Design
The design is a two-arm cluster-randomised control trial with randomisation at the level of the speech and language therapist. Randomisation will be stratified by site. There will be three participant groups: speech and language therapists, parent-child dyads, and SLT managers of services that are participating.
Participating SLTs allocated to the intervention arm will deliver Palin STSC(8-14) and those allocated to the control arm will deliver the usual intervention provided to school-aged children who stammer within that NHS service. SLTs allocated to the Palin STSC(8-14) arm will receive training in the approach immediately. It is possible that their approach to children who stammer will be influenced by this training and so those SLTs allocated to the treatment as usual control arm will have the option to attend training after all study measures have been completed. Potential child participants will be identified by SLTs and will receive intervention based on their assigned SLTs allocation.
Children and parents will complete questionnaires at baseline and again six months later (see Figure 1). SLTs complete a questionnaire at the start and end of their involvement in the trial. In addition, semi structured interviews with a purposive sample of SLTs, parents, children and SLT managers will be conducted at the start and end of the trial (all groups) and at the end of therapy (children, parents and SLTs). This trial protocol is in line with Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 (see SPIRIT checklist in Additional File 1).
Setting
This is a multi-site trial. NHS speech and language therapy services in England and Wales which offer, or which have potential to offer, therapy to children who stammer aged 8-14 can take part in the study.
Intervention
Since the onset of the pandemic in March 2019, there has been an increase in access to services through telehealth. To reflect this change, either intervention may be accessed during face-to-face appointments or via a video platform.
Treatment as Usual
There is no defined or accepted standard practice for this client group and little evidence of what constitutes Treatment as Usual and so the intervention that children receive will vary across services. It may include: generalist therapy, group therapy, reviews of progress, referral to a specialist service or no treatment at all. It is anticipated that the goals and content of therapy will also differ and may include: general advice, fluency strategies, stammering modification, confidence speaking, and/or increased environmental support.
Palin Stammering Therapy for School Children (8-14)
The primary aim of the intervention is for the child to become a more confident and effective communicator. In order to achieve this, the following goals are targeted:
- maximising the child’s fluency and/or reducing struggle speaking
- enhancing communication skills in the family
- increasing confidence and participation in speaking situations
- reducing child’s and parents’ worry about stammering
- developing parents’ skills and confidence to support the child
Sessions start with a Solution Focused Brief Therapy session to explore goals and expectations from therapy24. Therapy focuses on the development of skills across three main areas: (1) family communication skills, (2) strategies to address the cognitive/emotional needs, and, (3) direct fluency work. The approach assumes that the child is only able to make changes in these areas with the support and input of at least one parent/carer and so they are involved in the therapy throughout. Palin STSC (8-14) comprises 10 sessions, as this ‘dosage’ level was shown to be efficacious in our preliminary study23.
Participants
There are three groups of participants: SLTs who deliver the intervention, parent and child dyads and SLT managers. We seek to recruit a total of 30 SLTs to the trial (i.e. 15 per intervention arm), with each therapist providing on average 2 children (and no more than 4). We aim to recruit in total, 60 children who stammer between the ages of 8-14. With the expectation of attrition of around 20%, this would result in a total of around 50 completers. We will recruit 6 SLT managers to be interviewed.
Recruitment
SLTs: SLTs will be recruited from NHS speech and language therapy services who receive referrals for children who stammer and are aged between 8-14 years old. The services can be located anywhere within England or Wales in line with the trial’s ethical approvals from the Health Research Authority.
SLTs will be made aware of the study through personal contact, social media posts, notifications through Clinical Research Networks, SLT special interest groups and through the Royal College of Speech and Language Therapists. No previous expertise or experience in stammering is required. SLTs and managers will be sent SLT participant information sheets and a trial summary in response to any expression of interest.
Child and parent participants: The participating SLTs will be responsible for identifying eligible children who are on the caseload or who are referred to their service for stammering intervention. Figure 2 summaries the process of recruitment and participation for children and parents. Once identified by a SLT, the parent will be approached by the SLT to be told about the service’s involvement in the trial and to request their consent to be contacted by the study research team. The study team will then contact the parent to screen for eligibility and initiate the informed consent process. The Participant information sheets, for both children and parents, as well as copies of the consent and assent forms will be sent for consideration and two weeks will be given for the family to consider participating.
SLT managers: A purposive sample of 4-6 managers of services that are participating in the trial will be invited to take part in semi-structured interviews.
Consenting
SLTs: SLTs will be offered time with the research team to discuss the trial and to ask questions. If the SLT wishes to proceed and the service manager supports this, then the research team will liaise with the local R&D team to open the trial at that site. SLT consent will be taken remotely via the trial database.
Children and parents: At the end of a two-week consideration period, the family will be contacted by the research team to provide an opportunity to ask questions. The parent consent form and child assent form will be completed remotely via the database.
Inclusion criteria:
The inclusion criteria for each participant group is as follows:
SLTs:
- Not currently delivering Palin STSC (8-14) and not previously trained in the approach
- Have the potential to identify and provide intervention (either Palin STSC (8-14) or Treatment as Usual) to at least 2 children within the duration of the study.
Children:
- Aged between 8 years 0 months and 14 years 11 months at the start of the study
- Diagnosed / recognised as stammering by speech and language therapist, parent and self.
- Want to receive stammering therapy
- Have at least one parent who is able to attend therapy and participate in the study
Parents:
- Have a child who is participating in the study
SLT managers:
- Have management responsibility for the SLTs involved in the study
- Had a role in the decision about whether the service would take part in the trial
Exclusion criteria
SLTs:
- Currently working with families on a weekly basis using principles of Palin STSC (8-14)
- Unable to provide weekly family therapy sessions if allocated to Palin STSC(8-14) due to service or resource limitations.
Children:
- have received therapy in the previous 6 months or are currently receiving speech and language therapy
- have attended the Michael Palin Centre for assessment or therapy
- do not have a parent who consents to take part in the study
- are involved in any other research study which involves an intervention or multiple assessments
- are accessing formal counselling through psychological services and the addition of Palin STSC is contra-indicated
Parents:
- Their child has been excluded for any reason
- There are any contra-indications for therapy. This might include any significant issues in the family which would be taken into account for the timing of any therapy e.g., bereavement or birth of a baby.
Randomisation and allocation process
SLTs: The randomisation list has been generated by Norwich Clinical Trials Unit data management team using an online sequence generation service, which is reviewed and approved by the study statistician. Speech and language therapists (SLTs) are randomly allocated into one of the two treatment arms, Palin STSC (8-14) or TAU, using blocked randomisation with a block length of 2, stratified by service.
Following consent and baseline data collection, a member of the trial team responsible for trial enrolment randomises the SLT via a web-based randomisation system integrated into the study database. An immediate allocation is provided by the system to the person entering the data. Automated email alerts are sent to the Site PI and the Research Assistant informing them of the allocation.
Child and parent dyads: Once baseline assessments are completed, families will be allocated to a SLT in accordance with local SLT service procedures. For some, the SLT is predetermined by clinic or school attended, for others, families are allocated on the basis of SLT availability (i.e., when a therapist has a free assessment appointment, the next family on the list are allocated to that appointment and therapist). Processes were monitored by the research team to avoid the potential of bias in allocation processes.
Procedure
SLT Training
TAU: SLTs in the TAU arm will receive training on the research process and procedures.
Palin STSC(8-14): SLTs in the Palin STSC(8-14) arm will also receive research protocol training, along with training in the Palin STSC(8-14) assessment and therapy methods. This has been developed into a package of 13 video recorded sessions which take approximately 20 hours to complete (comparable to the original 3-day face to face training programme). In addition, SLTs will be able to attend up to three remote question and answer sessions with an SLT who is specialist in the programme.
Fidelity Assessment
SLTs from both arms will complete an online ‘session summary checklist’ following each therapy session to detail the logistical information (location, attendees, session number, length of time etc.) and therapeutic elements / topics covered. Those from the Palin STSC(8-14) arm will be used to check the topics covered in each session for fidelity purposes. In addition, SLTs delivering Palin STSC (8-14) will video record each therapy session. A random sample of two sessions from each child will be evaluated by an SLT who is a specialist in the programme, using a treatment fidelity checklist to be developed as part of the trial.
Measures
Basic demographic data on age, gender, ethnicity will be collected alongside stammering history information (time since onset, family history, previous therapy history). Selected quantitative measures have been chosen for their psychometric properties and clinical utility. Measures will be used to evaluate the feasibility outcomes: to ascertain recruitment and retention, treatment fidelity, treatment outcome, acceptability of the research and therapy methods. The study also includes measures for variables that are predicted to influence stammering impact and therapy outcome, since this will form part of the full scale trial. Some of the assessments completed are directly relevant for the implementation of therapy and the results will be provided to the therapists for their clinical decision making needs (those provided to therapists in the Palin STSC(8-14) arm are marked *). Each therapist providing TAU will identify in advance any assessments that are part of usual practice and will receive the results of those. The online and postal questionnaires will take each parent and child approximately 1.5 hours to complete.
Recruitment and retention rates
Recruitment and attrition rates will be established for SLTs, children and parents. The numbers of children approached, numbers consenting and attrition rates will be recorded. Participants are free to withdraw at any time without giving reasons and without prejudicing any further treatment that would be usually available. Participants will be asked if they are willing to say why they are withdrawing so that this information may be used in the context of whether the therapy and trial methods are acceptable. Permission will be sought to use data to the point of withdrawal but this may be withdrawn up until the time of analysis.
Child completed measures:
Speech Sample*
A video recording of the child reading and talking during conversation, and analysed by the research assistant using the Stammering Severity Instrument Fourth Edition25 (SSI-4), provides a standardised clinical measure of stammering severity. The conversation section will be analysed using the CHILDES26 system to examine the child’s use of language and turn taking skills. This will be conducted at both baseline and follow up. (4 minutes).
Satisfaction with fluency27
Adapted from the Satisfaction with Communication in Everyday Speaking Situations Scale27 and completed on the on-line database. Children rate “how happy are you with your speech at the moment?” on a nine-point scale at both baseline and follow up (1 minute).
Behavior Assessment Battery28
The Behavior Assessment Battery28 includes: Behavior Checklist, Speech Situation Checklists and Communication Attitude Test completed on paper. The Behavior Checklist: requires a yes/no response to a series of potential features of stammering. The child records whether or not each is typical of their own speech. Speech Situation Checklistconsists of two parts. In the first, the child rates the severity of the stammering in different speaking situations (SSC-SD) on a series of likert scales, and the second, the child rates the degree of fear about speaking in different situations (SSC-ER). Communication Attitude Test-Revised (CAT-R) evaluates the speech-related belief system of children who stammer using 35 items with Yes/No response. These assessments will be completed at both baseline and follow up (30 minutes).
The Overall Assessment of the Speaker’s Experience of Stammering (School age/Teenage versions) (OASES-S/T)29*
The OASES-S29 evaluates the impact of stammering on an individual’s life across four sections: general information, reactions to stammering, communication in daily situations, and, quality of life. For participants over the age of 12, the teenage version will be used. The OASES-S or OASES-T will be completed at both baseline and follow up (15 minutes).
Strengths and Difficulties Questionnaire (SDQ)30.
This is a screening tool for behavioural, emotional and social development with 5 subscales: emotional symptoms, conduct problems, hyperactivity/ inattention, peer relationship problems and pro-social behaviour. Children aged 11-14 years will complete an adolescent version at both baseline and follow up (10 minutes).
Revised Children’s Anxiety and Depression Scale (RCADS)31
This is a 47-item, youth self-report questionnaire with subscales including: separation anxiety disorder (SAD), social phobia (SP), generalized anxiety disorder (GAD), panic disorder (PD), obsessive compulsive disorder (OCD), and major depressive disorder (MDD). There are parent and child versions for each to complete at both baseline and follow up (10 minutes).
Early Adolescent Temperament Questionnaire-Revised (EATQ-R)32
Designed to assess temperament in 9-15 year olds to gain insight into positive and negative reactivity and regulation. There are two questionnaires, one for parents and one for young people. For children aged 8 the Child Behaviour Questionnaire33 will be used (20-30 minutes). In the definitive trial temperament will be considered as a potential predictor variable for outcome. The EATQ-R32 will be completed at baseline only.
Child Health Utility 9D (CHU-9D)34
This is a series of 9 questions for the child to rate how they feel and their ability to participate in daily activities. The scores from this assessment can be converted to Quality of Life Years and will be used in the health economic evaluation and it will be completed at both baseline and follow up (10 minutes).
Completed by parents:
Palin Parent Rating Scales (Palin PRS)35*
This standardised assessment consists of 19 rating scales and is completed by parents to measure their perspective across three factors: the impact of the stammer on the child, the severity of the stammer and the impact on the parents, parents’ knowledge about stammering and their confidence to support their child. It will be completed at both baseline and follow up (5-10 minutes).
Strengths and Difficulties Questionnaire (SDQ)30
Parent version of the SDQ described above will be completed at both baseline and follow up (5-10 minutes).
Revised Children’s Anxiety and Depression Scale (RCADS)31
Parent version of the RCADS described above will be completed at both baseline and follow up (10 minutes).
Early Adolescent Temperament Questionnaire-Revised (EATQ-R)32
Parent version of the EATQ-R described above. Parents of children aged 8 complete the Temperament in Middle Childhood Questionnaire (TMCQ)33. This will be completed at baseline only (40 minutes).
Quality of Life in a Child’s Chronic Disease Questionnaire (QLCCDQ)36
This is a 15 item Likert scale concerning the impact of a child’s impairment on a parent’s life. Parents will rate the degree to which their daily lives have been affected by the child’s condition in the previous two weeks, at both baseline and follow up (10 minutes)
Service Use Schedule
This questionnaire has been developed for the purposes of this study and explores the use of different services for the health economic evaluation. It will be completed at both baseline and follow up (10 minutes).
Completed by speech and language therapists
Confidence and knowledge about working with children who stammer
A questionnaire has been developed for the purposes of this study to explore therapists’ knowledge and practices for working with school age children who stammer in the NHS. The SLTs confidence in working with this client group will also be explored. It is anticipated that these factors will improve with the trial for therapists in both treatment arms and so this will be completed at baseline and at the end of the trial. Those in the Palin STSC(8-14) treatment arm will also complete this at the end of their training in the programme (15 minutes).
Session record form
SLTs will record the content of each therapy session using a checklist, along with attendance and the homework completed by the family. This will inform the degree to which SLTs are completing the programme and the engagement of the families, providing data for the treatment fidelity evaluation and acceptability of the therapy methods (10 minutes per therapy session).
Data Collection
Each parent and child dyad will complete the questionnaires at baseline and again approximately 6 months later. Study data will be collected and managed using REDCap electronic data capture tools hosted at the University of East Anglia37. REDCap (Research Electronic Data Capture) is a secure, web-based application designed to support data capture for research studies, providing: 1) an intuitive interface for validated data entry, 2) audit trails for tracking data manipulation and export procedures, 3) automated export procedures for seamless data downloads to common statistical packages, and 4) procedures for importing data from external sources.
Prior to COVID, all measures were completed online into a REDCAP database platform developed and managed by NCTU with the exception of the Speech Situation Checklists, Communication Attitude Test28, Strengths and Difficulties Questionnaire30, and Stuttering Severity Instrument-425 which were completed in person with the Research Assistant. Following COVID, the Speech Situation Checklists, Communication Attitude Test28, Strengths and Difficulties Questionnaire30 were changed to postal completion and the Stuttering Severity Instrument-425 was collected via video link with the Research Assistant.
All SLT measures were completed online.
Process evaluation
As recommended by the Medical Research Council guidance for RCTs38 we will include a qualitative process evaluation of the trial and the intervention. We hope to identify the barriers and facilitators to taking part in the trial, as well as modifications that can be made to improve the experience of taking part in this research.
A purposive sample of 4 SLTs, 4 children and 4 parents from each treatment arm will be recruited to take part in semi-structured qualitative interviews at the start and end of therapy, and at the end of the trial. We will seek to recruit children of different ages, from different services, with varying levels of stammering impact, and parents with differing levels of knowledge and confidence. We will select therapists from different speech and language therapy services and from both Palin STSC(8-14) and Treatment as Usual groups. These participants will complete semi- structured interviews at the three time points (total: 72 interviews).
We noticed that when we first spoke to SLTs and SLT teams that they were very enthusiastic and would suggest that multiple SLTs would be available to take part. However, it was often the case that by the time the R&D approvals were in place and recruitment of SLTs started, the numbers reduced considerably compared to the numbers originally proposed. Further, during conversations with SLTs during their baseline interviews and in their training sessions, it became apparent that the support and attitudes of different managers varied but crucial for successful participation. We therefore amended the protocol to include interviews with managers to identify the drivers that influenced decisions about whether a service or individual therapist could take part in the trial and the amount of time they would be given to do so. Understanding the facilitators and obstacles from managers’ perspectives will inform how managers are approached and the information that they receive in the full-scale trial.
We will interview a small sample of 4-6 service managers of participating sites at two time points: the opening of their service as a research site and at the end of the trial. We will aim to understand more about why the service managers agreed for their service and staff members to take part in the trial, their expectations of service participation and their experience of having their service involved. All interviews will be conducted over the phone and audio recorded.
Once the Palin STSC(8-14) training package is completed, SLTs will provide feedback about the quality and content of the training, with recommendations for any changes. This is particularly important given the changes made to the method of training that were made as a result of the pandemic (15 minutes).
Analysis plan
The primary statistical analysis will be directed towards the aims of the feasibility study outlined under the aims and objectives section. Particularly, the values under objective 1, including recruitment and attrition rates, will be calculated with an appropriate 95% confidence interval to inform the design of a future study. Similarly, we will aim to quantify and summarise treatment ‘doses’ actually delivered and adherence.
Although not designed as an efficacy trial, we will make an initial estimate of treatment effect on the above-named efficacy variables using linear models with appropriate link and error terms. These will adjust for known baseline prognostic variables (decided prior to analysis) and will be presented with 95% confidence intervals. data from all participants as randomised in the trial will be included in the ‘intention to treat’ analyses. To avoid attrition bias, outcome data from all participants will be included regardless of protocol adherence.
For the process evaluation, all interviews will be transcribed and analysed using constant comparisons from grounded theory methodology to explore themes regarding process variables, fidelity and acceptability for both the therapy and the trial methods.
In terms of health economics, estimation of cost-effectiveness, within a health-technology assessment, is an iterative process39. Here we aim to monitor levels of resource-use and quality of life (QoL), to inform the decision as to how costs and benefits can be measured as part of a future, more definitive study. We will therefore estimate the completion rates for both the aforemtioned CHU-9D (responses to the CHU-9D can be converted into QALY (Quality Adjusted Life Year) scores)40 and service use schedule, and seek to identify big cost drivers.