Cognitive computer training in children with attention deficit hyperactivity disorder (ADHD) versus no intervention: study protocol for a randomized controlled trial

Cognitive training is rooted in cognitive rehabilitation, based on the concept that direct training can result in a reorganization of neural functions. Among other effects, neuroplasticity allows the central nervous system to learn new skills, remember information and reorganize neural networks in response to external stimulation [33]. The basic mechanisms involved are neurogenesis, programmed cell death and activity-dependent synaptic plasticity [33]. Childhood is a period of changes in the brain’s anatomical structure and synaptic connections. A child’s brain is more susceptible to the environmental impact than the adult’s brain due to increased plasticity [33]. Thus, injuries and some neurological diseases are overcome by children faster and easier than by adults. Several studies indicate that the peak of brain plasticity is reached within the first 7 years of life [34], although the potential is likely to be lifelong. For example, a functional magnetic resonance imaging (fMRI) open trial of young healthy adults found that training working memory resulted in an increased brain activity in the dorsolateral, prefrontal, and parietal association cortex, indicating plasticity of the neural system [35]. These cortical areas are overlapping the prefrontal regions, which are likely implicated in the pathology of ADHD [36, 37]. Despite the hypothesis that children under the age of 7 have better neuroplasticity and therefore may benefit more from cognitive training as compared to older children, we have not identified any studies investigating the effect of cognitive training in different age groups.

Cognitive training is typically delivered in a computerized format and is aimed at training cognitive functions that are deficient in a patient population by using a special kind of computer games. A rapidly growing number of randomized trials support the hypothesis that cognitive dysfunctions can be trained in children with ADHD [38‐41]. Most ADHD trials with children have focused solely on working memory training and findings have been somewhat inconsistent [42‐44]. Overall, working memory training shows effects on verbal and spatial working memory [45], and these effects are generalized to improved sustained attention up to 6 months follow-up [42]. Some few studies have shown improvements in academic abilities, but there is no consensus yet as several newer studies had negative results [46‐48]. Working memory, combined with response inhibition training, has shown significant improvements on symptoms, spatial working memory, ability to ignore distracting stimuli and sustained attention as rated by a significant other [41]. Training of executive functions improved parent-rated executive functions and ADHD behavior when compared to waiting-list condition [49].

Fewer trials have focused on the attention training in children with ADHD that results in a significant improvement in trained and untrained attention and vigilance [39, 50‐52], a measure of school performance and a significant reduction in parent and teacher observation of inattention [39, 40, 53]. In addition, the effects on inhibition and working memory have been found [52], and significant changes in inattentiveness, behavior and executive functions measured by parent ratings on Behavior Rating Inventory of Executive Functions (BRIEF) [54].

Structural and functional correlates of cognitive training have been shown in several small studies. Enhanced activity in neural structures closely related to ADHD pathology [55] and increase of focal volumetric gray area in bilateral middle frontal cortex and right inferior-posterior cerebellum after attention and executive functions training [56]. Cognitive training has also been shown to induce neurochemical changes at the synapse in dopamine function after training [57].

In conclusion, randomized trials and observational studies suggest that cognitive training of children with ADHD has some beneficial effects. However, the empirical evidence in this field is still insufficient as most trials have a high risk of systematic errors (bias) mainly due to lack of blinding, incomplete outcome data, and selective outcome reporting. Further, most trials have small sample sizes, which result in an increased risk of imprecision. As children with ADHD have impairments in many different cognitive functions, there is a need for randomized trials to examine effects of broader cognitive training, rather than focusing on only one or two domains, for instance, working memory, response inhibition or sustained attention. It is important to validate and extend existing knowledge on the effects of cognitive training for patients with ADHD.

Hence, to overcome some of these limitations, the present trial will use ACTIVATE™ a cognitive computerized program that aims to improve eight different cognitive functions. We will include a sample of children and adolescents with ADHD, and in addition to considering the ratings of clinical symptoms by parents and teachers, we will measure the outcome with an objective, valid and reliable cognitive test battery. The trial is, to our knowledge, the first to examine the effect of cognitive training on the outcome of Cambridge Automated Neurocognitive Test Battery (CANTAB) in children with ADHD.

The primary objective is to investigate whether computer training with the games embedded in ACTIVATE™ (http://​denmarkstudy2.​c8sciences.​com/​?​language=​da) has a positive effect on specific cognitive functions. The secondary objectives are to investigate whether there is an effect on ADHD symptoms and functional outcome. Exploratory objectives are to investigate the effects at 12 and 24 weeks after training and to investigate whether younger children benefit more from training than older children.

Participants are included in three sites in southern Denmark: the Child Psychiatric Departments of Aabenraa (including Augustenborg), Kolding and Odense.

The three sites are part of the same organization, Region of Southern Denmark, and are under the same leadership. All children and adolescent mental health hospitals in Denmark are state-owned, and everyone is eligible to get treatment. Referral from the treating physician or school psychologists is required. No children who are being treated by private practicing child and adolescent psychiatrists will be included in the study. One site (Odense) is part of a university hospital, and the two other sites are part of regional hospitals. Eventual differences between sites will be assessed using data on demographics.

Enrollment of children into the trial is done consecutively throughout the calendar year. The vast majority of children will be enrolled during school year. A few participants will be enrolled during school vacations, but as the intervention is home-based, this will likely not affect the adherence.

All children who are newly referred with ADHD symptoms to one of the Child Psychiatric Departments or currently in treatment with ADHD-medication will be invited to participate in the trial and will be offered an individual information session, after which their custodian can give their informed consent. The diagnostic assessment is done in a two-step model: In Step 1 parents, a teacher and children over 11 years of age complete an online questionnaire, including the Strength and Difficulties Questionnaire (SDQ) [58, 59] in conjunction with the psychiatric diagnostic interview Development and Well-being Assessment (DAWBA). The DAWBA is a valid hybrid between a structured and a semi-structured interview for the diagnosis of child and adolescent psychiatric disorders according to both the ICD-10 and DSM-IV [60, 61]. DAWBA’s sensitivity is 92 % in a clinical sample and its specificity is 89 % for all psychiatric diagnoses in a community sample [60]. Parents and teachers answer structured and open-ended questions regarding diagnostic symptoms using the online DAWBA-platform. A child and adolescent psychiatrist rates this information clinically. Children fulfilling diagnostic criteria for ADHD based on this rating of DAWBA proceed to Step 2, which includes a confirmatory clinical interview with parents at the hospital, using the Kiddie-Schedule for Affective Disorders and Schizophrenia (K-SADS, ADHD section) [62]. K-SADS is a semi-structured clinical interview of parents and children and is the most widely used psychometric instrument for the diagnostic investigation of children in clinical research. All children with confirmed ADHD are assessed with Reynolds Intellectual Assessment Scales (RIAS) [63] to ensure that all participants have an IQ above 80. Finally, children are included in the trial if they comply with the following inclusion and exclusion criteria.

Children are included if the following inclusion criteria are fulfilled: ADHD diagnosis after DAWBA interview [60] and verification with clinical interview K-SADS, ADHD section parent interview [62]; age between 6 and 13 years, both inclusive; the patient has access to a computer and the internet; and informed consent.

Patients fulfilling any of the following exclusion criteria will not be included: comorbid conduct disorder, autism spectrum disorder, depression or schizophrenia; head injury or verified neurological disease; intelligence quotient (IQ) < 80; motor or perceptual handicaps that prevent computer use; medical condition that requires primary treatment, and no informed consent from custodian (Fig. 1).

Both the intervention group and the control group will receive treatment as usual (TAU). TAU consists of clinical assessment and treatment. Clinical assessment includes intelligence tests, cognitive testing, school observations and parent and teacher questionnaires. TAU may involve psycho-education, parent training, advising the parents and school, and for some children, medication. Parallel to the trial, the participating children will undergo a regular assessment and treatment procedure at the clinic. It is the treating specialist, who is independent of the trial and blind to the child’s randomization status, who will consider possible medical treatment, independent of the child’s participation in the trial. Children in medical treatment are asked not to change their medication dose during the 8 weeks of intervention.

In addition to TAU, the intervention group will use the computer program ACTIVATE™ (http://​denmarkstudy2.​c8sciences.​com/​?​language=​da). ACTIVATE™ includes an engaging computer program aimed to train multiple cognitive functions simultaneously: sustained attention, working memory, speed of information processing, response inhibition, cognitive flexibility, category formation, pattern recognition and multiple simultaneous attention. ACTIVATE™ consists of three different games: “Catch the ball,” “Butterflies,” and “What comes next.” These are described below.

In all three games, the child is earning points, which are converted into coins. At the end of all three games, the child comes to a “garden” where s/he can purchase different things. The child can decide to buy things for the garden, such as plants and animals, or a car, a zoo or a house. The child can buy things after each game or can save coins to buy more expensive things later. The games are designed in a manner to be interesting and rewarding for children. All participants are doing the same kind of exercises. All participants start at the same very basic level. Progression and the level of difficulty in the games depend on the child’s performance. Hence, the level of difficulty is therefore dynamically adjusted during the trial, according to the abilities of each child or cognitive phenotype.

Training with ACTIVATE™ is home-based for 40 minutes per day, 6 days per week, for 8 weeks, resulting in a cumulative training of a maximum of 32 hours. ACTIVATE™ records each time the participant logs on and is measuring compliance, time on task and progress. All participants randomized to the intervention group are introduced to the program at the clinic. In case of any problems with the program, the participants can contact the principal investigator. C8 is also providing IT support. Parents are given verbal and written instructions that the child should use a computer with an external mouse (not an iPad or a laptop with an integrated mousepad), that training should be performed in a quiet setting, and that using headphones is mandatory. In addition, the parent are instructed to help the child remember and engage in training and to supervise the child during training sessions, to ensure adherence. We are in touch with parents, and they can contact us any time. Parents are given the instruction to supervise the child and make sure they are doing the training. There were no restrictions on the time of the day the training should be performed. However, we inform parents that most children usually like to do the exercises in the late afternoon or early evening and that parents should ensure that sessions do not conflict with school or family schedules.

The control group will only receive treatment as usual.

Randomization is performed centrally by the Copenhagen Trial Unit. Participants are randomized 1:1 to the intervention group or control group. The allocation sequence is computer-generated with a varying block size kept unknown to the investigators, and it is stratified by trial site (“Aabenraa,” “Kolding,” or “Odense”) and use of medication (“yes” or “no”). Allocation is performed by the investigator telephoning the Copenhagen Trial Unit, giving a personal PIN code as well as information about the participant (strata, participant ID number etc.), and the participant is then allocated to an intervention group.

To examine whether the randomization sufficiently reduced the risk of systematic group differences between children in the intervention and the control arm, we will compare the distribution of history of scholar retention events, and pharmacological treatment (dose and type of medication) in the child, mean parental age, socioeconomic status and level of education, and the number of people living at home.

Due to the nature of the trial, it is not possible to blind the participants and their parents. However, we will employ blinding in all other possible areas. Investigators conducting the cognitive testing with CANTAB will be blind to the participants’ group allocation. The statistical analyses will be performed blinded with the two intervention groups coded as, for example, X and Y. The analyses will be presented blinded to the Steering Committee, who will draw two conclusions: one assuming that X is the intervention group and Y is the control group, and one assuming the opposite. After this, the blind will be broken.

To reduce the risk of biasing the rating of outcomes caused by unblinding information on group allocation, we chose an objective computerized primary outcome measure on the CANTAB. Still, the clinicians performing the CANTAB are blinded to group allocation. The treating physicians are not directly connected to the trial and do not assess or provide information on any trial outcomes. Due to regulations by the ethics committee, we were not allowed to inform the treating physician about included children to avoid that influencing the treatment in the outpatient clinic. Hence, as these clinicians were responsible for the treatment as usual (TAU) in both groups, they were also blinded to the group allocation of the child to ensure that this did not affect the TAU.

For an overview of all outcomes and assessments, please see Table 1.

Each CANTAB assessment lasts between 70 and 90 minutes. We aim to collect all cognitive assessments between 8:30 am and 2:00 pm to avoid a time of the day that would have an impact on the cognitive performance. While the child is being assessed, the parent questionnaire data are collected. If the child is unable to complete the assessment in one session, the assessment can be split up.

CANTAB is a nonverbal computerized cognitive test battery with solid psychometric properties [64‐66] (Cambridge Cognition Limited, 2011). The primary outcome is the sustained attention test from the CANTAB: “Rapid Visual Information Processing (RVP) probability of hit,” assessed at the end of the intervention.

The following secondary outcomes will be assessed at the end of the intervention:

The following exploratory outcomes will be assessed at the end of the intervention:

Further, all outcomes will be assessed again 12 and 24 weeks after the end of the intervention (Fig. 2).

This trial is being conducted in accordance with the protocol approved by the Danish Data Protection Agency (ID.nr. 2008-58-0035) and the Regional Scientific Ethical Committees for Southern Denmark (nr. S20120096). The protocol is in accordance with the latest version of the Declaration of Helsinki. No significant deviation from the protocol will be implemented without prior review and approval by the regulatory authorities unless it may be necessary to eliminate an immediate hazard to the trial participants. In this case, the deviation will be reported to the regulatory authorities as quickly as possible.

A child can participate in the trial if the written consent of both parents is obtained. The patients’ treatment as usual will not be affected, including the use of medications, by their participation in this trial. When testing with CANTAB, patients will be asked to postpone any ADHD medication for 24 hours before testing, when medical treatment can affect efficacy measures. Trial participants will receive a gift certificate worth DKK 400 for participation. Transportation costs will be reimbursed.

The processing of personal data will be respected. There are no known risks associated with the use of computerized cognitive training. The method has been tested in many studies with patients with schizophrenia (see reviews [69, 70]) and in children with ADHD [38‐40, 53]. No adverse events have been reported.

Although parents may have agreed to participate, they can always withdraw their child from the trial without further explanation. Pulling a child out of the trial, will not affects his or her further treatment. Patients can choose to stop at any time in the trial. Patients who were randomized will be included in the intention-to treat analyses unless they completely withdraw consent. In this case, all data regarding this participant will be deleted.

If the trial results indicate that this intervention reduces specific cognitive deficits in children with ADHD without causing any adverse reactions effects, our interpretation will be that the intervention can be an important part of a treatment plan as cognitive dysfunctions are very common in children with ADHD. Furthermore, if we find improvement in the BRIEF measures and ADHD-RS, this would suggest that the effects of the intervention could be generalized to behavior in an everyday setting.

The intervention group compliance will be monitored via the computer program that records patients log on, which games they have played, and for how long.

Aida Bikic is the initiator of the trial and the investigator psychologist, research coordinator and PhD student. Participants are being randomized in the Child and Adolescent Psychiatric setting of Augustenborg and Aabenraa, Odense and Kolding. The trial has received grants from Region of Southern Denmark Psychiatry Research, The Region of Southern Denmark’s PhD pool, Child and Adolescent Psychiatric Department Aabenraa, TrygFonden (J.nr. 7-12-1137) and the University of Southern Denmark. C8 Sciences allowed us to use the ACTIVATE™ program at no charge in this trial. None of the funders has any role in the development of the trial design, trial conduct or trial reporting.