Cost–Effectiveness of Rivaroxaban Versus Heparins for Prevention of Venous Thromboembolism After Total Hip or Knee Surgery in Sweden

Lars Ryttberg; Alex Diamantopoulos; Fiona Forster; Michael Lees; Anina Fraschke; Ingela Björholt

Disclosures

Expert Rev Pharmacoeconomics Outcomes Res. 2011;11(5):601-615. 

In This Article

Results

Results of the cost–effectiveness analyses in THR patients are reported in Table 5; 35 days of rivaroxaban was associated with higher incremental cost over 14 days of enoxaparin (SEK119). This was attributed to the extended duration of prophylaxis in the rivaroxaban arm as indicated by the prophylaxis cost: SEK1771 versus SEK886 for rivaroxaban and enoxaparin, respectively. Nevertheless, an extended duration of rivaroxaban prophylaxis was associated with higher incremental quality-adjusted life-years (QALYs; 0.0040) versus 13 days of enoxaparin, which resulted in an incremental cost per QALY gained of SEK29,378.

In patients who underwent TKR surgery, rivaroxaban was associated with cost savings over enoxaparin of SEK873. Moreover, the rivaroxaban arm had marginally higher incremental QALYs (0.0029) compared with enoxaparin. Therefore, rivaroxaban dominated enoxaparin in the TKR indication (Table 6).

In the comparison of rivaroxaban versus dalteparin, the model predicted similar results in both indications. In THR, rivaroxaban prophylaxis was associated with higher cost (SEK143), and in TKR was associated with cost savings (-SEK880). Because the clinical input was the same as in the comparison with enoxaparin, the QALY estimates remained as in the comparison with enoxaparin. The result was an incremental cost per QALY gained with rivaroxaban versus dalteparin of SEK35,436 for THR, while in TKR rivaroxaban dominated.

The main drivers of cost–effectiveness were the improved health outcomes with rivaroxaban, with resulting reductions in the cost of treating events and gain in QALYs, and reduced costs due to the oral administration of rivaroxaban. Although the single most substantial cost was the cost of prophylaxis as determined by the assumed duration of prophylaxis and different drug prices, the reduction in the costs of treating symptomatic events and administering subcutaneous LMWH partially or completely offset the higher cost of prophylaxis. In THR, where extended-duration rivaroxaban prophylaxis was compared with approximately 2 weeks of enoxaparin, the incremental cost of SEK885 when only prophylaxis costs are included was reduced to an incremental cost of SEK119 when cost offsets from venous thromboembolic events and long-term complications avoided were included.

Similarly, in TKR patients, where all prophylactic agents were compared over the same duration, the incremental cost with rivaroxaban versus enoxaparin when only prophylaxis costs are included was SEK171. This became a cost saving of SEK290 already during the prophylaxis period when costs linked to the administration and dosing (treatment scheme) of enoxaparin were considered. Cost savings further increased to SEK873 with rivaroxaban when cost offsets from venous thromboembolic events and long-term complications avoided were included.

The importance of prophylaxis cost in the THR comparison was evident in the first sensitivity analysis where the cost was removed; the result changed from showing a low incremental cost per QALY gained with rivaroxaban versus enoxaparin to dominance for rivaroxaban. Changing the discounting rates caused only very minimal change to the model results. Moreover, increasing or decreasing the duration of hospitalization produced similar estimates to the base case (Table 7).

A separate analysis was performed based on a different THR study that presented evidence of efficacy and safety for an equal duration of prophylaxis (35 days).[8] Keeping all other treatments and long-term complication costs and utilities the same, rivaroxaban was associated with a small incremental QoL benefit against LMWHs (0.0009) and dominated against enoxaparin (cost savings: SEK45).

Furthermore, the cost of hospitalization and nurse administration, the unit cost of event treatment and the cost of long-term complications were tested in sensitivity analysis. The largest impact was observed in changes to the diagnosis and treatment cost for DVT and PE. In the latter two cases in THR, the result changed from dominance to an incremental cost of SEK68,815 per QALY gained. Changes to the utility values had little impact on the overall model results (Table 7).

Sensitivity analysis in the TKR population also demonstrates that model results were robust to input changes. Rivaroxaban retained dominance over enoxaparin and dalteparin for all variations (Table 8).

Given that the model assumed the same efficacy for dalteparin as for enoxaparin, the sensitivity analysis results moved in the same direction in the comparisons of rivaroxaban versus dalteparin.

Figure 3 presents the results of the PSA in the cost–effectiveness plane for the THR population. The majority of samples plot in the first quadrant, indicating that rivaroxaban was both more effective and more expensive than enoxaparin. There was a proportion of samples falling in the fourth quadrant (over 30%), indicating that rivaroxaban dominated enoxaparin. In Figure 4, the cost–effectiveness acceptability curve presents the probability of rivaroxaban being cost effective when different threshold values for the cost/QALY were applied. The cost–effectiveness acceptability curve crosses the vertical axis at a point higher than the origin (over 30%), indicating that there was a proportion of samples where rivaroxaban was cost effective even with a cost–effectiveness threshold of zero. This reflects the dominating samples of the PSA, for which the probability of rivaroxaban being cost effective at a zero threshold was 30%.

Figure 3.

Cost–effectiveness scatter plot: rivaroxaban versus enoxaparin (total hip replacement).
SEK: Swedish crowns.

Figure 4.

Cost–effectiveness acceptability curve: rivaroxaban versus enoxaparin (total hip replacement).
CE: Cost–effectiveness; SEK: Swedish crowns.

Figure 5 presents the results of the PSA in the TKR population. All samples plot in the fourth quadrant, indicating that rivaroxaban dominated the comparator. The results of this and the one-way sensitivity analyses indicated that the baseline results were robust to changes in the values of key variables.

Figure 5.

Cost–effectiveness scatter plot: rivaroxaban versus enoxaparin (total knee replacement).
SEK: Swedish crowns.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....