BackgroundImpaired patient-reported shoulder function and pain, external-rotation strength, abduction strength, and abduction range-of-motion (ROM) is reported in patients with subacromial impingement (SIS). However, it is unknown how much strength and ROM improves in real-life practice settings with current care. Furthermore, outcomes of treatment might depend on specific rehabilitation parameters, such as the time spent on exercises (exercise-time), number of physiotherapy sessions (physio-sessions) and number of corticosteroid injections, respectively. However, this has not previously been investigated. The purpose of this study was to describe changes in shoulder strength, ROM, patient-reported function and pain, in real-life practice settings, and explore the association between changes in clinical core outcomes and specific rehabilitation parameters.MethodsPatients diagnosed with SIS at initial assessment at an outpatient hospital clinic using predefined criteria’s, who had not undergone surgery after 6 months, were included in this prospective cohort study. After initial assessment (baseline), all patients underwent treatment as usual, with no interference from the investigators. The outcomes Shoulder Pain and Disability Index (SPADI:0–100), average pain (NRS:0–10), external rotation strength, abduction strength and abduction ROM, pain during each test (NRS:0–10), were collected at baseline and at six month follow-up. Amount of exercise-time, physio-sessions and steroid-injections was recorded at follow-up. Changes in outcomes were analyzed using Wilcoxon Signed-Rank test, and the corresponding effect sizes (ES) were estimated. The associations between changes in outcomes and rehabilitation parameters were explored using multiple regression analyses.ResultsSixty-three patients completed both baseline and follow-up testing. Significant improvements were seen in SPADI (19 points, ES:0.53, p < 0.001) and all pain variables (median 1–1.5 points, ES:0.26–0.39, p < 0.01), but not in strength and ROM (ES:0.9–0.12, p > 0.2). A higher number of physio-sessions was significantly associated with larger improvements in external rotation strength (0.7 Newton/session, p = 0.046), and higher exercise-time was significantly associated with decrease in average pain (−0.2 points/1,000 min, p = 0.048).DiscussionPatient-reported function and pain improved after six months of current care, but strength and ROM did not improve. This is interesting, as strengthening exercises is part of most current interventions. While two significant associations were identified between self-reported rehabilitation parameters and outcomes, the small gains per physio-session or 1,000 min of exercise-time reduces the clinical relevance of these relationships. Collectively, the findings from this study indicate room for improvement of the current rehabilitation of SIS, especially with regard to core clinical outcomes, such as strength and range of motion.