AFFILIATIONS
1 University of Missouri Kansas City, Kansas City, Missouri, United States
2 University of Missouri Kansas City, Kansas City, Missouri, United States
3 University of Missouri Kansas City, Kansas City, Missouri, United States
4 University of Missouri Kansas City, Kansas City, Missouri, United States
5 University of Missouri Kansas City, Kansas City, Missouri, United States
6 University of Missouri Kansas City, Kansas City, Missouri, United States
7 University of Missouri Kansas City, Kansas City, Missouri, United States
8 University of Missouri Kansas City, Kansas City, Missouri, United States
9 University of Missouri Kansas City, Kansas City, Missouri, United States
10 University of Missouri Kansas City, Kansas City, Missouri, United States
ABSTRACT
Background
Hydroxyurea is the most established disease-modifying pharmacotherapy for sickle cell disease (SCD), with demonstrated reductions in vaso-occlusive crises (VOC), acute chest syndrome (ACS), and mortality. Despite longstanding guideline recommendations for hydroxyurea in HbSS and HbS/β°-thalassemia, contemporary national utilization patterns remain incompletely characterized. We evaluated hydroxyurea prescription, temporal trends, demographic variation, and acute healthcare utilization among adults in these genotypes.
Methods
We conducted a retrospective observational study using Epic Cosmos, a national, deidentified electronic health record (EHR) database aggregating data from participating U.S. healthcare organizations using Epic as EHR, from January 1, 2015, through December 31, 2025. Adults aged ≥18 years with active ICD-10-CM diagnoses of HbSS or HbS/β°-thalassemia were included. The cohort was restricted to these genotypes because of the strong and broad guideline recommendations for hydroxyurea use during the study period. HbSC and HbS/β?-thalassemia were excluded because their clinical severity and indications for hydroxyurea were more heterogeneous, selective and not universal.
Documented hydroxyurea prescription was ascertained from medication records. The primary outcome was the proportion of eligible patients with any documented hydroxyurea use during the study period. Annual utilization was defined as the proportion of patients with hydroxyurea documented in each calendar year. Utilization was stratified by current age, sex, and recorded race. Exploratory acute-care outcomes included ≥1 emergency department (ED) encounter, a VOC, ACS and mean inpatient length of stay (LOS). Temporal trends were assessed using the Cochran–Armitage test, and the annual change in odds of documented use was estimated using binomial logistic regression. Categorical comparisons were performed using chi-square testing.
Results
Among 70,056 eligible adults, 33,480 (47.8%) had documented hydroxyurea prescription, whereas 36,576 (52.2%) had none. Annual utilization increased from 25.6% in 2015 to 36.3% in 2025, an absolute increase of 10.7 percentage points (P for trend <0.001). The odds of documented use increased by 4.9% per calendar year (OR, 1.049; 95% CI, 1.047–1.051). Utilization declined progressively with increasing age: 61.2% among patients aged 18–29 years, 51.7% at 30–39 years, 46.0% at 40–49 years, 37.3% at 50–64 years, 23.1% at 65–74 years, 9.7% at 75–84 years, and 3.1% at ≥85 years (P for trend <0.001). Utilization was higher among men than women (52.5% vs 44.2%). Within recorded race categories, hydroxyurea prescribing was documented in 28,851 of 57,148 Black/African American patients (50.5%), 165 of 411 Asian patients (40.1%), 1,511 of 4,866 White patients (31.1%), and 3,821 of 9,304 patients with other or unselected race (41.1%). Hydroxyurea users more frequently had an ED encounter (75.5% vs 57.8%), VOC (12.0% vs 2.4%), and ACS (11.4% vs 2.8%; all P<0.001) compared to documented hydroxyurea non-users. Mean inpatient LOS was 5.9 days among users and 6.4 days among nonusers.
Conclusions
Despite improvement over the past decade, fewer than half of adults with HbSS or HbS/β°-thalassemia had any documented hydroxyurea use, and annual utilization reached only 36.3% by 2025. Markedly lower utilization among older adults and women identifies important gaps in disease-modifying therapy delivery. The inverse age gradient may partly reflect survivor selection, as older adults surviving with SCD may have milder disease severity.
The higher frequencies of recorded ED encounters, VOC, and ACS among hydroxyurea users likely reflect confounding by indication and differential healthcare contact. Patients with more severe disease are more likely to receive hydroxyurea and interact frequently with the healthcare system, whereas nonusers may have milder disease, higher unmeasured baseline fetal hemoglobin, less consistent follow-up, or care outside participating organizations. Medication records indicate prescribing but do not confirm medication adherence or actual use. Hydroxyurea users had a modestly shorter inpatient LOS, although causality cannot be inferred from this. Longitudinal studies incorporating disease severity, baseline fetal hemoglobin, treatment timing, adherence, and healthcare engagement are needed to identify modifiable barriers and optimize hydroxyurea delivery among patients with HbSS or HbS/β°-thalassemia.
DISCLOSURES
Varshini Srinivasan, MBBS: No relevant conflicts of interest to declare.