Factors associated with longitudinal hydroxyurea adherence trajectories among individuals with sickle cell disease in Texas Medicaid: A group-based trajectory modeling study.

BACKGROUND: Hydroxyurea, first-line therapy in sickle cell disease (SCD), has suboptimal adherence ranging from 10%-36%. Thus, examining longitudinal adherence trajectories may help identify critical timepoints and individual subgroups for targeted interventions. OBJECTIVES: To identify distinct longitudinal hydroxyurea adherence trajectories among individuals with SCD enrolled in Texas Medicaid and to examine sociodemographic and clinical factors associated with trajectory group membership. DESIGN: Retrospective, longitudinal cohort study. METHODS: A retrospective analysis of Texas Medicaid data (01/01/2016-08/31/2023) included individuals with ≥3 medical claims for SCD, ≥1 prescription claim for hydroxyurea without a claim in the pre-index period, aged 2-63 years on the index date (first hydroxyurea claim), and continuously enrolled in Texas Medicaid for 6 months pre- and 12 months post-index. Proportion of days covered (PDC) was used to estimate adherence. Group-based trajectory modeling (GBTM) was used to identify hydroxyurea adherence trajectories. Multinomial logistic regression was used to examine factors associated with hydroxyurea adherence trajectory group membership. RESULTS: Of included individuals (N=1,246; mean age=18.1±12.7, 50.8% female), GBTM revealed five membership groups: consistent adherence (16.1%; mean PDC=89.2±8.5%), fluctuating adherence (28.7%; mean PDC=52.6±12.8), sharp decline and fluctuating adherence (16.5%; mean PDC=26.2±8.8), rapidly decreasing adherence (22.2%; mean PDC=25.6±10.3), and early and consistent nonadherence (16.5%; mean PDC=8.4±1.3). Potential intervention timepoints were identified at months 2, 3, and 5, when adherence declined markedly. Age groups 16-21, 22-34, and ≥35 years (reference: 2-15 years), and an increasing number of vaso-occlusive crises (VOC) in the pre-index period were significantly associated with higher odds of being in the less-adherent trajectory groups compared to the consistent adherence trajectory group: adjusted odds ratio (aORs) range: age groups=2.2-8.7; VOC=1.2-1.4. CONCLUSION: Unique adherence trajectory patterns with critical timepoints for intervention were observed. Individualized and timely interventions should be tailored to adults, especially those who are transitioning to adult care (22-34 years), and those experiencing frequent VOC.