BackgroundDespite the broad popularity of a numeric rating scale (NRS) its psychometric properties are not well known. The objective was to determine if there is any difference in the discrimination ability of the NRS when used for measuring pain severity separately in different body regions.MethodsCross-sectional survey study of 630 professional musicians. Item Response Theory (IRT) was used to define the psychometric properties of the NRS.ResultsThe discrimination ability of the pain NRS was dependent on the body area to which it was applied. The discrimination was low 0.5 (95% CI 0.4. to 0.7) for the hand region and perfect for the shoulder and upper part of the neck– 3.2 (95% CI 1.2 to 5.2) and 10.5 (95% CI 10.0 to 10.9), respectively. Both shoulder and neck NRSs showed a great shift towards higher levels of pain severity meaning that the ability of the NRS to discriminate low levels of pain is poor. NRS scores obtained from all other regions did not demonstrate any discrimination ability.ConclusionsThe pain NRS might have different psychometric properties depending on the body area to which it is applied. Overall, the modest discrimination ability of the pain NRS implies that it should be used in screening questionnaires with some reservations.