Development of the ‘COuld it Be RA’ (COBRA) tool to facilitate early identification of people at risk of developing rheumatoid arthritis in primary care

Objectives We aimed to develop a new complex intervention, the ‘COuld it Be RA’ (COBRA) tool, to support the implementation of a clinical prediction model to identify people likely to be anti-cyclic citrullinated peptide (CCP) positive and at risk of rheumatoid arthritis in primary care. Methods The COBRA tool was developed using the UK Medical Research Council and National Institute for Health and Care Research complex intervention research framework. This study involved three sequential phases with primary care clinicians: a qualitative descriptive study, clinician consultation engagement workshops and a think-aloud interview study. Ethical approval was obtained for all three phases. Results Sixteen primary care clinicians participated in semistructured interviews to identify barriers and facilitators. An initial list of nine candidate components for the intervention, including design considerations, was developed. During phase 2 workshops with eight participants, four components were prioritised as ‘Must have’ or ‘Should have’: the clinical decision support system (CDSS); guidance on using the CDSS/associated actions; evidence for the CDSS; patient education resources. A COBRA tool prototype incorporating these components was developed. Twelve participants tested the prototype during think-aloud interviews. Key perceived benefits of the COBRA tool included supporting clinicians’ decision-making and reducing unnecessary anti-CCP testing. Over 40 changes were made to the COBRA tool. Conclusion Our research included the views of clinicians and PPI representatives and was underpinned by a complex intervention research framework. This was critical to understanding barriers and facilitators to implementing the clinical prediction model in primary care and developing the COBRA tool.