Self‐reported bio‐psycho‐social factors partially distinguish rotator cuff tendinopathy from other shoulder problems and explain shoulder severity: A case‐control study

Objective Examine how rotator cuff (RC) tendinopathy differed from other shoulder problems (OSP) by measuring a variety of self-reported bio-psycho-social factors, and establish which explain severity. Methods A validated online survey battery was used to collect self-reported biopsychosocial variables in an international population. Diagnostic group and severity were the dependent variables. Multiple logistic and linear regression analyses were utilised to generate explanatory models for group differences and severity after group comparison and univariate regression analysis. Results 82 people with RC tendinopathy (50 female, 42.8 ± 13.9 years) and 54 with OSP (33 female, 40.2 ± 14.1 years) were recruited. Both groups had comparable severity results (Shoulder Pain and Disability Index = 37.3 ± 24.5 vs. 33.7 ± 22.5). Seven factors individually differentiated RC tendinopathy from OSP. The multivariable model included 4 factors: activity effect on pain (OR(95%CI) = 2.24(1.02–4.90)), previous injury in the shoulder (OR(95% CI) = 0.30(0.13–0.69)), activity level (moderate OR(95% CI) = 3.97(1.29–12.18), high OR(95% CI) = 3.66(1.41–9.48)) and self-efficacy (OR(95%CI) = 1.12(1.02–1.22)) demonstrating acceptable accuracy. The second multivariable model for RC tendinopathy severity included one demographic, three psychological and two biomedical variables (β(range) = 0.19–0.38) and explained 68% of the variance. Conclusion Self-reported bio-psycho-social variables may be beneficial for further detailed clinical assessment as they partially distinguish RC tendinopathy from OSP, even when the groups have comparable overall pain and functional problems. Moreover, these variables were shown to be substantially associated with RC tendinopathy severity variance, implying that the clinical evaluation might be improved, perhaps by pre-consultation online data collection. The models should be validated in the future and considered alongside data from physical and imaging examinations