Work Packages details
Academy Strategic Focus Area I: Raise Awareness about the importance of research using real world MS data
We believe the perspectives of people with MS are central to establishing a trustworthy ecosystem for reusing health data for research and for learning health systems. We aim to launch a multi-faceted motivation campaign to engage people with MS in decisions about the use of their data. We collaborate closely with the European Patient Form and their “Data Saves Lives” initiative.
An informed and engaged patient community will lead to:
- increased compliance to data collection procedures (more specifically when it comes to patient reported outcome measures)
- increased patient-driven advocacy for the inclusion of patient relevant real world data in (regulatory) decision making.
Academy Strategic Focus Area II: Build a multi-stakeholder MS Data Community: Stakeholder engagement meeting
We believe a multi-stakeholder synergic approach is required to jointly tackle the socio-technical challenges related to scaling-up real world MS data. Therefore, we aim to regularly bring together all relevant stakeholders involved during a multi-stakeholder engagement meeting. During 2019, this meeting will be organised the 20th of November 2019 in Baveno (Italy).
The primary objectives of this first stakeholder meeting are to:
- inform all stakeholders on the MSDA initiative, partners and strategy
- discuss challenges and opportunities of the project
- provide a platform to inform all stakeholders on the value of real world MS data
- motivate collaborative research by sharing testimonials and data sharing success stories in MS and other diseases.
Academy Strategic Focus Area III: Promote trustworthy and transparent practices in the use of real world MS data:
We aim to support greater understanding within the community about reuse of real-world data through MSDA educational sessions. Examples of educational topics are: Data Protection, General Data Protection Regulation (GDPR), legal basis and the appropriate role of consent, Data Quality assessment and improvement, Interoperability standards and common data models, Federated architectures, distributed querying, information security, …
Toolbox Strategic Focus Area I: Cataloguing and publishing descriptions of data sources:
Improved awareness of existing and planned cohorts and registries is needed. We developed a web-based catalogue that provides this strategic oversight and allows end-users to browse metadata of MS data cohorts and registries. Our first version of the MSDA Catalogue digitalizes the questionnaire used in the most recent European Mapping Exercise. It is hosted by the European Medical Information Framework Catalogue and will be a moderated community besides many other disease specific communities (incl. Alzheimer Disease, Rare Diseases, …). We encourage you to register for the EMIF-catalogue and request access to the MSDA community.
Following data is collected (list is not exhaustive):
- Organisational information (Name, Custodian information, MS society involvement, contact details, funding info)
- Background/Purpose (Start year, purpose, target population, sources of data collection)
- Inclusion criteria (patient inclusion criteria, centre inclusion criteria)
- Documentation Process (who collects data? How? Identity system? Language?)
- List of variables (personal data, basic disease data, relapses, disability, cognition scales, treatments, MRI, Paraclinical measures, Patient-derived measures, depression, fatigue, co-morbidities, socio-economic data, societal services, healthcare utilization)
- Quality control (quality control mechanisms, data coverage, representativeness, trigger data entry)
- Governance (Approval, informed consent, access to data)
- Current state of registry (number of patients, number of visits)
Toolbox Strategic Focus Area II: Develop a standardised MS data dictionary and support local harmonisation efforts
Observational databases differ in both purpose and design. Each database is situated in different organizational settings and uses different terminologies or parameter denotations for describing their content (parameters/variables). Here, we aim to maximize reusability of MS registry/cohort data by harmonizing structure and variables by developing promoting the adoption of a “common data model”and developing a standardised data dictionary of commonly used MS data concepts. A common data model (CDM) aims to achieve both syntactic* and semantic interoperability**. We built on top of the “OMOP CDM”. OMOP CDM is the common data model, originating from the Observational Medical Outcome Partnership (OMOP), which is now being implemented and updated by the OHDSI (Observational Health Data Sciences and Informatics) community.It is being used within the two large Innovative Medicine Initiative (European Medical Information Framework (EMIF)/European Health Data and Evidence Network (EHDEN))) projects which are the models for the MSDA infrastructure. The maturity, the scalability of and the community behind the OMOP CDM makes it the perfect candidate for the CDM. We aim to facilitate and support the local mapping by data sources to this CDM.
*Syntactic interoperability represents a common, agreed upon data structure. -> Grammar
**Semantic interoperability represents the common understanding of the syntactic interoperable data structure. -> Vocabulary
The following opportunities of the SwitchBox are formulated:
- Reduce time and costs for harmonisation efforts: Indeed, we aim to support multiple MS collaborative research projects, prospective as well as retrospective data collections. Therefore, a particular data source should harmonize their data only once
- Increased and guaranteed quality of the data (quality checks are included in the mapping procedure)
- Our harmonisation strategy is reproducible & transparent. Indeed, detailed information of strategy is documented and publicly available
- Multi-language: It remains possible to keep using the local vocabulary for local use
- Flexible & Scalable: our CDM can grow when the list of variables grows
Toolbox Strategic Focus Area III: Develop a federated data network to allow group level queries on harmonized datasets:
With the MSDA cohort explorer, we aim to allow group level queries on harmonized datasets from multiple MS registries & cohorts to facilitate and support collaborative research. We envision to develop and implement a “federated data network”. Easily put, a federated data network allows local querying of different cohorts or registries (=data does not leave the local storage unit), but the aggregated data is combined in a central platform.
A federated data network has several advantages:
- Data remains under the control of the data owner
- Locally required legal and ethical approvals apply
- No patient level data leaves the owner’s site, only aggregated counts, thereby ensuring patient privacy
- GDPR – ‘Privacy by Design’
- Analysis is “brought to the data” rather than creating central data repository
- Use of common data model allows for efficient search / analysis across multiple data sets
- Requires close collaboration with data owners which builds trust
We are in the process of defining the legal frame for data sharing and look forward to your input. However, some general principles are already agreed upon.
- The MSDA Cohort Explorer can only be used, for assessing the feasibility of a study and for conducting research, by bona fide research organisations and for the objective of discovering new knowledge intended for the public good and to be made publicly accessible (i.e. published)
- Data sources will always have autonomy over which data are made accessible and for which types of research, will always determine ethical acceptability and scientific validity and must be transparent about their data
- Data users must adhere to the ethical rules and privacy protection policies of each data source, may only use the data for the specific agreed research purposes and must acknowledge the sources of the data they have used.