Table 2.
Adjusted Annual Part D Drug Costs and Adherence to Three Diabetes-Related Drug Classes among All Subjects
| Total Drug Costs (All Part D Drugs) in 2006 | Coverage Gap (TDC U.S.$) | No Gap or Generic-Only Gap Coverage (TDC U.S.$) | Relative Cost | (95% CI) |
|---|---|---|---|---|
| Integrated MAPD (gap versus no gap) | 1,750 | 1,802 | 0.97 | (0.96, 0.99) |
| Network MAPD (gap versus generic-only gap coverage) | 1,974 | 2,048 | 0.96 | (0.94, 0.99) |
| Out-of-Pocket Expenditures (All Part D Drugs) in 2006 | Coverage Gap (OOP U.S.$) | No Gap or Generic-Only Gap Coverage (OOP U.S.$) | Relative Cost | (95% CI) |
| Integrated MAPD (gap versus no gap) | 806 | 279 | 2.89 | (2.85, 2.93) |
| Network MAPD (gap versus generic-only gap coverage) | 949 | 834 | 1.14 | (1.10, 1.17) |
| Adherence to Three Drug Classes in 2006 | Coverage Gap (% Adherent) | No Gap or Generic-Only Gap Coverage (% Adherent) | Odds Ratio | (95% CI) |
| Oral diabetes drugs | ||||
| Integrated MAPD (gap versus no gap) | 61.7 | 66.0 | 0.83 | (0.79, 0.88) |
| Network MAPD (gap versus generic-only gap coverage) | 60.0 | 59.5 | 1.01 | (0.97, 1.05) |
| Hypertension drugs | ||||
| Integrated MAPD (gap versus no gap) | 66.3 | 71.5 | 0.78 | (0.74, 0.83) |
| Network MAPD (gap versus generic-only gap coverage) | 64.7 | 64.5 | 1.00 | (0.96, 1.05) |
| Lipid drugs | ||||
| Integrated MAPD (gap versus no gap) | 66.0 | 73.9 | 0.69 | (0.65, 0.73) |
| Network MAPD (gap versus generic-only gap coverage) | 58.6 | 56.2 | 1.05 | (1.00, 1.10) |
Notes. This table reports adjusted annual total drug costs and out-of-pocket expenditures in 2006, and relative costs for the gap versus supplemented gap groups, estimated using a one-part general linear model with log transformed costs. All comparisons are within individual health systems: in the Integrated MAPD, we compared having a coverage gap versus no gap; in the Network MAPD we compared having a coverage gap versus generic-only coverage during the gap. These estimates are adjusted for individual beneficiary characteristics, including prior year costs (in 2005) summarized by a propensity score of having a coverage gap. The table also reports the adjusted percent of beneficiaries in each group adherent to each of the three chronic drug classes, and the odds ratio of adherence for gap versus supplemented gap groups, estimated using logistic regression models and adjusted for beneficiary characteristics summarized by a propensity score. The difference in adherence to lipid drugs in the Network MAPD is not significant at the p<.05 level.
CI, confidence interval; OOP, out-of-pocket drug expenditures; TDC, total drug costs.