A Systematic Study of Assessment of Attention-Deficit/Hyperactivity Disorder (ADHD)

ADHD is one of the most prevalent psychiatric conditions, consuming a large proportion of resources in psychiatric care, often accompanied by long waiting lists to receive proper assessment. The number of ADHD cases has increased, possible due to heightened awareness of the condition. There are large prevalence differences, potentially due to variations in assessments procedures. Many clinicians and parents view the diagnostic process as too extensive, taking time from treatment and interventions. In addition, assessments may be perceived as too focused on diagnostic criteria to be fully helpful. Systematic research on how assessment procedures can be optimized is essentially lacking. It is largely unknown whether brief protocols including medical history, diagnostic interview, and rating scales differ from comprehensive protocols that also encompass neuropsychological testing regarding validity, reliability, patient satisfaction and cost-effectiveness. Further, feasible biomarkers (e.g. heart rate variability, pupil dilation and the pupillary light reflex) of the autonomic nervous system have been proposed as indicators of diagnostic status. The aim of this study is to gain knowledge about diagnostic processes to enable valid, reliable, and cost-effective ADHD assessments. Using a randomized controlled trial design (N = 240 children, 8-17 years, referred to child and adolescent psychiatric units), differences between a brief and a comprehensive ADHD assessment protocol regarding assessment outcome, reliability, validity, patient satisfaction, and future outcome taking gender into account will be examined. The investigators will explore diagnostic sensitivity and specificity of the included assessment instruments and estimate cost-effectiveness of the brief and comprehensive protocols to enable policy makers to make informed decisions. The project will provide important knowledge for patients and clinicians, and inform our understanding of mechanisms underpinning ADHD.

Group A will receive a brief assessment protocol and Group B will receive a comprehensive assessment protocol.

Masking is not possible. The assessors that will review the material for reliability and validity purposes will be blind to diagnostic status of the participants.

The brief protocol will contain a minimum according to guidelines for assessing ADHD: review of medical records, a validated diagnostic interview (MINI-KID), a structured medical history, a pedagogical statement including suspicion of intellectual disability, and rating scales for symptoms such as ADHD, oppositional defiant disorder, autism, anxiety, and depression directed to children (≥ 13 years), parents, and teachers.

The comprehensive protocol will extend the brief protocol by adding an approximately three-hour long battery of neuropsychological tests (WISC-V and CPT 3) and biomarkers (heart-rate variability, pupil dilation and the pupillary light reflex) assessed during the CPT 3.

The certainty score will be assigned by the assessor on a 1 to 10-point scale, where 6 to 10 index the assessor's degree of certainty of the child meeting criteria for ADHD and 1 to 5 the assessor's certainty of the child not meeting criteria for ADHD

A second assessor will review the assessment material (blind to diagnostic status) and assign diagnostic status to each case. S/he will be given all information included in the assessment, except for diagnostic status and asked to assign diagnosis/es as well as a certainty score on a 1 to 10-point scale, where 6 to 10 index the assessor's degree of certainty of the child meeting criteria for ADHD and 1 to 5 the assessor's certainty of the child not meeting criteria for ADHD.

An experienced clinician will assign diagnostic status using Longitudinal Expert All Data (LEAD; i.e. review of all available material) roughly one year post assessment, after follow-up data has been collected. For this purpose, all available material will include information included in the assessment, medical records following the assessment, and symptom ratings at follow up. S/he will assign diagnosis/es and a certainty score (on a 1 to 10-point scale, where 6 to 10 index the assessor's degree of certainty of the child meeting criteria for ADHD and 1 to 5 the assessor's certainty of the child not meeting criteria for ADHD), blind to original diagnostic status.

Children (> 13 years) and all legal guardians will rate satisfaction with the assessment process right after the assessment, using a satisfaction scale ranging from 1 to 5, where 5 indicates high satisfaction with the assessment.

Children (> 13 years) and all legal guardians will rate satisfaction with the assessment process at follow up, using a satisfaction scale ranging from 1 to 5, where 5 indicates high satisfaction with the assessment.

To enable cost-effectiveness analyses, a health economist will build a model based on the following: Child (> 13 years) and parent ratings of child quality of life at baseline and follow-up using the Child Health Utility D9 (CHUD-9), on a scale from 1 to 5, where 1 indicates high quality of life. CHUD-9 will be used to estimate quality adjusted life years (QALYs). Personnel resources and overheads will be estimated for each protocol, and other inputs (such as risks of negative school outcomes and related costs, as well as costs related to ADHD treatment) will be collected from the literature. Consumption of healthcare and school resources for children, as well as productivity losses associated with absenteeism and presentism at school will be estimated from the Treatment Inventory of Costs in Patients (TIC-P), which will be filled out by the caregivers and teachers.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for the diagnostic interview, Mini Internationell Neuropsykiatrisk Intervju (MINI-kid), using formal diagnosis as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for the Adult ADHD Self-Report Scale Adolescent version (ASRS-A), in which symptoms are rated on a scale from 1-5, where 5 indicate high symptoms, with formal diagnosis as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for the working memory index from Wechlser Intelligence Scale for Children (WISC-V), on a scale from 1 to 19, where 19 indicated high ability for workning memory. Formal diagnosis will be used as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for the Conners Continous Performance Test (CPT-3). The CPT-3 renders a T-score ranging from 0-100, where 50 is the mean and 15 the standard diviation. Higher scores indicate better attention abilities. Formal diagnosis will be used as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for heart rate variability assessed during the Conners Continous Performance Test (CPT-3). Formal diagnosis will be used as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for pupil dilation assessed with an eye-tracker during the Conners Continous Performance Test (CPT-3). Formal diagnosis will be used as the external criterion.

Discriminative validity (positive predictive value, sensitivity, specificity, and area under the curve) will be calculated for the pupillary light reflex. Formal diagnosis will be used as the external criterion.

Inclusion Criteria: Age 8-17 years Referral for ADHD assessment Exclusion Criteria: Suspected intellectual disability Substance abuse Psychosis Severe depression Parent not fluent in Swedish Child not living with legal guardian Child having protected identity