Confidence Regions on Optimal Drug or Treatment Combination

  • M. J. de Matos Barbosa
  • Rosa O. Reis
Conference paper
Part of the Lecture Notes in Medical Informatics book series (LNMED, volume 45)


With Toxic drugs, to find the optimal dosage levels, when two or more drugs are combined in the treatment, is a rather complex problem. The best way to analyse this kind of problem is to apply Response Surface Methodologies (RSM) which can be presented through Contours of Continuous Response. In theses contours, the darker shade or the higher digits represent the optimal treatment region and whithin its bondaries will lie the best dosage levels for each drug in the combination. However, some constrains must be set in order to avoid significant toxicity. And a confident region of 100(1-α)%, which includes the optimal treament level, is indeed the best solution to help the analysis of the problem. In this paper, we are concerned mainly how to calculate such confidence regions. But a brief explanation about the whole process will be given.


Response surface methodologies Optimization methods Multi Logistic Regression Asymptotic properties Maximum-likelihood estimation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Barbosa M.J. (1990)-The use of contour plots for the inteipretation of multi-drug combinations. Lecture Notes in Medical Informatics N2 40pp 45–47Google Scholar
  2. Box G.E. and Hunter J.S. (1954)-A confidence region for the solution of a set of simultaneous equations with applications to experimental desiign. Biometrika N2 41, pp 190–199.MathSciNetMATHGoogle Scholar
  3. Cox D.R. (1972) Regression models and life tables. J.Royal Stas. Society N2 34, pp 187–220.Google Scholar
  4. Hoerl AJE. (1959) Optimum solution of many variable equations.Chem.Eng.Prog. N2 55, 69–82.Google Scholar
  5. Wampler G.L., Carter W.H. and Williams V.R. (1978)-Combination chemotherapy aniiiving at optimal treatment levels by incorporating side effects constrains. Cancer Treat. Rep. N2 62, pp 333–340.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • M. J. de Matos Barbosa
    • 1
  • Rosa O. Reis
    • 1
  1. 1.Department of Biomathematics and Medical Informatics, Faculty of MedicineUniversity of CoimbraPortugal

Personalised recommendations