Precision computerised cognitive behavioural therapy (cCBT) for adolescents with depression: a pilot and feasibility randomised controlled trial protocol for SPARX-UK

Across all CAMHS/MHST sites (Patient Identification Centres [PICs] and research site), the procedure will be as follows: the initial approach will be from a member of the patient’s usual care team (i.e. CAMHS/MHST clinician/practitioner), and information about the trial will be given to potential participants. All individuals conducting initial patient identification at sites will be given the inclusion/exclusion criteria for the study. The usual care team will provide the adolescent and their parent/guardian a participant information sheet. Once the study has been discussed and information sheet given to the adolescent/parent/guardian, a member of the patient’s usual care team will send their contact details confidentially by email to the research team with the family’s consent. The parent/guardian participant information sheet will also have a QR code and website link to an online consent to contact form specifying their preferred mode of contact (e.g. telephone or Microsoft Teams). Parents/guardians who provide consent to contact will be approached by a member of the research team who will explain the study process and ascertain initial screening eligibility over the phone/Microsoft Teams to determine the presence of any obvious exclusion criteria. Consent will be obtained via videoconferencing to provide some assurance of identity. It is a requirement that parents/guardians consent to the trial as they will be completing the Development and Well-Being Assessment (DAWBA) [41] and other measures (side effects questionnaire). However, there will be a dual consent process for participation, with under 16 years old providing child assent along with their parent/guardian providing written consent. Those who are 16 years and over can provide their own written consent.

Members of the clinical care team, or clinical research study officers at study sites, will be asked to record the numbers of patients approached (and to record the reason for nonparticipation). For those who consent to contact, parents’/patient’s contact details will be provided to the research team. Reasons for nonparticipation will be recorded (where given) for CONSORT purposes.

For participants identified and recruited from GP practices, the procedure will be as follows: utilising an approved search with filters to identify potentially eligible participants, staff at GP practices will conduct a search of their database which will generate a list of potential participants. At the practice, a member of staff will use a letter/email/SMS template to send to families, which will include information about the study and a flyer with brief information about the trial or a participant information sheet. Parents/guardians will be able to provide written consent to contact via a QR code or website link to an online consent to contact form. Parent/guardians who provide consent to contact will be approached by a member of the research team who will explain the study process and ascertain some screening eligibility over the phone/Teams to determine the presence of any obvious exclusion criteria. Consent will be done via videoconferencing to provide some assurance of identity.

Another means of participant recruitment for the trial will be in using MyHealthE — an online portal for the automated screening of referred families using NHS CAMHS data. MyHealthE uses a secure text/email system through which primary caregivers are invited to register and complete validated clinical screening measures using an online portal. These are then automatically coded using standard algorithms to subsequently allow the research team to identify potentially eligible participants. MyHealthE seeks parents’ permission to be contacted with invitations to take part in research studies. By consenting to research contact, parents also give their permission for NHS trust-approved researchers to review their children’s medical records to establish eligibility. In some of the participating organisations, clinical triage and rapid screening for mild to moderate levels of anxiety and depression will occur as part of routine care using MyHealthE.

Members of the SPARX-UK research team will regularly log in to the MyHealthE researcher portal (this portal only includes information about children whose parents gave consent to contact) to check for any new cases flagged as “eligible”. Researchers will identify potential participants by reviewing cases flagged up as “eligible” by MyHealthE using a standard algorithm. The “eligible” flag will be treated as a potential participating family, who the research team will then contact with an invitation to take part in the study.

The NIHR BioResource is a recallable resource of over 250,000 volunteers, who have agreed to take part in health-related research. We will be approaching selected potential participants from two trials within the NIHR BioResource: Genetics Links to Anxiety and Depression (GLAD) and the DNA, Children + Young People’s Health Resource (D-CYPHR). Participants in the GLAD trial have been referred through clinics or self-referral, with all participants having mild to moderate anxiety and depression. GLAD participants are all 16 years and over, whilst the D-CYPHR trial participants are 0–16 years. NIHR BioResource will create a list of potential participants based on our inclusion/exclusion criteria, which is then input into a secure database. An automated message is sent to selected parents. Once a parent expresses an interest in taking part by completing consent to contact, members of the research team will make contact to arrange an initial screening appointment.

After the research team receive consent to contact and the associated contact details, they will arrange a telephone/videoconferencing screening appointment with the parent/guardian or adolescent (if they are 16 years and over). This screening approach has shown to be effective in previous trials of similar interventions [42, 43]. At this point, the following details will be recorded: contact details, age of depression onset and brief clinical history, previous contact with healthcare services, previous/current medications or therapy for depression, access to Internet/PC/Mac/laptop/smartphone, other diagnoses, and inclusion/exclusion criteria. The researcher will outline the time commitment involved in the trial at this stage. Adolescents meeting any exclusion criteria or not meeting all inclusion criteria at this time point will not be invited to the baseline assessment. If the adolescent is not eligible for the study or does not wish to attend a baseline appointment, the researcher will record the reason for nonparticipation. Reasons for not attending a baseline assessment will be recorded for CONSORT reporting.

Families who meet the initial eligibility screening criteria and who are willing to attend a Microsoft Teams baseline appointment will be asked to complete a DAWBA [41] assessment online prior to the baseline appointment. Only the parent/guardian will complete the DAWBA.

Adolescents and one of their parents/guardians will be invited to attend a remote baseline appointment via Microsoft Teams. At this appointment, the details collected over the telephone/videoconference for the patient will be checked. The researcher will complete the consent process with the adolescent and their parent/guardian, and they will be consented into the trial. During this assessment, the researcher will send a live link to participants on Microsoft Teams where the participant will have an opportunity to read the information sheet again and then complete the consent form. The researcher will complete a paper or online version of an intellectual disability screening measure (Child and Adolescent Intellectual Disability Screening Questionnaire [CAIDS-Q]) [44] and the PHQ-A with adolescents to confirm eligibility. The researcher will then complete paper or online versions of all baseline measures with eligible participants. The participant will be given their study ID at the point of randomisation, at which point they are enrolled into the study. A log will be kept matching screening identification (IDs) with study IDs. Randomisation will occur via REDCap at or just after the baseline appointment (i.e. within 2 weeks), prior to starting the trial. The method of randomisation is block randomisation, performed independently of method of referral.

As these sources will not have a suitably qualified NHS healthcare professional to confirm the presence of mild to moderate depressive symptoms, the chief investigator (CI), who is a consultant child and adolescent psychiatrist, will review screening and assessment data to confirm eligibility. If eligible, randomisation will be confirmed by the research assistant. This eligibility decision will be made within 2 weeks of the information being presented. A log will be kept recording the assessment outcome.

The SPARX-UK trial obtained ethical approval from the South West-Cornwall & Plymouth Research Ethics Committee on 15 December 2022 (Ethics Ref.: 22/SW/0149) and has been registered with ISRCTN, trial number 15124804.

A schematic diagram of the trial design is shown in Fig. 1.

Study participants will be adolescents aged 11 to 19 years with mild to moderate levels of depression.

Feasibility measures: This includes feasibility of the intervention and feasibility of the trial measures. These data will be collected throughout the trial by researchers and used to inform the feasibility of conducting a future RCT.

Feasibility of intervention measures: This includes adherence to the intervention and qualitative information on acceptability and feasibility.

Feasibility of trial measures: This will consist of numbers approached/numbers consented and randomised, errors in randomisation, drop-out rates at each time point, outcome measure completion rates, instances of unblinding, and protocol deviations.

Pilot outcome: The variance of the change in Patient Health Questionnaire modified for adolescents (PHQ-A) [38] from baseline. PHQ-A is adapted from the PHQ-9 and modified for adolescents. It is a 9-item measure with each measure rated on a 4-point scale that assesses the severity of depressive disorders and episodes) in children and adolescents. The measure is completed by the adolescent. Each item asks the adolescent to rate the severity of their depression symptoms during the past 2 weeks. The total score can range from 0 to 27, with higher scores indicating greater severity of depression. There are four additional items which relate to difficulties performing everyday tasks, suicidal thoughts, and attempts; however, these are not scored. The questionnaire has established validity and reliability [38]. A 5-point change is viewed as clinically significant [46]. The PHQ-A was chosen as the primary outcome measure, as this was the preferred option of our patient and public involvement (PPI) group over other commonly used measures of depression.

Revised Child Anxiety and Depression Scale (RCADS) [47]: The RCADS is a 47-item self-report questionnaire that measures symptoms of depression and anxiety in children and adolescents. The RCADS consists of six subscales helpful in screening children and adolescents for high prevalence disorders, including separation anxiety disorder, social phobia, generalised anxiety disorder, panic disorder, obsessive compulsive disorder (OCD), and major depressive disorder. A Total Anxiety Scale score (sum of the five anxiety subscales) and a Total Internalizing Scale score (sum of all six subscales) are derived, with higher scores indicating increased symptom severity. The RCADS has demonstrated good structural validity, reliability, and convergent and discriminant validity [47]. The RCADS is a widely used instrument within CAMHS and MHSTs.

EQ-5D-Y and EQ-5D-Y (Proxy version) [48, 49]: The child-friendly EQ-5D version (EQ-5D-Y) was introduced by the EuroQol Group in 2009 as a more comprehensible instrument suitable for children and adolescents and is based on the EQ-5D-3L. The EQ-5D-Y comprises the following five dimensions: mobility, looking after myself, doing usual activities, having pain or discomfort, and feeling worried, sad, or unhappy. Each dimension has three levels: no problems, some problems, and a lot of problems. The adolescent is asked to indicate his/her health state by selecting the most appropriate statement in each of the five dimensions. This decision results in a 1-digit number that expresses the level selected for that dimension. The digits for the five dimensions can be combined into a 5-digit number that describes the adolescent’s health state. The EQ VAS records the adolescent’s self-rated health on a vertical visual analogue scale where the endpoints are labelled “The best health you can imagine” and “The worst health you can imagine”. The EQ-5D-Y (Proxy version) asks the caregiver (the proxy) is to rate the child’s/adolescent’s health-related quality of life in their (the proxy’s) opinion. Both measures have established validity and reliability [48, 49] and will be used in this study to measure health related quality of life.

Clinical Global Impressions Scale (CGI)-Severity/Improvement [50]: The CGI provides an overall clinician-determined summary measure that considers all available information, including a knowledge of the patient’s history, psychosocial circumstances, symptoms, behaviour, and the impact of the symptoms on the patient’s ability to function. The CGI comprises two companion one-item measures evaluating the following: (a) severity of psychopathology from 1 to 7 and (b) change from the initiation of intervention on a similar 7-point scale. The CGI-S asks one question: “Considering your total clinical experience with this particular population, how mentally ill is the patient at this time?” which is rated on a 7-point scale (from 1 = normal, not at all ill to 7 = amongst the most extremely ill patients). The CGI-I consists of one item: “Compared to the patient’s condition at admission to the project this patient’s condition is…” and is rated from 1 = very much improved since the initiation of intervention to 7 = very much worse since the initiation of intervention. The questionnaire has established validity and reliability and is widely used in clinical research [51].

Concomitant interventions: To assess what other mental health treatments/interventions the adolescent is accessing during the study period, parents/guardians will be asked to complete a short questionnaire, which asks about current diagnoses and treatments/interventions/medication in progress. This will be completed at baseline as part of the demographic questionnaire completed with the researcher and then again at 8–10-week follow-up with the researcher via telephone/Microsoft Teams.

Usage data: Indices such as number of modules completed, time spent using SPARX, and number of logins will be collected via an in-built SPARX feature, and time spent with e-coaches (if in supported arm) will be recorded in REDCap. These data will be collected throughout the trial and will be used to measure adherence and engagement as part of the process evaluation.

Adverse events/side effects will be recorded by the PHQ-A and on a modified version of the side effects scale developed by Hill and Taylor [52]. An adverse event will be recorded where there is clinically meaningful deterioration from baseline on the PHQ-A. On the PHQ-A, a > = 5-point increase (i.e. moving one severity category) is regarded as clinically meaningful [46], and thus, an adverse event will be recorded when this occurs or where there is an increase of one, or more, point from baseline to follow-up (4 weeks and/or 8–10 weeks) on item 9 or a positive response (i.e. “yes”) on any of the two additional suicide items from baseline to follow-up. The Hill and Taylor [52] scale consists of 17 short items relating to common side effects (such as headaches, anxiety, sleep, and low mood). The participant is asked to respond on a 5-point scale ranging from “not at all” to “all the time” to describe the presence of each item. A score on any item that is equal to or greater than 2 (“about half the time”) and greater than their baseline score is recorded as an adverse event on this scale. All adverse events/side effects will be captured at baseline (to ascertain presence of these symptoms prior to the intervention), mid-intervention, and 8–10-week follow-up.

Development and Wellbeing Assessment (DAWBA) [41]: The DAWBA is a package of interviews and questionnaires completed by parents/guardians only to reduce the assessment burden on the adolescent. This approach has been commonly taken in other trials with children and young people and their parents [42, 53]. The DAWBA is designed to generate ICD-10 and DSM-IV/DSM-5 psychiatric diagnoses for children and young people. The DAWBA computer algorithm estimates the probability of having a psychiatric disorder in bands of <. 1%, 0.5%, 3%, 15%, 50%, and > 70% based on large community-based population studies [41], and the top two levels have been shown to reliably indicate the presence of a clinician-rated diagnosis and can be used as an alternative to clinician-rated diagnoses in research studies [54]. The DAWBA has established validity and reliability [41]. The parent-reported DAWBA starts with a Strengths and Difficulties Questionnaire (SDQ) [55]. If the participant is later enrolled in the trial, this will be used as their baseline SDQ. DAWBAs that score people as being likely to have any conditions will be second reviewed by the chief investigator (a medical practitioner) to ascertain that they should be excluded from the trial. It must be noted that the DAWBA is not being used in this trial to diagnose depression but to establish possible comorbid disorders and determine study eligibility (i.e. whether to exclude potential participants from the study).

Child and Adolescent Intellectual Disability Screening Questionnaire (CAIDS-Q) [44]: The CAIDS-Q is being utilised to determine the presence of intellectual disability at baseline. The questionnaire contains seven items (two on literacy, one on telling the time, one on friendships, two on previous contact with specialist services and current educational support, and one on tying laces). They are answered in a yes/no format by someone who knows the person well, and some items can be tested directly with the adolescent (depending on age/communication). A total score is calculated which is converted to a percentage score. A cutoff (by age group) indicates if the child is likely to have an intellectual disability or not. The questionnaire has established validity and reliability [56].

The process evaluation will follow the Medical Research Council’s (MRC) guidelines for evaluating complex interventions [57]. It will explore the components suggested in MRC guidelines, namely reach, dose, and fidelity of implementation of intervention, and make recommendations for adaptations for the future RCT. It will also examine the contextual factors and potential mechanisms underlying participant behaviour change and probe for any unexpected consequences.

Interviews will be conducted with adolescents assigned to either version of SPARX (target n ≥ 30) and parents/guardians of adolescents (n ≥ 30) after they have completed their 8–10-week follow-up. This sample size will be determined based on the model of information power where we will aim to achieve both breadth and depth of views [58]. Purposive sampling will be used so that a diverse range of views are voiced, including diversity in terms of ethnicity, socioeconomic status, and adherence to the intervention (i.e. interviewing those who completed 0 level to treatment completers). Interviews with e-coaches supporting the intervention (target n ≥ 3) and clinicians recruiting to the study (n ≥ 5) will be conducted during trial delivery. Online feedback from participants will also be analysed together with usage indices recorded as part of the online system such as total time spent with e-coach, number of modules completed, and number of logins. A brief online questionnaire will be given to participants who drop out of the trial early to gain a more holistic overview of engagement with SPARX. A full process evaluation protocol will be published.

Table 2 outlines the completion time points for each outcome measure and who completes them.

No significant risks to physical safety are anticipated; however, changes in depressive symptoms may occur in adolescents during the trial and cause psychological harm. Although a formal data monitoring committee is not required due to the minimal physical risks associated with the intervention, we will have an independent scientific advisory board, which serves a similar function to a Trial Steering Committee (TSC), throughout the trial. To ensure participant safety, adverse events will be recorded at baseline, 4 weeks, and 8–10 weeks. Any related or unexpected serious adverse events will be reported to the research ethics committee.

The current project sits within a programme of work with its own PPI group of young people with lived experience of accessing services for depression/mental health conditions, called Sprouting Minds. The initial study grant was reviewed and received feedback from Sprouting Minds. Since January 2022, a subset of Sprouting Minds have met regularly to advise and guide the research. The group comprises one parent and three young people with lived experience relevant to the study. PPI has shaped the design of the trial and its dissemination in the following ways:

The PPI group mainly contributes remotely to enable involvement from members who are not geographically close, and all members are paid for their time in line with National Institute for Health and Care Research (NIHR) involvement guidelines. Their advice and input will be sought throughout the trial including participation in facilitating and analysis of the interviews with adolescents and parents/guardians, co-creating lay summaries, and other dissemination materials including involvement in conference presentations and as co-authors on publications.

Summary statistics will be presented for this trial using Microsoft Excel. No interim analyses are planned. Summary statistics, proportions, and standard deviations will be given to estimate the change in PHQ-A, variance, and recruitment rates (per month) in different settings and whether there is compliance (defined as the percentage completing four levels or more of SPARX). The primary pilot outcome is the variance in change in the primary outcome measure (PHQ-A). We will not stratify at this point because we are estimating variance, not effect, and we are assuming they are similar, although this will be checked. A full analysis plan will be made available before unblinding of data, and anything that is not planned will be considered data dependent and considered speculative [59].

In-depth semi-structured qualitative interviews with adolescent participants receiving the online intervention and their parent/guardian will capture general feedback on their experiences of taking part in the trial as well as satisfaction and acceptability of the intervention as part of the process evaluation. We will explore, with both participants receiving the intervention and intervention supporters (i.e. the e-coaches), their experiences of receiving and supporting the intervention online, mechanisms of impact, barriers, and facilitators to taking part in the intervention and in their continued involvement. Referring clinicians and e-coaches will also be interviewed to gain their experiences of being involved in the trial.

All interviews will be audio-recorded and transcribed by a University of Nottingham-approved external transcription company or by a member of the research team. Transcripts will be anonymised before analysis. Interviews will be analysed using reflexive thematic analysis [60], and, more broadly, themes will be organised using the framework method [61]. All analysis will be conducted using NVivo 12. Coding and organisation of codes will be cross-checked within the research team to ensure validity.