Thoracic movement screening in adults with cystic fibrosis: reliability of the Manchester musculoskeletal screening tool

Hodgson, N, Taylor, J, Ashbrook, J, Goodwin, P ORCID: https://orcid.org/0000-0001-6533-0949, Bright-Thomas, R and Caunt, J (2022) Thoracic movement screening in adults with cystic fibrosis: reliability of the Manchester musculoskeletal screening tool. Physiotherapy Theory and Practice: an international journal of physical therapy, 38 (12). pp. 2195-2201. ISSN 0959-3985

Preview

Accepted Version Download (285kB) | Preview

Abstract

Objectives: The Manchester Musculoskeletal Screening Tool (MMST) is used internationally to screen for pain, postural changes, and urinary incontinence in adults with cystic fibrosis (CF). The tool has been validated for the outcome measures of pain and incontinence but not for the thoracic movement section. The aim of this study was to assess intra (single rater) and inter-rater (between rater) reliability of the thoracic movement screen section of the MMST. Methods: This is a prospective reliability study. Digital videos of thoracic movement were taken of adults with CF during their annual musculoskeletal screening at a large UK Adult CF Center. Twelve physiotherapists independently watched the videos and scored the movements on two occasions, 2 weeks apart, using MMST. Cohen’s kappa and Krippendorff alpha were used to establish intra- and inter-rater reliability. Results: Intra-rater reliability using Cohen’s kappa calculation ranged between 0.35 and 0.93. Eleven out of 12 physiotherapists had a moderate-substantial reliability score as assessed by the Landis Koch criteria. Percentage agreement for each physiotherapist ranged from 67%-97%. The inter-rater reliability was poor (Krippendorff alpha score = 0.422 (CI: 0.24–0.60)). Conclusion: The thoracic section of the MMST is reliable in adults with CF to highlight changes in posture and thoracic mobility that may go undetected or under-reported by the patient when repeated by the same clinician. However, the inter-rater variability is high, and it should not be considered reliable when carried out by different clinicians over time.