AUTHORS
Elnara Fazio-Eynullayeva, MA, Brenna Brady, PhD2, Megan Richards, PhD3, F. Leila Jerome Clay, MD, MCTS4, Biree Andemariam, MD5
AFFILIATIONS
1 Novo Nordisk Inc., Plainsboro, New Jersey, United States
2 Merative, Ann Arbor, Michigan, United States
3 Merative, Ann Arbor, Michigan, United States
4 Novo Nordisk Inc., Plainsboro, New Jersey, United States
5 University of Connecticut School of Medicine, Farmington, Connecticut, United States
ABSTRACT
Introduction:
Sickle cell disease (SCD) is a group of rare hereditary blood conditions affecting approximately 100,000 individuals in the United States. The landscape of disease-modifying treatments (DMTs) for SCD expanded in the past decade, but there is a lack of evidence on treatment utilization, persistence, and economic outcomes among patients with SCD. To address this gap, this study examined treatment adherence, treatment persistence, healthcare resource utilization (HCRU), and costs among US patients newly initiating DMTs for SCD.
Methods:
This retrospective cohort study used the Merative™ MarketScan® Multi-state Medicaid Database from January 1, 2017, through December 31, 2023. Treatment utilization of DMTs (hydroxyurea [HU], crizanlizumab, L-glutamine, or voxelotor) was evaluated in patients with SCD (≥1 inpatient claim or ≥2 outpatient claims 30-364 days apart with a diagnosis based on ICD-10 Clinical Modification codes D57.X, excluding D57.3) with ≥6 months of continuous pre-index enrollment and ≥12 months of continuous post-index enrollment. Among patients with ≥1 SCD claim who newly initiated 1 of the 4 study DMTs, treatment persistence (days from treatment initiation to the first medication gap >90 days) was evaluated over a variable post-index period defined by the duration of continuous eligibility. Treatment adherence, calculated as the proportion of days covered (PDC), along with all-cause and SCD-related HCRU and costs were assessed among patients who persisted on the index DMT for ≥3 months and reported over the period of persistence. HCRU variables included the number of inpatient visits, emergency department (ED) visits, outpatient office visits, outpatient other visits, and pharmacy fills. Outcomes were reported separately for each DMT cohort. No between-cohort statistical comparisons were conducted as cohorts were not mutually exclusive.
Results:
Of 19,896 patients with SCD on Medicaid, 3,790 (19.0%) had ≥1 claim for HU, 288 (1.4%) for crizanlizumab, 447 (2.2%) for L-glutamine, and 509 (2.6%) for voxelotor; of these, 77.2%, 97.6%, 95.5%, and 98.2% were new initiators, respectively. The persistence analysis included 2,211 patients on HU, 299 on crizanlizumab, 426 on L-glutamine, and 451 on voxelotor. Persistence averaged less than 1 year (HU: 271 days; crizanlizumab: 212 days; L-glutamine: 181 days; voxelotor: 222 days). Across cohorts, more than half of patients discontinued treatment (HU: 70.7%; crizanlizumab: 72.6%; L-glutamine: 81.5%; voxelotor: 57.2%). A total of 1,155 patients on HU, 177 on crizanlizumab, 210 on L-glutamine, and 278 on voxelotor persisted on treatment for ≥3 months and were included in the adherence, HCRU, and cost analyses. Even among persistent patients, the proportion with PDC ≥80% was low across cohorts (HU: 37.7%; crizanlizumab: 57.6%; L-glutamine: 43.3%; voxelotor: 58.6%). All-cause inpatient visits and outpatient visits were high, with >50% of patients having an admission in both the pre- and post-index periods across cohorts. Pre- and post-index average total all-cause costs varied across cohorts (HU: $4,474 vs $4,751; crizanlizumab: $10,087 vs $15,587; L-glutamine: $6,008 vs $7,724; voxelotor: $10,512 vs $17,051). SCD-related inpatient visits decreased slightly from pre- to post-index periods across cohorts, and SCD-related ED visits decreased in all cohorts except voxelotor. SCD-related outpatient office visits increased among patients treated with HU and crizanlizumab but decreased among those treated with L-glutamine and voxelotor. SCD-related pharmacy fills increased across all 4 cohorts. Pre- and post-index average total SCD-related costs were higher for crizanlizumab ($8,228 vs $12,341; cost ratio: 1.5) and voxelotor ($8,180 vs $14,247; cost ratio: 1.7) than HU ($3,220 vs $3,664; cost ratio: 1.1) and L-glutamine ($5,385 vs $6,353; cost ratio: 1.2).
Conclusion:
Among patients with SCD on Medicaid, DMT adherence was suboptimal and persistence averaged less than 1 year. SCD-related HCRU was high overall with no consistent pattern of improvement from pre- to post-index periods. This study revealed that, despite initiation of DMTs, patients with SCD experience treatment continuity gaps and have high HCRU and costs.
DISCLOSURES
Elnara Fazio-Eynullayeva, MA: Novo Nordisk Inc. - Current equity holder in publicly-traded company; Novo Nordisk Inc. - Current Employment