Background/Purpose: Psoriasis (PsO)/Psoriatic arthritis (PsA) often requires treatment with systemic immunosuppressive agents, some of which may increase hospitalized infection risk. Few population-based studies to date have evaluated the incidence of hospitalized infections this population.

Methods: We used the US Medicare data from 2006-2011 to identify a large cohort of PsA and PsO patients. We defined PsA and PsO patients as those with >1 rheumatologist-diagnosis code for psoriatic arthritis (ICD 9 696.0), or >1 dermatologist-diagnosis code for psoriasis (ICD 9 696.1) respectively, followed by a prescription for etanercept (ETA), cyclosporine (CIC), ustekinumab (UST), adalimumab (ADA), methotrexate (MTX) or ultraviolet light (UV) therapy. Patients had at least 6 months of continuous Medicare enrollment prior to the first date of exposure to these therapies. We excluded patients with organ transplantation, human immunodeficiency virus infection, advanced kidney and liver disease, or cancer with a 183-day period prior to cohort inception. We used validated-claims based algorithms to identify hospitalized infections among all exposure groups. Patient exposures were censored at time of serious infection, death, end of study, loss of coverage, or 90 days following end of treatment exposure whichever came first. Pairwise propensity scores (PS) were calculated and used to control for potential differences between comparator treatments. We calculated crude incidence rates for exposure groups, and used Cox-proportional hazard regression models to calculate hazard ratios for hospitalized infection between exposure groups while adjusting for PS quintile.

Results: We identified 10,261 PsA individuals and 31,052 PsO individuals. Within the PsA cohort, we identified 185 hospitalized infections for an overall incidence rate of 36.2 (95% CI 31.1-41.8) per 1,000 py. The rate of hospitalized infections ranged from 13.2 (95% CI 4.3-41.0) per 1,000 py for the UV group to 38.7 (95% CI 28.8-52.0) per 1,000 py for the ETA group. In Cox modeling, incidence rates were similar between exposure groups, with the exception of patients starting ETA as compared to UV therapy (HR 3.1 [95% CI 0.9-1.9]) where a non-statistically significant trend was noted. Within the PsO cohort, there were 1,198 hospitalized infections yielding an overall incidence rate of 37.0 (95% CI 35.0-39.1) per 1,000 py. The rate of such infections ranged from 27.2 (95% CI 23.1-31.9) per 1,000 py for the UV group to 40.5 (95% CI 36.9-44.5) per 1,000 py for the MTX group. After adjustment for possible confounders, patients on ADA had higher incidence rates as compared to those on CIC (HR 1.4 [95% CI 1.0, 2.0]), or UV therapy (HR 1.4 [95% CI 1.1-1.9]), respectively. Patients on MTX were also at higher risk for infection than those using UV therapy (HR 1.3 [95% CI 1.1, 1.6]). Incidence rates between all other exposure comparisons were similar.

Conclusion: Among Medicare enrollees with PsA or PsO, rates of hospitalized infections varied across therapies but were largely similar. PsA patients were at similar risk no matter their therapy, although PsO patients starting ADA or MTX were at higher risk for infection than those using UV therapy.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/risk-of-hospitalized-infection-in-a-psoriasispsoriatic-arthritis-cohort/