Reproducibility of Hypermobility Assessment Scales for Children When Performed Using Telehealth versus In-Person Modes

Abstract

Aims Evaluate reproducibility of hypermobility assessments using in-person versus telehealth modes. Methods Hypermobility of 20 children (7–12 years) was evaluated using the Beighton Score, Upper Limb Hypermobility Assessment Tool (ULHAT), and Lower Limb Assessment Score (LLAS) via in-person and telehealth modes. Agreement between the two modes was examined using percentage of exact agreement (%EA and %EA ± 2), Limits of Agreement (LoA) and Smallest detectable change (SDC). Reliability was calculated using intra-class correlation coefficients (ICCs). Results Agreement between modes for total Scores was best for the Beighton (%EA = fair, %EA ± 2 = good), then the ULHAT (%EA = poor, %EA ± 2 = excellent), and LLAS (%EA = poor, %EA ± 2 = fair). Total scores for all scales showed wide LoA, large SDC (25-31%), and fair to good reliability (ICC = 0.54–0.61). Exact agreement for Generalized Joint Hypermobility classification was excellent for the Beighton (≥7/9 threshold) and fair for the ULHAT and LLAS (≥7/12 threshold). Percentage of individual test items with good/excellent agreement was highest for the Beighton (78%, 7/9 items), then the ULHAT (58%, 14/24) and LLAS (42%, 10/24). Conclusion Total Scores of hypermobility scales showed low exact agreement between in-person and telehealth, but fair–excellent agreement within two points. Classification using the Beighton ≥7/9 threshold was excellent. Research is recommended to increase accuracy of online assessments.