Attention deficit/hyperactivity-disorder (ADHD) is a neuro-developmental disorder with symptoms frequently occurring in early childhood, however, there is a parenting program that was developed specifically for the management of ADHD in preschool children, noted below [1,2]. ADHD is one of the most common psychiatric disorders in childhood [3] and the most common reason for referral to child mental health services (CAMHS) [4]. Studies indicate that ADHD is as common in preschoolers as it is in school-age children, with population based prevalence rates ranging between 2 and 5% [5,6]. Impairment is equally common [7], including marked impairments in relationships with parents, siblings, and peers [8-11]; social and preacademic skills [12]; and neuropsychological functioning [13]. Preschool children with ADHD are at greater risk of placement in special educational classes and use more special needs services [14]. Young children with ADHD are more likely to suffer physical injury and accidental poisoning related to impulsive behaviors [7]. ADHD symptoms in preschoolers show persistence over time [15,16]. Lahey et al [17] found that 4-6 year old children who met full criteria for ADHD were highly likely to continue to meet criteria 3 years later. Preschool hyperactivity is associated with long-term economic burden. Evidence from a longitudinal health economic analysis of costs incurred across childhood, adolescence, and young adulthood found significantly higher costs for this group of individuals in areas of mental health, education, social service, and the criminal justice system [18]. In a large register-based study, it was found that childhood ADHD, by the time the child is 10, reduces parental socioeconomic status by lowering labor supply and earnings, and reduces parental relationship stability [19]. Left untreated, preschool ADHD is associated with long-term poor outcomes for the patient [20]. Hence, it is of central public priority generally, and for CAMHS specifically, to offer evidence-based and cost-effective treatments at an early stage of development in an attempt to reduce the long-term risks and change the negative trajectories that ADHD presents to the individual and to society [2].

Although medication is recommended as the front-line treatment for older children and adolescents with ADHD, international treatment guidelines do not recommend pharmacological therapies for preschool ADHD [21,22]. Indeed, available data suggest that stimulant medication such as methylphenidate is less efficacious in the preschool years and side effects are more common [23]. Long-term effects on growth and brain development are currently not definitively established. Because of these factors, psychosocial behavioral treatments for ADHD, including behavioral parent training and behaviorally based day-care interventions, are recommended as first-line treatments for ADHD in the preschool years [21]. A number of psychosocial interventions have been applied and evaluated for the treatment of preschool ADHD. A recent systematic review found that behavioral parent training is valuable as a treatment for preschool ADHD [24]. Furthermore, Sonuga-Barke et al [25] found that trials evaluating behaviorally based parent interventions had the largest effects for preschool children with ADHD.

Whereas many behaviorally based interventions are generic in nature, in that they were originally developed to help parents manage preschoolers’ oppositional and challenging behavior, the New Forest Parenting Programme (NFPP) [1] is a parenting program that was developed specifically for the management of ADHD in preschool children. In addition to behavioral strategies, it includes ideas for games and activities aimed at targeting some of the self-regulatory and attention deficits that cause impairment in the condition. Supporting the child’s development through parental scaffolding is a key component of the NFPP, and everyday play scenarios in the home constitute opportunities for parents to train and improve the child’s ADHD symptoms. A single practitioner delivers NFPP to families in their home during eight weekly sessions.

Each session lasts approximately 90 minutes and is either for the parent only (five sessions) or the parent and child together. In an evaluation of the NFPP, Sonuga-Barke et al [1] found that the intervention reduced ADHD symptoms as reported by parents and improved behavior during a play observation task, as rated by blinded observers. Similarly, Thompson et al [26] reported that NFPP improved parent-reported ADHD symptoms in a sample of preschool children with ADHD.

At present, the evidence of the effect of NFPP is based on efficacy research carried out by the expert team of researchers and clinicians who developed the program originally, with therapists receiving support from lead investigators, and with children who have not been referred for treatment through regular community pathways. Yet, it is unexplored if the findings from the original research can be replicated in different clinical and cultural contexts, including in a different language. This leaves open the question of whether effects shown in specialist research led services can be translated in the everyday practice of established clinical CAMHS [27]. In a review of implementation studies [28], it was shown that there is a risk that interventions will lose impact and potency when implemented in everyday clinical settings. Hence, it is emphasised that the transition of efficacious treatments has to be carefully planned and adapted to the clinical setting in which the treatment is being implemented [27,29].

Evidence-based parent training interventions specifically designed for preschool ADHD, such as the NFPP, are not currently provided by Danish CAMHS. The development and implementation of effective psychosocial treatments for young children with ADHD represent an important health policy objective. The trial described in this protocol is designed to evaluate the effectiveness of the NFPP in the treatment of young children diagnosed with ADHD referred through established clinical pathways to child psychiatry departments in Denmark. It is the first time that the NFPP has been tested in a European country outside the United Kingdom, and the first time the NFPP has been tested on children referred through established clinical pathways to tertiary child psychiatry services, to our knowledge. For these purposes, the NFPP program has been adjusted to fit the clinical and cultural context of Danish child and adolescent psychiatry services.

The objective of the trial is to examine whether the NFPP can be implemented effectively as a treatment for preschool ADHD in everyday clinical settings of existing services. To do this, we will examine ADHD and other outcomes at the end of treatment and also at a 6-month follow-up.

A randomized controlled trial will be carried out to investigate the effectiveness of the NFPP against a treatment as usual condition (TAU) in the treatment of ADHD in a clinical sample of young children referred to, assessed, and diagnosed at tertiary, preschool child psychiatry services in Denmark. The effectiveness of the NFPP intervention will be examined at three time points: T1 (baseline), T2 (week 12, post intervention), and T3 (6 month follow/up).

The study is approved by the Ethics Committee for Central Danish Region (No: 1-10-72-140-12), and is approved by the Danish Data Protection Agency. The trial is registered at ClinicalTrial.gov identity no: NCT01684644, September 4, 2012.

Participants will be enlisted from the regional, hospital-based child psychiatry departments for preschool children at the three participating centers in the trial: Centre for Child and Adolescent Psychiatry-Risskov; Centre for Child and Adolescent Psychiatry-Herning; and Centre for Child and Adolescent Psychiatry-Glostrup. The departments are highly specialized, tertiary CAMH services, and lie at the end of the referral pathway for young children with mental health problems in Denmark. Prior to their referral to specialist child psychiatry services, children will usually have received an initial assessment and intervention by primary care professionals, such as an early years educational psychologist, a community pediatrician, and social services.

A total of 140 children (age 3-7) referred through official referral routes to a general child psychiatry service for preschool children and their parents will be recruited into the trial. A clinical psychologist or a specialist in child and adolescent psychiatry initially screens all referrals for eligibility. Referrals indicating noneligibility, due to direct descriptions of exclusion criteria in the referral letter, receive standard clinical assessment from the preschool child psychiatry team. At this point, all other referrals are regarded as eligible and parents are sent a letter including a personal access code asking them to complete the Internet DAWBA [4] or the Preschool DAWBA (DAWBA-preschool for 2-4 year olds) [30]. The DAWBA generates computerized diagnostic profiles, which are scored by child and adolescent psychiatrists and clinical psychologists trained in DAWBA rating. The DAWBA is employed at all three participating sites in an attempt to homogenize diagnostic practices and is used to compliment and inform the clinical, diagnostic assessment process.

In conjunction with the DAWBA, eligible children will undergo standard clinical assessment at the referral site. For referrals to preschool child psychiatry services in Denmark, this standard clinical assessment usually consists of: interview with parents regarding the child’s developmental history and current difficulties; child cognitive assessment and neuropsychological tests or neuropsychological tests; child medical examination; and, at Risskov and Herning, a semistructured observation of the child at the child’s day-care facility. Results from clinical assessments and results from the Internet DAWBA are presented and discussed at weekly multidisciplinary clinical team meetings where a specialist doctor in child and adolescent psychiatry issues final clinical diagnoses.

Parents of children meeting the research criteria are invited to join the study, and are included in the trial, once written consent is obtained. Figure 1 shows the recruitment flow according to Consolidated Standards of Reporting Trials (CONSORT) guidelines [31].

Since outcome may be dependent on treatment center and regional variations in referral pathways, as well as gender and age of the child, the study uses stratified allocation to treatment condition to ensure a balanced number of participants over conditions. Randomization is based on the defined strata as well as on the assigned treatment condition of earlier randomized participants within the strata. Parents will be randomized to one of the two arms using an allocation ratio of 1:1 for NFPP and TAU. Randomization to NFPP and TAU (1:1) will be generated by a Web-based randomization computer program within the Internet data management service Trialpartner [32], which allows for on-the-spot randomization of participants into an arm of the study. Randomization is done in blocks of size four or six and in 12 strata defined by center, gender, and age of the child (3-5 and 6-7 years). Once the parent has provided written informed consent to participate in the study, the administrator completes participant registration on the Internet in Trialpartner and enters a randomization request, at which point the randomization program randomly assigns the treatment condition, for example, NFPP or TAU, to the respective participant ID number. Trialpartner simultaneously generates an automated email message for the administrator and the main investigator of the study, in which participant ID, site, and treatment allocation is outlined. Parents are informed of their respective treatment condition following completion of the T1 visit, so that the first assessment is not influenced by participant’s knowledge of treatment condition. Information about treatment allocation is delivered to parents in a telephone call by the researcher who initially informed the parent of the study. 

All T1, T2, and T3 assessment schedules administered to parents and to day-care staff are set up electronically in Trialpartner [32] by the The Data Management Unit in The Central Denmark Region and are administered for completion on the Internet to parents and day-care staff by a research assistant. All data are stored and managed by The Data Management Unit in The Central Denmark Region. Data will remain locked until all participants have completed T3.

Research assistants carrying out research assessments and observations at the three time-points will be blinded. Parents are instructed not to reveal treatment allocation to research assistants. All raters of videotaped observations will be blinded. Research assistants and raters are postgraduate psychology students and newly qualified psychologists. Parents, NFPP therapists, NFPP supervisor, and TAU mental health professionals will not be blinded. Efforts will be made to keep teachers blind to treatment allocation. Yet full blinding cannot be guaranteed, as participants may spontaneously talk about treatment to teachers.

The primary endpoint is parent ADHD ratings at T2. Sample size has been estimated at 126 children to obtain power at 0.80, ES=.50 (Cohen’s d). The usual convention that is generally proposed and accepted for randomized trials is that power in a study ought to be 0.80 when ES is estimated at 0.50 [33]. An estimated effect size of 0.5 is deemed reasonable for a trial evaluating the effect of parenting interventions in the treatment of ADHD. By comparison, a recent meta-analysis of parent training interventions for ADHD obtained an ES of 0.4 [34].

For a design comparing two groups, for example, NFPP versus TAU, and allowing for a participant drop out of 10%, a total of 140 children, for example, a sample of 70 children per group is needed to attain a power of 0.80 [35]. A drop out rate of 7% was reported in a Dutch parent training trial implemented in routine clinical CAMHS with a similar age group [36]. In this light, the present trial has chosen to operate with a potential drop out rate of 10%. The preschool child psychiatry sites at Risskov and Herning are expected to enroll 2/3 of participants and the Glostrup site is expected to enroll 1/3 of participants for the trial.

Measures administered in this trial are listed in Table 1.

The data will be analyzed on an intention to treat basis. All outcomes will be analyzed with a repeated measure model with a randomization arm, gender, age (above or below 5 year), center, year, and a random level for each child as covariates. We will estimate the effect of the intervention at T2 (intervention, TAU) in terms of the difference between the change from T1-T2 and its effect at T3 in terms of change between the T3-T1. Missing data will be imputed assuming Missingness At Random in relevant analysis models including additional control variables if relevant. Outcomes with more than fifty percent missing will not be analyzed. To examine the effect of outliers, a sensitivity analysis will be performed excluding observation with residuals exceeding 2.5*SD. In addition, four other sensitivity analyses will be performed for each analysis exploring how sensitive the results are for reasonable deviations from the Missing At Random assumption. The missing data will be predicted based on the analysis of the observed data. In the first sensitivity analysis, we will reanalyze the data including the predicted value for the missing values, but add .2*SD to the missing data in the intervention group only. In the second sensitivity analysis, we will subtract .2*SD from the missing data in the intervention group only. In the third, we add .2*SD to the missing data in the TAU group only, and, in the last, subtract .2*SD from the missing data in the TAU group only.

The results for the primary outcome as a supplement will be split according to gender, single parenting (yes/no), oppositional defiant disorder/conduct disorder-comorbidity, and maternal ADHD symptoms. A number of scientific articles will be generated on the basis of collected data from this trial, including publications outlining moderator and mediator analyses. Results will be reported according to the CONSORT statement for nonpharmacological interventions [31].

Time-Line Funding for this trial was granted in 2011 and in 2014. Enrollment of participants started in August, 2012, and the data collection is expected to be finalized at the end of December, 2015. The primary outcome paper will be submitted for publication mid-year 2016.

This trial addresses the need to investigate the effect of nonpharmacological treatments for young children with ADHD in everyday clinical practice. It is the first trial of the NFPP to be based on the recruitment of children referred directly to a child psychiatry department through official community pathways, to our knowledge. In this sense, this randomized trial tests the effectiveness of the NFPP with the children, families, and clinicians in the clinical setting for which the intervention is ultimately intended, and compares directly to the usual clinical care on offer for young children diagnosed with ADHD. The trial is also the first evaluation of the NFPP outside an English speaking context, to our knowledge.

Given the controversies surrounding the ADHD diagnoses in young children, and the challenge that confronts assessment of preschool children with ADHD symptoms, diagnostic procedures require particular attention [59]. This trial is the first to include young children with a formal clinical diagnosis of ADHD following standard clinical and DAWBA assessment, to our knowledge. The clinical assessment ensures thorough medical, psychological, and psychosocial evaluation of the child and the child’s environment, and the adjunct systematic DAWBA evaluation ensures standardization in the assessment procedure for children entering the trial.

To reduce the chance of bias, the trial includes measures involving direct observation of children and of parenting by blinded raters as well as ratings by naïve informants (ie, day-care staff) along with parent ratings. This will allow for the investigation of rater effect on child outcome, and facilitate the exploration of whether child outcome is stable across settings.

Identifying efficacious treatments for young children with ADHD is important, but finding evidence from trials in settings created for research (eg, university labs, specialist clinics) or with participants recruited through advertisements does not guarantee that a treatment will work well in everyday clinical practice, across different populations, clinical contexts, cultures, and countries [27]. The current trial has carefully adapted the NFPP to fit clinical and cultural needs and commenced an evaluation of its effectiveness compared to the usual treatment interventions offered to young children with ADHD and their families referred through official community pathways. To these ends, the trial will provide important information about the effectiveness and replicability of the NFPP in the treatment of young children with ADHD in an everyday CAMHS setting. This level of investigation is needed to shrink the gap between intervention research and clinical practice [60].

ADHD is a disorder that presents with major personal and social costs. The immediate and long-term impact and adverse outcomes of ADHD for individuals and their families are considerable [61]. The benefits of early intervention for ADHD is an area of current scientific and public health focus, but is not fully documented [62]. It is therefore important that carefully implemented effectiveness trials of evidence-based treatments for young children are translated and implemented into different cultures with a view to enhance outcome for a broader population of children with ADHD and their families.