Sample Size Estimation when Comparing More than Two Treatment Groups


A large amount of literature has been published on methods for determining sample size and power for clinical trials. Although formulae exist to determine sample size for different designs, in practice the methods are rarely used. Instead, sample size estimates are usually based on simple formulae derived for the comparison of two means or binomial proportions.

This paper describes some useful theory for sample size estimation when comparing more than two treatment groups. Where the theory is not applicable, it is demonstrated that simulation can be used as an alternative approach.

This is a preview of subscription content, access via your institution.


  1. 1.

    Armitage P, Berry G. Statistical methods in medical research. 2nd ed. Oxford: Blackwell; 1987:182.

    Google Scholar 

  2. 2.

    Bland JM. An introduction to medical statistics. Oxford: Oxford Medical Publications; 1987:160.

    Google Scholar 

  3. 3.

    Fleiss JL. The design and analysis of clinical experiments. Chichester: Wiley; 1986:371–376.

    Google Scholar 

  4. 4.

    Frison L, Pocock SJ. Repeated measures in clinical trials: Analysis using mean summary statistics and its implications for design. Stat Med. 1992;11:1685–1704.

    CAS  Article  Google Scholar 

  5. 5.

    Pocock SJ. Clinical Trials. Chichester: Wiley; 1983:128–229.

    Google Scholar 

  6. 6.

    Montgomery SA, Asberg M. A new depression rating scale designed to be sensitive to change. Br J Psychiatry. 1979;134:382–389.

    CAS  Article  Google Scholar 

  7. 7.

    Lindley DV, Scott WF. New Cambridge Elementary Statistical Tables. Cambridge: Cambridge University Press; 1984.

    Google Scholar 

  8. 8.

    Banner NR, Lloyd HM, Hamilton RD, Innes JA, Guz A, Yacoub MH. Cardiopulmonary response to dynamic exercise after heart and combined heart-lung transplanatation. Br Heart J. 1989;61:215–223.

    CAS  Article  Google Scholar 

  9. 9.

    Lucki I, Rickeis K, Gesecke MA, Geller A. Differential effects of the anxiolytic drugs, diazepam and buspirone, on memory function. Br J Clin Pharacol. 1987;23:207–211.

    CAS  Article  Google Scholar 

  10. 10.

    Rowbotham DJ, Nimmo WS. Effects of cisapride on morphine-induced delay in gastric emptying. Br J Anaesth. 1987;59:536–539.

    CAS  Article  Google Scholar 

  11. 11.

    Tatsuta M, Isihi H, Yamamura H, Yamamoto R, Taniguchi H. Enhancement by tetragastrin of experimental induction of gastric epithelium in the duodenum. Gut. 1989;30:311–315.

    CAS  Article  Google Scholar 

  12. 12.

    Laubscher NF. Normalizing the noncentral t and F distributions. Ann Math Stat. 1960;31:1105–1112.

    Article  Google Scholar 

  13. 13.

    Cohen J. Statistical power analysis for the behavioral sciences. New York: Academic Press; 1977.

    Google Scholar 

  14. 14.

    Ruberg SJ. Dose response studies. Some design considerations. J Biopharma Stat. 1995;5:1–14.

    CAS  Article  Google Scholar 

  15. 15.

    Ruberg SJ. Dose response studies. Analysis and interpretation. J Biopharma Stat. 1995;5:15–12.

    CAS  Article  Google Scholar 

  16. 16.

    Abelson RP, Tukey JW. Efficient utilisation of non-numerical information in quantitative analysis: General theory and the case of simple order. Ann Math Stat. 1963;34:1347–1369.

    Article  Google Scholar 

  17. 17.

    Yateman N, Skene A. The use of simulation in the design of two cardiovascular survival studies. Stat Med. 1993;12:1365–1372.

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Alan Phillips BSc, PhD.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Phillips, A. Sample Size Estimation when Comparing More than Two Treatment Groups. Ther Innov Regul Sci 32, 193–199 (1998).

Download citation

Key Words

  • Sample size
  • Analysis of variance
  • Dose-response
  • Simulation