What question did this study set out to answer?

To develop a decision-theoretic framework for designing optimal clinical trials by minimizing the consequences of assigning patients to inferior medical treatments.

June 1, 1963Journal of the American Statistical Association

A Model for Selecting One of Two Medical Treatments

View Full Paper

Ask AI

Bookmark

Authors

TCTheodore ColtonCosmos Corporation (United States)

Discussion

Member takes

Implication

Theoretical modeling study demonstrates optimal clinical trial designs across fixed and sequential patient cohorts, highlighting cost-effective treatment selection strategies.

Key Points

To develop a decision-theoretic framework for designing optimal clinical trials by minimizing the consequences of assigning patients to inferior medical treatments.
Formulated a single-cost objective function based on the loss incurred from administering an inferior therapy to patients across a total disease population of N.
Applied minimax, maximin, and Bayesian optimization frameworks to evaluate fixed sample size trials and boundary positions in sequential trials.
Derived analytical solutions for the optimal sample size in fixed-sample trials and optimal stopping boundaries in sequential trial designs.
Identified design trade-offs between fixed and sequential approaches under differing optimization criteria to minimize total patient risk.

Cite This Study

Theodore Colton (1963) studied this question.

synapsesocial.com/papers/6a1547cf5347fbb1739f8654 https://doi.org/10.2307/2283274

View Full Paper

Ask AI

Bookmark

Implication

Theoretical modeling study demonstrates optimal clinical trial designs across fixed and sequential patient cohorts, highlighting cost-effective treatment selection strategies.

A Model for Selecting One of Two Medical Treatments

Discussion

Member takes

Implication

Key Points

Cite This Study

A Model for Selecting One of Two Medical Treatments

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider