenidate and amphetamines-may trigger a rebound worsening of symptoms, manifested in short-term fluctuations, including decreased attention, low mood, and even motor disturbances[59]. Additional studies have linked treatment discontinuation among children and adolescents to increased risk of symptom deterioration, declines in quality of life, and elevated engagement in high-risk behaviors[55].

In addition to treatment discontinuation, the development of pharmacological tolerance may also contribute to symptom fluctuation by diminishing the efficacy of medication over time. Notably, several studies have identified cases of short-term tolerance to methylphenidate, with patients exhibiting a rapid decline in therapeutic response within days to one year. This often necessitates medication switches or temporary withdrawal to “reset” treatment effects[60]. However, long-term follow-up studies suggest that medication benefits may gradually diminish over time, rather than remain consistently sustained, thereby increasing the risk of symptom recurrence.

The MTA study, which compared the efficacy of medication management, behavioral therapy, combined treatment, and community care, found that approximately 66% of children receiving pharmacological treatment showed diminished treatment effects after 2 years, with medications no longer producing significant symptom relief. This attenuation may be influenced by factors such as declining medication adherence, developmental changes, and naturalistic variations in treatment over time[61].

Further follow-up suggested that continuous use from childhood to adolescence may not improve symptom outcomes, but may be associated with greater growth-related costs[62]. In addition, the study reported that most patients exhibited cyclical patterns of remission and relapse during follow-up, with sustained recovery observed in only 9.1% of cases[6]. These findings suggest a close association between changes in medication efficacy and fluctuation in symptom severity. Similarly, a meta-analysis involving over 9000 adult patients reported that while pharmacological interventions produced moderate effects, these benefits tended to diminish over time with continued treatment[63].

LIMITATIONS OF EXISTING RESEARCH

Although an increasing number of studies have explored the genetic and environmental mechanisms underlying ADHD symptom fluctuation-particularly G × E-findings remain inconsistent. These may partly reflect methodological challenges rather than true heterogeneity in symptom trajectories. Many studies rely on candidate gene approaches, which may fail to capture the polygenic nature of ADHD.

Moreover, environmental exposures are often assessed retrospectively or at a single time point, with limited application of longitudinal and developmentally informed designs. This methodological constraint hinders a deeper understanding of the dynamic interplay between genetic and environmental influences over time. Although initial research has begun to link G × E effects to neurobiological mechanisms-such as brain network connectivity and neurotransmitter regulation-comprehensive, integrative models remain scarce.

In addition, the potential impact of emerging treatment strategies, such as nonstimulant medications and digital adherence tools, on symptom fluctuation remain largely unexplored. Taken together, these gaps highlight the need for more rigorous, developmentally sensitive, and multidimensional approaches to advance both theoretical frameworks and clinical applications in the field of ADHD.

CONCLUSION

ADHD symptomatology fluctuates dynamically across developmental stages, environmental contexts, and intervention strategies[64]. This fluctuation is not a pathological aberration but a core feature of the disorder. Such a perspective necessitates a shift away from static, typological diagnostic frameworks toward dynamic, trajectory-based models. Accordingly, future research and clinical practice should adopt an individualized, developmentally informed perspective through which ADHD may be understood as a neurodevelopmental condition evolving through ongoing gene–environment interactions.

In light of the above, in-depth analysis of the long-term outcomes associated with distinct symptom trajectories may offer critical insights for the development of precision interventions. Given the transient nature of remission in many cases, leveraging individualized temporal data could help detect early signals of symptom fluctuation, enabling sustained monitoring and timely response to relapse following remission[65].

In the context of fluctuating ADHD presentations, it is imperative to identify the factors that modulate phenotypic expression and to conceptualize the fit between the individual and their environment as a key framework for long-term assessment and intervention[6]. Accordingly, clinical strategies should extend beyond symptom reduction to enhance the development of regulatory capacity and remain adaptable to state-dependent needs. This requires flexible adjustment of intervention intensity and modality in alignment with each individual’s developmental trajectory.

The development of future longitudinal research models is increasingly necessary to integrate biological, psychosocial, and ecological variables. A comprehensive understanding of the temporal complexity of ADHD may be key to shifting from static management toward more dynamic, precise, and individualized approaches to intervention.