Summary

The steps that are involved in a medical study are:
  1. Select the population for the study - the more participants, the better (if it can be funded).
    2. The population is randomly split into two, and one half get a placebo.
  2. Track the effects on the participants.
    3. The patients are monitored periodically for signs of illness, and side effects of the medication.
  3. Use the Kaplan-Meier analysis method.
    4. This compensates for "dropouts" who leave the experiment in the middle.
  4. Determine the statistical relevance of the outcome.
    5. Typically, results are quoted "in a Margin of Error". This implies that there is a one-in-twenty chance that the true results lie outside of the interval.
  5. Compare and combine results with other similar studies (if they exist).
    6. This improves the statistical precision of the summed result.
There is an excellent tutorial in the British Medical Journal that covers some of these topics in much more detail.

If you benefitted from this site, please consider supporting us.




Previous

Last Modification - August 14, 2004