Use of validated, up-to-date core outcome sets with good measurement properties is essential for conducting clinical trials with the minimum of unnecessary duplication of research effort, time, and resources. However, development of core outcome sets can be a long process, where inefficiencies can occur when multiple Working Groups replicate the same work for similar rheumatic diseases. Developing an OMERACT core outcome set requires two essential and sequential components: deciding what to measure (core domain set) and then deciding how to measure each of the endorsed domains (core outcome measurement set). The word ‘domain’ is similar to the words ‘concept’, ‘attribute’, and ‘construct’ that have also been used in the literature to describe those things that are being measured [1]. Potential domains are generated through scoping reviews and qualitative work, and then a consensus process determines those that are deemed the highest priority to measure in clinical trials and longitudinal observational studies. A consensus process requires all those participating to share a common understanding of the domain under discussion. This requires a clear definition of a particular or target domain. By target domain we mean the specific concept that will be measured in a clinical trial and as part of a core outcome set. We use the term ‘target domain’ to distinguish this specific concept from one which is less specific – what we term a ‘broad domain’. For example, within the broad domain of pain, different target domains such as pain intensity or pain interference can be measured and therefore, they need to be clearly defined [2]. The target domain definition should be as clear as possible, yet not simple (i.e., it must not be vague or unclear [1,[3], [4]].

OMERACT has developed an explicit method to describe target domains using a detailed, stepwise approach. The target domain is defined by detailing the breadth and depth of the important elements that are essential to capture to measure the target domain [2]. This work resulted in the development of a detailed definition for each domain that is recorded on the OMERACT Domain Definition report, which OMERACT working groups complete for each of their core domains [1]. The detailed target domain definition recorded on this report then becomes the “gold standard” for the critical first step in the OMERACT instrument selection process of what an instrument should capture when the concept match and content validity is evaluated in the first stage of the instrument selection process [5,6]. In this paper when we use the term ‘domain’ we mean the target domain.

Guidance from organizations involved in core outcome set development suggest that sufficient detail about the definition of the domain is needed to communicate clearly what is being measured [7,8]. Experience within OMERACT Working Groups has shown that creating and agreeing on detailed domain definitions is a challenging and time-consuming task that can present a significant barrier to making progress towards core outcome set development. Core outcome set developers are often faced with many domains, some shared across diseases. A systematic review of core outcome set development studies [9] highlighted the absence of standardized definitions for domains. This is a problem for two reasons: 1) it introduces variability in how domains are defined across different core outcome sets, and 2) it hinders efforts to evaluate and select instruments to match a domain ([10,11]).

Within existing OMERACT core outcome sets, certain patient-reported domains were found to be common across the different rheumatic diseases [15]. Previous work had reviewed and listed all domains that are part of published OMERACT core domain sets [12], and we updated the list to capture recently endorsed core domain sets. From this list, we identified the top five most reported patient-reported life impact domains: pain, physical function, fatigue, patient global assessment, and health-related quality of life. At a prior OMERACT conference in May 2014, both pain intensity and pain interference were deemed important constructs to be measured in clinical trials of chronic pain in rheumatological conditions [13]. Therefore, we decided to specify pain intensity and pain interference as two distinct domains needing detailed definitions. We termed the six domains ‘foundational domains’ due to their consistent importance across multiple OMERACT core domain sets of different rheumatological conditions. Various definitions of these foundational domains were used in the existing core domain sets.

To address the challenge of the significant amount of time and resources needed to develop a core domain set, OMERACT aims to provide working groups with the option of selecting from a set of foundational detailed target domains that have been endorsed by the OMERACT community. Working groups can review what they have found when generating domains through literature searches and qualitative work and decide whether the OMERACT-recommended definition for these foundational domains meets their needs. It is our intent that sharing these definitions and having them widely available will accelerate core outcome set development.