Association of the sense of coherence with physical and psychosocial health in the rehabilitation of osteoarthritis of the hip and knee: a prospective cohort study

Background: According to Antonovsky's salutogenic concept, a strong sense of coherence is associated with physical and psychological health. The goal of this study was to analyze the association of Antonovsky's sense of coherence with physical and psychosocial health components in patients with hip and knee osteoarthritis before and after in-and outpatient rehabilitation. Methods: Prospective cohort study with 335 patients, 136 (41%) with hip and 199 (59%) with knee osteoarthritis. The outcome was measured by Short Form-36 (SF-36), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and the Sense of Coherence (SOC-13). Baseline scores of the SF-36 and WOMAC scales and the observed effect sizes after rehabilitation were correlated with the baseline SOC-13. These correlations of the SF-36 scales were compared to the Factor Score Coefficients for the Mental Component Summary of SF-36, which quantify the factor load on the psychosocial dimension. Predictive impact of the baseline SOC-13 for the SF-36 and WOMAC scales (baseline scores and effect sizes) was then determined by multivariate linear regression controlled for possible confounders. Results: At baseline, the SOC-13 correlated with the WOMAC scores between r = 0.18 (stiffness) and r = 0.25 (pain) and with the SF-36 scores between r = 0.10 (physical functioning) and r = 0.53 (mental health). The correlation of these SF-36 correlation coefficients to the Factor Score Coefficient of the SF-36 Mental Component Summary was r = 0.95. The correlations for the effect sizes (baseline -> discharge) with the baseline SOC-13 global score were all negative and varied between r = 0.00 (physical functioning) and r = -0.19 (social functioning). In the multivariate linear regression model, the explained variance of the SF-36 scores by the baseline SOC-13 increased continuously from physical to psychosocial health dimensions (from 12.9% to 29.8%). This gradient was consistently observed for both the baseline scores and the effect sizes. The results of the WOMAC were consistent with the physical health scales of SF-36. Conclusions: The sense of coherence was associated with psychosocial health dimensions but hardly with physical health. The higher the load of a scale on the psychosocial dimension the higher was its correlation to the sense of coherence. This is in contrast to the idea of Antonovsky who predicted high associations with both mental and physical health.