Objectives: This study aimed to identify phenotypes of people with frozen shoulder (FS) based on baseline measures including pain intensity, disability, pain-related psychological factors, and measures of central pain processing. A secondary objective was to evaluate whether these baseline phenotypes were associated with distinct longitudinal trajectories of pain intensity and disability over 9 months.
Methods: People with FS were assessed at baseline for clinical (pain intensity and disability) and psychophysical variables (pain catastrophizing and hypervigilance, self-reported symptoms related to central sensitization, primary and secondary allodynia and hyperalgesia, and pain modulation). A latent profile analysis was conducted using baseline clinical and psychophysical variables. Linear mixed models were then used to examine longitudinal trajectories of pain intensity and disability across phenotypes over 3, 6, and 9 months.
Results: A total of 149 people with FS were recruited. Two distinct phenotypes were identified, of which phenotype 2 (n=70) presented higher intensity of pain, greater disability, and higher scores for pain catastrophizing, hypervigilance, and central sensitization-related symptoms at baseline compared to phenotype 1 (n=79). Over 9 months, trajectories differed by phenotype. Phenotype 2 exhibited a greater absolute reduction in both pain and disability, yet phenotype 1 remained lower at all time points, including 9 months, for both outcomes.
Discussion: This study highlights two distinct phenotypes in people with FS based on baseline clinical and psychophysical variables. These phenotypes showed distinct trajectories of pain and disability over 9 months, supporting the relevance of baseline phenotyping for understanding clinical heterogeneity in FS.
Keywords: central pain processing; chronic pain; frozen shoulder; latent profile analysis; pain-related psychological factors; phenotyping.
Copyright © 2026 Wolters Kluwer Health, Inc. All rights reserved.