A randomised route to better government


An evaluator-general would help policymakers understand which policies work, and which to drop.

  • Andrew Leigh
  • The Canberra Times

In the mid-1990s, researchers embarked on a massive clinical trial.

More than 16,000 post-menopausal women volunteered to be assigned randomly to a treatment group, receiving hormone-replacement therapy, or to a control group.

Many expected that the study would back the common view – based on observational studies – that taking estrogen plus progestin was good for these women’s health.

Five years in, the trial’s data and safety-monitoring board stopped the study.

Not only did the…

Leave a Reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.