Purpose: To determine the changes in each of the five dimensions of the EuroQol 5-dimension index associated with community-based physiotherapy. Methods: Four thousand one hundred and thirty-six patients that received community-based musculoskeletal physiotherapy across five NHS centres completed the EQ-5D on entry into the service and upon discharge. Patients were categorised on symptom location and response to treatment based on their EQ-5D index improving by at least 0.1 (“EQ-5D responders”). For each symptom location, and for responders and non-responders to treatment, the mean (± SD) were calculated for each dimension pre- and post-treatment as well as the size of effect. Results: The mobility dimension improved (p < 0.05) in all symptom locations for EQ-5D responders (d = 0.26–1.58) and in ankle, knee, hip and lumbar symptoms for EQ-5D non-responders (d = 0.17–0.45). The self-care dimension improved (p < 0.05) in all symptom locations for EQ-5D responders (d = 0.49–1.16). The usual activities dimension improved (p < 0.05) across all symptom locations for EQ-5D responders (d = 1.00–1.75) and EQ-5D non-responders (d = 0.14–0.60). Despite the pain/discomfort dimension improving (p < 0.05) across all symptom locations for both EQ-5D responders (d = 1.07–1.43) and EQ-5D non-responders (d = 0.29–0.66), the anxiety/depression dimension improved (p < 0.05) from higher starting levels in EQ-5D responders (d = 0.76–1.05) with no change seen for EQ-5D non-responders (d = − 0.16 to 0.06). Conclusions: Clinicians should not assume that a patient presenting with pain but expressing high anxiety/depression is unlikely to respond to treatment, as they may show the best HRQoL outcomes. For patients presenting with pain/discomfort and low levels of anxiety/depression, the EQ-5D index is perhaps not a suitable tool for sole use in patient management and service evaluation.