The tyranny of power: is there a better way to calculate sample size?

The author explains the need to change the method by which sample sizes in medical studies is calculated. He references research papers about the so-called power calculations published in September 1972. He notes that power calculations are founded on significant tests. On the other hand, the author argues, sample size calculations should be based on the scope of a confidence interval, not power. Also explored is a requirement by medical journals for results to be presented with confidence intervals.