Page 124 - The diagnostic work-up of women with postmenopausal bleeding

Chapter 6
Diagnostic strategy consisting of SIS and hysteroscopy
In this hypothetical strategy, we select women based on a positive or inconclusive result on the SIS to have an indication for hysteroscopy (see Figure 1). Costs per patient in the hysteroscopy group were statistically significantly higher compared to the expectant management group (mean difference € 634, 95% CI 408; 996).
The ICER for cancers detected shows that to detect one case of (pre) cancer € 8,913 should be invested in the SIS + hysteroscopy group as compared with the expectant management group (Table 3).The CEA curve (Figure 3b) shows that at willingness-to-pay values of € 5,000 or 10,000 or 20,000/case of recurrent bleeding prevented the probability that the intervention is cost-effective in comparison with usual care is 0, 0.68 and 0.97, respectively. At a willingness to pay of 16,000 € / detected (pre) cancer extra, the probability of cost-effectiveness is 0.95.
Discussion
Principal findings
In this study, we evaluated the cost-effectiveness of hysteroscopy in comparison to expectant management in women with postmenopausal bleeding, a thickened endometrium and a benign result of endometrial sampling to prevent recurrent bleeding and to diagnose an endometrial (pre) cancer.The results show that costs in the intervention group were statistically significantly higher in the hysteroscopy- group compared to the expectant management group, although the effect difference in the number of women with recurrent bleeding was not statistically significant different between the two groups.
Based on the CEA curves, hysteroscopy is not considered cost-effective to prevent recurrent bleeding.However,a direct diagnostic work-up with a hysteroscopy led to a statistically significant increase in the number of detected (pre) cancers. Detection of one case of (pre) cancer needs almost an investment of € 10,917 in the hysteroscopy group as compared with the expectant management group. The CEA curve showed the probability for hysteroscopy alone to be cost-effective in comparison with expectant management is 0.95 at a willingness-to-pay of € 19,500 /detected pre (cancer) and for hysteroscopy preceded by a SIS € 16,000 / detected (pre) cancer.We can conclude that a strategy using SIS to select women for hysteroscopy the costs to detect one case of (pre) cancer can be, depending on the probability to be cost-effective, lowered with € 2,000 to € 3000.
122