Desktop version

Home arrow Health arrow Quantitative Methods for HIV/AIDS Research


Benefits and Limitations of Adaptive Design

This section describes the relative advantages, limitations, and feasibility of various adaptive designs.

Possible Benefits

As indicated by Chow and Corey (2011), the use of adaptive design methods in clinical trials entails the following possible benefits: (1) they allow the investigator to correct wrong assumptions made at the beginning of the trial;

  • (2) they help the investigator to select the most promising option early;
  • (3) they make use of external information (e.g., recent publications in the literature); (4) they enable the investigator to react earlier to surprises (either positive or negative); and (5) they may shorten the development time and consequently speed up the development process.

The use of adaptive design methods in clinical research and development may provide the investigator a second chance to modify or reevaluate the trial after seeing data from the trial itself at interim. However, despite the flexibility and possible benefits of adaptive design methods in clinical trials, it should be noted that valid statistical inference is often difficult, if not impossible, under complicated adaptive designs.


Both Chow et al. (2005) and Gallo et al. (2006) indicated that adaptive design methods must not undermine the validity and integrity of a trial (see also Emerson 2006). Validity is defined as (1) minimization of operational biases that may be introduced when applying adaptations to the trial, (2) correct or valid statistical inference, and (3) results that are convincing (e.g., accurate and reliable) to the broader scientific community. Integrity is defined as (1) minimizing operational biases, (2) maintaining data confidentiality, and (3) assuring consistency during the conduct of the trial, especially when multistage adaptive design is used. As indicated in the FDA draft guidance (FDA 2010a), bias introduced through adaptive design methods can adversely affect decision-making during the conduct of a trial. However, adaptations based on blinded analyses at interim, if the blinding is strictly maintained, can largely reduce or completely avoid bias.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics