Time-dependent resource use and costs associated with different states of disease in patients diagnosed with HER-2-positive metastatic breast cancer.

G.W.J. Frederix, J.L. Severens, A.M. Hovels, J.G.C. van Hasselt, J.A.M. Raaijmakers, J.H.M. Schellens

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Adequate reflection of disease progression and costs over time is essential in cost-effectiveness analyses based on health state-transition models. However, costing studies normally investigate the burden of metastatic breast cancer (MBC) without explicitly examining the impact of specific-disease states on health care costs over time. The objective of this study was to assess time-dependent costs of different health states of human epidermal receptor-2 (HER-2) positive MBC and the factors contributing to these costs. In the Netherlands, HER-2-positive MBC patients were identified in three different hospitals. Resource use was collected during 24 months, which was linked to unit costs and related to time with respect to date of MBC diagnosis, disease progression and death for each individual patient. Subsequently, monthly costs for different health states were calculated. Finally, a nonlinear mixed-effect modelling approach was used to provide a quantitative description of the time course of cumulative progression costs. Costs during stable disease were constant over time with a mean of $4,158. In contrast, monthly costs for progressive disease demonstrated a change over time with the largest costs in the first 2 months after diagnosis (p <0.005). The developed mixed-effect model adequately described cumulative cost-time course and associated variability. During the last months of life, costs varied over time, with the last month of life as the most expensive one with a mean of $5,811 per patient per month. To reflect costs of HER-2-positive MBC accurately in Markov models, costs for stable disease can be defined time independent, however, costs of progressive disease should be defined time dependent, and costs related to the final months of life should be modelled as such. The mixed-effect model we have developed could now be considered for adequate description of the time-dependent cost of progressive disease.
Original languageUndefined/Unknown
Pages (from-to)489-95
Number of pages7
JournalBreast Cancer Res Treat
Volume139
Issue number2
Publication statusPublished - 2013

Keywords

  • Farmacie/Biofarmaceutische wetenschappen (FARM)
  • Epidemiology
  • Farmacie(FARM)
  • Biomedische technologie en medicijnen
  • Ziekenhuisstructuur en organisatie van de gezondheidszorg
  • Public Health

Cite this