Objective: To investigate and compare clinical features and US signs of inflammation in joints and entheses in patients with psoriasis with and without musculoskeletal pain, and the additional value of US in classification of PsA. Furthermore, to explore the association between such findings and patient-reported outcomes (PROs) and the performance of screening-questionnaires for identifying patients with PsA. Methods: Patients with psoriasis (n = 126) recruited from a nationwide survey were evaluated at one of four rheumatology departments. The evaluation included clinical examination, laboratory tests, radiography, greyscale and colour Doppler US of 48 joints and 12 entheses, PROs, and four screening questionnaires for PsA. Patients were classified with Classification for PsA (CASPAR), US-modified CASPAR, and US-only criteria. Results: When subgroups of self-reported pain (63%), no pain (29%) and diagnosed PsA (9%) were compared, patients with pain had higher tenderness-related clinical scores (tender joints, entheses and FM points) and US greyscale sum-scores, compared with 'no pain' patients. PROs were negligibly moderately correlated with pain-related clinical scores (Spearman's rho = 0.11-0.59, all patients), and negligibly weakly with US sum-scores (rho = 0.01-0.34). More patients could be classified as PsA when US synovitis/enthesitis was included as an entry criterion (US-modified CASPAR, 66% of all patients) compared with conventional CASPAR (35%) or US-only criteria (52%). Sensitivities of screening questionnaires were low for fulfilment of CASPAR (0.23-0.66), US-modified CASPAR (0.17-0.57), and US-only (0.20-0.57) criteria. Conclusion: Self-reported pain in psoriasis is related to US inflammation. US-modified CASPAR criteria identified almost twice as many patients as conventional CASPAR criteria. Screening questionnaires showed limited value.