%0 Journal Article
%T Improving domain definition and outcome instrument selection: Lessons learned for OMERACT from imaging.
%A D'Agostino MA
%A Beaton DE
%A Maxwell LJ
%A Cembalo SM
%A Hoens AM
%A Hofstetter C
%A Zabalan C
%A Bird P
%A Christensen R
%A de Wit M
%A Doria AS
%A Maksymowych WP
%A Oo WM
%A Østergaard M
%A Serban T
%A Sloan VS
%A Terslev L
%A van Rossum MA
%A Conaghan PG
%A Boers M
%J Semin Arthritis Rheum
%V 51
%N 5
%D 10 2021
%M 34452758
%F 5.431
%R 10.1016/j.semarthrit.2021.08.004
%X Imaging is one of the most rapidly evolving fields in medicine. Unfortunately, many imaging technologies have been applied as measurement instruments without rigorous evaluation of the evidence supporting their truth, discriminatory capability and feasibility for that context of use. The Outcome Measures in Rheumatology (OMERACT) Filter 2.1 Instrument Selection Algorithm (OFISA) is used to evaluate such evidence for use of an instrument in a research setting. The objectives of this work are to: [1] define and describe the key conceptual aspects that are essential for the evaluation of imaging as an outcome measurement instrument and [2] describe how these aspects can be assessed through OFISA.
Experts in imaging and/or methodology met to formalize concepts and define key steps. These concepts were discussed with a team of patient research partners with interest in imaging to refine technical and methodological aspects into comprehensible information. A workshop was held at OMERACT2020 and feedback was incorporated into existing OMERACT process for domain and instrument selection.
Three key lessons were identified: (1) a clear definition of the domain we want to measure is a necessary prerequisite to the selection of a good instrument, (2) the sources of variability that can directly influence the instrument should be clearly identified, (3) incorporating these first two lessons into OFISA improves the quality of every instrument selection process.
The incorporation of these lessons in the updated OMERACT Filter (now 2.2) will improve the quality of the selection process for all types of outcome measurement instruments.