Managed Care - February 2009 - (Page 41) between high use of certain medications with better diabetes HEDIS scores. For the first time, the relationship between utilization of specific classes of medications and declining rates of patients in poor diabetes control was demonstrated (Weiss 2007). That study examined the relationship among health plans of HEDIS-measured HbA1c control and the prescription share of four categories of diabetes products: analog insulin, human insulin, branded oral products, and generic oral products. We reported a strong positive relationship between better HEDIS scores and higher analog insulin prescription share (P<0.0001). The purpose of the current study was to investigate whether these results could be corroborated with a more robust and recent data set. generic oral products consisted of all single source and multisource oral products respectively captured in the IMS prescription database. The prescription share of each of the four drug classes was calculated based on all prescription diabetes products in the PlanTrak database. Therefore, the prescription shares of the four classes within each health plan totaled 100 percent. METHODS To identify the level of HbA1c control for a given health plan, this study used the HEDIS measurement “percent of diabetes patients having poor HbA1c control” as reported in the NCQA Quality Compass product (NCQA 2008). These measurements were matched to the plan’s prescription share (based on prescription volume) in each calendar year for four groupings of antidiabetic products: (1) analog insulin, (2) human insulin, (3) single-source branded oral products, and (4) multisource generic oral products. A health plan was included in the study for 2004 or 2005, provided the plan had a valid HbA 1c “poor control” HEDIS measurement and at least 2,000 total diabetes prescriptions in that year. Prescription volume for each product group was reported by IMS HealthCare (Plymouth Meeting, Pa.) for individual health plans in its PlanTrak report. Analog insulins included the insulin lispro, insulin aspart, and insulin glargine product lines. Human insulin included the insulin isophane (Humulin) and human insulin isophane suspension (Novolin) product lines.Branded and ANALYSIS Univariate (simple) and multivariate regression analysis was performed between the health plan’s (HbA1c)-based HEDIS score and its prescription share of each drug category. The effect of plan size and region of the United States was also explored via multivariate regression. In order to assess the effect of copayments on HEDIS score we matched HEDIS score to the percentage of diabetes prescriptions in the lowest copayment tier in each plan and performed a linear regression across health plans to determine the relationship between HEDIS score and percentage of diabetes prescriptions with the lowest copayment. RESULTS A total of 194 eligible health plans were identified for 2005, and 211 health plans were identified for 2006. Table 1 summarizes the prescription volume from eligible plans for each TABLE 1 year. The total diabetes prescription volume for eligible plans was about 21.2 million in 2005 and about 24.5 million in 2006. Combined branded and multisource generic oral products represented slightly more than two thirds of total diabetes prescriptions (68 percent and 68.1 percent for 2005 and 2006, respectively).Analog insulin prescription share increased from 20.7 percent in 2005 to 23 percent in 2006, while human insulin share decreased from 11.3 percent to 8.9 percent during the same time period. The oral multisource generic market share increased slightly from 48.1 percent in 2005 to 49.4 percent in 2006, while branded oral market share decreased from 19.8 percent in 2005 to 18.6 percent in 2006. The mean HEDIS score for percentage of patients with diabetes enrolled in eligible plans who were not in control improved from 29.6 percent ± 7.7 in 2005 to 28.8 percent ± 7.13 in 2006 (P=NS). A statistically significant correlation (p = 0.002 and 9 percent) HEDIS scores among health plans in both 2005 and 2006 (Table 2). That is, a higher analog insulin prescription share was associated with a lower (fewer patients with HbA1c level > 9 percent) HEDIS score. The slope Summary of prescription volume and share for eligible plans 2005 N = 194 plans TRx volume Analog insulin Human insulin Total insulin Brand orals Generic orals Total orals Total 4,404,688 2,395,031 6,799,719 4,211,152 10,223,185 14,434,337 21,234,056 TRx share 20.7% 11.3% 32.0% 19.8% 48.1% 68.0% 100.0% 2006 N = 211 plans TRx volume 5,651,117 2,182,978 7,834,095 4,573,738 12,127,876 16,701,614 24,535,709 TRx share 23.0% 8.9% 31.9% 18.6% 49.4% 68.1% 100.0% TRx = Total prescriptions (new and refill); N = number of plans in analysis FEBRUARY 2009 / MANAGED CARE 41
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.