Posted by Gary Schwitzer in Health care journalism
Among the headlines that turn up on a Google search today:
Coffee and tea to the rescue in minimizing the spread of flesh eating bacteria
Tea Time; Drinking Beverage May Half MRSA Superbug Nostril Infection
Coffee and tea: MRSA fighters
Tea and coffee help fight superbug
Reminder: these headlines are based on an observational study, which can’t prove cause and effect. So all these statements – or even hints – of a causal link – are wrong.
If you need help understanding the limitations of observational studies, read our primer on association versus causation.
Kip Hansen posted on July 16, 2011 at 12:30 pm
Read the original study in the Annals of Family Medicine. They claim to find the effect in those who drink ONE or more cups of tea or coffee PER MONTH. — but for coffee only in the statistically transmogrified data.
That’s one powerful effect!
It appears again that advanced statistics in the hands of the statistically innumerate can produce the desired results from almost any body of numbers.