Exploring correlations between Conners’ Continuous Performance Test and subjective measures of attention deficit hyperactivity disorder symptoms in a paediatric clinical sample

Abstract

Introduction:

Attention deficit hyperactivity disorder (ADHD) is a childhood-onset neurodevelopmental condition characterised by inattention and/or hyperactivity–impulsivity. The ADHD symptoms are often evaluated and quantified using various assessment tools, such as the Conners’ Continuous Performance Test II (CCPT-II), ADHD Rating Scale (ADHD-RS), Child Behaviour Checklist (CBCL), Clinical Global Assessment Scale (CGAS) and Clinical Global Impression Scale (CGIS). This study sought to compare CCPT-II with parent- and clinician-rated rating scales (ADHD-RS, CBCL, CGAS and CGIS) in measuring the core ADHD symptoms within the paediatric ADHD population.

Methods:

The data, gathered from a large-scale randomised controlled trial involving 172 children aged 6–12 years with ADHD, was pooled, and a Pearson correlation analysis was conducted.

Results:

No significant correlations were observed between CCPT-II and ADHD-RS, as well as the various subscales of CBCL, CGAS and CGIS.

Conclusion:

While CCPT-II may offer insights into ADHD symptomatology, its relationship with parent- and clinician-rated rating scales such as ADHD-RS, CBCL, CGAS and CGIS appears limited. Further research is warranted to elucidate the nuances of these assessment tools and their roles in evaluating ADHD.

INTRODUCTION

Attention deficit hyperactivity disorder (ADHD) is a childhood-onset neurodevelopmental condition characterised by inattention, hyperactivity and impulsivity.[1] Global estimates suggest that up to 8% of young persons have an ADHD diagnosis.[2] In Singapore, ADHD ranks as the disorder associated with the third highest degree of burden among youths aged 14 and below,[3] and is the most common problem referred to local specialist mental health service providers. Given the wide heterogeneity of symptoms, varying manifestations and high comorbidity rates, there is a need for reliable diagnostic markers, so that proper and timely treatment can be administered.

The symptoms of ADHD have negative and persistent implication on daily functioning and development,[4,5,6] with symptoms persisting into adulthood for 30%–80% of children diagnosed with ADHD.[7,8,9] Untreated and/or poorly managed ADHD is further associated with poorer academic achievement,[10,11] behavioural problems,[12] relationship difficulties,[13,14] low self-esteem[15] and other mental health conditions.[16] With appropriate assessment and early identification of ADHD, these negative outcomes can be minimised.[17,18] However, the process of diagnosing and assessing ADHD can be challenging due to the complexity and heterogeneity of its symptoms.

SUMMARY BOX

What is known?

A multi-informant approach is crucial for ADHD diagnosis and treatment planning. Consistency between the Conners’ Continuous Performance Test II (CCPT-II) and subjective measures enhances ADHD assessment, while discrepancies highlight assessment tool limitations.

What is new?

Our findings prompt several important considerations regarding the assessment and understanding of ADHD symptoms in the paediatric population and shed light on CCPT-II’s validity and clinical utility.

What is the impact?

While CCPT-II alone may be limited, it supplements other tools. Exploring alternative measures can provide a holistic view of ADHD presentation and its impact on individuals. This research highlights the need for a nuanced approach to ADHD assessment and treatment in children, potentially impacting current clinical practices.

The symptoms of ADHD often manifest in diverse domains of functioning and are observed by various individuals across at least two different settings, such as home and school. Therefore, a multi-informant approach is recommended as part of a comprehensive clinical assessment, which is essential for accurate diagnosis and effective treatment planning.[19] By gathering information from various sources, it allows for validation and cross-verification of ADHD symptoms. Consequently, consistency across symptom reports can strengthen the accuracy of the diagnosis. The diagnostic process includes a clinical assessment, which is often complemented with parent-rated rating scales such as the ADHD Rating Scale (ADHD-RS),[20] Child Behaviour Checklist (CBCL),[21] in combination with the clinician-rated Clinical Global Assessment Scale (CGAS),[22] the Clinical Global Impression Scale (CGIS),[23] as well as the Conners’ Continuous Performance Task-second edition (CCPT-II).[24]

While clinical interviews and rating scales completed by various informants can provide valuable information, they can be influenced by factors such as clinician and reporter bias, parenting styles, family dynamics and cultural beliefs, leading to individual differences in perception and interpretation of the symptom’s presentation and severity.[25,26,27] Therefore, an objective, non-invasive neurocognitive test such as CCPT-II can improve the accuracy of the diagnostic process.

A continuous performance test requires individuals to view a stimulus sequence on a screen and respond to target stimuli, while withholding response to non-target stimuli. Existing literature suggests that CCPT-II can be a valuable tool in assessing specific core features of ADHD, such as deficits in sustained attention and response inhibition, and is supportive of its validity as a measure of ADHD symptoms.[28,29,30] For instance, researchers found that cognitive deficits assessed by CCPT-II are correlated with parent-reported ADHD symptoms and day-to-day behavioural difficulties.[31] Similarly, CCPT-II results were also predictive of functional impairment and clinical severity as rated by clinicians.[32] Conversely, there are studies that reported limitations when using CCPT-II. Several studies have found that CCPT-II seems to focus on specific cognitive domains and may not fully capture the heterogeneity of ADHD symptoms.[33,34,35] A systematic review and meta-analysis conducted by Arrondo et al.[36] also found that at the clinical level, CCPT-II as a stand-alone tool only exhibited a small to moderate ability to differentiate ADHD from non-ADHD participants.

With mixed findings from existing literature,[37] there remains a need to examine how CCPT-II findings align with subjective reports from parents and clinicians. Consistency between CCPT-II and parent- and clinician-rated rating scales can enhance the evidence and its clinical utility as a measure for ADHD symptoms, while discrepancies may also highlight potential limitations in the assessment tools employed. By elucidating the association between CCPT-II and subjective measures, this can provide insights into the validity and clinical utility of CCPT-II in capturing the multifaceted nature of ADHD symptoms. Therefore, with the aspiration that the findings from our study can contribute to existing literature and the advancement of assessment protocols, this study sought to examine the validity of CCPT-II relative to other subjective methods of assessing severity of ADHD symptoms in children with ADHD.

This study hypothesised that the response error scores on CCPT-II would be positively correlated with the ADHD symptoms subscales on the parent-rated ADHD-RS and CBCL. More specifically, (a) omission errors on CCPT-II, which reflect difficulties in sustained attention, will be positively correlated with the inattention subscale of the parent-rated ADHD-RS and the attention problem domain of CBCL; (b) commission errors on CCPT-II, which reflect difficulties in inhibition and impulse control, will be positively correlated with the hyperactive–impulsivity subscale of the parent-rated ADHD-RS and the attention deficit hyperactivity problems domain of CBCL; and (c) perseveration errors on CCPT-II, which reflect difficulties in attention, impulsivity and behavioural dysregulation, will be positively correlated with the total score of the parent-rated ADHD-RS and the aggressive behaviour domain of CBCL.

Taking into consideration that CCPT-II errors capture and reflect core difficulties of ADHD, such as sustained attention and inhibition of impulsive responses, which can bring about significant functioning impairment, we hypothesised that, CCPT-II omission, commission and perseveration errors will all be negatively correlated with clinician-rated CGAS and positively correlated with clinician-rated CGIS, indicating poorer overall functioning and greater clinical severity.

METHODS

The data used in this study were collected as part of a larger randomised controlled trial that investigated the effectiveness of a brain–computer interface-based attention training program in improving inattentive symptoms in children with ADHD.[38] The original study was supported by a grant from the National Medical Research Council of Singapore (grant number CIRG11nov087) and was approved by the ethics review board of the National Healthcare Group, Singapore. Permission to use the data for the purpose of this study was obtained from the authors.

Participants

A cohort of 172 participants, aged 6–12 (mean ± standard deviation 8.6 ± 1.54) years, was recruited from Child Guidance Clinic, an outpatient psychiatric clinic at the Institute of Mental Health, Singapore. These participants were clinically assessed and diagnosed with ADHD by their attending child psychiatrist based on the Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition, text revised (DSM-IV TR), and referred to participate in a randomised controlled trial. Upon receiving the referral, the study team members reviewed the participants’ medical records to verify the absence of any pre-existing comorbidities. Subsequently, parents completed the Computerised Diagnostic Interview Schedule for Children Version IV, and children who met the criteria for significant inattentive symptoms (i.e. either predominantly inattentive or combined subtype of ADHD) were enrolled. Exclusion criteria included severe coexisting psychiatric conditions, known sensorineural deficit, epileptic seizures and intellectual disabilities.

Materials

Upon enrolment to the randomised controlled trial, a battery of assessment and rating scales were administered to the child participants and their parents to establish the baseline level of ADHD symptomology. Participants were also reviewed by a blinded clinician, who rated their functioning on CGAS and CGIS based on interviews with parents and information gathered from teachers. The battery of assessment and rating scales are discussed below.

Conners’ Continuous Performance Test-second edition

The CCPT-II is a 14-min computerised task designed to assess sustained attention. Child participants were instructed to press the spacebar as quickly as possible for any letter that appeared (targets), except for the letter ‘X’ (non-target). The outcome measures examined were as follows:[39]

1. Omission errors: Errors involving a failure to respond to targets. Higher scores are associated with inattention during the task.

2. Commission errors: Errors involving a response to the non-target. Higher scores are associated with either inattentiveness or impulsivity during the task.

3. Perseveration errors: Errors involving a response in the absence of a stimulus. Scores reflect anticipatory responses outside of context, slow responses to a previous stimulus, or random responding.

4. Reaction time: This refers to the participants’ mean response time to the stimuli.

5. Reaction time variability: This refers to the participants’ responding speed consistency.

6. Hit reaction time: This refers to the participants’ mean response time for all non-perseverative responses.

ADHD Rating Scale

The ADHD-RS is an 18-item rating scale that measures the severity of ADHD symptoms in children based on the DSM-IV TR and comprises two subscales: the inattention subscale and the hyperactive–impulsive subscale. Parents rated the severity of ADHD symptoms on a 4-point Likert scale (0 = never or rarely, 3 = very often). The scores from both subscales add up to a maximum of 54, and higher scores indicate increased clinical severity.

Child Behaviour Checklist

The CBCL is a 113-item rating scale that yields information on behavioural and emotional problems in children and adolescents. Parents rated their child’s behaviour on a 3-point Likert scale (0 = not true, 2 = very true or often true), and the scores are summarised into the following domains: anxious-depressed mood, withdrawn-depressed mood, somatic complaints, social problems, thought problems, attention problems, rule-breaking behaviour, aggressive behaviour, internalising problems, externalising problems and total problems, as well as the DSM-oriented scores, which include affective problems, anxiety problems, somatic problems, attention deficit hyperactivity problems, oppositional defiant problems and conduct problems. The T scores were used to determine if the child’s behaviour falls within a clinical range, and higher T scores indicate more significant impairment to overall functioning.

Clinical Global Assessment Scale

The CGAS is an adaptation of the Global Assessment Scale for adults and is used to reflect the general functioning of children and adolescents aged 4–16 years. The clinician assessed the participants based on their social and psychological functioning and assigned them a score between 1 and 100 (1 = most functionally impaired, 100 = superior functioning). The scores place the participants in one of ten categories (1–10 = needs constant supervision, 91–100 = superior functioning), and scores above 70 indicate normal functioning.

Clinical Global Impression Scale

CGIS was used as an overall assessment of the current severity of the participants’ ADHD symptoms. The clinician rated the severity on a 7-point Likert scale (1 = normal, not at all ill, 7 = among the most extremely ill patients). A score of ‘0’ is allocated if the participant was not assessed.

Statistical analysis

Data collected at baseline timepoint from both the intervention group and the waitlist-control group were pooled. Pearson correlations were computed to assess the relationship of CCPT-II with parent-rated ADHD-RS, CBCL, clinician-rated CGAS and CGIS, respectively. The magnitude of association was determined based on the guidelines proposed by Cohen: 0.1 < r ≤ 0.3 = small correlation, 0.3 < r ≤ 0.5 = medium correlation, r ≥ 0.5 = large correlation.[40] All analyses were conducted on the IBM SPSS Statistics version 26 (IBM Corp, Armonk, NY, USA).

RESULTS

No significant correlations were found between CCPT-II and parent-rated ADHD-RS, ADHD subscales of CBCL, clinician-rated CGAS and CGIS [Table 1]. Commission errors on CCPT-II was found to have a weak negative correlation with the thought problems subscale on parent-rated CBCL: r(159) = –0.173, P = 0.028.

Table 1.

Correlations between CCPT-II and ADHD-RS, CBCL, CGAS and CGIS.

Variable	Omissions	Commissions	Reaction time	Reaction time variability	Hit reaction time	Perseverations
Parent ADHD-RS Inattention Subscale	0.035	−0.061	0.063	0.069	0.063	0.02
Parent ADHD-RS Hyperactivity–Impulsive Subscale	−0.087	−0.075	0.075	−0.03	0.075	−0.006
Parent ADHD-RS Total	−0.039	−0.078	0.079	0.015	0.079	0.006
Clinician ADHD-RS Inattention Subscale	−0.025	−0.092	0.067	0.009	0.067	−0.042
Clinician ADHD-RS Hyperactivity–Impulsive Subscale	−0.087	−0.12	0.099	−0.023	0.099	0.007
Clinician ADHD-RS Total	−0.07	−0.124	0.098	−0.011	0.098	−0.016
CBCL Attention Problems Subscale	−0.02	−0.114	0.036	−0.021	0.036	−0.062
CBCL Attention Deficit Hyperactivity Problems	−0.079	−0.149	0.099	−0.019	0.099	0.011
CGAS	−0.016	0.081	−0.104	−0.009	−0.104	0.046
CGIS	−0.0-3	−0.04	0.049	−0.013	0.049	−0.009

Note: P<0.05 is considered statistically significant. ADHD: attention deficit hyperactivity disorder, AHD-RS: ADHD Rating Scale, CBCL: Child Behaviour Checklist, CCPT-II: Conners’ Continuous Performance Test II, CGAS: Clinical Global Assessment Scale, CGIS: Clinical Global Impression Scale

DISCUSSION

The present study examined the relationship between CCPT-II, an objective measure of ADHD, and various subjective measures obtained from parent- and clinician-rated assessment tools (i.e. ADHD-RS, CBCL, CGAS and CGIS). Contrary to our initial hypotheses, no significant correlations were found between CCPT-II scores and parent-rated ADHD-RS, CBCL or clinician-rated CGAS and CGIS. These findings prompted several important considerations regarding the assessment and understanding of ADHD symptoms in the paediatric population.

First, as discussed, ADHD is a heterogeneous condition that has diverse symptom presentations and manifestations as well as comorbidities. Parent- and clinician-rated assessment measures and CCPT-II are differing assessment modalities, and the information captured may relate to subtle but discrete aspects of ADHD. In addition, the rating scales are prone to subjectivity in interpretation by parents and clinicians, and hence distinct from those assessed by CCPT-II. Similarly, performance on CCPT-II can be affected by extraneous factors such as the child’s motivation and level of cooperation. Current international clinical practice guidelines concur that a thorough clinical assessment remains the gold standard for the diagnosis of ADHD, and the clinician needs to carefully assess how information from these various assessment tools can help the evaluation process.[41]

Second, unique characteristics of CCPT-II may have further contributed to the insignificant outcome. During the trial, the study participants did not receive co-intervention such as stimulant medication; so, it was unlikely that their CCPT-II performance could have been affected by the short-term effect of medication. Typically, CCPT-II is done in a sterile environment, which rarely mirrors the distracting conditions that are typical of the child’s daily life and are thought to significantly impact functioning, thus leading to possible false negatives and deviations from ratings provided by parents who observe the child in typical daily settings. Furthermore, with the child knowing it is an assessment, his/her level of motivation to complete the tasks may also be affected. The Hawthorne effect may temporarily influence the level of arousal and the child’s behaviour during CCPT-II evaluation.[42] Conversely, the child may also have lower levels of motivation to provide good effort on CCPT-II. Clinicians also need to consider other factors like the child’s mood, intelligence and even computer skills, which may also variably and unpredictably influence the CCPT-II test performance.

The CCPT-II may be sensitive to some cognitive deficits, but less sensitive to broader behavioural and emotional difficulties observed by caregivers.[35] It is also important to remember that CCPT-II is a snapshot assessment of the child in a brief period compared to the caregivers’ observations. Discrepancy between objective and subjective measures highlights the importance of considering multiple sources of information in ADHD assessment. The CCPT-II can provide data on cognitive functioning, while subjective ratings offer insights into observed behaviours and functioning impairments in real-world settings (e.g. home and school). The clinical utility of CCPT-II alone may be limited, and thus should not be the only tool used for assessing ADHD. However, it can still be useful as a supplement. In addition, exploring alternative measures or incorporating additional assessments could provide a more comprehensive understanding of ADHD presentation and its impact on individuals. Thus, it is important for the clinician to understand what information various assessment tools can provide and integrate these assessment modalities to provide a comprehensive assessment.

One limitation of this study is that in the original study population, children with predominantly hyperactive and impulsive subtype of ADHD were excluded. Thus, the results cannot be generalised to this population and are only limited to children with predominantly inattentive or combined subtype of ADHD. There is also limited generalisability to children outside the age range (e.g. preschool children and adolescents). It would be worthwhile looking into future studies, as early intervention is vital to reducing and limiting functional impact. Our study also included predominantly boys; this is consistent with studies and clinical experience, which suggest that boys outnumber girls in clinical samples.[43,44,45] This study also focused on those with ADHD but who do not have other significant psychiatric comorbidities. However, in the real-world setting, many ADHD children have comorbidities, which can also influence the CCPT-II result significantly.[46,47] Hence, it is important to evaluate for common comorbidities during ADHD assessment, so that results from assessment tools can be better understood and used.

Even though those with intellectual disability were excluded, it would be useful to have information on the intelligence quotient (IQ) levels of the participants, as we know that adaptive functioning in ADHD can be influenced by IQ levels.[48,49] This may confound CCPT-II or subjective rating scales. We also acknowledge the likelihood of parental reporting bias, as we recruited this sample after a diagnosis of ADHD was made by the child’s attending clinician. In addition, voluntary participation in this study could suggest a more cooperative and motivated sample of individuals; hence, this sample may not be generalised to those who did not agree to be included, for instance, those with behaviour problems who may not cooperate with the test. Finally, as ADHD symptoms need to be present in at least two different settings for the diagnosis to be made; additional information is usually obtained from their teachers in our age group of 6–12 years. Following this, it would be worthwhile to evaluate for any correlation between CCPT-II and teachers’ ratings.

Future studies can consider recruiting a more diverse sample of participants to enhance generalisability — for example, those in the preschool and adolescent age group and those with predominantly hyperactive–impulsive symptoms. There are also various other versions of continuous performance test that have tried to improve on the aforementioned limitations, and thus strive to emulate real-world situations and/or capture more domains. Future studies can, therefore, consider comparing these different objective measures with one another, such as CCPT 3 and Conners Continuous Auditory Test of Attention, or with subjective rating scales.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.