Neurodevelopmental disorders (NDDs), estimated to affect 5%-9% of children [1,2], are a heterogeneous group of chronic conditions that affect the central nervous system. Children with NDDs are a vulnerable population that experiences significant health disparities compared to typically developing (TD) children, including high rates of sleep difficulties. In children with NDDs, sleep difficulties contribute to morbidity and impair cognition, emotional regulation, and neurobehavioral functioning [3]. Prevalence rates of sleep disturbances have exceeded 90% in children with certain NDDs, with insomnia being the most frequently cited sleep disorder [4]. Sleep disturbances can have a negative impact on children’s quality of life (QoL) and exacerbate stress in their caregivers [3].

Despite limited safety and efficacy data in children, pharmacotherapies, both over-the-counter and prescription, are often given to treat insomnia in this population [5]. However, pediatric sleep experts recommend that psychosocial interventions be implemented first [6]. Specifically, the intervention would start with psychoeducation and healthy sleep practices, and be followed by psychological interventions such as bedtime fading or modified extinction [7-9]. Then, if psychosocial interventions are ineffective or unfeasible, medications become an option [7-9].

The existing literature has demonstrated the efficacy of treating insomnia in TD children with psychosocial interventions [10,11]. While there is less research on children with NDDs, the available evidence also demonstrates the effectiveness of these interventions. A review of nonpharmacological interventions for sleep problems in youths with chronic health conditions concluded that the effectiveness of psychosocial interventions was not exclusive to specific health conditions but had comparable effects across children with a variety of mental and physical health conditions, including NDDs [12]. Psychosocial interventions for pediatric insomnia have also been shown to improve parents’ sleep and increase their feelings of competence and control [13,14].

Existing psychosocial interventions are typically delivered in conventional settings (eg, hospital clinics, outpatient professionals’ offices), consisting of weekly training sessions for parents [15-18] and several visits with health care providers [19,20]. Hence, time constraints and required travel expenses to health facilities may be access barriers for families. Reaching families in rural or remote communities and those who are financially constrained or lack insurance coverage for private psychological and other necessary clinical services are significant challenges in the delivery of behavioral interventions [21].

Economic evaluations can help guide policy making and resource allocation decisions. Little is known about the cost-effectiveness of eHealth (internet-based in this case) behavioral interventions for sleep problems in children. Many parents are interested in internet-based interventions for their children’s sleep problems [22]. An internet-delivered treatment program may be an effective tool to address and overcome barriers to accessing pediatric insomnia care [22,23]. The cost-effectiveness of internet-based sleep interventions for children in comparison to usual care has not been determined for children with NDDs [22].

The value for money of internet-based interventions can be assessed using cost-effectiveness analysis (CEA) and cost-consequence analysis (CCA) [24]. A CEA incorporates cost and a single measure of effectiveness into a metric represented by an incremental cost-effectiveness ratio (ICER) [25]. An ICER is calculated by dividing the difference in cost by the difference in outcome measure between 2 interventions, often treatment and control conditions in randomized controlled trials (RCTs). A CCA reports various costs and outcomes to enable decision makers to decide which are most relevant to their context and perspective [24]. It can capture costs to the user, system, and society and portray health, nonhealth, and nonpatient impacts [24].

An economic evaluation will be performed alongside the Better Nights, Better Days for Children with Neurodevelopmental Disorders (BNBD-NDD) RCT (ClinicalTrial.gov ID NCT02694003), which seeks to determine the efficacy of the eHealth behavioral intervention compared to usual care. BNBD-NDD is a parent-implemented eHealth behavioral program delivered to parents to enable accessible and effective care for insomnia in children with NDDs. The web-based nature of BNBD-NDD allows it to be readily scalable to reach many users. BNBD-NDD is transdiagnostic because it targets children ages 4 to 12 years with diagnoses of mildly to moderately severe attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), cerebral palsy (CP), and fetal alcohol spectrum disorder (FASD). Although these 4 conditions present a broad range of cognitive, physical, and social characteristics, they are all associated with higher than normal rates of insomnia [26,27].

This protocol describes the BNBD-NDD economic evaluation, consisting of a CEA to evaluate the cost-effectiveness of BNBD-NDD in children with NDD compared with usual care and a CCA to examine outcomes beyond QoL. A CEA will be undertaken using the parent-reported Pediatric Quality of Life Inventory (PedsQL) questionnaire total scores, which will assess the children’s QoL. The QoL measure is used in economic evaluation because it allows comparison of the health outcomes of competing alternative interventions over time and across populations [28]. In addition, to illustrate a more comprehensive picture of BNBD-NDD’s impacts, a CCA will also be conducted to examine outcomes beyond the children’s health-related quality of life (HRQoL), including children’s sleep, parents’ daytime functioning, and psychosocial health.

Understanding the added value of BNBD-NDD is essential to informing the intervention's scale and spread. Thus, this protocol aims to complement the BNBD-NDD RCT and describe the economic evaluation from both family and societal perspectives.

Participants were recruited for the BNBD-NDD trial nationwide in Canada and stratified based on the parent-reported primary NDD diagnosis of their child. Potential parent participants were targeted via three groups: (1) parents of children with NDDs, (2) external stakeholders and organizations, and (3) the media. Internet-based tools were primarily used for recruitment, notably this study’s website and social media, as well as print materials and presentations to health care professionals.

Participants must (1) reside in Canada and (2) be caregivers of children (3) ages 4 to 12 years who have been diagnosed with (4) ADHD, ASD, CP, or FASD by a health care provider and (5) meet the research criteria for insomnia. Insomnia criteria (ie, sleep onset disturbance) were characterized by at least 2 of 3 symptoms (>20 minutes to fall asleep, parent remaining in room for sleep onset, and more than 2 reunions) occurring 2 to 4 times per week for at least a month [29]. Additionally, these caregivers must have (6) regular access to a high-speed internet connection, (7) an email account, and (8) be proficient in English for everyday tasks.

The BNBD-NDD RCT was powered to detect a group by time interaction across the 3 data collection time points (baseline, 4-month, and 8-month) and between the BNBD-NDD and usual care groups. The primary sleep outcome was the Disorders of Initiating and Maintaining Sleep (DIMS) [22] variable. To achieve a power of 0.80 with an α of .05, 39 participants per group would have to complete the DIMS measure at 8 months. Based on this analysis and the anticipated postenrollment dropout rate of 50% from a previous BNBD RCT, an estimated 160 participants were targeted for enrollment at baseline.

Participants were randomized using a block stratified approach with 1:1 allocation to either the intervention (BNBD-NDD) or the usual care group and were not blinded to their assigned group. Stratification was by parent-reported primary NDD diagnosis.

Participants were randomized to either the intervention group or the usual care group. Both groups may use alternative resources and other programs or services during this study. The intervention group was provided access to the BNBD-NDD program for 8 months after randomization (the duration of this study). As such, participants in the intervention group were regarded as receiving BNBD-NDD plus usual care. The usual care group could access BNBD-NDD after completing all follow-up assessments at 8 months.

BNBD-NDD is an eHealth program for parents or caregivers of children with NDDs and insomnia. The program is parent-implemented, incorporating transdiagnostic, evidence-based behavioral approaches, such as sleep education, stimulus fading, and scheduled awakenings, and presents NDD-specific strategies. The intervention is accessible via desktops, laptops, tablets, and mobile devices.

BNBD-NDD includes 5 sessions for participants to work through sequentially, with the recommended completion time ranging from 5 to 10 weeks. The sessions cover the following topics: (1) about BNBD-NDD; (2) healthy sleep practices; (3) settling to sleep independently; (4) night waking, napping, and early morning awakenings; and (5) looking back and ahead. Participants may access and complete the program at times that are convenient for them. Treatment fidelity and adherence were evaluated through process measurements and user statistics gathered by the intervention platform software. Participants may withdraw from the program at any time and are invited to provide reasons for withdrawal through open-ended questions.

Costs and outcomes are weighed differently by various decision makers. A way to determine the most pertinent perspective is by asking whom the analysis is intended to inform [30]. As BNBD-NDD is designed as a direct-to-user product, with the program’s users and beneficiaries being the parents and their children with NDDs and insomnia, this economic evaluation aims to inform them of the costs and effects of the program. The reference case costing will be completed from the perspective of the family payer (in this case, the parents). In accordance with recommendations from the Canadian Agency for Drugs and Technologies in Health, the reference case analysis will evaluate the QoL outcome for children, the primary target beneficiaries of the program [25]. In the nonreference case analysis from a societal perspective, the outcomes for caregivers (ie, daytime fatigue, psychosocial health), specific measures of children's sleep (ie, DIMS score), and societal costs (ie, treatment utilization [TU]) will also be examined.

In the BNBD-NDD RCT, data collection occurred at baseline, 4 months, and 8 months. The reference case CEA and secondary CCA will use an 8-month time horizon, focusing on baseline and 8-month data; 8-month data are used to best capture the longer term costs and effects.

PedsQL is an instrument that measures generic (rather than disease-specific) HRQoL and has been shown to have appropriate psychometric properties when used with children with NDDs [31]. It can be used to evaluate HRQoL in healthy children, youths, and those with medical conditions [32]. PedsQL uses a modular approach with 4 multidimensional scales on physical, emotional, social, and school functioning [33]. Parents rate each item using a 5-point Likert scale from 0 (never) to 4 (always) [31]. A Total Scale Score is calculated by adding all of the items and dividing the total by the number of items answered in all scales [33]. Mean Scale Scores are computed by dividing the sum of the items by the number of items answered in the scale [33]. Depending on the age of the participant's child, 3 parent-report versions were used (for ages 2-4 years, 5-7 years, and 8-12 years) to accommodate the age range of this study sample. The difference between the intervention and usual care groups’ average Total Scale Score in parent-proxy reports of their children’s PedsQL at baseline and 8 months will be evaluated in the reference case CEA. Secondary analyses will examine each of the 4 modules and age group trends; it is noted that inconsistent reporting of the school functioning section may occur due to age and the time of year of data collection (off-school period). The scoring and missing data (where applicable) will be completed in accordance with the PedsQL user manual [34].

In the CCA, the children’s sleep and parental outcomes are analyzed by comparing baseline and 8-month data and looking at changes in sample averages. Based on parent-reported questionnaire data, the DIMS score quantifies the severity of children's sleeping problems. The Single Item Fatigue Impact Scale [35] and the Depression, Anxiety, and Stress Scales [36] are used to evaluate the daytime fatigue and psychosocial health, respectively, of parents. Single Item Fatigue Impact Scale is adapted from the 40-item Fatigue Impact Scale [35]. The Depression, Anxiety, and Stress Scales have demonstrated adequate construct validity and high reliability [36].

Multiple imputation techniques are used to address missing data. This method works well for longitudinal data and is robust to violations of the nonnormality of the variables analyzed. Participants who do not complete both the baseline and the 8-month follow-up will not be included in the analysis. Where applicable, the measures’ user guides on missing data will be followed. For the PedsQL, the means of completed items in the scale will be imputed when 50% or more of the items are completed [34].

As all costs and outcomes assessed are within 1 year, a discount rate will not be applied [25].

All study participants provided informed consent to participate in the BNBD-NDD RCT. Ethical approval for the BNBD-NDD trial (REB-1020681) was granted by the IWK Health Centre Research Ethics Board and the associated economic evaluation by the IWK REB and the University of Calgary Conjoint Health Research Ethics Board (REB22-0694). This study complies with all applicable standards and procedures of the IWK REB, the Personal Health Information Act, the Personal Information International Disclosure Protection Act, and the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans.