The article concerns what we can determine about the FDAs process when it find problems in a clinical trial. As an article addressing that issue, it appears (af

The article concerns what we can determine about the FDAs process when it find problems in a clinical trial. As an article addressing that issue, it appears (after a brief look) to be pretty decent. What it isn't good for is figuring out what percentage of clinical trials have problems found in them. "Fifty-seven published clinical trials were identified for which an FDA inspection of a trial site had found significant evidence of 1 or more of the following problems: ...." But 57 out of what? The opening says, "Every year, the US Food and Drug Administration (FDA) inspects several hundred clinical sites ...." Then it notes that the data were gathered from FDA documents gathered over a period of 5.75 years. Picking a number out of the air, say the FDA inspects 500 trials each year. We would expect 2875 inspections over 5.75 years, and 57 is less than 2% of that. I would call that a low rate. OTOH, that "2%" figure is based on some unfounded assumptions. I don't think this article tells us much at all regarding the reliability of scientific publications.