Jonathan Karnon1, Jackie Brown. 1. Health Economics Research Group, Brunel University, Uxbridge, Middlesex, United Kingdom. jon.karnon@hsl.gov.uk
Abstract
BACKGROUND: There remains uncertainty around the appropriate choice of adjuvant therapies to offer postmenopausal women with node-positive early breast cancer. OBJECTIVE AND STUDY DESIGN: To present the results derived from a discrete event simulation (DES) model that compared tamoxifen plus chemotherapy versus tamoxifen alone in node-positive postmenopausal women diagnosed with early breast cancer. METHODS: The data populating the model were mainly derived from the existing literature, which was analysed to specify probability distributions describing the uncertainty around the true value of each input parameter. The specified probability distributions facilitated the stochastic analysis of the decision model, whereby distributions of the model's outputs [aggregate costs and quality-adjusted life years (QALYs)] were estimated. RESULTS: The baseline results show that the addition of chemotherapy to tamoxifen in this patient group is relatively cost effective (under pound 4000 per additional QALY), but the distribution of the incremental cost-effectiveness ratio shows a wide range, including 10% of observations in which tamoxifen dominates tamoxifen plus chemotherapy. CONCLUSIONS: The results demonstrate the intuitive nature of stochastic evaluations of healthcare technologies, which may ease decision-makers' interpretation of cost-effectiveness results.
BACKGROUND: There remains uncertainty around the appropriate choice of adjuvant therapies to offer postmenopausal women with node-positive early breast cancer. OBJECTIVE AND STUDY DESIGN: To present the results derived from a discrete event simulation (DES) model that compared tamoxifen plus chemotherapy versus tamoxifen alone in node-positive postmenopausal women diagnosed with early breast cancer. METHODS: The data populating the model were mainly derived from the existing literature, which was analysed to specify probability distributions describing the uncertainty around the true value of each input parameter. The specified probability distributions facilitated the stochastic analysis of the decision model, whereby distributions of the model's outputs [aggregate costs and quality-adjusted life years (QALYs)] were estimated. RESULTS: The baseline results show that the addition of chemotherapy to tamoxifen in this patient group is relatively cost effective (under pound 4000 per additional QALY), but the distribution of the incremental cost-effectiveness ratio shows a wide range, including 10% of observations in which tamoxifen dominates tamoxifen plus chemotherapy. CONCLUSIONS: The results demonstrate the intuitive nature of stochastic evaluations of healthcare technologies, which may ease decision-makers' interpretation of cost-effectiveness results.
Authors: C Hürny; J Bernhard; A S Coates; M Castiglione-Gertsch; H F Peterson; R D Gelber; J F Forbes; C M Rudenstam; E Simoncini; D Crivellari; A Goldhirsch; H J Senn Journal: Lancet Date: 1996-05-11 Impact factor: 79.321
Authors: J P Gérard; M Héry; D Gedouin; A Monnier; M J Goudier; J P Jacquin; F Plat; E Cabarrot; D Serin; M Namer Journal: Drugs Date: 1993 Impact factor: 9.546
Authors: M Schumacher; G Bastert; H Bojar; K Hübner; M Olschewski; W Sauerbrei; C Schmoor; C Beyerle; R L Neumann; H F Rauschecker Journal: J Clin Oncol Date: 1994-10 Impact factor: 44.544
Authors: F J Cummings; R Gray; D C Tormey; T E Davis; H Volk; J Harris; G Falkson; J M Bennett Journal: J Clin Oncol Date: 1993-01 Impact factor: 44.544
Authors: Jonathan Karnon; Thomas Delea; Stephen R D Johnston; Robert Smith; Jane Brandman; Jennifer Sung; Paul E Goss Journal: Pharmacoeconomics Date: 2006 Impact factor: 4.981