ADHD and the QbTest: Diagnostic Validity of QbTest

Abstract

Objective: We assess the diagnostic accuracy of the QbTest, which measures the cardinal symptoms of ADHD. Method: The study group comprised 182 children (mean age about 10 years), of whom 124 had ADHD and 58 had other clinical diagnosis of which 81% had ASD. Results: Only QbTest parameters for inattention and hyperactivity differentiated between ADHD and other clinical diagnoses at the p ≤ .01 level, not for measures of impulsivity. Sensitivity ranged from 47% to 67% and specificity from 72% to 84%. Positive predictive value ranged from 41% to 86%, and negative predictive value from 43% to 86%. Area under the curve varied from .70 to .80. Conclusion: The ability of the individual QbTest parameters to identify ADHD was moderate. The test’s ability to discriminate between ADHD subtypes was unsatisfactory.

Keywords

QbTest inattention hyperactivity impulsivity attention deficit/hyperactive disorder ASD diagnosis

Introduction

Over the past 20 years, many millions of children globally have been identified, diagnosed, and treated because of ADHD. It is recommended (National Collaborating Centre for Mental Health (UK) [NCCMH], 2009) that the diagnosis (Diagnostic and Statistical Manual of Mental Disorders [4th ed.; DSM-IV]; American Psychiatric Association, 1994) be made on the basis of a full clinical and psychosocial assessment, a full developmental and psychiatric history, and independent observer reports and that the diagnostic criteria of the DSM-IV (ADHD) or International Statistical Classification of Diseases and Related Health Problems (10th rev. [ICD-10]; hyperkinetic disorder, World Health Organization, 1992) be met. Many subjective instruments with the purpose of “guiding” the diagnostic work-up have been developed over the years, particularly self-rating scales or informant rating scales, for example, the Swanson, Nolan and Pelham-IV (SNAP) (Bussing et al., 2008), and the Conners parent and teacher rating scales (Conners, Parker, Sitarenios, & Epstein, 1998; Conners, Sitarenios, Parker, & Epstein, 1997). Clinical instruments and neuropsychological tests such as Continuous Performance Tests (CPTs), proposing to objectively measure the “endophenotypes” of ADHD (e.g., “inattention,” “hyperactivity,” and “impulsivity”), have entered the arena in the last decades. The idea is appealing to clinicians who struggle with interrater disagreement or contradictions found in information from parents and teachers.

CPTs have been widely used as an objective measure of attention in children with ADHD. As the name implies, CPTs require the participant to pay attention over a relatively long period of time by monitoring a sequence of auditory and/or visually presented stimuli and to respond whenever a previously designated target stimulus appears. As it is impossible to compensate once the stimulus presentation is over and as the stimuli are presented rather rapidly and over an extended period of time, even brief periods of inattention can be assessed. There are several CPTs on the market with reliable psychometric qualities. For example, a study by Soreni, Crosbie, Ickowicz, and Schachar (2009) showed that Conner’s CPT (CCPT) demonstrated higher test–retest reliability than more subjective instruments such as behavioral questionnaires. Halperin, Sharma, Greenblatt, and Schwartz (1991) showed that CPTs had adequate split-half and test–retest reliability.

Meta-analyses of validity have shown that children with ADHD generally perform poorly on CPTs compared with those without ADHD (Losier, McGrath, & Klein, 1996). However, McGee, Clark, and Symons (2000) showed that CCPT did not differentiate between children with ADHD from clinical controls. Zelnik, Bennett-Back, Miari, Goez, and Fattal-Valevski (2012) examined conditional probabilities for the Test of Variables of Attention (TOVA) finding a sensitivity of .91 and specificity of .21, that is, 79% false positives when used as a diagnostic instrument. Edwards et al. (2007) evaluated CCPT regarding validity and utility in assessment of ADHD in a study of 104 children 6 to 12 years of age. The CCPT performed better than a random test in classifying ADHD, but the receiver operating characteristic (ROC) analysis showed the accuracy of the CCPT to be low.

The CPTs mentioned above measure many of the core symptoms in ADHD, but “hyperactivity” is not included. A CPT test measuring aspects of motor activity is the QbTest. Sharma and Singh (2009) reported that the results of the QbTest agreed with ADHD diagnosis in 90% of cases. They estimated sensitivity at .96 and specificity at .81, respectively. Vogt and Shameli (2011) evaluated the clinical utility of the QbTest and found that it provided an increased robustness of the clinical diagnosis. There is a clear need to further evaluate the diagnostic accuracy of the QbTest in the assessment of ADHD and to (a) study QbTest’s ability to identify ADHD in clinical groups, and (b) assess its power to separate ADHD subtypes from each other. The present study was launched with these two objectives.

Method

Procedure

The study was conducted between 2005 and 2009 at the Child Neuropsychiatry Clinic (CNC), a statewide regional and specialized clinic for the assessment of ADHD, autism, and other neurodevelopmental disorders, in Gothenburg, Sweden. A total of 322 children between 6 and 12 years of age were referred to CNC with suspected ADHD, autism, or another neurodevelopmental disorder during this period and were all given the QbTests as part of a neuropsychiatric assessment. All the children were comprehensively evaluated including examination by a child psychiatrist/pediatric neurologist, and a neuropsychologist. The diagnosis of ADHD and other neurodevelopmental disorders was performed according to “gold standard” in clinical setting including assessment by a multi-professional team using LEAD procedure (Longitudinal, Experts, All, Data; Spitzer, 1983). The diagnosis was based on behavioral criteria according to the DSM-IV on the basis of all available information except that collected from the QbTest. The diagnostic decision was not made on the basis of results obtained at the QbTest. Different rating scales were used as guidelines in the diagnostic work-up, but the clinician’s judgment was decisive. Exclusionary criteria were as follows: (a) ongoing medication with central stimulants at the time of the assessment (n = 85), (b) not valid QbTest (e.g., computer failure, noncompliance; n = 32), (c) Wechsler Intelligence Scale for Children-IV (WISC-IV) or Wechsler Preschool and Primary Scale of Intelligence-III (WPPSI-III) full scale IQ at or below 70 (n = 18), and (d) syndromal medical disorder diagnosis including 22q11 deletion syndrome (n = 4) or Ehler Danlos syndrome (n = 1). This means that 182 children were left for inclusion in the study. Retrieval of data was made by one of the authors (J.K.) who examined all CNC-files and registered diagnoses, IQ, comorbidity, QbTest scores, and ongoing medication at the time of the assessment.

Participants

As already mentioned, 182 children met criteria for inclusion in the study; these including the ADHD group (n = 124), with a mean age of 10.3 ± 1.7 years and the non-ADHD clinical comparison (CC) group (n = 58) with a mean age of 10.8 ± 1.8 years (Table 1). The boy:girl ratios were 3:1 in the ADHD group and 9:1 in the CC group. In the ADHD group, 88 children were diagnosed with ADHD combined subtype, 30 with inattentive subtype, 2 with hyperactive/impulsive subtype, and 4 were diagnosed as ADHD-NOS (ADHD–not otherwise specified). As shown in Table 1, the most frequent comorbidities in the ADHD group were developmental coordination disorder (DCD; 32%), dyslexia (31%), and autism spectrum disorder (ASD; 28%). In the CC group, the majority of the participants had ASD (81%). Borderline intellectual level (IQ = 70-84), language disorder, tics, and depression/anxiety disorder were found in small subgroups in both the ADHD and CC groups. No significant differences were found between groups regarding age or full scale IQ. Number of comorbid diagnosis averaged 2.4 (SD = 1.4) in the ADHD group and 1.4 (SD = 0.6) in the CC group. For assessment of effects of ASD comorbidity on QbTest performance, we divided the ADHD group (n = 124) in two subgroups: ADHD + ASD (n = 35) and ADHD−no ASD (n = 89).

Table 1.

Demographical and Clinical Descriptive of the Sample.

	ADHD n = 124	CC n = 58	p
	n (%)	n (%)	p
M age (SD)	10.3 (1.7)	10.8 (1.8)	n.s.
Mean full scale IQ (SD)	89.5 (13.2)	92.2 (14.6)	n.s.
Male/female	97/27 (78/22)	53/5 (91/9)	.001
Autism spectrum disorders	35 (28)	47 (81)	.001
Tic disorders	5 (4)	7 (12)	n.s.
DCD	39 (32)	4 (7)	.001
Borderline intellectual functioning	12 (10)	9 (16)	n.s.
Dyslexia	38 (31)	6 (10)	<.01
Language disorder	11 (9)	6 (10)	n.s.
Depression/anxiety disorder	6 (5)	4 (7)	n.s.

Note. CC = clinical comparison; DCD = developmental coordination disorder.

Instrument

The QbTest is a computerized CPT including measures of inattention/impulsivity combined with a motion tracking device recording activity measures. The QbTest measures the three cardinal symptoms of ADHD; inattention, hyperactivity, and impulsivity presented in the test report as cardinal parameters—QbInattention, QbActivity, and QbImpulsivity. The test report is calculated by the QbTest software. There is no total ADHD score provided in the QbTest report.

The test developer assumed that some parameters measure the same underlying construct. By using statistical procedure of factor analysis (principal–component analysis) to identify parameters that are strongly correlated with each other, three cardinal parameters were created to aid in the test result interpretation. Each parameter included in the cardinal parameters is weighted differently depending on their correlation (factor loading) to the cardinal parameter (Knagenhjelm & Ulberstad, 2010). QbActivity includes data from the parameters Time Active, Distance, Area, and Micro Events registered during the second half of the test. QbInattention include the parameters Omission Errors, Reaction Time, and Reaction Time Variation during the second half of the test. QbImpulsivity is computed from Commission Errors, Normalized Commission Errors, and also includes Anticipatory responses. The cardinal parameters are weighted by a component score coefficient. During the attention task test, a high-resolution infrared camera monitors the head movements of the participant responding to stimuli appearing on the computer screen. The results are compared with a norm group of the same age and gender. The raw score is transformed and presented as Q-score, equivalent to Z-score, and percentiles in the test report (Knagenhjelm & Ulberstad, 2010). The norm sample, according to the QbTest manual, comprised 426 children aged between 6 and 12 years from six different schools situated in cities of different sizes and level of urbanization in the central parts of Sweden. The sample was representative of Swedish demographics in terms of parent ethnicity, parent marital status, possession of car in the household, and the free-time activities that the child took part in.

Cardinal parameters

QbInattention provides an index of inattention based on Omission Errors, Reaction Time, and Reaction Time Variation (see Points 1-3 below).

QbActivity provides an index of the patient’s ability to regulate motor activity. It is based on Time active, Distance, Area, and Micro Events (see Points 4-7 below).

QbImpulsivity provides an index of impulsivity based on Commission Errors, Normalized Commission Errors, and Anticipatory Responses (see Points 8-10 below).

Attention parameters

1. Reaction Time, measured in milliseconds, is the average time between stimulus presentation and correct button press.

2. Reaction Time Variation is the standard deviation of the Reaction Time and reflects Reaction Time consistency

3. Omission Errors is a measure of registered omitted responses to the targets.

Activity parameters

4. Time active is a measure of the test participant’s movement during the test period and is calculated on a second to second basis. Each second with movement more than 1 cm is counted.

5. Distance indicates the amount of total activity for the test period and is measured as the distance in meters that the marker on the headband has traveled.

6. Area reflects how vivid the movements are during the test, measured as the surface covered by the marker on the headband reflector.

7. Micro Events reflect the degree of activity during the test period, which is measured by quantifying a change in position greater than 1 mm since the last micro event.

Impulsivity parameters

8. Commission Errors occur when a response is registered on a nontarget stimulus.

9. Normalized Commission Errors is measured by examining the percentage of Commission Error rates relative to the percentage of correct responses to target.

10. Anticipatory is a response detected just before or just after a stimulus is presented (“guesses”).

Statistical Analysis

The Mann–Whitney U test was used for group comparison. The Pearson two-tailed test was used to correlate number of comorbidities and QbTest cardinal parameters. The level of significance was set at p ≤ .01. For evaluation of the diagnostic utility of QbTest cardinal parameters, we used ROC curves and computed the area under the curve (AUC). We also calculated sensitivity (Se), specificity (Sp), positive predictive value (PPV), and negative predictive value (NPV)—with a 68% prevalence of ADHD in this clinical sample. This required both the use of a reference standard, “gold standard” for diagnosis and a cutoff score for the QbTest. Our “gold standard” was the clinical diagnosis of ADHD and the QbTest cutoff was a q-score of 1.25 as recommended by the test developer. For all statistical analyses, we used SPSS statistical package system version 20.

Ethics

The study was approved by the Ethics Committee at the University of Gothenburg.

Results

QbTest Cardinal Parameters in ADHD Total Group and CC Group

The cardinal parameters QbActivity and QbInattention median scores were significantly higher, indicating more deficits, in the ADHD total group than in the CC group (Table 2). No difference was found between the ADHD total group and CC group regarding QbImpulsivity.

Table 2.

Comparisons of ADHD Group (n = 124) With CC (n = 58) in QbTest Parameters Using Mann–Whitney U Test.

QbTest parameter	ADHD total	CC
	n = 124	n = 58
	Q-score (SD)	Q-score (SD)
QbActivity	1.7 (1.2)*	0.5 (1.2)
QbInAttention	1.2 (1.1)*	0.2 (0.9)
QbImpulsivity	1.1 (1.6)	0.4 (1.3)
Activity
Time active	1.1 (0.8)*	0.6 (1.1)
Distance	1.8 (1.5)*	0.6 (1.3)
Area	1.9 (1.5)*	0.8 (1.5)
Micro Events	1.5 (1.0)*	0.6 (1.1)
Inattention
Reaction time	0.6 (1.3)*	0.1 (1.1)
Reaction time variation	2.0 (2.0)*	0.1 (1.8)
Omission error	1.5 (1.0)*	0.6 (1.3)
Impulsivity
Commission error	0.8 (1.1)	0.0 (1.0)
Normalized commission	0.8 (1.4)	0.1 (1.4)
Anticipatory	0.9 (1.1)*	0.2 (1.0)

Note. CC = clinical comparison.

p = .01.

QbTest Standard Parameters in ADHD Total Group and CC Group

Group medians were significantly higher in the ADHD total group in 8 out of 10 ordinary test parameters (Table 2). No differences were found between ADHD total and CC group regarding Commission Error and Normalized Commission. All ordinary test parameters measuring activity were significantly different between ADHD group and CC group.

Diagnostic Accuracy of Cardinal Parameters in ADHD Total Group

The cardinal parameters ability to correctly classify those with and without ADHD is presented as AUC in Table 3. In the ADHD total group (n = 124) QbInattention and QbActivity showed a moderate overall capacity (AUCs = .76 and .74, respectively) to identify true positives. QbImpulsivity was found to have a relatively weak overall capacity (AUC = .62). Sensitivity ranged between .42 and .66 and specificity between .72 and .83. PPVs ranged from .76 to .86 and NPVs from .37 to .50.

Table 3.

Clinical Utility of QbTest for Identifying ADHD and ADHD Subtypes With Cutoff Set at Recommended 1.25 Q-Score.

	95% Confidence interval
	AUC	Se	Sp	PPV	NPV
ADHD total (n = 124)
QbActivity	.74 [.66, .82]	.63 [.56, .73]	.74 [.61, .84]	.84 [.75, .91]	.50 [.39, .61]
QbInAttention	.76 [.69, .84]	.48 [.39, .57]	.83 [70, .91]	.86 [.75, .93]	.43 [.34, .53]
QbImpulsivity	.62 [.53, .70]	.42 [.33, .51]	.72 [.59, .83]	.76 [.64, .86]	.37 [.28, .46]
ADHD-Combined (n = 88)
QbActivity	.74 [.66, .83]	.67 [.56, .76]	.74 [.61, .84]	.80 [.68, .88]	.60 [.47, .71]
QbInAttention	.77 [.69, .85]	.51 [.40, .62]	.83 [.70, .91]	.82 [.69, .90]	.53 [.42, .63]
QbImpulsivity	.62 [.53, .71]	.44 [34, 55]	.72 [59, 83]	.71 [57, 82]	.46 [36, 57]
ADHD-Inattentive (n = 30)
QbActivity	.73 [.63, .84]	.60 [.41, .77]	.74 [.61, .84]	.55 [.37, .71]	.78 [.65, .88]
QbInAttention	.76 [.66, .86]	.47 [.29, .65]	.83 [.70, .91]	.58 [.37, .77]	.75 [.62, .85]
QbImpulsivity	.62 [.50, .74]	.37 [.21, .56]	.72 [.59, .83]	.41 [.23, .61]	.69 [.56, .80]

Note. AUC = area under the curve; Se = sensitivity; Sp = specificity; PPV = positive predictive value; NPV = negative predictive value.

Diagnostic Accuracy of Cardinal Parameters in ADHD Subtypes

In the ADHD combined group (n = 88), QbInattention and QbActivity showed a moderate overall capacity (AUCs = .77 and .74, respectively) to identify true positives. QbImpulsivity was found to have a slight overall capacity (AUC = .62). Sensitivity ranged between .44 and .67 and specificity between .72 and .83. PPVs ranged from .71 to .82 and NPVs from .46 to .53 (Table 3).

Regarding predominantly inattentive subtype (n = 30), AUCs for QbInattention and QbActivity was moderate (AUCs = .73 and .76), whereas QbImpulsivity was slight (AUC = .62). Sensitivity ranges were between .37 and .60 and specificity between .72 and .83. PPVs ranged from .41 to .58 and NPVs from .69 to .78.

Due to small sample size in the hyperactive/impulsive group (n = 2), analysis of diagnostic accuracy was not carried out in this group as well as in the ADHD-NOS group (n = 4).

Comorbidity and QbTest Results

We found a positive correlation between number of comorbid diagnosis and QbActivity (r = .195, p = .05) but not regarding QbInattention and QbImpulsivity.

Comparison Between the CC Group and ADHD–No ASD and ADHD + ASD on Cardinal Parameters Using Mann–Whitney U Test

The CC group performed significantly better than both the ADHD + ASD and ADHD−no ASD groups regarding two of the cardinal parameters, QbInattention and QbActivity (p = .01; Table 4). However, only the ADHD + ASD group showed significantly higher results regarding QbImpulsitivity than the CC group (p = .01). Compared with the ADHD−no ASD group, the ADHD + ASD performed better regarding QbInattention.

Table 4.

Comparison of CC Group (n = 58) With ADHD−No ASD (n = 89) and ADHD + ASD (n = 35) Groups in QbTest Cardinal Parameters Using Mann–Whitney U Test.

Cardinal parameter	ADHD total n = 124
	ADHD–no ASD (n = 89)	ADHD + ASD (n = 35)	CC (n = 58)
	Q-score (SD)	Q-score (SD)	Q-score (SD)
QbActivity	1.6 (1.2)*	1.7 (1.2)*	0.5 (1.2)
QbInAttention	1.5 (1.1)*	0.9 (0.9)*	0.2 (0.9)
QbImpulsivity	0.8 (1.5)	1.6 (1.9)*	0.4 (1.3)

Note. CC = clinical comparison; ASD = autism spectrum disorder.

p = .01.

Discussion

The QbTest differentiated between clinical groups with and without ADHD at a group level. Three quarters of all the activity, inattention, and impulsivity parameters clearly separated ADHD from non-ADHD on a group-wise basis. However, the cardinal parameter QbImpulsivity only showed a trend toward separation of the two groups (p < .05).

The overall capacity for the different QbTest parameters to accurately identify individual cases of ADHD in the clinical sample was only moderate. The AUC was similar for the ADHD-Combined and ADHD-Inattentive subgroups indicating that the test has similar capacity independent of ADHD subtype. This is in line with previous studies that have shown that behavior ratings of inattentive symptoms are more related to objective measures of hyperactivity than to those of inattention (Günther, Kahraman-Lanzerath, Knospe, Herpertz-Dahlmann, & Konrad, 2012). Thus, the individual test parameters could not discriminate between ADHD subtypes. With cutoff set to 1.25 Q-score, as recommended by the manufactor, sensitivity ranged from .47 to .67, and specificity from .72 to .84. Results which replies findings from Sharma and Singh (2009).

When analyzing the effect of comorbidity as expressed by number of comorbid disorders on Qb results, we found that activity was influenced by the load of comorbidity. This result highlights the role of motor control problems as a signal that should lead to a comprehensive assessment including not only attentional deficits but also other comorbid disorders.

Our results are representative of a clinical sample with a high prevalence of ADHD and other neurodevelopmental disorders and have implications for specialized clinics for ADHD assessments. It is not necessarily representative of clinics in primary care as it is heavily weighted with ASD. In clinical settings, high rates of PPV, presumably above 85% to 90%, are desired for instruments chosen for guidance in the diagnostic processes as diagnostic tests are meant to provide the clinician with some surety that an illness/condition is present. The individual QbTest parameters do not appear to have these characteristics in this clinical study. In our study, the PPV for the cardinal parameters ranged from .76 to .86 and for the NPV from .37 to .50. In other words, among those who had a positive test result, the probability of having ADHD was 76% to 86% (depending on which cardinal parameter is analyzed), and for those who had a negative test result, the probability of not having ADHD was 37% to 50%. This emphasizes that ADHD assessment needs to be performed by experienced clinical practitioners, that interpretation of the QbTest result should be made with caution, and that the clinical evaluation should still remain the gold standard for diagnosis. The QbTest captures the core symptoms of ADHD at a group level and also includes objective activity measures, which no other test does. Like other neuropsychological tests, it provides a setting for relevant clinical observation. This indicates that the administration of the test and evaluation of test results should be in the hands of experienced clinicians. Although this study showed moderate validity for the individual QbTest parameters, the intended use of the test is to complement the clinical interview and validated symptom rating scales to improve assessment precision in ADHD.

Further on, the clinician needs to pay attention to comorbid ASD. It is well known that almost half the children with ASD suffer from hyperactivity, inattention, and impulsivity (Murray, 2010; Sturm, Fernell, & Gillberg, 2004; Yoshida & Uchiyama, 2004). This study highlights the similarities of CPT profiles in children with ADHD and children with ASD, with or without coexisting ADHD. However, when comparing children with ADHD (and no ASD) to those with ADHD combined with ASD or to the CC group (where a majority had ASD but no ADHD), we found interesting differences in impulsivity among the groups that the comorbid ADHD + ASD group were more impulsive than the ADHD group and CC group. We also found that the comorbid ADHD + ASD group received lower scores on inattentive test parameters, whereas all groups performed similar on QbInattentive cardinal parameter. This raises the question whether the clinical presentation of ADHD in children with coexistence of ADHD and ASD is different from in children with only ADHD. Inhibitory control is an important component in executive functions, and individuals with ASD are reported to encounter inhibitory control deficits (Geurts, van den Bergh, & Ruzzano, 2014). The higher impulsivity score seen in the ADHD + ASD group might be a double effect of inhibitory deficit due both to ASD and ADHD. The difference in inattention might be explained by different attentive styles, for example, that individuals with ASD may exhibit selective inattention to social stimuli or to unmotivated activities, whereas sustaining focus on activities with limited connection to social stimuli (as in a computerized test) is easier.

To sum up, the individual QbTest parameter ability to differentiate between ADHD and CC group was good, but the ability to identify ADHD in a clinical sample was moderate and the ability to discriminate between ADHD subtypes was unsatisfactory. However, analyzing QbTest performances in different clinical groups (including ADHD) might give valuable information on clinical presentation that might explain more than the broad diagnostic categories.

Strengths and Limitations

The strengths of the study is that the study group is representative of patients who will receive the test in practice and that both the study group and the comparison group had received the same reference test, a clinical diagnostic work-up using ADHD criteria according to DSM-IV.

A limitation in the study is that although the ADHD diagnosis was not based on results from the QbTest, the results were known to some of the clinicians, whose information could have contributed to the final diagnosis. Another limitation is that no interrater reliability tests regarding diagnoses were carried out.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Nicklas Hult is a clinical psychologist, working as a neuropsychologist at a child and adolescent clinic in Gothenburg and is also a PhD student at the Gillberg Neuropsychiatry Centre, at the University of Gothenburg.

Josefin Kadesjö is a clinical psychologist working both as a school psychologist and a clinical psychologist at a child and adolescent clinic in Gothenburg.

Björn Kadesjö is a paediatrician at the Child Neuropsychiatric Clinic at Queen Silvias Children’s Hospital, Gothenburg.

Christopher Gillberg is professor Gillberg Neuropsychiatry Centre at the University of Gothenburg and child psychiatrist at the Child Neuropsychiatric Clinic at Queen Silvias Children’s Hospital, Gothenburg.

Eva Billstedt is an associate professor at Gillberg Neuropsychiatry Centre, University of Gothenburg and clinical psychologist at the Child Neuropsychiatric Clinic at Queen Silvias Children’s Hospital, Gothenburg.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Bussing

Fernandez

Harwood

Hou

Garvan

C. W.

Eyberg

S. M.

Swanson

J. M.

(2008). Parent and teacher SNAP-IV ratings of attention deficit hyperactivity disorder symptoms: Psychometric properties and normative ratings from a school district sample. Assessment, 15, 317-328. doi:10.1177/1073191107313888

Conners

C. K.

Parker

J. D. A.

Sitarenios

Epstein

J. N.

(1998). The Revised Conners’ Parent Rating Scale (CPRS-R): Factor structure, reliability, and criterion validity. Journal of Abnormal Child Psychology, 26, 257-268.

Conners

C. K.

Sitarenios

Parker

J. D. A.

Epstein

J. N.

(1997). Revision and restandardization of the Conners Teacher Rating Scale (CTRS-R): Factor structure, reliability, and criterion validity. Journal of Abnormal Child Psychology, 26, 279-291.

Edwards

M. C.

Gardner

E. S.

Chelonis

J. J.

Schulz

E. G.

Flake

R. A.

Diaz

P. F.

(2007). Estimates of the validity and utility of the Conners’ Continuous Performance Test in the assessment of inattentive and/or hyperactive-impulsive behaviors in children. Journal of Abnormal Child Psychology, 35, 393-404. doi:10.1007/s10802-007-9098-3

Geurts

H. M.

van den Bergh

S. F.

Ruzzano

(2014). Prepotent response inhibition and interference control in autism spectrum disorders: Two meta-analyses. Autism Research, 7, 407-420. doi:10.1002/aur.1369

Günther

Kahraman-Lanzerath

Knospe

E. L.

Herpertz-Dahlmann

Konrad

(2012). Modulation of attention-deficit/hyperactivity disorder symptoms by short- and long-acting methylphenidate over the course of a day. Journal of Child and Adolescent Psychopharmacology, 22, 131-138. doi:10.1089/cap.2010.0146

Halperin

J. M.

Sharma

Greenblatt

Schwartz

S. T.

(1991). Assessment of the Continuous Performance Test: Reliability and validity in a nonreferred sample. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 3, 603-608.

Knagenhjelm

Ulberstad

(2010). QbTest technical manual. Stockholm, Sweden: Qbtech AB.

10.

Losier

B. J.

McGrath

P. J.

Klein

R. M.

(1996). Error patterns on the Continuous Performance Test in non-medicated and medicated samples of children with and without ADHD: A meta-analytic review. Journal of Child Psychology and Psychiatry, 37, 971-987.

11.

McGee

R. A.

Clark

S. E.

Symons

D. K.

(2000). Does the Conners’ Continuous Performance Test aid in ADHD diagnosis? Journal of Abnormal Child Psychology, 28, 415-424.

12.

Murray

M. J.

(2010). Attention-deficit/hyperactivity disorder in the context of autism spectrum disorders. Current Psychiatry Reports, 12, 382-388. doi:10.1007/s11920-010-0145-3

13.

National Collaborating Centre for Mental Health (NCCMH). (2009). Attention deficit hyperactivity disorder: Diagnosis and management of ADHD in children, young people and adults. Leicester and London: The British Psychological Society and the Royal College of Psychiatrists.

14.

Sharma

Singh

(2009). Evaluation of the role of Qb testing in attention deficit hyperactivity disorder. Archives of Disease in Childhood, 94(Suppl. 1), A72.

15.

Soreni

Crosbie

Ickowicz

Schachar

(2009). Stop signal and Conners’ continuous performance tasks: Test–retest reliability of two inhibition measures in ADHD children. Journal of Attention Disorders, 13, 137-143. doi:10.1177/1087054708326110

16.

Spitzer

R. L.

(1983). Psychiatric diagnosis: Are clinicians still necessary? Comprehensive Psychiatry, 24, 399-411.

17.

Sturm

Fernell

Gillberg

(2004). Autism spectrum disorders in children with normal intellectual levels: Associated impairments and subgroups. Developmental Medicine & Child Neurology, 46, 444-447.

18.

Vogt

Shameli

(2011). Assessments for attention-deficit hyperactivity disorder: Use of objective measurements. The Psychiatrist, 35, 380-383. doi:10.1192/pb.bp.110.03214

19.

World Health Organization. (1992). ICD-10 classifications of mental and behavioural disorder: Clinical descriptions and diagnostic guidelines. Geneva, Switzerland: Author.

20.

Yoshida

Uchiyama

(2004). The clinical necessity for assessing attention deficit/hyperactivity disorder (AD/HD) symptoms in children with high-functioning pervasive developmental disorder (PDD). European Child & Adolescent Psychiatry, 13, 307-314.

21.

Zelnik

Bennett-Back

Miari

Goez

H. R.

Fattal-Valevski

(2012). Is the test of variables of attention reliable for the diagnosis of attention-deficit hyperactivity disorder (ADHD)? Journal of Child Neurology, 27, 703-707. doi:10.1177/0883073811423821