You can't do that, I'm afraid. Too many uncontrolled variables still. What if placebo rates change seasonally? Also, you cannot properly double blind the test -

You can't do that, I'm afraid. Too many uncontrolled variables still. What if placebo rates change seasonally? Also, you cannot properly double blind the test - the experimenters all know about the prior study, and know which cohort they are treating. To be fair, your attitude is quite common in cash-strapped university departments - but it leads to bad science.