Core outcome sets for research and clinical practice

Abstract

Background: This masterclass introduces the topic of core outcome sets, describing rationale and methods for developing them, and providing some examples that are relevant for clinical research and practice.

Method: A core outcome set is a minimum consensus-based set of outcomes that should be measured and reported in all clinical trials for a specific health condition and/or intervention. Issues surrounding outcome assessment, such as selective reporting and inconsistency across studies, can be addressed by the development of a core set. As suggested by key initiatives in this field (i.e. OMERACT and COMET), the development requires achieving consensus on: (1) core outcome domains and (2) core outcome measurement instruments. Different methods can be used to reach consensus, including: literature systematic reviews to inform the process, qualitative research with clinicians and patients, group discussions (e.g. nominal group technique), and structured surveys (e.g. Delphi technique). Various stakeholders should be involved in the process, with particular attention to patients.

Results and conclusions: Several COSs have been developed for musculoskeletal conditions including a longstanding one for low back pain, IMMPACT recommendations on outcomes for chronic pain, and OMERACT COSs for hip, knee and hand osteoarthritis. There is a lack of COSs for neurological, geriatric, cardio-respiratory and pediatric conditions, therefore, future research could determine the value of developing COSs for these conditions.

Keywords: Core outcome set; Effectiveness; Interventions; Musculoskeletal pain.

Figure 1

OMERACT conceptual framework of core areas for outcome measurement in the setting of clinical trials. The choice of specific domains within a core area depends on the context for which the core outcome set is developed, e.g. domains can be generic- or disease-specific, or time-specific.

Figure 2

Development process of a core outcome set from a set of core outcome domains to reach consensus on core outcome measurement instruments (i.e. core outcome measurement set). As depicted, this process allows core set developers to establish a preliminary core outcome measurement set when not all domains are covered by at least one applicable measurement instrument.