A COMPARISON OF THE CONSTANT SHOULDER SCORE AND OXFORD SHOULDER SCORE IN PATIENTS WITH CONSERVATIVELY TREATED PROXIMAL HUMERAL FRACTURES

Abstract

Introduction: Scoring systems for assessment of shoulder function are invaluable tools in determining changes in a patient’s condition. We utilised two commonly used assessment tools in patients with conservatively treated proximal humeral fractures to establish their behaviour in this patient group.

Methods: OSS and Constant Scores were collected prospectively at 3 and 12 months post injury, for 103 consecutive patients treated conservatively for proximal humeral fractures. Comparison of the scores was undertaken by creating scattergraphs, calculating Pearson’s correlation coefficient and producing Bland and Altman plots. Sensitivity to change was calculated using paired t-tests. Linear regression analysis was finally performed to predict Constant Score from the OSS.

Results: 177 sets of scores were collected. The scores correlated well with a correlation coefficient (r) of 0.84 (p< 0.001,n=177). This relationship was equally strong at 3 (r=0.77 (p< 0.001,n=94)) and 12 months (r=0.87 (p< 0.001,n=83)) and demonstrated a clear relationship between the scoring systems. Bland and Altman plots showed good agreement between the scores. Both scores were sensitive to change over time (OSS (t(81)=6.14,p< 0.001), Constant (t(80)=−10.27,p< 0.001)). Regression analysis produced a regression equation (R2=0.70) of: Constant Score=99.3-(1.67 times OSS). This level of model fit was statistically significant (F(1,175)=412.8,(p< 0.001))

Conclusion: This study provides information about the behaviour of two frequently used functional scoring systems in patients with proximal humeral fractures. Based on our finding we feel that these scores are appropriate assessment tools in these patients. The OSS may also be considered as an alternative for assessing longer term follow up as, being solely subjective, it is easier to administer and correlates well with the Constant Score.