Background Patients (pts) treated for RA have higher incidence rates of infections compared with the general population, which can complicate the clinical management of RA.1 However, there is a paucity of data in the literature on medical resource utilization associated with infection in RA pts in clinical practice.
Objectives The primary objective of this analysis was to compare resource use (proportion of pts with durable medical equipment [DME] use, hospitalization and emergency room [ER] visits) in RA pts with vs without infections in clinical practice. The secondary objective was to compare the pt-reported outcomes (PROs) of physical functioning (mHAQ) and quality of life (EQ-5D) in RA pts with vs without infections in clinical practice.
Methods Pts enrolled in the Brigham and Women's Hospital Rheumatoid Arthritis Sequential Study (BRASS) registry, established in 2003, were analysed. BRASS mostly comprises pts with established RA who were evaluated semi-annually for multiple clinical PROs and resource utilization parameters. Infections/opportunistic infections were recorded annually by a rheumatologist. The current analysis is based on the first 5 years of pt follow-up in BRASS. Pts with any infections at baseline or follow-up were categorized as Infection=Yes; those without were categorized as Infection=No. To control for any intra-class correlation of the panel data in BRASS, the generalized estimating equation method was utilized for bivariate outcomes of hospitalizations, ER visits and use of DME; mixed models were used for continuous outcomes of mHAQ and EQ-5D. Variables for the fixed effects included baseline age, sex, disease duration, DAS28 (CRP), BMI, osteoporosis, renal disease, diabetes, infection and corticosteroid use.
Results Overall, 1137 (84.9%) BRASS pts were included in the current analysis. Of these, 462 (40.6%) had at least one infection during follow-up. In general, the two groups were comparable, although pts with infections were older (mean [SD] age 57.4 [13.2] yrs vs 55.2 [14.4] yrs for Infection=Yes and No, respectively), had a higher number of comorbidities (mean [SD] 1.3 [1.2] vs 0.9 [0.8]), and lower physical functioning (mean [SD] mHAQ 0.49 [0.51] vs 0.39 [0.43]). After controlling for baseline covariates, RA pts with infections (vs those without) had higher odds for hospitalizations (odds ratio [OR]=2.27; 95% CI 1.802, 2.868), ER visits (OR=2.27; 95% CI 1.802, 2.868) and DME use (OR=1.29; 95% CI 1.078, 1.546). The multivariate models for mHAQ and EQ-5D indicated no significant difference in physical functioning between pts with vs without infections (mean difference in mHAQ=0.0157; p=0.153) and marginal decrease in quality of life for pts with vs without infections (mean difference in EQ-5D= –0.009; p=0.053).
Conclusions Both univariate and multivariate results indicate that RA pts with infections tend to have higher healthcare utilization than those without infections. Use of DME was 29% more likely, while hospitalizations and ER visits were twice as likely in pts with infections. Reducing infections in RA patients through implementation of new ACR draft guidelines could reduce healthcare utilization.
Doran MF, Gabriel SE. J Rheumatol 2001;28:1942–3
Disclosure of Interest E. Alemao Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, S. Joo Shareholder of: Bristol-Myers Squibb, Employee of: Bristol-Myers Squibb, M. Frits: None declared, C. Iannaccone: None declared, N. Shadick Grant/research support from: Bristol-Myers Squibb, Crescendo Biosciences, Amgen UCB, AbbVie, M. Weinblatt Grant/research support from: Bristol-Myers Squibb, Crescendo Bioscience, UCB, Consultant for: Bristol-Myers Squibb, Crescendo Bioscience, UCB, AbbVie, Roche, Janssen, Pfizer, Lilly, Amgen