A US Claims Database Analysis Estimating Risk of All-Cause Mortality in Patients with Generalized Pustular Psoriasis

Abstract

Introduction: Generalized pustular psoriasis (GPP) is a rare, chronic, neutrophilic skin disease characterized by recurring flares of widespread erythema, edema, coalescing pustules, and possible systemic symptoms. GPP flares may require hospitalization and can be life-threatening. There are limited data describing the mortality burden of GPP in the US. This study compared all-cause mortality among patients with GPP with matched patients with plaque psoriasis (PsO), and the general population without GPP or PsO.
 Methods: Inovalon Insights real-world claims data were used to identify 5 cohorts (All GPP, Comorbid GPP+PsO, GPP Only, PsO Only, General Population) based on the International Classification of Diseases, Tenth Revision (ICD-10) diagnosis codes from Jan 1, 2016, to Dec 31, 2019. GPP Only, PsO Only, and Comorbid GPP+PsO were mutually exclusive. All GPP included GPP Only (excluding PsO) and Comorbid GPP+PsO. General Population comprised qualifying patients without a medical claim for GPP or PsO during pre- and post-index periods. All-cause mortality was assessed both 365 days post-index diagnosis and during a maximum follow-up period (until study period ended or patient experienced an event). Greedy caliper propensity score matching was used to match GPP patients 1:2 to PsO and General Population using index year, age, gender, insurance type, region, and Charlson Comorbidity Index (CCI). All-cause mortality risk was assessed using Cox proportional hazard models.
 Results: Patients were primarily female and commercially insured. The GPP Only cohort was older at index and had higher CCI scores than PsO Only and General Population (p<0.001), highlighting the higher severity of GPP. At 365 days follow-up, All GPP had a significantly higher mortality risk than General Population (HR 4.93, 95% CI 2.24-10.88) and PsO Only (HR 2.31, 95% CI 1.32-4.04). GPP+PsO had a significantly higher mortality risk than PsO Only (HR 2.67, 95% CI 1.15-6.20). At 365 days follow-up, GPP Only had a numerically higher mortality risk than PsO; risk was significantly higher at the later timepoint. At maximum follow-up, the mortality risk for All GPP was almost 4 times higher than General Population (HR 3.98, 95% CI 2.92-5.43) and 1.5 times higher than PsO Only (HR 1.49, 95% CI 1.20-1.85). In GPP+PsO and GPP Only, the mortality risk was almost 1.5 times the risk of PsO Only (HR 1.41, 95% CI 1.05-1.90; HR 1.49, 95% CI 1.10-2.03); all comparisons at maximum follow-up were statistically significant.
 Conclusion: This study characterizing all-cause mortality in patients with GPP demonstrated a higher risk of mortality in GPP than matched PsO and General Population cohorts. These results fill a gap in the existing literature and reinforce the need for awareness of the mortality burden and the comorbidity burden in GPP