Background Development of a minimum core dataset (MCD) for a specific disease presents an invaluable way of collecting data in a consistent manner, allowing for clear definitions for individual items and standardised tools for data collection.
Objectives To outline the process in developing a national MCD for observational cohort studies in rheumatoid arthritis (RA) that could be adapted for other rheumatic conditions.
Methods The process involved three main stages, centering on two multi-stakeholder meetings and a teleconference (TC). Individuals with diverse ranges of expertise and backgrounds participated, including clinical and academic rheumatology, epidemiology, stratified medicine, health economics, RA patients, OMERACT, BSR, ARUK and Clinical Studies Group (CSG) for RA. Using the OMERACT Filter 2.0 methodology aims and objectives, definition of scope, identification of important and feasible research questions and selection of key domains were discussed. Two item-specific workshops were held in parallel with the meetings.
Results Stage 1 of this process focused on establishing an important research question of a general nature in RA which could be answered by a MCD alone. 31 items were identified as potential candidates for a MCD, based on four key OMERACT-recommended domains: Life Impact (e.g. QoL, pain, function,); Pathophysiological manifestations (e.g. Disease Activity); Resource use (e.g. drug use, joint surgery); Date of death. 14 items achieved consensus as essential and adequate for achieving meaningful long term outcome data for simple research questions. Stage 2 involved primarily email communications. All MCD members were asked to vote on the relative importance of the items which were considered essential but lacked consensus, and the most appropriate measurement instruments for inclusion. Comments were collated and recirculated to MCD members. Consensus was reached on 20 items, including demographics, and at least one item in each of the four key domains. The group took into account the needs of clinical engagement with both general hospital and academic settings, and patient perspectives. In stage 3, a TC meeting agreed on the essential items and most appropriate measurement tools, and also identified items suitable for add-on studies. For three essential items, consensus was lacking for a measurement tool: participation/work/social, fatigue and comorbidity. For fatigue and comorbidity, two dedicated workshops were undertaken during this process, both of which recognised the importance of these items and identified lack of consensus on their measurement tools. Further work is currently in progress, including an online voting survey to agree on the most suitable tools for data collection on these two items.
Conclusions The development of a MCD for observational studies has involved three main stages and two workshops, and achieved consensus on 20 essential items. Further work is required to agree on measurement tools for fatigue, comorbidity and participation/work/social. Novel challenges and opportunities were identified as part of this process, which could be adapted for MCDs in other disease areas and to answer more complex questions at both national and international level. Paradigms taken from interventional clinical trials may need to be re-evaluated in light of this.
Disclosure of Interest : None declared