Design: Retrospective single-centre cohort
Population: 72 patients (74 shoulders) after locking plate fixation of displaced proximal humerus fractures
Primary outcome: Agreement between Constant-Murley score and Subjective Shoulder Value
Effect: Mean difference −14.2 points (95% CI −17.3 to −11.2); pain the only independent predictor of poor self-rating

An Indian cohort assessed 74 shoulders at least a year after locking plate fixation of displaced proximal humerus fractures, comparing the clinician-derived Constant-Murley score with the patient's own Subjective Shoulder Value.

Patients rated their shoulders on average 14 points lower than the Constant score (95% CI −17.3 to −11.2). Only pain was independently associated with a poor self-rating; range of motion and strength were not. In shoulders without avascular necrosis, malreduction and a missing medial hinge were associated with more pain.

This is a small, single-centre study, but it makes a useful point: a respectable Constant score can hide a dissatisfied patient, and pain is what patients weigh most.

Record a patient-rated score alongside the Constant score at follow-up.
Ask about pain specifically, not only about movement.
Aim to restore the medial hinge and an anatomical reduction at fixation.
Investigate persistent pain for malreduction, screw cut-out or avascular necrosis.

Why it matters

Surgeon-rated success after fracture fixation can overstate how the patient feels.

The statistics, in plain English

The 14-point gap means the two scores measure different things, not that one is wrong. The associations with malreduction and the medial hinge come from simple two-way comparisons without adjustment and are exploratory.