Individual child factors affecting the diagnosis of attention deficit hyperactivity disorder (ADHD) in children and adolescents: a systematic review

Attention deficit hyperactivity disorder (ADHD) is a neurodevelopmental condition characterised by three core symptoms: persistent and excessive levels of inattention, hyperactivity and impulsivity [1, 2]. The Diagnostic and Statistical Manual of Mental Disorders (DSM-V) [1, 2] currently classifies ADHD into three primary subtypes: predominantly inattentive (ADHD-I), predominantly hyperactive-impulsive (ADHD-HI), and a combined subtype (ADHD-C). Although ADHD is associated with a wide range of psychosocial difficulties and impairment [3], it is increasingly recognized that children with ADHD are neurodivergent, with ADHD-related traits existing on a spectrum. Aspects of ADHD can be adaptive rather than impairing, with positive traits associated with ADHD including cognitive dynamism, non-conformism, creativity, and flexibility [4, 5]. Nonetheless, ADHD is often comorbid with other mental disorders such as oppositional defiant disorder (ODD), conduct disorder (CD), anxiety disorders, autism spectrum disorder (ASD), mood disorders, and learning and intellectual difficulties [6, 7]. While acknowledging that there are strengths associated with ADHD, it also predicts a wide range of adverse outcomes, including academic and vocational under-achievement [8], social difficulties [9], proneness to accidents and injuries [10], reduced quality of life [11], increased criminality [12], and increased risk for premature mortality [13]. Therefore, the early and accurate detection of ADHD in children is crucial to mitigate potential negative outcomes.

ADHD is relatively common, estimated to affect 5–8% of school-age children, depending on study sampling methods, location, and the diagnostic instruments used [14‐16]. Ideally, all children with ADHD would be identified early on and receive an appropriate diagnosis. However, there are wide variations in ADHD diagnosis rates that cannot be explained by unequal healthcare access or differences in ADHD symptom severity [17], raising concerns about the potential over- or under-diagnosis of ADHD. The process of obtaining an ADHD diagnosis typically involves multiple steps [18]. In the current review, the receipt and timing of the diagnosis of ADHD are understood as encompassing the entire diagnostic process. While the specifics of obtaining an ADHD diagnosis can vary across countries and settings, the process generally follows these steps: the detection of ADHD symptoms, referral for specialist assessment, comprehensive evaluation, and ultimately a formal diagnosis by a clinician. Importantly, diagnosing ADHD involves more than just identifying symptoms; it typically begins with parents or teachers recognising potential problems. Parents may seek initial consultation with a GP, who may then refer the child to a specialist. Similarly, teachers who observe ADHD symptoms can initiate a referral. Specialists then collaborate with parents, educators and the child to conduct a thorough evaluation before making a formal diagnosis, and providing appropriate support if needed. While this review will be confined to child-level factors, it is recognised that these factors both shape and are shaped by their socio-cultural environment. Receipt of an ADHD diagnosis may be the result of a complex interplay between interconnected factors operating across multiple levels, thus findings will be interpreted in relation to the family, school, health and broader systems in which individual child-level factors are embedded [19]. Consistent with the ‘goodness of fit’ model [20], whether or not ADHD traits are adaptive or maladaptive depends on the environmental context. Supportive home, school and work environments can enable children and adolescents with ADHD to express related strengths, mediating and/or compensating for ADHD symptoms that might otherwise cause distress and impairment [5], as such if the threshold for diagnosis is reached, this may not occur until later in life, potentially leading to delayed diagnosis.

Concerns have been expressed regarding the potential delay or failure to identify ADHD in children [e.g., 21‐23, as it prevents children and their families from accessing support and achieving their full potential, which may lead to cascading adverse outcomes across domains [3], such as the development of comorbid difficulties such as substance abuse [6]. Dalsgaard et al. [13] reported an elevated premature mortality rate among individuals with a delayed diagnosis of ADHD. The negative impact of ADHD can be diminished by careful monitoring and appropriate interventions, including medication [24], and various non-pharmacological interventions [25, 26]. Receipt of an ADHD diagnosis has also been shown to improve children and families’ understanding of the condition, thus providing them coping and problem-solving skills [27]. With neurodiversity affirming practices and engagement-focused family therapies, an ADHD diagnosis can empower children to strengthen or express aspects of ADHD that are adaptive in some contexts, contributing to the development of positive self-concept [28, 29]. Given the substantial personal, familial, and societal costs of ADHD [30, 31], an increased understanding of the factors influencing the receipt and timing of ADHD diagnosis may help to promote timely and accurate identification, diagnosis and service access.

There is a large gap between the age of onset and age of diagnosis in European children [22], and many children with ADHD remain undiagnosed across many countries [23, 32]. This under-diagnosis appears to be affecting certain groups of children disproportionately, with the child’s socioeconomic status (SES) [33], gender [23, 34], and race/ethnicity [35, 36] affecting the diagnosis of ADHD. A meta-analysis featuring data from 102 studies from all world regions showed that the male-to-female prevalence ratio was higher in clinical (4:1) than population studies (2.4:1) [32]. Girls are estimated to be diagnosed five years later than boys on average [37]. Children from minoritized ethnic and linguistic communities are also less likely to receive a diagnosis than White children, despite having more pronounced symptoms [38, 39].

There is also evidence suggesting that ADHD is over-diagnosed in specific groups of children [38, 40]. Rowland et al. [41] revealed that 10% of elementary students in rural North Carolina have been misdiagnosed with ADHD diagnosis when they did not meet diagnostic criteria. Likewise, within a cohort exhibiting adequate functioning, White children had a greater likelihood of receiving an ADHD diagnosis than children from minoritized ethnic backgrounds [42]. Moreover, children with a more ‘prototypical’ ADHD presentation, such as boys with behavior problems, may be over-diagnosed, even by qualified specialist clinicians [40]. It has been speculated that overdiagnosis could be the result of ADHD-related marketing from pharmaceutical companies [43], the increase in misinformation about ADHD on social media platforms leading to inaccurate self-diagnosis [44], or by the desire for parents, teachers and the medical profession to manage the disruptive behavior of children that fail to live up to social norms, demands and expectations [45].

There is concern over the over-diagnosis of ADHD due to the potential for the unnecessary medicalisation of children [46]. The economic burden of ADHD medication is substantial, and the long-term safety of different medications used to treat ADHD in children is largely unknown [47]. Medications prescribed for ADHD have been linked to weight loss, hepatotoxicity, an increased risk of suicidal ideation [48] and serious cardiovascular events [49]. However, it is important to note that medications are not incentivised for prescribers in most countries outside of the USA, which may contribute to cross-cultural differences in prescription rates. Child-related factors influencing ADHD diagnoses may differ across different settings, countries, and time periods, given different health and social care policies and systems.

Both the underdiagnosis and overdiagnosis of ADHD can present significant challenges. With the heterogenous prevalence rates of ADHD within [e.g., 17, 50 and between countries [51, 52], coupled with high levels of comorbidity with other mental disorders [7], a better understanding of the child sociodemographic and clinical factors contributing to the receipt and timing of ADHD diagnoses is needed. A past review of factors influencing the barriers and facilitators in the pathway towards an ADHD diagnosis [53] was limited to papers published before 2012. There have been advances in the recognition and treatment of ADHD, and with the rise in ADHD prevalence in the past few decades which may be due to increased awareness and detection or overdiagnosis [23, 54], a more updated review is warranted. Kappi and Martel [55] investigated barriers in seeking mental health services for ADHD in children, but focused on factors related to caregiver help-seeking rather than individual child factors that may form a barrier to diagnosis. French et al. [56] reviewed barriers to the recognition of ADHD, but restricted study to factors specific to the primary care setting. Different child characteristics, including age, gender, ethnicity, SES, symptom type and severity, and comorbid diagnoses have been investigated in relation to an ADHD diagnosis, but to our knowledge, no systematic review has been conducted synthesizing these factors. Therefore, the aim of this study is to identify child-level clinical and socio-demographic factors that are related to the identification and diagnosis of ADHD.

The systematic review protocol was pre-registered with PROSPERO (registration number: CRD42023392902), and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [57] standards were followed at all stages of this review.

This systematic review identified 16 child-related factors associated with the diagnosis of ADHD, namely: ADHD subtype, ADHD symptom severity, level of impairment, comorbid mental disorders, disruptive behavior, internalizing symptoms, social and cognitive functioning, physical health, gender, age, relative age, race/ethnicity, SES, insurance coverage, urbanicity of residence and family structure. As expected, findings indicated that greater impairment may aid recognition and actions conducive to the timely diagnosis of ADHD. Similarly, elevated ADHD symptoms played a facilitative role in ADHD diagnosis. This pattern aligns with the intuitive notion that heightened symptomology translates to increased impairment, thereby expediting recognition and progress along the diagnostic pathway [102]. This facilitation is also evident for ADHD diagnosis in children with more severe externalizing symptoms and poorer social functioning, which is likely attributable to the cumulative impact of these comorbid symptoms alongside ADHD, intensifying distress and impairment [7]. These cumulative difficulties may become increasingly problematic and noticeable to caregivers, teachers, and clinicians, motivating them to actively seek assistance [103, 104]. Current findings also indicated that a diagnosis of ADHD-C, and to a lesser extent ADHD-HI, tended to accelerate the diagnostic process, particularly the steps preceding contact with specialist services. This trend may relate to other clinical factors such as ADHD symptoms, disruptive behavior problems, social and cognitive functioning, and impairment. The impulsive-hyperactive symptoms in children with ADHD-C and ADHD-HI often interfere with social functioning and may therefore draw attention from parents and teachers much earlier [105]. For instance, a vignette study by Moldavsky et al. [106] revealed that children with ADHD-I were deemed to be less impaired and received the least referrals than those with more pronounced hyperactive-impulsive symptoms. Greater concern in relation to child functioning appears to amplify the visibility of ADHD, prompting faster problem recognition, referral, and subsequent diagnosis.

Nonetheless, not all studies found that ADHD symptom severity, disruptive behavior and impairment are not always significantly related to enhanced ADHD diagnosis. These inconsistent findings may be attributable to the differences in study methodology, context, cross-cultural variation, and/or the number and type of confounding variables controlled for in analyses. Potential mediating factors, such as child cognitive ability, health, service engagement, attitudes towards help-seeking, and SES may have influenced study results. Current review findings suggest that the specific relationship between a child’s presentation and an ADHD diagnosis often varies across different stages of the diagnostic process, and these effects may be contingent on the informants involved. Parent and teacher ratings of severity and impairment may reflect preconceived expectations regarding child behavior, or the strain they experience in relation to child behavior [107, 108]. For instance, Gadow et al. [109] found that teacher-rated intelligence and academic performance may be biased, with teacher report reflecting a child’s classroom behavior rather than their cognitive ability. Parent and teacher perceptions of ADHD may be interpreted in the context of their daily experience, concerns, and understanding of ADHD, thus their report may reflect their own distress levels, a strained relationship, or comorbid difficulties rather than the disorder itself [109].

The significance of different informants may also vary at different stages of the diagnostic process. Parent-reported symptom severity appears to influence stages where parents have greater involvement, such as early problem identification [87], but not the time taken by clinicians to make a diagnosis [99]. Sayal et al. [90] found that HKD criteria, used as a proxy for symptom severity, was not linked to parental recognition, acceptance of referral, or contact with specialists. In contrast, parent-reported impairment and family strain was related, underscoring the influential role of parental perceptions in the early stages of the diagnostic process. However, Lecendreux et al. [110] demonstrated that subthreshold ADHD symptoms in early childhood predicted an ADHD diagnosis in adolescence, and children with subthreshold symptoms face heightened risk for later adverse outcomes [111]. These findings emphasize the need for early detection and support for children who show subclinical levels of ADHD symptoms.

Cultural context also plays an important role. A study conducted in Japan [100] produced conflicting findings compared to other included studies conducted in Western countries, particularly regarding the effects of child impairment and disruptive behavior. These discrepancies potentially indicate cultural differences in parental attitudes and behaviors related to seeking help for child behavioral issues [112]. Factors such as stigmatization of mental health problems, reliance on informal social networks, and cross-cultural differences in perceptions of what constitutes normal child behavior, could explain the possible hesitation or reluctance among Japanese parents to actively seek help [113]. Arya et al. [97] investigated reasons for parental delays in seeking help in India and found that common beliefs hindering help-seeking for ADHD included perceiving their child as merely being ‘naughty’, attributing symptoms to normal developmental processes, and viewing behavior problems as stemming from a lack of strict parenting. In contrast, studies in Western countries such as Denmark and the USA, suggested that parents often cite more systemic barriers and healthcare-related stigma, including insufficient information about where to seek help, financial impediments, and negative expectations regarding professional responsiveness or willingness to initiate interventions or provide referrals to services [87, 114].

The role of comorbid mental disorders also produced mixed results. Different types of comorbid symptoms and conditions can differentially affect ADHD diagnosis and they also appear to be sensitive to contextual factors. Internalizing symptoms and disruptive behavior problems may complicate the diagnostic process by overshadowing ADHD symptoms [115]. For instance, the mood instability, inner tension, racing thoughts, and internal restlessness stemming from ADHD might be misconstrued as symptoms of mood or anxiety disorders [115, 116], which results in ADHD-related difficulties being overlooked. On the other hand, symptoms of comorbid mental health problems may act as facilitators, as the opportunity for clinical evaluation plays a pivotal role in the ADHD diagnostic process, regardless of the initial reason for assessment or contact with specialists. This is demonstrated in Barry et al.’s [86] study, where the recognition and referral of ADHD symptoms was influenced by whether children were already in receipt of services but not the severity of symptoms of comorbid disorders. Thus, the impact of comorbid mental health problems on the diagnostic process may either hinder or facilitate a diagnosis, depending on factors such as pre-existing service contact, and the overall presentation of the child, including the degree of overlap with symptoms of comorbid mental disorders.

Unexpectedly, despite ASD being frequently comorbid with ADHD, its relationship with ADHD diagnosis was not explored in the included papers. While Sikov et al.’s [84] study included autism, it was combined with other non-ADHD diagnoses as a factor. They found that children had significantly higher odds of an ADHD diagnosis if they had a comorbid diagnoses of a mental disorder, including ASD, even when inattention scores were adjusted. The presence of ASD symptoms can intensify overall impairment, as individuals experiencing both ADHD and ASD symptoms have been reported to be more severely impaired compared to those with ADHD or ASD alone [117, 118], which can prompt caregivers to seek assistance. Similarly, as discussed earlier, the identification of ASD can trigger a comprehensive clinical evaluation, potentially uncovering any additional challenges during the assessment process and facilitating ADHD diagnosis. On the other hand, given the significant overlap between ASD and ADHD symptoms such as deficits in executive functions and difficulties in social interaction [119, 120], diagnosing one condition over the other can be challenging. Individuals with ASD may also exhibit attention deficits, albeit more dependent on the specific stimuli and intrinsic motivation, complicating differential diagnosis. This symptom overlap can lead to ADHD symptoms being misconstrued as ASD symptoms by clinicians [120]. However, while some studies have investigated these two diagnoses together, they predominantly focus on how ADHD symptoms or diagnosis affect ASD diagnosis rather than the other way round [e.g., 121, 122. Therefore, exploring how the presence of an ASD diagnosis or how ASD symptoms influence the identification or detection of ADHD is an important area for future research.

With regards to sociodemographic factors, our findings indicate that boys had a consistently higher likelihood of receiving an ADHD diagnosis, or being diagnosed at an earlier age than girls, irrespective of clinical severity. This skewed gender ratio has been attributed to potential biases within diagnostic instruments, which may be more aligned with the male phenotypical expression of ADHD [34], leading to an underestimation of symptoms and severity in girls [123]. Alternatively, this gender-biased pattern may be rooted in boys’ greater propensity for exhibiting co-occurring disruptive behavior than girls [124], prompting parents and educators to seek assistance and evaluation [106, 125]. Girls are also more likely to present with the inattentive subtype and internalizing symptoms, which often go unnoticed [126]. The impairment caused by ADHD symptoms, particularly the inattentive subtype, may not become apparent until the later school years when academic demands increase and greater independence is expected from teachers. In contrast, boys may be more readily recognized due to more severe externalizing symptoms, which are deemed more impairing and disruptive in the classroom setting. Research indicates that parents and teachers display gender biases in recognizing problems, seeking help, and making referrals. Mowlem et al. [79] found that externalizing symptoms exhibited a stronger association with clinically diagnosed ADHD in girls than in boys, suggesting that when children’s symptoms deviate from gender role expectations, girls are more likely to garner specialist or parental recognition. Evidence has shown that gender-related disparities also exist in the timing of help-seeking in mental health conditions. Assistance is generally sought more for boys in childhood and early adolescence, while parents and teachers tend to seek help for girls during late adolescence [127], highlighting the complex relationship between gender-related societal norms, biases, and differences in recognising ADHD at different developmental stages.

Concerning race/ethnicity, most studies were conducted in the USA, and revealed that being Black or Latinx was associated with a reduced likelihood of receiving a diagnosis of ADHD than being White. However, there is a notable lack of analysis regarding the relationship between racial/ethnic groups other than Black and Latinx, and ADHD diagnosis in our included study. Specifically, groups such as Asians, Native Americans, and Alaskan Native children were often grouped into an ‘other’ racial/ethnic category, limiting the ability to differentiate between them. A recent study indicated that Asian children also have lower ADHD prevalence rates compared to White children [128], suggesting that racial/ethnic disparities may extend beyond Black and Latinx child populations. Therefore, future studies exploring the impact of race/ethnicity on ADHD identification and diagnosis across a broader range of racial/ethnic groups is warranted.

Moreover, while the evidence for this disparity among Black and Latinx children was consistent, there is a lack of understanding of the specific points in the diagnostic process where race/ethnicity plays a role and the reasons behind it. Several suggestions have been made regarding potential contributing factors. Some families expressed reservations about reporting ADHD symptoms or seeking care due to concerns related to stigma or racism [87]. Healthcare providers may have implicit biases and hold stereotypes against minoritized ethnic groups, potentially influencing their ADHD evaluation process and diagnostic decision-making [129]. Language may form a barrier, especially for Latinx children. Families whose English is not the primary language face challenges in accurately reporting ADHD symptoms to healthcare providers in English-speaking nations [130]. The lack of diagnostic tools in non-English languages further exacerbates this gap in care [131]. Therefore, ensuring culturally relevant care in the families’ preferred language is crucial [132]. Multicultural competency training interventions for clinicians and healthcare providers is essential to mitigate biases and equip them with the necessary awareness, skills and sensitivity in clinical and diagnostic settings [30, 128], and to foster an inclusive culture that encourage help-seeking behavior among parents from diverse backgrounds [133].

Findings for the relationship between SES and the identification of ADHD were inconsistent, potentially due to the diversity of countries in which studies were conducted, as the impact of SES on recognition and help-seeking may be contingent on the healthcare system and cultural norms in each country [127]. In the USA, the high cost of healthcare may result in inadequate ADHD diagnosis in lower SES households due to barriers in accessing healthcare. Barry et al. [86] identified engagement difficulties as a barrier for low SES families, noting challenges in schools reaching out to discuss ADHD symptoms. Bussing et al. [87] also reported that parents from low SES families in the US were more likely to cite financial barriers to accessing ADHD services than families of higher SES. However, in countries with universal healthcare, barriers to access for low SES families may be less pronounced.

Divergent findings relating to SES may also reflect variation in its definition and assessment [134]. While parental education is a common SES indicator, its impact on ADHD diagnosis may differ from other SES variables, as it may be more closely linked to caregivers' perceptions than access to services. Education and knowledge associated with parental education can shape attitudes and beliefs about ADHD, thereby influencing help-seeking behavior [135]. Highly educated parents may possess enhanced communication skills and health literacy, leading to an increased ability to navigate the healthcare system successfully [136]. The type of insurance, which was categorised separately as ‘Insurance’ rather than ‘SES’ in this review, may also serve as a proxy for SES. This categorization decision was driven by the heterogenous use of insurance coverage in the included studies. Some studies considered the presence/absence of insurance, while others compared public, private, commercial, Medicaid, and military insurance. Some studies treated insurance as a distinct variable outside of SES [e.g., 87, while some conceptualized it as an SES factor [e.g., 64. However, most studies showed no significant effects of SES on ADHD identification despite variations in study design, settings, and the conceptualization and assessment of SES.

Although some studies examined the impact of child-level factors as contributing to the over- or under-diagnosis of ADHD, they did not investigate whether these differences primarily stem from underdiagnosis or overdiagnosis. For example, some authors posited that the predictive relationship between race/ethnicity and ADHD diagnosis primarily resulted from the underdiagnosis of Black and Latinx children [e.g., 35, while others attributed it to an overdiagnosis of White children [e.g., 42. Coker et al. [35] attempted to investigate both possibilities, but their focus was on the effects at the medical treatment level rather than identification. Consequently, reports concerning under- and over-diagnosis should be interpreted with caution, and it future studies should use designs that contrast false-positive with false-negative diagnoses to address this research gap [137].