Trial design
The PExMS project follows the guidance of the Medical Research Council(30) for the development and evaluation of complex interventions and adheres to the SPIRIT checklist (see additional file 1). The process has a development as well as a feasibility and piloting phase.
- Development: The PExMS-website was developed within a qualitative interview study, where PEx were collected (see ‘Intervention description’).
- Feasibility and piloting: First, people with RRMS (see ‘Intervention description’) will be asked about their experiences with the PExMS website (e.g. usability, acceptance, layout, navigation) and the outcome measurement procedures.
Second, in a two-arm pilot RCT, we will compare a combination of the PExMS-website and evidence-based patient information with evidence-based patient information alone. The pilot trial is intended to test recruitment, outcome measures, data collection (sampling and timing), randomisation procedures and acceptability of the intervention for pwMS and practitioners and gain exploratory results that could be used for sample size estimations in future RCTs. We want to assess whether the intervention was conducted as intended (fidelity) and assess the quantity of the implemented intervention (dose).
A mixed-methods process evaluation will be applied. In the recruitment phase, we will perform semi-structured telephone interviews with all responsible persons in participating centres to assess perceptions of strengths, weaknesses, risks, and opportunities of the study.
After completing the study, centres which succeeded in recruitment and centres that dropped out prematurely or failed in recruitment will be interviewed and will be asked to fill out a process evaluation questionnaire to assess acceptability of the recruitment, randomisation and consent procedure and the intervention itself.
Additionally, approximately 10 patients, namely maximum users and minimum users of the PExMS-website and those having dropped out will be interviewed in each group and will be requested to complete a process evaluation questionnaire.
Study setting
The study will be conducted in different participating neurological practices, clinics or rehabilitation centres located in Germany.
Eligibility criteria
Eligibility criteria are RRMS, being ≥18 years old and considering immunotherapy for mild/moderate or (highly) active courses. Participants will be excluded if they are diagnosed with SPMS or PPMS, have major cognitive deficits and/or poor German language skills. Also, prior participation in the development phase of the PExMS-website is an exclusion criterion.
Interventions
Explanation for the choice of comparators
The control group will receive access to the DECIMS (Decision coaching in MS)-Wiki (www.wiki2.kkn-ms.de), an evidence-based patient information website focusing on MS immunotherapies for four weeks. DECIMS-Wiki is considered to provide excellent factual information with figures, but without PEx and is regarded as the gold standard. Further information is described elsewhere.(31) Site visits will be recorded and tracked with the explicit consent of the participants. At the end of the four-week period, study participants will be invited to complete follow-up questionnaires.
Intervention description
Development of an intervention website with PEx
The intervention is a multimedia PExMS-website providing short videos, audio recordings and written excerpts of 50 pwMS. The development of the intervention followed largely recommendations for standardised qualitative research provided by international DIPEx (Database of Individual Patients' Experience of illness) association. This collaboration of researchers and health professionals use standardised qualitative research methods to understand PEx and tries to provide ‘balanced’ information from original interview data. However, this project is not affiliated to DIPEx. Figure 1 shows the process of the website’s development.
Figure 1: Flow diagram showing the process of developing a website with PEx
Data for the website was generated in a qualitative interview study with RRMS patients (≥ 18 years old) across Germany. A maximum variation sampling strategy(32) was used aiming to cover experiences with all licensed MS drug treatments. A problem-centred interview guide according to Witzel(33) was applied involving open-ended and closed questions on PEx with diagnosis, talking about MS diagnosis with others, MS in everyday life and experiences with management approaches (DMT, lifestyle intervention, alternative medicine, rehabilitation) as well as decision-making processes. Interviews were audio- and videotaped, and afterward transcribed. We analysed qualitative data thematically(34) using deductive and inductive categories. All themes representing experiences with MS in everyday life and with therapies were organised in sub-themes and illustrated by text, audio, and video clips to provide the basis for the PExMS-website.
Feasibility testing of the PExMS-website will be conducted with people with RRMS (≥ 18 years old) not previously involved in the development phase of the PExMS-website. They will receive access to the PExMS-website for one week for evaluation of the preliminary version of the website. We will record log information (e.g. date, number of logins and file name of the webpage) by the webserver to assess users’ navigation on the website. Then two focus group discussions with 5-8 RRMS patients per group will be conducted asking for experiences with the website (e.g. usability, acceptance, layout, navigation) and suggestions for improvement.(19) The focus groups will be audiotaped and transcribed verbatim. Moreover, outcome measures, which are described in the following, will be tested. We will analyse this data descriptively using SPSS Statistics™ Version 25 and use the results for the optimisation of the content and presentation of PEx on the website. The final tool will be used as an intervention in the pilot trial.
The intervention group
The intervention group will receive access to the PExMS-website and the DECIMS-Wiki for four weeks. As in the control group, the site visits are tracked. At the end of the four weeks, pwMS will be invited to complete follow-up questionnaires.
Criteria for discontinuing or modifying allocated interventions
Studies investigating PEx as a patient information tool found no adverse events.(14, 22, 35) Therefore, we do not expect any relevant harm of the intervention and did not establish criteria for discontinuing or modifying the allocated intervention for participants. Nevertheless, study participants may withdraw their participation at any time upon request.
Strategies to improve adherence to interventions
All participating neurological practices, clinics or rehabilitation centres will receive a study folder including all relevant documents. Also, regular phone calls and telephone conferences will be conducted for arising questions, updates and for sharing experiences.
Usage of the intervention will be monitored by tracking the site visits. PwMS will be contacted by regular email reminders by a member of the coordinating centre to use the intervention. In case of non-use, participants will be reminded by email or telephone.
Outcomes
Primary outcomes
Primary outcomes will be assessed by mixed-methods consisting of interviews and process evaluation questionnaires, which includes the following overall feasibility and process-related measures:
- Number of participants consented, recruited, randomised, withdrawn and retained
- Numbers with completed outcome measures or lost to follow-up
- Acceptability of the intervention in terms of frequency and duration of usage as well as experience, motivation, barriers and facilitators to use the intervention – from the pwMS and participating study centre perspective
- Acceptability of the study procedures in terms of barriers and facilitators to implementation – from the pwMS and participating study centre. We will also ascertain the number and type of adaptations made to implementation strategies including information on how and why.
Secondary outcomes
We hypothesise that PEx will lead to a positive effect on decision self-efficacy.(23) This will be assessed by the 11-item ‘Decision Self-Efficacy Scale’ (DSES)(36), which measures self-confidence in decision-making on a five-point-Likert scale. It ranges between 0 (not at all confident) and 4 (very confident). For the total score, items are summed, divided by 11 and multiplied by 25. A total score of 0 means ‘extremely low self-efficacy’ and a score of 100 means ‘extremely high self-efficacy’. The Decision Self-Efficacy Scale shows adequate internal consistency achieving a Cronbach’s alpha of 0.84.(37)
To assess how useful a decision support intervention is in preparing pwMS to communicate at a consultation visit and to make a decision concerning a therapy, the 10-item ‘Preparation for Decision Making (PrepDM)’(30) scale will be used. It has a five-point-Likert scale format and is designed to be administered after the consultation visit to discuss treatment options. Higher scores show a higher perceived level of preparation for decision-making. Cronbach’s alpha ranges from 0.92 to 0.96.(38)
It is assumed that patients seeing PEx develop greater MS risk knowledge, which we will assess with a short version of the 19-item ‘Risk Knowledge in Relapsing Multiple Sclerosis 2.0’ (RIKNO).(39, 40) Item difficulty is high ranging from 0.07 to 0.79. RIKNO 2.0 has good internal consistency achieving a Cronbach’s alpha of 0.73.(40)
‘Informed choice’(41) will be assessed by the Multidimensional Measure of Informed Choice (MMIC) as in previous trials.(9, 42). It comprises three dichotomous dimensions: risk knowledge (good, poor), attitude towards or against immunotherapy or change of immunotherapy (positive, negative) and choice (uptake or non-uptake of the intervention under consideration) leading to eight types of choices. PExMS is expected not to change the percentage of informed choices. Besides, the distribution of decisions made for or against immunotherapy is expected to remain unchanged.(22, 35)
The 4-item ‘SURE’ (Sure of myself; Understand information; Risk-benefit ratio; Encouragement) scale(43) addresses decisional conflict in patients. It will assess patients’ perception of uncertainty about decision making for a therapy. The reliability of SURE with a Cronbach’s alpha of 0.54 is moderate.(44)
We will assess patients’ preferences for involvement in treatment decisions by the 5-item Control Preferences Scale (CPS)(45). In a validation study, a Cronbach’s alpha of 0.72 was attained.(46)
PEx may give pwMS confidence in active participation and management in medical care.(47) Therefore, the 13-item ‘Patient Activation Measure 13’ (PAM13-D)(48) will be used which is a reliable and valid measure of patient activation with a four-point-Likert scale. It shows good internal consistency achieving a Cronbach’s alpha of 0.87.(48)
The eHealth Impact Questionnaire ‘eHIQ’ (47, 49) measures attitudes of users towards a website which they recently viewed. It is divided into the 11-item eHIQ-Part 1 and the 26-item eHIQ-Part 2. The former represents general attitudes towards using the internet to access health information. The second part is related to the effects of using a specific health-related website on three subscales: 1) confidence and identification, 2) information and presentation and 3) understanding and motivation. Both answering formats range from 1 (‘strongly disagree’) to 5 (‘strongly agree’). The eHIQ shows good internal consistency achieving a Cronbach's alpha of ≥ 0.77. A German translation will be validated and used. Moreover, identifying with others who show their experiences on a website, knowing that other pwMS are handling similar problems and learning how they manage difficult issues can reduce the feeling of isolation, improve the sense of social support and group membership.(14, 25) Reformulated items from the subscale “confidence and identification” of the eHIQ will be used to assess social support.
‘Affective forecasting’ is defined as the patient’s perceived ability to make predictions of future feelings about difficult health procedures or disabling health conditions. We hypothesize that PEx could have an impact on patients’ ‘affective forecasting’.(23) As no validated questionnaire exists for this, a preliminary scale with 5 to 10 items will be developed through a review of relevant literature. We will conduct cognitive interviewing using the think-aloud methodology after showing approximately five patients clips from the website with PEx and the preliminary scale. The clips will show how other patients have coped with emotions during a certain situation as well as the intensity and duration of an emotional state, which the patients experienced. Pilot testing of the revised scale will be conducted applying again cognitive interviewing for face and content validity.
Demographic/clinical variables/covariates
As a control parameter, the ‘Hospital Anxiety and Depression Scale’ (HADS) will be administered exploring symptoms of anxiety and depression.(50) We will also measure patient-reported disability of pwMS using the ‘Patient Determined Disease Steps’ (PDDS)(51). The ‘Stage of Decision Making’ (52) questionnaire asks for the patient’s readiness to engage in decision-making, progress in decision-making and susceptibility to considering or re-considering options. It may be useful in screening out pwMS who may not benefit from decision aid interventions.
For a list of outcomes and a schedule of assessments, see table 1.
For a list of outcomes and a schedule of assessments, see table 1.
Table 1: Study assessment
|
Study period
|
|
Enrolment
|
Pre-study baseline / allocation
|
Study / post-allocation
|
TIMEPOINT
|
-t1
|
0
|
t1
|
t2
|
ENROLMENT:
|
|
|
|
|
Eligibility screen
|
X
|
|
|
|
Informed consent
|
X
|
|
|
|
Randomisation
|
|
X
|
|
|
Sociodemographic data and MS-related data
|
|
X
|
|
|
ASSESSMENTS:
|
|
|
|
|
PDDS
|
|
X
|
|
|
HADS
|
|
X
|
X
|
|
DSES
|
|
X
|
X
|
|
PrepDM
|
|
|
X
|
|
RIKNO
|
|
X
|
X
|
|
Informed choice
|
|
X
|
X
|
|
SURE
|
|
X
|
X
|
|
CPS
|
|
X
|
X
|
|
PAM 13-D
|
|
X
|
X
|
|
Affective forecasting
|
|
X
|
X
|
|
Social support
|
|
|
X
|
|
Stage of Decision Making
|
|
X
|
|
|
Control Preferences Scale
|
|
|
X
|
|
eHIQ-Part 1
|
|
X
|
|
|
eHIQ-Part 2
|
|
|
X
|
|
Patient treatment decision/Uptake
|
|
|
|
X
|
t1 = after four weeks usage of intervention or comparator website(s); t2 = after physician encounter
|
Participant timeline
Figure 2 shows the flow of enrolment, interventions, assessments and visits for pwMS.
Figure 2: Study flowchart
Sample size
To our knowledge, there is no previous research using decision aids with patient narratives and addressing decision self-efficacy. However, since the main aim of the pilot RCT is not to give a formal assessment of efficacy, but rather to evaluate trial procedures and processes as well as to get estimates of parameters for the main RCT sample, the sample size justification is based on rules of thumb for pilot RCTs as given by Sim and Lewis. The authors recommend for a two-armed pilot RCT a total sample size of 55, which would minimise the overall sample size for the main RCT for small to medium standardised effect sizes. (53-55) Additionally, pilot RCTs on decision support tools in MS showed comparable sample sizes. (31, 56)
Recruitment
Participants will be recruited by 2–3 neurological practices, clinics or rehabilitation centres in Germany. Potential participants will be invited to read the patient information sheet before deciding whether to take part in the study. Informed consent will be obtained by a physician or trained scientific personal in the participating neurological practices, clinics or rehabilitation centres.
Once consent has been obtained, participants will get a link and an individual access key to the study registration webpage, which is QuestBack Unipark (2019), an online survey tool. There, participants will be asked for socio-demographic and MS-related data.
Assignment of interventions
Sequence generation
PwMS will be randomly allocated in blocks with variable block length at a 1:1 ratio to the intervention or the control group. The concealed allocation sequence will be computer-generated by the randomisation sequence of QuestBack Unipark.
Concealment mechanism
The randomisation will be automatically performed via QuestBack Unipark and will be therefore concealed.
Implementation
PwMS will be enrolled in the participating neurological practices, clinics or rehabilitation centres. The allocation sequence and participants’ assignment to interventions will be automatically generated by QuestBack Unipark.
Blinding
Blinding of the pwMS is pursued but is only possible to a limited extent due to the type of intervention. We will explain in our information and consent form that we intend to assess whether pwMS find different health information websites helpful on a range of self-reported outcomes. Researchers involved in study recruitment, administration of interventions and data analysis will be blind to the allocation of intervention and control group. Since we do not expect relevant adverse events, no procedure for unblinding is planned.
Data collection and management
Plans for assessment and collection of outcomes
Data from the Pilot RCT will be collected at four time points using web-based questionnaires from QuestBack Unipark (see ‘Outcomes’). This promotes the quality of data due to the avoidance of data entry and coding errors by third persons, reduced administrative burden, prevention of item nonresponse and automatic application of skip patterns. Patients’ treatment decisions will be assessed after the intervention by a physician of the participating neurological practices, clinics or rehabilitation centres and documented by physicians or study nurses of the study centre.
Plans to promote participant retention and complete follow-up
PwMS will be contacted via regular email reminders by a member of the coordinating centre when it is time to fill in the questionnaires and will be asked to complete these within a specified period. Those who miss the completion will be reminded by email or telephone. PwMS who withdraw from the study will be requested whether they agree to continue to complete the questionnaires related to the primary and secondary outcomes.
Data management
PwMS will enter their data in QuestBack Unipark. When the data collection is complete, it will be exported from QuestBack Unipark. The variables will be then renamed using a previously prepared codebook and initial data checks will be conducted in IBM SPSS Statistics 25 by a statistician of the coordinating study centre. Pseudonymised data will be stored at the Institute of Neuroimmunology and Multiple Sclerosis at the University Medical Center Hamburg-Eppendorf. The analysis and usage of the data by the study principal investigator and his staff is done in pseudonymised form. The data collected during the study will only be made available for the scientific community in anonymised form. The same applies to the publication of the study results.
Confidentiality
All data collected during the study will be kept strictly confidential. All electronic data from the pilot RCT will be pseudonymised and there will be no possibility to link data to persons without access to the code list. Through the personal access of patients to the PExMS-website or DECIMS-Wiki based on a personal account (sent via email), it will be possible to individually track the use of the websites (e.g. frequency, use of different parts), which will be used to analyse user behaviour and the feasibility of the study design. User's IP address will be stored by the system but shielded to anyone but the system administration. A personal account name will be used to connect users and pseudonyms.
Analysis
The feasibility and process evaluation findings will inform the design of a future fully powered, randomised controlled trial to formally evaluate the effectiveness of the intervention. Therefore, we seek a drop-out rate < 20 % and to collect > 60% of patient-reported outcomes (questionnaire response rate). Frequencies will be determined for categorical variables and summary statistics (mean and standard deviation or median and quartiles, as appropriate) will be calculated for quantitative variables. Exploratory formal hypothesis testing using ANCOVA models will be performed since this is a pilot RCT. Estimated differences in outcome measures of efficacy adjusted for baseline values will be calculated. No interim analyses and stopping guidelines are planned. Initially, all analyses will be performed on complete cases. If necessary, sensitivity analyses will be conducted with imputed datasets derived by multiple imputation or mean value imputation.
Qualitative data gained through interviews will be managed using MAXQDA Analytics Pro 2018 and analysed thematically following Braun and Clarke.(34)
Oversight and monitoring
Composition of the coordinating centre and trial steering committee
The project is mentored by an advisory panel of seven persons, which includes representatives of MS patients, researchers, and neurologists. It is a key component of the DIPEx research methodology. The panel is involved in interview guide development, sample selection strategies, identification of missing topics and characteristics in interview data as well as in helping with arising issues. The panel will also review the materials prepared for the PExMS-website.
PwMS are involved in content development of the intervention, reported previously. Before uploading the multimedia files on the PExMS-website, participants will be given transcriptions of the media and asked again for permission for uploading. The preliminary version of the PExMS-website will be evaluated by pwMS. Dissemination of results to participants will be conducted via letters/newsletters and websites of MS self-help organisations.
Adverse event reporting and harms
There is no guarantee that MS patients will personally benefit from this study. Studies investigating PEx as an intervention found no adverse events.(14, 22, 35) Since we do not expect relevant adverse events, no reporting is planned. Nevertheless, HADS as a control parameter will be applied to control for anxiety and depression.
Plans for communicating important protocol amendments to relevant parties
Important protocol modifications will be communicated to the ethical committee, the advisory panel as well as to the trial registry and principal investigators.
Dissemination plans
The results of our study will be presented at national and international conferences and meetings and published in peer-reviewed journals.