The influence of age and unreported symptoms on the Oxford Shoulder Score.

Abstract

One potential limitation of interpreting the Oxford Shoulder Score (OSS) in longitudinal studies is that the observed score may be influenced by age and other variables, which may change over time. The purpose of the present study was to investigate the influence of increasing age and unreported non-shoulder upper limb and neck symptoms on the OSS. We collected OSS data from a sample of our ethnically diverse local population. All subjects indicated whether they suffered from any neck, shoulder, elbow or wrist symptoms for which they had not sought a medical opinion. Those reporting no symptoms formed the asymptomatic group. We found a significant decline in OSS with increasing age in the whole study population, as well as in both the asymptomatic and symptomatic groups with previously unreported symptoms: Spearman correlation coefficient = -0.27, -0.28 and -0.33 respectively (p < 0.001). The median OSS in the asymptomatic group was 48 [interquartile range (IQR) 48 to 48]. This was significantly higher than the symptomatic group, with a median OSS of 46 (IQR 40 to 47) (p < 0.001). We found the OSS to be affected by non-shoulder upper-limb and neck pathology as well as age. Within the limitations of the OSS, the differences we found do not exceed the minimal important change.