Confirm and Explore: A Stepwise Approach to Clinical Study Designs


In general, the scientific learning process begins with exploration. Scientists learn from exploratory findings, generate a new hypothesis, and then perform experiments to confirm the hypothesis. After they do more exploration, and learn more, scientists continue with this process. The process is to explore first, then confirm, next explore and confirm again, and so on. The same is true for the entire drug development process.

However, in designing an individual phase 2 or 3 clinical trial, this thinking process needs to be reversed because each study is designed to address a specific clinical question. This question is formulated as the primary objective of the study. Clinical trials are designed to address this well-defined study objective. Hence this part is a confirmatory step. After the primary objective is achieved, scientists then explore all available data and see what additional information can be generated from the study.

In this article, we discuss the thinking process of clinical trials. It is clear that the clinical trial design follows a stepwise approach that first considers the confirmatory step by testing a pre-specified, well-defined statistical hypothesis, and then moves to the exploratory (or learning) step. We discuss a few examples to illustrate how this thinking process can be applied in designing some of the phase 2/3 studies.

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


  1. 1.

    Ting N. Introduction and new drug development process. In: Ting N. ed. Dose Finding in Drug Development. New York: Springer: 2003;1–17.

    Google Scholar 

  2. 2.

    Fricke J, Varkalis J, Zwillich S. et al. Valdecoxib is more efficacious than rofecoxib in relieving pain associated with oral surgery. Am J Therapeutics. 2002;9(2):89–97.

    Article  Google Scholar 

  3. 3.

    Tamhane AC, Hochberg Y, Dunnell CW. Multiple test procedures for dose finding. Biometrics. 1996;52:21–37.

    CAS  Article  Google Scholar 

  4. 4.

    Ruberg SJ. Dose response studies II. Analysis and interpretation. J Biopharm Stat. 1995;5(1): 15–42.

    CAS  Article  Google Scholar 

  5. 5.

    Dmitrienko A, Wiens B. West fall P. Fallback tests in dose-response clinical trials. J Biaphamn Stat. 2006;16:745–755.

    Article  Google Scholar 

  6. 6.

    Pinheiro JC, Bretz F, Branson M. Analysis of dose-response studies—modeling approaches. In: Ting N, ed. Dose Finding in Drug Development. New York: Springer; 2003;146–171.

    Google Scholar 

  7. 7.

    Weslfall PH, Krishen A. Optimally weighted, fixed sequence and gatekeeping multiple testing procedures. J Stat Plann Inference. 2001;99:25–40.

    Article  Google Scholar 

  8. 8.

    Wong WK, Lachenbruch PA. Tutorial in biostalis-tics: designing studies for dose response. Stat Med. 1996;15:343–359.

    CAS  Article  Google Scholar 

  9. 9.

    Hamlett A, Ting N, Hanumara C, Finman JS. Dose spacing in early dose response clinical trial designs. Drug Inf J. 2002;36:855–864.

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Dr. Naitee Ting PhD.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Ting, N. Confirm and Explore: A Stepwise Approach to Clinical Study Designs. Ther Innov Regul Sci 42, 545–554 (2008).

Download citation

Key Words

  • Phase 2/3 clinical study design
  • Dose selection
  • Proof of concept
  • Dose ranging