A Pilot Intervention to Improve Hydroxyurea Adherence in Youth with Sickle Cell Disease through Community Health Workers: The "Habit" Study

Background: Chronic disease regimens for youth require self-management skills for a daily medication habit. Incomplete adherence to hydroxyurea (HU) is a treatment barrier for sickle cell disease (SCD) (Walsh KE, Pediatrics 2014). HU induction of fetal hemoglobin (HbF) is dose-dependent but not a standardized target (Ware RE, Hematology Am Soc Hematol Educ Program, 2015). In a previous 2-site retrospective analysis of youth with SCD, large deviation from historical personal best (PBest) HbF, a clinic-based form of maximum dose (Green NS, PBC, 2016), was common and correlated with increased acute hospital use. Here we hypothesized that: 1) A HU habit can develop through education and support of youth-parent dyads from a structured intervention of community health workers (CHW), augmented by text messaging; 2) Improved used will increase HbF; 3) The intervention is feasible and acceptable. Methods: We conducted a 2-site R21-funded randomized controlled feasibility trial (RCT), "Hydroxyurea Adherence for Personal Best in Sickle Cell Treatment: HABIT" (Smaldone A, Contemp Clin Trials, 2016)." Youth- parent dyads (youth ages 10-18) were recruited for the 6-month intervention. Eligibility required a ≥15% drop from PBest HbF on stable HU dose over the preceding year. Intervention dyads received a 6-month structured intervention of education and support by CHWs with home visits and text messaging, versus educational materials. Primary outcomes were HbF, % deviation from PBest HbF and HU pharmacy refills. Results: Using 2:1 randomization, 28 parent-youth dyads participated (18 intervention, 10 control): mean age 14.3±2.6 years, 42.9% female. At baseline, no differences were seen in HbF (10.5% vs. 13.4%, p=0.25), but deviation from PBest HbF was higher in the control group (-42.6%) than the intervention (18.1%, p=0.009). Using a growth model for % deviation from PBest HbF, the intervention group demonstrated a modest absolute improvement in deviation from PBest (2.2% per month higher in the intervention group), peaking at month 4 (p=0.31), and modest gain in mean HbF (3 versus 1.8). Four subjects exceeded their PBest. Compared to the year before intervention, pharmacy refills improved for the intervention group: 64.6% year prior, 80.7% (months 0 to 3), 85.9% (months >3 to 6), while the control group was unchanged (79.2% year prior, 83.6% months 0 to 3, 79.4% months >3 to 6). Intervention feasibility and acceptability by dyads were excellent. Conclusions : The 6-month pilot RCT "HABIT" may improve HU adherence measured by improved HbF, deviation from PBest and pharmacy refill, was feasible and acceptable, and suggests utility as an intervention. Based on these findings, a multi-center HABIT RCT of 1-year duration powered to test efficacy is planned. Funded by 5R21NR013745 (PIs Green, Smaldone) ![Figure][1] Figure Disclosures No relevant conflicts of interest to declare. [1]: pending:yes