Association of health care utilization and costs with patient-reported outcomes in patients with ankylosing spondylitis.

Abstract

BACKGROUND: Interventions for ankylosing spondylitis (AS) have improved patient-reported outcomes (PROs) in clinical studies. However, limited data exist associating these improvements with health care resource utilization (HCRU) or cost savings. Few studies have evaluated the economic impact of patient-reported physical status and related disease burden in patients with AS in the United States. OBJECTIVE: To assess the association of PRO measures with HCRU and health care costs in patients with AS from a national US registry. METHODS: This cohort study included adults with a diagnosis of AS enrolled in the FORWARD registry from July 2009 to June 2019 who completed at least 1 questionnaire from January 2010 to December 2019 and completed the Health Assessment Questionnaire Disability Index (HAQ-DI) (0-3) and/or Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) (0-10). Patient-reported data for demographics, clinical characteristics, and PROs were collected through questionnaires administered biannually and reported from the most recent questionnaire. Patient-reported HCRU and total health care costs (2019 US dollars) for hospitalizations, emergency department (ED) visits, outpatient visits, diagnostic tests, and procedures were captured during the 6 months prior to the most recent survey completion. The relationship between HAQ-DI or BASDAI and HCRU outcomes was assessed using negative binomial regression models, and the relationship between HAQ-DI or BASDAI and the cost outcomes was evaluated using generalized linear models with γ distribution and log-link function. RESULTS: Overall, 334 patients with AS who completed the HAQ-DI (n = 253) or BASDAI (n = 81) were included. The mean (SD) HAQ-DI and BASDAI scores at the time of patients' most recent surveys were 0.9 (0.7) and 3.7 (2.3), respectively. HAQ-DI score was positively associated with number of hospitalizations, ED visits, outpatient visits, and diagnostic tests, whereas BASDAI was not associated with HCRU outcomes. Overall annualized mean (SD) total health care, medical, and pharmacy costs for patients with AS were $44,783 ($40,595); $6,521 ($12,733); and $38,263 ($40,595), respectively. Annualized total health care, medical, and pharmacy costs adjusted for confounders increased by 35%, 76%, and 26%, respectively, for each 1.0-unit increase in HAQ-DI score (coefficient [95% CI]: 1.35 [1.15-1.58], 1.76 [1.22-2.55]; both P < 0.01 and 1.26 [1.04-1.52]; P < 0.05, respectively); BASDAI score was not significantly associated with cost outcomes. CONCLUSIONS: Higher HAQ-DI scores were associated with higher HCRU and total health care costs among patients with AS in FORWARD, but BASDAI scores were not. These findings indicate that greater functional impairment may impose an increased economic burden compared with other patient-reported measures of AS. DISCLOSURES: A. Ogdie has received consulting fees from Amgen, AbbVie, Bristol Myers Squibb, Celgene, CorEvitas (formerly Corrona), Gilead, Janssen, Lilly, Novartis, Pfizer, and UCB and has received grant support from the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases, Rheumatology Research Foundation, National Psoriasis Foundation, Pfizer (University of Pennsylvania), Amgen (FORWARD), and Novartis (FORWARD). M. Hwang has received consulting fees from Novartis and UCB and has received grant support (5KL2TR003168-03) from the University of Texas Health Science Center at Houston Center of Clinical and Translational Sciences KL2 program. P. Veeranki and J. Shafrin were employees of PRECISION-heor at the time of this analysis. A. Portelli and S. Sison are employees of PRECISION-heor. S. Pedro does not have anything to disclose. N. Kim was a postdoctoral fellow at the University of Texas at Austin and Baylor Scott and White Health, providing services to Novartis at the time of this study. E. Yi is an employee of Novartis. K. Michaud received grant funding from the Rheumatology Research Foundation at the time of this analysis. This study was funded by Novartis Pharmaceuticals Corporation, East Hanover, NJ.