INTRODUCTION
With the emerging concept of attention-deficit/hyperactivity disorder (ADHD) as a lifelong problem [1], the prevalence of ADHD in the adult population is increasing [2]. According to data from Korean Health Insurance Review and Assessment Service, the rate of diagnosis of adult ADHD in Korea increased 10.1-fold from 2008 to 2018 [3]. The steep increase in the prevalence of adult ADHD in Korea can be attributed to increased awareness of the disorder and the extension of insurance coverage for ADHD medications in adults [3]. From 2015 to 2018, the number of adults newly prescribed medication for ADHD tripled in Korea [4]. Clinically, there is a growing trend of patients self-diagnosing with adult ADHD based on internet resources and seeking medication for this condition.
While patients may benefit from the treatment of adult ADHD, concerns about overdiagnosis and stimulant abuse persist [5-7]. Approximately 20% of college students in the US have been prescribed stimulants [8] and the non-prescribed use of stimulants is also prevalent [9]. Young adults experiencing stressful competitive academic environments and workplace burnout commonly experience cognitive problems that mimic ADHD symptoms [10-12], and they are thus tempted to use stimulants [6,13-15]. The potential for the diagnosis of adult ADHD to serve as a justification for stimulant use, which could detract from patient welfare, highlights the need for careful diagnosis and prescription in clinical psychiatry.
Adult ADHD is typically diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [16] or through criteria-based diagnostic interviews such as the Diagnostic Interview for ADHD in Adults (DIVA-5) [17]. Several self-rating scales are also used in the diagnosis, including the Adult ADHD Self-Report Scale (ASRS) [18], Barkley Adult ADHD Rating Scale-IV [19], and Conners Adult ADHD Rating Scales (CAARS) [20]. For the prescription of stimulants in Korea, an evaluation by a psychiatrist based on diagnostic criteria and the results of one or more ADHD self-rating scale are required. However, self-rating scales are susceptible to unreliable responses from individuals feigning ADHD [21,22] and are also prone to false positives among those seeking secondary gains and those who are malingering [23]. Recall bias may further complicate diagnosis [24]. Since the scale scores rely on patient reports, diagnoses are susceptible to patient expectations. Moreover, some experts argue that the DSM-5 criteria are too broad [5], leading to overdiagnosis of ADHD. Many psychiatric conditions involve attention deficits with subjective symptoms overlapping with those of ADHD [25-27], making current diagnostic criteria and self-rating scales inadequate for differentiating ADHD from other conditions [5].
Comprehensive methods are used for diagnosing ADHD, including school reports, parent reports, self-reports, diagnostic interviews, and neurocognitive tests [28]. In childhood, the first two methods play a decisive role. However, these sources of information are often unavailable for adult patients, who typically seek hospital care independently. This absence poses challenges for clinicians in collecting objective information from adult patients. Consequently, objective findings from neurocognitive tests may play a crucial role in comprehensive diagnostics. Previous studies have identified impairments in attention, vigilance, working memory, and executive functions as the core psychopathology of ADHD [29,30]. Although neurocognitive test results of adult ADHD patients show discernable deviations from those in healthy controls, the clinical utility of these tests for diagnosing ADHD has been questioned due to their low sensitivity and specificity [31-33]. Studies aiming to differentiate cognitive issues caused by depression from those caused by ADHD using neurocognitive tests have not found significant differences in performance-based measures [34]. Furthermore, consensus is lacking on which neurocognitive domain is most relevant to adult ADHD.
Previous studies have investigated the neurocognitive characteristics of individuals diagnosed with adult ADHD using operational criteria. However, given the limitations of diagnostic criteria and self-rating scales, the populations included in these studies may not accurately represent the conceptual definition of ADHD, which differs from operationally defined ADHD. This discrepancy could limit our understanding of the core psychopathology or cognitive characteristics of patients who report attentional problems.
Therefore, our study adopted a novel approach to examine the neurocognitive profiles of adult ADHD. Rather than using diagnostic interviews and self-rating scales to identify ADHD patients, this study enrolled all patients who sought diagnosis and treatment for ADHD at the outpatient clinic. Subjects were categorized according to their clinical characteristics, and their neurocognitive profiles were examined. Additionally, we analyzed the neurocognitive characteristics associated with the severity of subjective ADHD symptoms.
METHODS
Study population and data collection
This study included 106 patients who visited Seoul National University Hospital (SNUH) and completed the “SNUH adult ADHD battery” between January 2020 and December 2022. Patient data were collected retrospectively from the Seoul National University Hospital Patients Research Environment system and electronic medical records. Data were collected on demographics (age, sex, and years of education), socioeconomic status (marital status, living arrangements, type of medical insurance, occupation, and grades at school), history of childhood ADHD and other psychiatric diagnoses, history of stimulant and other psychiatric prescriptions, family history of ADHD and other psychiatric disorders, chief complaint leading to hospital referral, current psychiatric medication, and neurocognitive test results. This study was approved by the Institutional Review Board of SNUH (IRB no. 2212-054-1384). The IRB waived the requirement for informed consent prior to data extraction.
SNUH adult ADHD battery
A computerized neurocognitive battery was constructed using resources available at SNUH [35-37], i.e., age-and-sex specific normative values for each test. The battery is composed of tests that measure attention (auditory/visual attention span, Conners Continuous Performance Test [CCPT], and Trail Making Test [TMT] A), verbal memory (immediate recall, delayed recall, and delayed recognition), visual memory (immediate recall and delayed recognition) and executive function (Wisconsin Card Sorting Test, Stroop test, and TMT B) (Table 1). All tests have been standardized and tested for validity and reliability in the Korean population, and the results are presented as T-scores. Self-rated depression (Center for Epidemiologic Studies Depression Scale [CES-D]) and ADHD symptom (Korean Adult ADHD Scale [K-AADHDS]) scales were also completed, and intelligence quotient (IQ) was measured using the Korean Wechsler Adult Intelligence Scale-Fourth Edition.
Components of the neurocognitive battery
Forward and backward auditory and visual attention span tasks were used to evaluate attention and working memory. Memory function, including encoding, storage, and retrieval, was evaluated in the verbal and visual domains. The TMT A/B was used to evaluate executive function and selective, focused, and divided attention. The Wisconsin Card Sorting Test and Stroop test were used to evaluate frontal lobe executive function.
The CCPT is widely used for diagnosing ADHD; this study used the third version of the instrument (CPT-3) [38]. The CPT-3 provides measures of different aspects of ADHD, including detectability (ability to discriminate between targets and non-targets), omission errors (failure to respond to the target), variability (response speed consistency), hit reaction time (HRT; mean response speed for correct responses), commission errors (incorrect responses to non-targets), perseveration (random, repetitive, or anticipatory responses), HRT block change (change in mean response speed across blocks), and HRT interstimulus interval (HRT ISI) change (change in mean response speed for various ISIs). Detectability, omission errors, commission errors, HRT, and variability are measures of inattentiveness. Measures of impulsivity include HRT, commission errors, and perseveration. HRT block change represents sustained attention, and HRT ISI change represents vigilance.
Self-reported ADHD scale
The K-AADHDS is the Korean adaptation of Murphy and Barkley’s Adult ADHD Scale [19], an 18-item questionnaire designed to assess inattention and hyperactivity/impulsivity. Its reliability and validity have been well-established for the Korean population [39]. While the original authors provided normative scores for this self-report measure [40], they noted that determining diagnostic thresholds for ADHD requires a careful approach and further research.
Self-reported depression scale
The CES-D was used to evaluate depressive symptoms. Originally developed by Radloff [41] and later translated into Korean [42], the CES-D is known for its high internal consistency, validity, and test-retest reliability. As a straightforward and effective tool for screening depression and exploring relationships among various variables, it is widely used in epidemiological studies [43].
Statistical analysis
The neurocognitive data were mainly analyzed in the following respects: 1) Deviation from norm values in all patients and subgroups and 2) Comparisons between the chief complaint of the ADHD group and the other chief complaint group; between patients with depressive symptoms (high CES-D score) and those without; between patients with and without a childhood (aged <12 years) diagnosis of ADHD; and between patients with and without a low IQ (below 85, the general cutoff point for borderline intellectual functioning).
Patients primarily reporting attentional problems and seeking diagnosis and treatment for adult ADHD were assigned to the “chief complaint of ADHD” group. Patients primarily reporting other symptoms (predominantly depressive mood) but also seeking a diagnosis of ADHD were assigned to the “other complaints” group.
Demographic and neurocognitive test results were compared between groups using the chi-square test and Student’s t-test. Bonferroni correction was used to correct for multiple comparisons. Correlations between self-reported ADHD scale scores and other neurocognitive scores were analyzed using Pearson correlation. Factors that showed correlations were included in stepwise multiple linear regression analysis to identify factors influencing self-reported ADHD scale scores. All statistical analyses were performed using SPSS software (ver. 25.0; IBM Corp., Armonk, NY, USA).
RESULTS
Patient characteristics
Table 2 summarizes the characteristics of the 106 patients included in the study (56.6% males; mean age at first visit, 26.6±8.0 years). Among the patients, 46 had a chief complaint of an attention problem, and 86 had CES-D scores above the cutoff for depression. The mean years of education was 13.8±2.1 and the unemployment rate was 31.1%. Among the patients whose school grades were recorded, more than half had low grades. One-fifth (21.7%) of the subjects were diagnosed with ADHD at childhood, and 18.9% had a history of stimulant therapy during childhood. In adulthood, 37.7% of the patients had a history of stimulant therapy and 23.6% were taking stimulants at the first visit.
Patients with chief complaints other than an attention problem had a higher probability of taking psychotropic medications (p<0.001). This was also seen for patients with high CES-D scores (p=0.002). Patients with a childhood ADHD diagnosis were younger (p<0.001) and less educated (p<0.001). They also had a more extensive family history of ADHD (p=0.028). Significantly more patients with a low IQ were unemployed (p=0.035).
Neurocognitive test results
Table 3 shows the patients’ average T-scores for the neurocognitive tests and the raw scores of the self-reported assessments. The majority of the T-scores were in the normal range. Verbal immediate recall was the only parameter for which the score was lower in the patients compared to the age-and-sexmatched norms (mean±standard deviation [SD]=40.57± 9.94). The patients outperformed the norm in visual immediate recall (mean±SD=67.20±8.95).
There was no significant difference in neurocognitive parameters between the chief complaint of ADHD and other complaints groups, except for the CES-D score (p=0.022). The T-score for HRT was significantly lower (p=0.037), and that of commission errors was higher, in patients with high CES-D scores (>16) compared to those with low scores (p=0.024). This indicates that more depressed patients were more inattentive and impulsive during the test. The K-AADHDS score was also significantly higher in depressed patients (p=0.001). Verbal (p=0.036) and visual memory (p=0.020) were significantly deficient in patients with a childhood diagnosis of ADHD. Patients with a low IQ performed worse on the attention span test, TMT, verbal memory, and Wisconsin Card Sorting Test, while there were no significant differences between the normal and low IQ groups in the CPT-3, Stroop test, or self-reported assessments.
Relationship between neurocognitive test results and self-rating scores
Overall, the K-AADHDS total score was significantly correlated with the CPT-3 and Stroop test subscale scores (Table 4). The K-AADHDS inattention and impulsivity/hyperactivity subscale scores showed a similar pattern of results. The inattention subscale score was correlated with the total IQ score (r=-0.328), perceptual reasoning subscale score (r=-0.376, p=0.048), and working-memory subscale score (r=-0.426, p=0.024). A similar pattern of correlations was found between the CES-D score and neurocognitive test results. The CES-D score was correlated with detectability (r=0.241, p=0.014), commission errors (r=0.295, p=0.003), and Stroop test performance. Moreover, the CES-D and K-AADHDS scores were highly correlated (r=0.367, p<0.001).
In multivariate linear regression, the CES-D total score and HRT ISI change were significantly associated with K-AADHDS variables (Table 5) (adjusted R2=0.430, β=0.457, p=0.002). Conversely, subjective inattentive symptoms of K-AADHDS were significantly associated with the CES-D total score (adjusted R2=0.278, β=0.547, p=0.001).
DISCUSSION
The results of this study indicate that in patients complaining of attention problems, many of the neurocognitive functions were not significantly impaired. The neurocognitive parameters also did not differ significantly according to the chief complaint during clinical visits. Individuals exhibiting more depressive symptoms showed more inattentive and impulsive responses on neurocognitive tests. IQ was related to deficits in various neurocognitive tests. Depressive symptoms and measures of vigilance were strongly associated with subjective ADHD symptoms.
The results indicate that patients purportedly suffering from ADHD do not exhibit significant attentional deficits in objective tests. Individuals with greater depressive symptoms had poorer performance in terms of HRT and made more commission errors. Previous studies using the CCPT suggested that it is not useful for distinguishing ADHD from other disorders [44,45]. Patients with depression tend to underperform on various neurocognitive tests, particularly those requiring sustained effort [46,47]. While attention and impulsivity are central to ADHD, they can also be influenced by mood.
This study found a correlation between changes in HRT ISI and self-reported ADHD symptoms. This aligns with previous studies that showed associations between HRT ISI and self-reported hyperactivity/impulsivity and inattentiveness [48,49]. However, vigilance can be affected by a variety of psychiatric conditions, including mood disorders, schizophrenia, and sleep-wake cycle disorders [50], with depression notably linked to decreased vigilance [51]. Given that vigilance can be affected by general psychiatric conditions, it is difficult to assert that vigilance is a specific characteristic associated with adult ADHD.
Previous studies have indicated that mood disorders frequently co-occur with adult ADHD [52,53]. Retrospective studies have found that 35%-50% of adult ADHD patients exhibit depressive symptoms [54]. These conditions may share a common risk factor; depression may develop as a result of ADHD-related impairments, or the observed overlap may stem from shared symptoms and diagnostic challenges [55-57]. The DSM-5 advises that the diagnosis of ADHD should be made when inattentive symptoms cannot be fully accounted for by other mental disorders. However, due to overlapping symptoms, it remains unclear whether these disorders are distinct entities or represent a single condition. The self-rating scales used in this study showed that individuals with profound depression perceived their attentional issues as more severe; this aligns with Harrison et al. [10], who found a significant association between depression and ADHD in university students. Bodenburg et al. [58], found that self-reported ADHD symptoms did not correlate with executive function or attention, but rather with a negative response bias influenced by negative cognitions in depressive patients. Our study showed no significant neurocognitive deficits in individuals reporting attentional problems. Moreover, neurocognitive test outcomes did not differ between individuals seeking treatment for depression versus attention. Clinically, patients reporting severe attentional problems in daily life but showing normal neurocognitive test performance may be experiencing a negative cognitive bias.
Qualitative studies report that patients perceive a positive impact of an ADHD diagnosis on identity and life aspects [59]. Korean patients experienced great relief upon learning their past difficulties were attributable to ADHD and that solutions exist [60]. In contrast, a diagnosis of depression often carries stigma and perceptions of personal weakness [61]. Thus, there is a tendency to attribute cognitive issues to ADHD and self-diagnose accordingly. These findings, alongside our study results, support the notion that late-onset ADHD might be considered a variant of depression, with clinicians potentially misdiagnosing cognitive changes associated with depression as ADHD due to their similar presentations [62]. Accordingly, the use of psychostimulants among students is closely tied to traditional middle-class values surrounding academic performance [63], suggesting that an ADHD diagnosis may medicalize underperformance in an achievement-oriented society [7].
Our results suggest that IQ affects various cognitive domains, including attention [64]. Adults with ADHD were found to have a lower full-scale IQ in a meta-analysis [65,66]. Individuals with intellectual disability may perform less well in neurocognitive tests [67]. The proportion of children meeting the DSM criteria for ADHD was significantly higher in a borderline intellectual functioning group [68]. However, diagnosing ADHD in individuals with intellectual disabilities remains controversial [69-71]. While stimulants may impact IQ scores, they do not necessarily improve intelligence [72]. Additionally, a lower IQ has been linked to a less favorable response to methylphenidate in ADHD patients [73,74]. The overlap between low intelligence and ADHD symptoms, coupled with the stigma associated with intellectual disability, may lead individuals to prefer a diagnosis of ADHD, which is perceived as a condition that can be improved. Therefore, assessing an individual’s intellectual function alongside ADHD is crucial for appropriate intervention.
Patients with a childhood diagnosis of ADHD have similar self-reported symptoms to those with adult ADHD, but show deficiencies in verbal and visual memory function. Working memory is gaining attention as a potential core deficit or endophenotype of ADHD, with deficits persisting into adulthood [75,76]. This study found that patients with and without a childhood diagnosis differed in working memory function, supporting the notion that early and late-onset ADHD are dissociable disease entities [62].
The results of this study raise a fundamental question: “What is adult ADHD?” The concept of ADHD has evolved over time, and operational diagnostic systems have changed accordingly [77]. The initial description of inattentive and problematic children by Still [78] differs greatly from the criteria outlined in the DSM-5. Operational diagnoses represent a consensus among specialists rather than an absolute depiction of reality, serving the needs of clinical medicine. Thus, the concept of adult ADHD is justifiable only to the extent that it facilitates effective treatment. Since the symptoms of ADHD are issues in daily living, evaluations are contextualized within everyday life scenarios. Our findings indicate that neurocognitive tests, despite being objective, may not be useful in adults. Treatment efficacy is assessed mainly through patient self-reports, which have inherent limitations. The therapeutic agents for ADHD, including stimulants and atomoxetine, are monoamine-augmenting drugs potentially effective in alleviating depression [79], which frequently co-occurs with ADHD. This may explain the success of treatment in adulthood ADHD.
Some scholars advocate for a deeper examination of the diagnostic criteria of ADHD and emphasize the importance of understanding the complexity of the condition [7]. A dimensional approach and spectrum model have been proposed for ADHD symptomatology [55,80]. Our findings indicate that diagnosing adult ADHD in clinical practice requires careful consideration, and further research and discussions are necessary to delineate the condition known as “adult ADHD.”
To our knowledge, this is the first study to analyze the characteristics of patients complaining of attention problems in a natural setting and to compare the characteristics of clinically distinct groups. Our approach can be distinguished from other studies that included operationally defined participants, which may elicit a systematic bias. By including all patients who visited the clinic with symptoms, this research offers a comprehensive view of attentive disorder, which is a new diagnosis surrounded by controversy. Additionally, this study integrated clinical information and the results of standardized tests covering various aspects of cognition.
However, there were some limitations to this study. First, it used data retrospectively collected at a single institution. SNUH is a tertiary hospital primarily treating patients referred from primary and secondary hospitals. Thus, the study population might not represent the broader patient population. Second, the results showed the importance of IQ in the neurocognitive function of attention, but only 34.9% of the patients underwent IQ testing. Due to the limited sample size, the associations of IQ with other neurocognitive tests and self-reported ADHD scales were low, and the contribution of IQ to ADHD symptoms might have been overlooked. Third, although the study collected data on current use of medication, the use of stimulants on the day of the test varied among participants, potentially affecting the study’s outcomes. Lastly, the data were insufficient to identify a causal relationship between adult ADHD and depression, suggesting the need for further research to explore their connection.
This study identified the neurocognitive characteristics of individuals visiting the hospital with symptoms of adult ADHD. This research may contribute to the understanding of the condition known as “adult ADHD” and its diagnosis.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
