The sensitivity and specificity of pain diagrams in rheumatic disease referrals

To determine patterns on pain diagrams and corresponding diagnoses in patients referred to a rheumatology clinic and their sensitivity, specificity, and positive and negative predictive values (PPV and NPV, respectively).

All new adult patient referrals from two rheumatologists over 6 years were reviewed and eligible if a pain diagram had been completed and they were not previously diagnosed with a rheumatic disease. Pain diagrams completed by the patient were organized into patterns based on the location of joint and/or soft tissue areas by two independent observers.

A total of 1101 patients were included. Five major patterns evolved: soft tissue (widespread pain or regional pain such as an entire arm) (n = 236), symmetrical articular (n = 647), asymmetrical articular (n = 136), monoarticular (n = 35) and back (n = 46); 480 had inflammatory arthritis, of whom 121 had RA, 35 PsA, 46 SpAs and 63 crystal arthropathy. FM or chronic pain also occurred in 63 and 25 had PMR. The sensitivity, specificity, PPV and NPV for polyarticular symmetrical pattern in RA was 87.6, 44.7, 16.4 and 96.7% and for inflammatory arthritis with symmetrical or asymmetrical pattern was 82.3, 37.4, 50.4 and 73.2%; and a back pattern in AS was 10.9, 98.9, 41.7 and 96.2%. Inter-rater reliability was high for monoarticular, symmetrical and asymmetrical patterns (intra-class correlation coefficient ≥ 0.777).

Pain diagram patterns may help to increase the likelihood of various rheumatic diagnoses including polyarticular pattern and inflammatory arthritis, and there was high inter-rater reliability. However, testing the value of pain diagrams in addition to a referral note is necessary to determine if they have added value.