Comparability of Patient-reported Outcome Measures and Clinical Assessment Tools for Shoulder Function in Patients with Proximal Humeral Fracture.

Abstract

Patient-reported outcome measures (PROMs) are gaining increasing importance in the context of quality management. Different PROMs and scoring tools are available to assess shoulder function after proximal humeral fracture (PHFx). In Europe, these include the Constant-Murley Score (CS), Neer Score (NS), Oxford Shoulder Score (OSS), University of California at Los Angeles Score (UCLA) and Disabilities of the Arm, Shoulder and Hands Score (DASH). In addition, the health-related quality of life can be assessed by the PROMs Short Form 36 (SF-36) and EuroQol (EQ-5D). Although all these test instruments assess shoulder function, the components to be answered objectively and subjectively vary and thus the possibility of independent assessment. The aim of the present study is to compare the correlation between the results of the different PROMs and the clinical screening tools in patients with PHFx. 76 patients who were treated with angular stable plate osteosynthesis for a proximal humeral fracture between 01/2001 and 12/2005 were included in this trial. The outcome was measured with PROMs or clinical scoring tools such as CS, NS, OSS, UCLA, DASH, SF-36 and EQ-5D and a correlation coefficient between those evaluation tools was calculated. In addition, a distinction was made between the two force measurement methods (wrist [HG] vs. deltoid muscle [DM]) for CS. The correlation of the results of CS and NS (HG: r = 0.85; p < 0.001/DM: r = 0.93; p < 0.001), CS and UCLA (HG: r = 0.83; p < 0.001/DM: r = 0.86; p < 0.001), NS and UCLA (r = 0.91; p < 0.001) as well as DASH and OSS (r = 0.88; p < 0.001) was strongly expressed. A good comparability of the results was demonstrated between CS and OSS (HG: r = 0.63; p < 0.001/DM: r = 0.66; p < 0.001) and between CS and DASH (HG: r = 0.62; p < 0.001/DM: r = 0.61; p < 0.001). The correlation of CS (HG/DM) and UCLA with the EQ-5D index was also good. Assessment of the physical components of SF-36 with CS, NS, OSS, UCLA and DASH showed a moderate to good association, while the mental components of SF-36 showed a low correlation (p > 0.05). The assessment of shoulder function after proximal humerus fracture showed a very strong correlation within the clinical questionnaires (CS/NS/UCLA) and the PROMs (OSS/DASH). A strong correlation also exists between the clinical questionnaires and the PROMs. There was only a moderate correlation with the EQ-5D. The moderate to strong correlation between the physical components of SF-36, with almost no correlation between the mental components of SF-36, indicates that the quality of life restriction is based on a physical, but not on a mental impairment.