Normalization Methods for Shoulder Strength in Baseball Athletes

Abstract

Background:

Arm injuries in baseball continue to rise. Previous studies have confirmed relationships between shoulder strength, injury risk, and performance in baseball athletes. Shoulder strength measures vary considerably within and among baseball players and are influenced heavily by anthropometric measures. Presently, the normalization of shoulder strength measures has not been tested adequately to meet necessary statistical assumptions.

Hypothesis:

At least 1 shoulder strength normalization method would meet 2 statistical assumptions: the intercept assumption and correlation assumption.

Study Design:

Cross-sectional study.

Level of Evidence:

Level 3.

Methods:

Isometric shoulder strength was tested in a cohort of uninjured baseball athletes on the dominant limb via handheld dynamometry: external rotation (ER), internal rotation (IR), and scapular plane abduction (SCAP). Five normalization methods were assessed using 2 statistical assumptions to determine the most appropriate normalization method for each shoulder strength measure: first, the regression line between a participant characteristic (body mass, height, ulnar length, and their respective combinations) and the unnormalized strength measure passes through the origin; second, normalizing shoulder strength eliminates its correlation with the participant characteristics. Statistical significance was set a priori at α < 0.05.

Results:

Body mass normalization was the only normalization method that satisfied the intercept and correlation assumptions for ER, IR, and SCAP strength (P > .05).

Conclusion:

Normalizing shoulder ER, IR, and SCAP strength in uninjured baseball athletes using body mass satisfied both statistical assumptions. When clinicians seek to compare shoulder strength in baseball athletes of differing anthropometric make-up, using body mass as the denominator is the most appropriate method for normalization.

Clinical Relevance:

The utilization of a statistically supported shoulder strength normalization method is warranted to account for anthropometric differences when comparing strength between athletes. This study adds a clinically relevant and reproducible method for normalizing isometric shoulder strength in uninjured baseball players.