Advertisement

Statistics and Computing

, Volume 9, Issue 3, pp 219–227 | Cite as

Assessing robustness of crossover designsto subjects dropping out

  • BY J. L. Low
  • S. M. Lewis
  • P. Prescott
Article

Abstract

In some crossover experiments, particularly in medical applications, subjects may fail to complete their sequences of treatments for reasons unconnected with the treatments received. A method is described of assessing the robustness of a planned crossover design, with more than two periods, to subjects leaving the study prematurely. The method involves computing measures of efficiency for every possible design that can result, and is therefore very computationally intensive. Summaries of these measures are used to choose between competing designs. The computational problem is reduced to a manageable size by a software implementation of Polya theory. The method is applied to comparing designs for crossover studies involving four treatments and four periods. Designs are identified that are more robust to subjects dropping out in the final period than those currently favoured in medical and clinical trials.

A-criterion crossover experiments dropout missing value MV-criterion Williams squares 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Ballinger, C., Pickering, R. M., Bannister, S. and McLellan D. L. (1995) Evaluating equipment for people with disabilities: user and technical perspectives of commodes. Clinical Rehabili-tation, 9, 157–66.Google Scholar
  2. Hall, F. M. (1972) An introduction to abstract algebra, vol. 1. Cambridge: Cambridge University Press.Google Scholar
  3. Herzberg, A. M. and Andrews, D. F. (1976) Some considerations in the optimal design of experiments in non-optimal situa-tions. J. Roy. Statist. Soc. B, 38, 284–89.Google Scholar
  4. Jacroux, M. (1983) Some minimum variance block designs for estimating treatment differences. J. Roy. Statist. Soc. B, 70–76.Google Scholar
  5. Jones, B. and Kenward, M. G. (1989) Design and Analysis of Cross-Over Trials. London: Chapman and Hall.Google Scholar
  6. Kiefer, J. (1975) Construction and optimality of generalised Youden designs. In A Survey of Statistical Designs and Linear Models (ed. J. N. Srivastava), pp 333–53. Amsterdam: North Holland.Google Scholar
  7. Low, J. L., Lewis S. M., McKay B. D. and Prescott, P. (1994) Computational issues for cross-over designs subject to dropout. In Proceedings in Computational Statistics, pp 423–28. Heidelberg: Physica-Verlag.Google Scholar
  8. Low, J. L., Lewis S. M., and Prescott. P. (1998) An application of Polya theory to crossover designs with dropout. Technical report number 294, Department of Mathematics, South-ampton University, Southampton.Google Scholar
  9. Gough, K. (1994) Discussion of Informative dropout in longitudi-nal data analysis by P. Diggle and M. G. Kenward. J. Roy. Statist. Soc. C, 43, 83.Google Scholar
  10. Matthews, J. N. S. (1988) Recent developments in crossover de-signs. International Statistical Review, 56, 117–27.Google Scholar
  11. Willan, A. R. and Pater, J. L. (1986) Carryover and the two-period crossover design. Biometrics, 42, 593–99.Google Scholar

Copyright information

© Kluwer Academic Publishers 1999

Authors and Affiliations

  • BY J. L. Low
  • S. M. Lewis
  • P. Prescott

There are no affiliations available

Personalised recommendations