Abstract Background Musculoskeletal disorders (MSDs) are among the most common pathologies in the general population. However, research into the prevalence of upper arm MSDs is hampered by a lack of uniformity in case definition, and by the absence of a gold standard for measurement. Furthermore, some sectors of the population have benefited from extensive research whilst others have largely been ignored. Study Design: Cross-sectional study. Objectives: to investigate the prevalence of shoulder MSDs in a working age female population not exposed to specific occupational risk factors such as heavy and/or repetitive work, assessing the differences in prevalence recorded by using three different standard measurement tools. Methods 302 working aged women were enrolled in this study (age 20–55 years). Each subject underwent three different assessments: standardized questionnaires for symptoms and disability and the SF36 health survey, a clinical assessment performed by a blinded orthopaedic specialist, and an imaging assessment by means of ultrasound (US) and Magnetic Resonance (MR) if indicated. Results According to the questionnaire 77 subjects (25.5%) complained of shoulder pain whilst 225 (74.5%) were asymptomatic. According to the clinical examination, 31 subjects (10.3%) resulted positive, whereas 271 subjects (89.7%) had normal shoulders. According to the imaging findings, 26 subjects (8.6%) had alterations to the anatomical structures of the shoulder, whilst 276 subjects (91.4%) had no detectable abnormalities in either shoulder. In all assessments, the prevalence increased with age (p = 0.001). Conclusion Depending on the outcome measure used, the prevalence of reported MSDs of the shoulder varies considerably. There is a striking difference between the prevalence of subjective reported symptoms and the standardized clinical/imaging examinations. However, the results of all the assessments did concur in one aspect; there was a significant trend of increased prevalence of shoulder MSDs with age. When looking at reported prevalence, this study shows the importance of noting the measurement method used before making comparisons, as it can vary considerably. The epidemic of shoulder pain reported is not indicative of an epidemic of shoulder pathology.