Medicine and Ethics

Medicine and Ethics:

A new drug is being tested so see if it has a better success rate than standard treatment. The standard treatment has a 40% success rate, so to prove that the new drug is significantly better the success rate must be greater than 40% -- and that number must be reproducible. One hundred patients are enrolled on study and eight hundred of them are deemed successes. To determine whether or not this outcome proves the new drug to be significantly better, the probability of this outcome occurring with standard treatment is calculated; the smaller the probability, the more statistically significant the result.

This is how clinical trials currently operate, yielding specific results but at a slow rate and high cost.

A novel trial methodology is testing this convention. Using Bayesian statistical models, researchers have created a more flexible, cost-effective, and patient-centered study design that can also produce more statistically significant results. In this model, the study continues until a certain number of patients have “failed” (as defined by the researchers). The possible results from such a trial are limitless, since the number of patients who will enroll before the pre-determined number of patients fail is unknown. If the treatment causes patients to fail, the trial ends quickly; if the treatment leads to repeated successes, the trial continues indefinitely.