Date of Award
2015
Document Type
Dissertation
Degree Name
Ph.D. in Pharmaceutical Sciences
Department
Pharmacy Administration
First Advisor
Yi Yang
Second Advisor
Yunhee Chang
Third Advisor
Benjamin F. Banahan
Relational Format
dissertation/thesis
Abstract
Objective: The objective of this study was to assess the prevalence and predictors of use of hydroxyurea (HU) and its impact on clinical and economic outcomes in pediatric patients with sickle cell disease (SCD) enrolled in Medicaid. Methods: A cohort of patients with SCD was identified in 2006 using ICD-9-CM codes from Medicaid claims from 40 US states. Patients who filled three prescriptions of HU in within 6 months in 2007 were identified as HU users. HU users were then matched with non-users and the impact of HU use on the presence of crises and economic outcomes including presence of hospitalizations and emergency department visits were assessed using conditional logistic regression stratified on matched pairs. Length of stay and medical costs were compared in the matched sample using generalized linear models. An additional clinical outcome, number of crises, was evaluated in the unmatched sample of HU users and non-users using conventional multivariable regression. Estimates obtained using this approach, were then compared with those obtained by minimizing selection bias using regional variation and physician preference-based instrumental variables (IVs). Results: Prevalence of HU use in children with SCD enrolled in Medicaid was found to be 10.72%. Age, gender, race, disease severity, previous office visits, presence of a comprehensive sickle cell center within the state of residence, and prior opioid use were all found to be significant predictors of HU use in this population (p<0.0001). HU users had a significantly greater likelihood of having a hospitalization (OR:2.09; 95% CI:1.28-3.43) and a longer LOS (β=0.49; 95% CI:0.14-0.84) compared to non-users. Even though the conventional multivariable model shothat HU users had a significantly greater number of crises compared to non-users (β=0.93; p<0.0001), analysis using IVs found no statistically significant relationship (β=-2.75; p=0.2013). Conclusion: HU use is not very prevalent among children with SCD enrolled in Medicaid. Based on the identified predictors, it seems that physicians follow guidelines when prescribing HU in this population. Since this study failed to corroborate the benefit associated with the use of HU on clinical outcomes and resource utilization, physicians should be wary in prescribing HU in this population.
Recommended Citation
Datar, Manasi, "Predictors Of Use Of Hydroxyurea And Its Impact On Clinical And Economic Outcomes Among Children With Sickle Cell Disease" (2015). Electronic Theses and Dissertations. 744.
https://egrove.olemiss.edu/etd/744
Concentration/Emphasis
Emphasis: Pharmacy Administration