Introduction

Attention deficit hyperactivity disorder (ADHD) is a highly comorbid [13] neurodevelopmental disorder. It has a worldwide prevalence of 3–5% in school age children [4], 80–85% of these individuals continue to be effected by their ADHD into adolescence [58] and 60% into adulthood [9].

Due to the symptoms and complexity of the condition, there are a number of important long-term difficulties associated with ADHD. These include low academic attainment [10, 11], which can persist into adulthood [12], poor executive functioning [13], poor social relationships, strained parent/child/sibling relationships [14] and problems with social interactions with peers [15]. This results in poorer quality of life and self-esteem in children and YP with ADHD [16, 17].

Children and YP with ADHD are reliant on clinicians and parents to help them to manage their condition. However, as they transit into adulthood, the support is not as readily available or indeed wanted by the individual [18]. It is, therefore, imperative that children and YP learn to self-manage their condition and indeed be educated about their condition and how to manage it [1923]. Individuals with ADHD often experience crises and access to their usual services may not be immediately available. However, the increasing sophistication and usage of technology may provide valuable resources to facilitate the self-management of ADHD for children and YP.

Over recent years, technological advances have meant that technology is more widely available and has become more popular and integrated into many lives. Society is also better connected with an estimated 46% of the worldwide population having an internet connection compared with 1% in 1995 [24]. As a result of this, a number of attempts have been made to harness technology to help manage ADHD in children and YP such as eye tracking [25], brain computer interface [26, 27] and a computerised test that quantifies ADHD core symptoms; the QbTest [28]. Technology has also been used for cognitive training in children and YP with ADHD [29]. However, these technologies are reliant on an administrator or a therapist. Other technologies have been developed to self-manage ADHD-related difficulties in children and YP that can be used independently of a therapist which, therefore, reduces the reliance on services. These include a handheld organisation device [30], a device to self-monitor ADHD symptoms [31], computer software to improve reading speed [32], and computer games that focus on mathematical ability [33] and the promotion of behavoural learning and organisation [34]. Although these studies report that technology has the potential to self-manage ADHD-related difficulties in children and YP, little is known about the level of evidence for these technologies. Therefore, this review will assess the level of evidence for currently available technologies for self-managing ADHD and related difficulties in children and YP.

Methods

The review protocol was registered with PROSPERO (CRD42017057715). The review was undertaken in accordance with the general principles recommended in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis [35].

Search methods

The following databases were searched in February 2017 from the last 5 years: MEDLINE, Web of Science (Core collection), CINAHL, the Cochrane Library, ProQuest ASSIA, PsycINFO and Scopus. Medical Subject Headings (MeSH) keywords used were attention deficit disorder with hyperactivity, hyperkinesis, attention deficit and disruptive behaviour disorders, conduct disorder, child behaviour disorders, disruptive impulse control and conduct disorders, adolescent, young adult, educational technology, technology, self-help devices, video games, internet, software, social media, mobile applications, self care and social support. Text terms used were attention deficit and disruptive behaviour disorders, attention deficit hyperactivity disorder, ADHD, ADDH, ADHS, hkd, attention, behaviour, dysfunctional, disorder, disrupt, defiant, impulsive, inattentive, inattention, hyperkinesis, damage, hyperactive, conduct, child, boy, girl, young person, YP, young people, adolescent, teen, youth, technology, assistive technology, self-help devices, game, website, download, forum, email, mobile app, condition management, manage, self-manage, support and support network. Terms were combined using Boolean logic (“AND”, “OR”). MeSH is specific recognised terms used for the purpose of identifying journal articles and books in electronic databases. Free text terms and synonyms are specific words that the search strategy looks for in the title and abstract.

A copy of the MEDLINE search strategy is presented in Appendix 1. Electronic citations were downloaded to Endnote software. The inclusion criteria are described in Table 1. Studies included in this review were from 2014 to 2016.

Table 1 Inclusion and exclusion criteria for this review

Due to the infancy of this research topic, any study design was accepted as appropriate to answer the research question. The research question is “What is the level of evidence that current technology that aims to self-manage difficulties associated with ADHD in children and young people is helpful?” The primary outcome measures (see Table 2) of this review are measures that assess ADHD related difficulties. 

Table 2 List of included outcome measures

Quality Assessment

Methodological quality of included studies was assessed using the Cochrane Risk of Bias tool (CRoB) [36] for RCT designs and the Downs and Black Instrument [37] for non-RCT designs. This CRoB tool addresses specific domains, namely, sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, and selective outcome reporting. The Downs and Black Instrument provides an individual score for each study with a maximum score of 32 [37] and assesses the way in which the studies report their findings, external validity, internal validity bias and selection bias.

Data extraction

Retrieved titles, abstracts, and/or papers were screened independently by 2 review authors (LP, JP) to identify studies that met the inclusion criteria. Disagreements were resolved between reviewers through discussion. A standardised form was used for data extraction using Microsoft Excel. Details of the study characteristics, including participants, the intervention, and comparator (where applicable) were recorded. Data extraction was carried out by reviewer LP and checked for accuracy by reviewer JP.

Outcome measurement assessment

It is vital that when undertaking a systematic review, the quality of the outcome measures used in each of the included studies is assessed. This is to ensure the validity and reliability of their results. To complete the outcome measure quality assessment, where possible, three domains should be considered for each outcome measure [38], (1) whether the psychometric properties of the scale have been assessed previously [39], (2) whether the clinimetric properties of the outcome measure have been thought through [4044], specifically the Minimally Clinically Important Difference (MCID) [43], and (3) whether the design and analysis of the outcome measure satisfies the requirements of measurement theory [4547]. We identified all outcome measures (N = 58) used across the 14 studies and reviewed each of them individually to assess whether they fulfilled the first domain described above. The MCID was not assessed for the included outcome measures and was, therefore, not assessed. The 58 included outcome measures are listed in Table 2.

Literature for each outcome measure, where applicable, was reviewed. We then examined each outcome measure to ascertain how the data were scored, collected and analysed within the results section of each study.

In line with the literature, all 58 outcome measures included were measures of difficulties related to ADHD.

Data synthesis

We have presented a narrative overview of the included studies with supporting evidence tables and text. A meta-analysis was not undertaken.

Appraisal of evidence

The results of the search varied from case studies to Randomised Controlled Trials (RCTs). The studies identified were appraised using the levels of evidence [48] to locate the best available evidence that involves the application of systematic, robust, transparent and explicit methodology [49]. The grading system (see Table 3) was created to highlight that varying study designs and methodologies are at risk of bias in their results. This is crucial as the study design may affect the validity and reliability of results due to the research method used. For example, when evaluating the effectiveness of an intervention, it is often considered that RCT evidence is the “gold standard”, the most reliable form of evidence due to the measures they take to reduce the influence confounding variables could potentially have on the results [50].

Table 3 Levels of evidence outlined by Weiss et al. [48]

Results

Search results

The electronic searches identified 7391 citations following de-duplication, including 9 additional citations that were identified through reference searches/other sources. We excluded 7331 citations at the title and abstract stages as they did not fit the inclusion criteria. We then obtained 60 citations as full-text articles. Of these, 50 were excluded at the full-text stage; details of these excluded studies with the reason for exclusion are shown in Appendix 2. 14 studies reported across 14 publications were included in the review (see Fig. 1). Four of these publications were obtained from a recent meta analysis [29], which examined the effects of cognitive training on ADHD symptoms, neuropsychological deficits, and academic skills in children and YP with ADHD [5154].

Fig. 1
figure 1

Studies included in this review

Quality assessment

Full details from the Cochrane risk of bias assessment are presented in Appendix 3. A summary of the RCT risk of bias assessment is presented in Table 4, non-RCT risk assessment in Table 5 and a summary of the outcome measurement quality assessment can be found in Appendix 5.

Table 4 RCT risk of bias summary
Table 5 non-RCT risk of bias summary

One of the eight included RCTs, one was considered to be at overall high risk of bias [34], five RCTs were judged as being at overall low risk [51, 52, 54, 55, 56] and two RCTs were considered to be at overall unclear risk of bias [53, 57]. All included RCTs were considered to be at a low risk of bias for selective reporting [34, 5157].

Non-RCT study designs were assessed using the Downs and Black Scale, as they were mainly exploratory interventional studies. Overall, the non-RCT studies obtained low scores on items covering external and internal validity, selection bias and statistical power. Studies obtained higher scores for the items covering reporting of results and study procedures. The maximum total score that could be obtained is 32. Of the six included non-RCT studies in this review, the lowest score was 6 [58] and the highest score was 11 [59, 60].

Quality assessment of measurement scales

Ten of the fourteen included studies [34, 5158, 63] used ordinal scales of measurement all with established psychometric properties. Twelve of the included studies [34, 51, 5362] used scales of measurement that did not have established psychometric properties. Four of the fourteen studies [5860, 63] did not perform any formal statistical analysis. The sample size for these four studies ranged from one to eight. Five of the fourteen studies [34, 52, 53, 56, 57] aggregated data used with ordinal scales, which may put findings at risk. Further details of the outcome measurement quality assessment can be found in Appendix 5. Description of the studies can be found in Table 6.

Table 6 Summary of study, participant and intervention characteristics and results

Discussion

This review set out to answer the question “What is the level of evidence that current technology that aims to self-manage difficulties associated with ADHD in children and young people is helpful?” The review found that the evidence demonstrates that technology shows promise in self-managing difficulties related to ADHD in children and YP. However, this claim is based on evidence that often consists of small sample sizes, use a wide variety of outcome measures (many of which are not validated) and provide little support for the importance of the role of psychoeducation in children and YP with ADHD that has been so widely reported and encouraged elsewhere [1923, 64]. For example, the European Guidelines suggest psychoeducation for parent/carer and child with ADHD as a first step to treatment [23]. One systematic review even stated that psychoeducation for YP with ADHD and their families could provide an expert understanding of their condition could lead to more positive individual choices [20].

Of the fourteen included studies in this review, the interventions assessed include two tablet devices [59, 60] two mobile applications [58, 62], the use of a Wii remote control [61], computer software [51, 54] and computer games [34, 52, 53, 5557, 63]. Following exclusions, outcome measure assessment (Appendix 5) and quality assessment of the fourteen included studies was conducted (appendices 3 and 4).

Only four of the sixteen papers included in the Cortese et al. [29] meta analysis were included in this review [5154] and one additional paper resulted from our search strategy before the Cortese review was screened for studies to include in this review [57]. Two of the unincluded papers presented in this meta analysis [29] did not report technological interventions [65, 66], one did not use validated outcome measures of ADHD-related difficulties [67], four reported technological interventions that were not for independent use [6871], two reported on participants who did not have a primary diagnosis of ADHD [72, 73] and two of the studies did not report including participants who had an official clinical diagnosis of ADHD [74, 75]. One of these papers reported participants who had a parent-reported ADHD diagnosis [75]. Papers included in this review all report on participants who have obtained a formal ADHD diagnosis. This is crucial to ensure that comparisons can be made across studies. Parent-reported diagnoses may not be as reliable as clinically reported diagnoses and therefore do not enable comparisons to be made and therefore the results from such studies should be interpreted with caution.

The interventions used in ten [34, 5157, 61, 62] of the fourteen included studies identified statistically significant results for some, not all, primary outcome measures included in this review. Statistically significant improvements included improved ADHD symptoms [54, 56, 57, 61], social skills [34], executive functioning [51, 52, 54, 55, 57] and educational outcomes [62]. Statistical significance was not observed for the quality of life [55] or self-efficacy [34] measures which interestingly, only featured in two of the included RCTs [34, 55]. Although a trend towards improved symptoms [58, 61] and executive functioning [63] was observed in three of the included studies [58, 61, 63], no formal statistical analysis was undertaken and the sample sizes were small ranging from one to eight participants. Therefore, these findings should be interpreted with caution.

As described fully in the quality assessment, one of the eight included RCTs were considered to be at overall high risk of bias [34], two were considered as having an unclear rick of bias [53, 57] and five were considered as having a low risk of bias [51, 52, 54, 55, 56]. This does not mean that interventions were not successful in improving ADHD-related difficulties. A number of conclusions could be drawn from this including the difficulty of blinding participants to an intervention as it is often impossible to conceal which arm participants are randomised to. It is also difficult to stop a potentially impulsive and hyperactive population to withhold their randomization allocation to an outcome measure assessor. Overall, the included non-RCTs obtained low scores on the Downs and Black scale. Out of a maximum score of 32, two studies scored eleven [59, 60], three scored nine [6163] and one obtained a score of six [58]. A number of conclusions could be drawn from this including low sample sizes and the non-RCT nature of the studies (thus obtaining low scores on items that assess whether or not participants and research staff are blinded).

Of the fourteen included studies in this review, five [34, 52, 53, 56, 57] aggregated data with ordinal scales, four used no formal statistical analysis [5860, 63] and two carried out statistical analysis when their sample sizes only consisted of two participants each [61, 62]. Clinimetric properties were not described for any of the primary outcome measures of this review. The lack of statistical significance across a number of outcome measures in this review could be a result of lack of statistical power due to small sample sizes and the inability to ascertain a clinically meaningful result.

The results from this systematic review should be generalised to a wider population of children and YP with ADHD with caution due to the low recruitment figures for five of included studies where n ranged from one to twelve [5862] and only two of the included RCTs had a sample size of more than one hundred [34, 53]. Observations of the lack statistical significance should also be interpreted with caution, given the level of evidence provided and the methodological quality of the existing evidence base.

This review included a small number of papers including 1040 participants overall with 170 being from one study alone [34]. Six of the selected studies recruited fewer than 20 participants [51, 58, 6163]. This could be for a number of reasons. It may have been difficult to engage with and recruit YP with ADHD to a research study, although this has not been our personal experience.

Additionally, ADHD severity and the presence of comorbidity can affect the level of impairment experienced by the individual, which can affect the way in which they respond to interventions. The included studies did not report the severity of the ADHD in their participants. However, one study [52] reported that their participants were participants were diagnosed with hyperkinetic disorder according to the International Statistical Classification of Diseases and Related Health Problems 10 (ICD-10) [76]. This diagnosis would have been based on narrower criteria than the DSV-IV as in the ICD-10 ADHD is diagnosed based on a minimum number of symptoms in all three dimensions (inattention, impulsivity and hyperactivity) [76] whereas the DSM-IV requires a minimum number of symptoms in one dimension [77]. This means that it is difficult to inter a significant improvement of ADHD-related difficulties.

There are a number of reasons evaluating a complex intervention with this population could remain challenging. For example, no ADHD diagnosis is the same. ADHD is a highly comorbid condition with a large number of potential-related difficulties. The extent to which each individual is impaired by their ADHD symptoms and related difficulties are also highly variable. In this review, four studies excluded participants who had specific comorbid diagnosis [5153, 55]; one study excluded participants with autism spectrum disorder and conduct disorder [55], one study excluded participants diagnosed with pervasive developmental disorders, Tourette’s disorder and those who show evidence of bipolar disorder and conduct disorder [52]. One study excluded those diagnosed with pervasive developmental disorder [51] and another study excluded those with any “clinically significant comorbid condition” [53]. These findings should be interpreted with caution as at least 65% of children and YP diagnosed with ADHD have a comorbid condition [1] therefore these participant groups are not representative of the wider ADHD population. These factors coupled with evaluating an intervention make it very difficult to ascertain a clinically significant improvement in this population following the use of an investigative intervention. It also means that it is difficult to control each arm of an RCT study design. It has therefore been suggested [78] that the integration of realist evaluation within an RCT design may be more appropriate for evidence-based medicine whereby “statistically significant benefits may be marginal in clinical practice” [79].

The results of the included studies were not combined for a meta-analysis due to the variety of the types and quality of data collected for the primary outcome measures. It would also be difficult to compare primary outcomes across the included studies accurately as there was a wide variety of measures assessing ADHD-related difficulties used, many of which lacked validity as a measure of ADHD-related difficulties in children and YP.

Despite the wide variety of outcome measures included in this review, none of them assessed ADHD knowledge and understanding. To self-manage ADHD, the Chronic Illness Model [80] states that psychoeducation with a collaborative care model enhances health outcomes [64]. Similarly, the Health Belief Model states that people are more likely to seek treatment if they have knowledge and understanding of their condition [81, 82]. It is important that ADHD psychoeducation delivery is conveyed to the individual and their parents in a culturally appropriate manner, via a reputable website and written and updated by reputable experts [64]. It has been suggested that psychoeducaiton for parents and the YP with ADHD is the first step to treatment [23]. A systematic review has emphasised the value of psychoeducation for children and YP with ADHD can lead to an expert understanding of their condition and lead them to making more positive individual choices [20]. Public Heath England [19] and the Mental Health Taskforce’s Five Year Forward View for Mental Health [22] states that early intervention avoids YP falling into crisis and expensive and longer-term interventions into adulthood. Therefore, it is vital that ADHD psychoeducation begins as early as possible following an ADHD diagnosis so that the YP can learn to accept and self-manage their condition in preparation for transition into adulthood.

Transition periods are particularly challenging for somebody diagnosed with ADHD and they present frequently throughout the course of a young person’s life. For example YP move to secondary education, undertake regular exams, have to navigate through puberty, sometimes move house and many transfer adult ADHD services. The latter is particularly challenging due to the nature of child and adult ADHD services being very different and providing support in very different ways. Child services provide more in person support and may involve more frequent appointments than adult services. Therefore, a smooth transition between services is vital for a YP with ADHD to minimise disruption [83]. Despite this evidence, none of the included studies provided psychoeducation as part of their interventions.

Future research should focus on the development and co-collaboration of an evidence-based intervention that may focus on psychoeducation for this population. Due to the majority of the included interventions in this review taking the form of computer games, perhaps an ADHD technological intervention with a psychoeducation focus should take a different form such as a website. Evidence suggests that to engage with this population, technological interventions should be interactive [84, 85]. Research in this area should also consider larger sample sizes and ADHD severity and the presence of comorbid conditions should be reported for participants and accounted for during analysis.

Outcome measures for all interventions for ADHD need to be carefully planned. They should include core symptoms but it is likely that these are not the realistic targets of this type of intervention and goal-orientated outcomes agreed with YP and families may be more relevant. Functional and quality of life outcomes need longer follow-up but in a chronic disorder have far more significance. Finally, advancements in technology and improvements of the suitability of interventions specifically designed for independent use to facilitate self-management could involve a psychoeducational component. Such technologies should be co-designed with stakeholders including children and YP with ADHD adopting a user-centred design methodology to ensure the technology is suitable for this population.