Background Specific comorbidity index can guide the management of psoriatic arthritis (PsA) patients and optimize their outcomes. So far, no disease-specific models are currently available for this purpose.
Objectives 1. identify comorbidities with greatest impact on PsA patients' health status. 2. develop and validate a prospectively applicable comorbidity index for classifying PsA patients according to their comorbid conditions which might alter their risk of hospitalization and mortality.
Methods This was a retrospective multicenter cohort analysis of PsA patients in a rheumatology clinical registry, assessing the effect of different comorbidities measured at patients' visits over 10-years period, on predicting future death and hospitalization. A weighted index that takes into account the number and seriousness of comorbid disease was developed in a cohort of 1707 PsA patients monitored over 10-years. Logistic and Cox Regression analyses were implemented to estimate the risk of mortality. Regression coefficients were used to develop the morbidity index score. ROC curve for the invented index was used to evaluate the discriminating ability of the index and identify different cutoff values that can delineate patients at different stages for risk of death. Disease activity parameters were considered.
Results PsA patients who had higher incidence of comorbid condition and were at high risk of hospitalization were men, with older age at disease onset, high BMI at baseline (p<0.05). The most prevalent comorbidities strongly associated with the 10-year death risk or hospitalization in PsA patients were: Cardiovascular (7 comorbidities), osteoporosis, falls, depression/ anxiety, diabetes mellitus, renal and liver diseases, lung and GIT affection, as well as infection (p<0.001). Multivariate regression analysis identified Multidimensional Disease Relapse score (including 5 parameters: DAPSA, PASI, Functional disability score, ESR and CRP) as independent predictor for disease status associated with the 10-year death risk or hospitalization (Wald χ2=9.2, p=0.002, OR=24.6). Binary regression analysis revealed that: male gender, cardiovascular diseases, evident fall risk, diabetes, infection, anxiety, and the multidimensional relapse score were significant independent factors affecting the 10-years outcome of the disease. A comorbidity index weighted according to the regression coefficient of the variables extracted through the logistic regression analysis was developed. The score ranges from 0 to 38. A cut off point of 14.5 was associated with a sensitivity of 97.5% and a specificity of 87%. Validation using ROC curve revealed AUC of 98.5%.
Conclusions The PsA-comorbidity index is a valid method for estimating risk of death in PsA patients. It enables the clinicians to include comorbidities assessment and management in their standard practice. It can be used to predict resource utilization, identify targets for reducing high costs, by prospectively identifying PsA patients at high risk. Rigorous application of systematic evaluation of comorbidities may permit earlier detection, which may ultimately result in an improved outcome of patients with PsA.
Disclosure of Interest None declared