Development and Evaluation of a Web-based Programme on Relapse Management for People With Multiple Sclerosis

Multiple sclerosis (MS) is a chronic inflammatory, degenerative disease of the central nervous system manifesting at first with relapses in about 85% of the cases. In Germany, intravenous mostly inpatient therapy with high-dose corticosteroids is the treatment standard of acute relapses. The treatment leads to short-term faster reduction of symptoms in about 25 of 100 treated patients (absolute risk reduction), but has no long-term benefits over placebo treatment. Also intravenous treatment is not superior to oral treatment. Therefore, informed decisions on relapse management are required. An earlier randomized controlled trial (RCT) showed that evidence-based patient information (EBPI) and education on relapse management leads to significantly more informed decisions and more relapses not treated or treated with oral steroids. POWER@MS2 builds on that evidence by transferring the content to a web-platform and making the intervention easily accessible and implementable. The intervention will be evaluated in a pragmatic double-blind (PwMS and outcome assessors) RCT. Participants will be randomized to an interactive web-based programme (intervention group) or a web-based programme with standard information on relapse management based on the contents of the German Multiple Sclerosis Society (DMSG). POWER@MS2 aims to demonstrate fewer glucocorticoid relapse therapies, and, in case of treated relapses, less intravenous and more orally administered therapies (primary endpoint: proportion of relapses not treated or treated with oral steroids). PwMS will be recruited via all participating centres (private neurology practices and MS outpatient departments (academic and community hospitals)) in Germany. In the federal state of Schleswig-Holstein all neurologists will be contacted. Interested and eligible PwMS will be provided with a study information sheet by the participating MS centres. Informed consent will be obtained by a physician in the MS centre. After giving their informed consent, contact data (address, email, telephone) will be forwarded to the study centre (UKE) via a secure communication platform or phone. After this, baseline data will be collected and group assignment will be performed by a block randomization procedure within the database secuTrial®. After successful randomization, PwMS will receive access (login) details to the intervention or control platform within 4-6 weeks. Participants will be monitored for at least 12 up to 36 months (on average 24 months). Every three months, measures are recorded by a standardized phone interview executed by the coordinating study centre. Recruitment will be completed after approximately nine months. The trial will end as soon as the last participant has reached 12 months of follow-up. All participants, who have not reached 36 month of follow-up, will be called for a final phone interview. We consider the specific risks for participating PwMS to be very low. No negative effects on the quality of life of PwMS as well as disability or other undesired events due to omitted or oral steroid administration are to be expected as previous studies showed. It is more likely that there will be positive effects for trained PwMS in terms of more autonomous decision-making and differentiated use of steroids. Participants will be educated on potentially dangerous side effects of steroid therapies and their early detection by an information sheet. To assist the physician in assessing whether oral medication is acceptable, participants will be issued a certificate with documented decision-making knowledge. As part of this study, all study participants will be contacted by phone every three months. This will also allow individual risk identification and the initiation of appropriate measures if required. Nonetheless, it could be possible that some participants feel harassed or pressured by the intervention or the permanent contact attempts. In order to detect possible adverse events, PwMS and physicians will be asked by questionnaires throughout the study as part of the process evaluation. As relevant adverse events are unlikely, no stopping rules will be applied. Nevertheless, safety measures are applied to control for anxiety, depression and disease specific quality of life. Furthermore, standard disease monitoring parameters will be collected (e.g. relapse rate, disability status) and discussed by the steering committee. Adverse effects will be documented in the secuTrial® system. Since the programme is accessed from home, there is little organisational and time expenditure. Study participants may leave the study at any time and may withdraw consent to study participation without negative consequences. Reasons will be asked for and, if provided, recorded. The unit of analysis for the primary endpoint is the occurrence of relapses. Eighty one relapses per group yield a power of 85% at a two-tailed significance level of 5% given proportions of 78% and 56% of orally treated or non-treated relapses in the intervention group and the control group, respectively, as observed in a previous study. It is expected that these relapse rate can be observed in a total of 170 patients with 1 to 2 years follow-up, corresponding to an annual relapse rate of 0.64. The dropout rate is expected be about 10%, as in the previous study. Therefore, 188 participants will be randomized. Data will be obtained at different time points using paper-based (primary choice) and (if requested) web-based questionnaires. First of all, informed consent and patient-related contact data will be obtained from interested PwMS in the MS centres and a copy will be send to the study centre (UKE) by regular mail. Baseline data will be obtained by participating MS centres before randomisation. From this point in time, the study centre (UKE), will take over the handling of the participants. To secure follow-up data, study participants will be contacted by the study centre via phone within the first two weeks to secure communication lines and to perform the phone interview. Student assistants will be trained on this interaction. Trial data will be collected throughout the course of the study as well as in the last follow-up in order to examine the intervention effect on the study outcomes. However, beyond 3-monthly phone interviews, major assessments will be only at baseline and after 12 months. All study relevant data will be entered in secuTrial® and provided online. PwMS and medical staff of centres as well as the central study team (defined health researchers and study assistants) can enter data. The preparation of the declaration of consent (including voluntariness) and the handling of all data collected within the scope of the study will be carried out in accordance with the recommendations of the Ethics Committee of the University of Lübeck and the EU General Data Protection Regulation (Datenschutz-Grundverordnung, DSGVO). Data protection concerns with regard to the intervention platform will be met by securing a protected web platform. The intervention programme as well as the control programme will be provided via a secure online platform that meets all legal requirements (SSL Encryption). All study data will be used and evaluated pseudonymously by the members of the coordinating centre and consortium partners involved. The publication of the study results and the provision of the data in an online resource will only take place in an anonymised form. Study participants will be informed about the results of the survey through a publication of the results on the DMSG website after completion of the study. The primary endpoint is evaluated using a generalized linear model with mixed effects and logit link function. Subject specific effects are modelled as random, whereas intervention group (IG vs. CG) and study centre are included as fixed effects in the model. The intervention effect is reported as odds ratio (OR) with 95% confidence interval (CI) and p-value testing the null hypothesis of no intervention effect (i.e. OR=1). Longitudinal assessments of quality of life and impairment are analysed by means of Gaussian linear models for repeated measures (so-called mixed model for repeated measurements (MMRM)) with intervention group (IG vs. CG), time, intervention-by-time interaction, and study centre as factors and baseline score as covariate. The error terms are assumed to follow a multivariate normal distribution with unstructured covariance. Least squares mean changes from baseline will be reported for both groups with 95% CI as well as the difference between the least squares intervention group means (IG vs. CG) with 95% CI and p-value testing the null hypothesis of no treatment effect. In subgroup and regression analyses, effects of age, gender, level of education, centre and impairment will be explored. Reasons for study withdrawal will be reported. All PwMS will be analysed in the group they were randomized to following the intention-to-treat principle. Early study discontinuations will be treated as independent right censoring in the primary analysis. In case of substantial or differential study discontinuations the validity of the independent censoring assumption will be explored in shared random effects models of the primary endpoint and time to study discontinuation. To handle missing data in baseline variables or follow-up assessments, multiple imputation models will be applied. All details of the statistical analyses including definitions of analysis populations will be prespecified in the statistical analysis plan.