Lies, Damned Lies, and Chinese Statistics

Here’s a story that should make anyone uncomfortable. A study by the Chinese government has found that 80% of the data in Chinese clinical trials are just made up. From Science Alert:

A Chinese government investigation has revealed that more than 80 percent of the data used in clinical trials of new pharmaceutical drugs have been “fabricated”.

The report uncovered fraudulent behaviour at almost every level, and showed that some pharmaceutical companies had hidden or deleted records of potentially adverse side effects, and tampered with data that didn’t meet their desired outcomes.

This sounds to me like a modern version of Lysenkoism. It’s the kind of thing you would expect when good (rather than true) results are rewarded and bad results punished.

In fairness most published research results here are bogus, too. Still, 80% is darned impressive. I’m sure that with a little effort the CCP can boost that to 90% or greater.

6 comments… add one
  • walt moffett Link

    Interesting that the Chinese government realized there was problem. Whether its a result of the current anti corruption drive is anyone’s guess. Wonder if any of these studies have been cited in seeking FDA approvals.

  • Wonder if any of these studies have been cited in seeking FDA approvals.

    That’s the real question. I’m guessing that there are people nervously checking their bibliographies.

    What I suspect is actually going on here is political infighting in the politburo. Somebody’s clients will be purged and replaced with somebody else’s who’ll engage in the same practices.

  • steve Link

    Their studies are false because of the fraud. Ours are not true because of the over reliance on the p value, not because of fraud. Not that we don’t have fraud, it just isn’t as common. It may be most common in diet studies.

    This is good to remember when libertarians say we should use data from other countries to move our drugs through faster, or just import them from other countries.

    Steve

  • walt moffett Link

    Steve, all good points, might also want to mention sample size and inability to reproduce results. Might I suggest country of origin labels for meds?

  • TastyBits Link

    Phoney US dollars being traded for phoney Chinese drugs – is this progress or what?

    The poor guy or gal, in their house with the corroded plumbing, crying over their dead dog can now take anti-depressant drugs that will likely cause them to kill themselves.

    Free-trade, ain’t it great?

  • Phoney US dollars being traded for phoney Chinese drugs – is this progress or what?

    I think that’s the best comment of the day.

Leave a Comment