Portal de Periódicos da CAPES

Cost‐effectiveness of hydroxyurea in reducing the frequency of pain episodes and hospitalization in pediatric sickle cell disease

2010; Wiley; Volume: 85; Issue: 10 Linguagem: Inglês

10.1002/ajh.21772

1096-8652

James R. Stallworth, Jeanette M. Jerrell, Avnish Tripathi,

Homelessness and Social Issues

In a cohort of children with sickle cell disease (SCD) and vaso-occlusive pain visits served through South Carolina's Medicaid system over a 6-year period (N = 523), we compared the number of vaso-occlusive pain or acute chest syndrome (ACS)/pneumonia episodes, and outpatient or acute service costs in those treated or not treated with hydroxyurea (HU). HU may be an underused intervention for SCD in this practice setting, for a variety of reasons. Treatment with HU varied greatly, appears to have been administered to more severely ill children, but was associated with a reduction in vaso-occlusive pain episodes, hospitalizations, and total costs of care within the HU cohort during a 2–3 year period of active HU treatment. Those receiving care through specialized SCD clinics were less likely to have pain or acute care episodes (RR = 0.79, P < 0.0001; RR = 0.90, P = 0.01). Compared with the non-HU cohort, the HU group evinced a significantly higher risk of experiencing vaso-occlusive pain episodes (RR = 3.32, P < 0.0001) and ACS/pneumonia episodes (RR = 2.66, P < 0.0001), and higher outpatient, inpatient/emergency, and total service costs (RR = 1.85, 2.11, 2.10, and P < 0.0001, respectively) over time. HU is clinically effective in reducing pain episodes, hospitalizations, and total care costs, but those receiving it might be more severely ill. In a cohort of children with sickle cell disease (SCD) and vaso-occlusive pain visits served through South Carolina's Medicaid system over a 6-year period (N = 523), we compared the number of vaso-occlusive pain or acute chest syndrome (ACS)/pneumonia episodes, and outpatient or acute service costs in those treated or not treated with hydroxyurea (HU). HU may be an underused intervention for SCD in this practice setting, for a variety of reasons. Treatment with HU varied greatly, appears to have been administered to more severely ill children, but was associated with a reduction in vaso-occlusive pain episodes, hospitalizations, and total costs of care within the HU cohort during a 2–3 year period of active HU treatment. Those receiving care through specialized SCD clinics were less likely to have pain or acute care episodes (RR = 0.79, P < 0.0001; RR = 0.90, P = 0.01). Compared with the non-HU cohort, the HU group evinced a significantly higher risk of experiencing vaso-occlusive pain episodes (RR = 3.32, P < 0.0001) and ACS/pneumonia episodes (RR = 2.66, P < 0.0001), and higher outpatient, inpatient/emergency, and total service costs (RR = 1.85, 2.11, 2.10, and P < 0.0001, respectively) over time. HU is clinically effective in reducing pain episodes, hospitalizations, and total care costs, but those receiving it might be more severely ill. Erythrocytes in children with sickle cell disease (SCD) become deoxygenated, dehydrated, and crescent-shaped, and tend to aggregate or stick to blood vessel walls, blocking blood flow within limbs and organs, causing painful episodes. These patients are frequently seen in emergency departments and hospitalized for these severe pain episodes [1]. SCD poses an enormous personal burden to these young patients and is also costly to the family and third-party payer, especially for low-income children [2]. Interventions designed to control vaso-occlusive pain episodes, and avoid hospitalizations may reduce the significant personal and economic burdens of the disease [3, 4]. Hydroxyurea (HU), a myelosuppressive agent, which raises the levels of Hb F [5] and of hemoglobin [6-8], effectively decreases the rate of painful vaso-occlusive and acute chest syndrome (ACS) episodes by ∼50% in adults [9-12], and is generally safe and well-tolerated in children older than 5 years of age [13]. There is strong evidence that HU reduces the frequency of hospitalization in children with SCD and moderate evidence that it decreases the frequency of painful crises [14]. Given the short-term safety profile of HU in children and its established efficacy in adults, HU is commonly used, off-label, in children with multiple painful episodes (≥3 per year) to reduce episode frequency and acute services utilization [15-17], even in children as young as 9 months. Results from a multicenter study of HU in sickle-cell anemia demonstrated that adult patients treated with HU had a 44% decrease in hospitalizations compared with those taking placebo, which accounted for the majority of cost savings in those taking HU and suggests that HU therapy is cost-effective [18]. However, investigators in an academic practice setting found that 70% of patients who were appropriate candidates for HU were not prescribed/taking the medication [19]. The purpose of this analysis is to examine the association between HU treatment and pain or ACS episodes and service costs over time in routine clinical practice for low-income pediatric patients. Medical and pharmacy claims for the calendar years January 1, 1998 through December 31, 2006 were used to identify a cohort of child and adolescent patients (ages 17 and under) enrolled in and eligible for Medicaid for a minimum of 9 months in each calendar year included in this analysis, who had a service encounter with a diagnosis of SCD (thalassemia and SC trait were not included). The selection date was the date of the first encounter in which the SCD diagnosis was made. This procedure yielded a total N = 2194 pediatric SCD cases during the 11-year period. South Carolina Medicaid medical claims were used to identify a service encounter, date of service, and the International classification of diseases, ninth clinical modification diagnosis codes related to that visit. Pharmacy claims identified the medication dispensed, and the date the prescription was filled. Demographics for each patient were summarized from an eligibility file. This study was approved by the University of South Carolina Institutional Review Board as exempt from human subject research guidelines (45 Code of Federal Regulations part 46). Primary or secondary diagnosis codes for sickle cell pain crises (282.62, 282.64, 282.69) and ACS or pneumonia (786.9, 517.3, 486) were counted separately for outpatient service visits, and for combined emergency visits and hospital days. The prescription of HU (DROXIA, HYDRIA) was coded from the Pharmacy file. Adherence with the prescribed HU was coded as days treated with the medication based on the Pharmacy file fills/refills and number of days per month for which the HU was dispensed. In this cohort, 175 patients were prescribed HU. An incidence density matching method was used to select the comparison group, which was not prescribed HU. Cases were matched (1 case: 2 controls) on age, gender, ethnicity, time in the Medicaid data set, and severity (mean number of vaso-occlusive pain visits during the baseline period), resulting in a control cohort of 348 without missing data (Table I). Pain visits, ACS visits, hospital use, and all cost variables were aggregated into four 6-month periods before the HU was prescribed (baseline) and eight 6-month periods after it was prescribed. Twelve time periods were also aggregated for the control cases. To examine the impact of the HU over time on frequency of pain episodes, ACS/pneumonia episodes, and the change in outpatient, inpatient/emergency, and total costs of SCD-related health services, a series of negative binomial regression models (for non-normally distributed count data; PROC GENMOD facility in SAS version 9.1; SAS Institute, Cary, North Carolina) was employed to calculate a ratio of the log rate of pain or ACS visits or costs per 6-month period that the patient was covered in the Medicaid data set. Furthermore, receipt of transfusions, prophylactic penicillin, or an adenotonsillectomy (dichotomously coded as yes/no), which might be associated with hypoxic episodes correlating with the onset of pain crises [20, 21], and the percentage of SCD services received from a specialty clinic for SCD were used as control variables to represent interventions or type of care that might also explain differences in pain or ACS episodes. Resulting model estimates were converted into rate ratios (RRs) and 95% confidence interval and P-values are used to report statistical significance. The 175 SCD patients treated with HU were prescribed the medication a mean of 870 days during the 4-year study period (i.e., 2.4 years of "active treatment"). As the large standard deviations in Table I indicate, however, there was much variability in these treatment regimens, for reasons that could not be ascertained from this data set. Compared with the control cohort, HU was associated with a significantly higher risk of experiencing vaso-occlusive pain episodes over time, that is, a mean of 13.1 pain visits per year for those prescribed HU versus 4.5 pain visits for those not prescribed HU. ACS/pneumonia episodes were also more likely in the HU group, that is, a mean of 1.7 ACS/pneumonia visits per year versus 0.6 ACS/pneumonia visits for those not prescribed HU. Within the HU cohort (Fig. 1, top line), however, there was a significant reduction over time in pain episodes. Hospitalization episodes declined during the 2–3 years of active HU treatment as well, from 4.1 episodes at the start of HU, to 3.0 episodes during the first 2 years, to 2.4 episodes at the end of 3 years of HU treatment (P ≤ 0.0001). Although higher outpatient, inpatient/emergency, and total service costs were also significantly more likely in the HU group (Table II), these costs substantially declined during the 2–3 year period of active HU treatment, from $12842 when HU was initiated, to $9367 in the first year, $10092 in the second year, and $8839 in the third year (P ≤ 0.0001). Mean Vaso-occlusive pain visits over time. Less than 10% of the total 2194 SCD cohort identified was prescribed HU. It is unclear whether the limited or inconsistent use of HU stems from access limitations to specialty care with a pediatric hematologist (9% non-HU and 12% HU), a lack of clear guidelines for its use, provider attitudes/knowledge or their concern over minimizing possible adverse reactions to it or family or cultural barriers to healthcare use [22-24]. Furthermore, up to 25% of SCD patients might be nonresponders to HU [25], but we have no way of exploring this possibility. Those receiving HU experienced more severe and frequent pain episodes over time than the control group, but the HU was associated with a reduction in their pain episodes, their hospitalization episodes, and their total costs during active treatment [18, 19]. These results shed additional light on long-standing concerns that targeting selected Medicaid enrollees with high emergency or inpatient expenditures for intervention will reduce overall expenditures [26]. This study has several strengths: the SCD cohort represents a large, heterogeneous group of children and adolescents; the long-term observational study provides information regarding important clinical correlates and their impact on pediatric SCD; and previous studies have found that although Medicaid databases provide much less detailed information on individuals and care than primary data collection, physician diagnoses, and utilization data correspond to clinical medical record reviews in the majority of the cases [27-30]. However, several limitations exist: the data were not gathered using a prospective, controlled design; structured clinical research interviews were not used to confirm any of the assigned medical conditions; these results report associations and, as a result, directions of causality cannot be inferred; there is no way to estimate how representative this Medicaid cohort is in relation to those in other states and service systems; and children and adolescents with SCD who dropped out of treatment or were periodically ineligible for Medicaid coverage are not represented in this data set and their outcomes may differ from those patients who remained in Medicaid over time. These results should provide additional incentives for the development of HU use guidelines with children for increasingly effective clinical use of the drug, and for pediatricians/primary health care providers diagnosing or suspecting a diagnosis of SCD for earlier referral to or frequent consultation from an experienced hematologist to achieve more optimal care of the SCD symptoms, and to decrease the personal burden of SCD to patients and their families. Data acquisition and analysis was supported by the Departments of Neuropsychiatry and Pediatrics, University of South Carolina School of Medicine. The authors thank Dr. Wilfred Karmaus for helpful suggestions on earlier drafts.