Abstract
Objective We aimed to identify and characterize distinct subgroups of patients with psoriatic arthritis (PsA) based on their responses to the Assessment of SpondyloArthritis international Society Health Index (ASAS HI), using latent class analysis (LCA).
Methods We performed an exploratory LCA on 17 dichotomous ASAS HI items in a cohort of patients with PsA (n = 90). Model adequacy was evaluated by log-likelihood, Akaike information criterion (AIC), Bayesian information criterion (BIC), entropy, and average posterior probabilities (AvePP). Clinical measures (Psoriatic Arthritis Impact of Disease [PsAID], Disease Activity Index for Psoriatic Arthritis [DAPSA], tender/swollen joint counts, treatment) were compared across classes. Sensitivity analyses with 3 to 4 classes assessed robustness.
Results A 6-class solution was retained, ordered from class 0 (lowest impact) to class 5 (highest impact). Mean ASAS HI ranged from 1.2 in class 0 to 11.6 in class 5, with parallel increases in PsAID and DAPSA. Conditional probabilities revealed distinct profiles: class 0 had minimal impairment, classes 1-2 showed predominantly physical function limitations of mild to moderate severity, class 3 combined physical and emotional function burden, class 4 was characterized by dominant participation and social impact, and class 5 displayed severe multidimensional impairment. All classes had AvePP > 0.86. Sensitivity analyses with K = 3 and K = 4 reproduced the same gradient of impact but provided less granular separation.
Conclusion LCA of the ASAS HI identified 6 clinically meaningful health impact profiles in PsA. These findings support the use of the ASAS HI as a multidimensional tool capable of capturing diverse patient experiences and may help inform individualized management strategies in PsA.
