What Did You Know, and When Did You Stop Knowing It? - Forbes

What Did You Know, and When Did You Stop Knowing It? - Forbes

Recently, a story (Analytical Trend Troubles Scientists) appeared in The Wall Street Journal that was critical of — what turns out to be — an increasingly common type of medical investigation: the observational study.

The WSJ story uses as its jumping-off point the (shocking? surprising?) discrepancy between two studies that both asked the same question and used the same data: does taking bisphosphonates (think Fosamax) increase your risk of stomach or throat cancer?

One study concluded that no, it does not significantly increase your cancer risk. The other said that yes, it does significantly increase said risk. ["Significantly" here means "probably", in the statistician's sense.] In neither case, though, did it find that your cancer risk would be particularly high: 0.1% vs 0.2% in 60/69-year-olds. So why the dustup?.....