GSK and Catch 22

Print Friendly, PDF & Email
June 5, 2014 | 2 Comments


  1. Here are the official answers GSK provides in 2014 on its website regarding suicide in adult patients on paroxetine (Paxil):

    The key one is the Medical Information Letter for healthcare professionals. It relies on a trick that David has explained a few times in public lectures. In a table on Page Three, they set out the rates of suicide or suicide attempt on Paxil vs. placebo in six different studies.

    Only one study is for major depression. That one shows 8 suicide incidents on Paxil and none on placebo. It’s still less than one percent of the patients – but it’s a stunner. Four other studies for people with other conditions are smaller, and show such tiny numbers of suicide incidents that there’s not much you can say.

    Then comes the trick – they insert a study on “intermittent brief depression” where the suicide-incident rates are huge, for both Paxil and placebo. This is a group of patients that I think in the US would usually be diagnosed with Borderline Personality Disorder. They have very unstable emotional lives, and it’s not unknown for one person to make four or five suicide attempts in a year. Mixing in this study allows them to come up with a “total” figure where the suicide rate on Paxil is very, very slightly lower than placebo.

    It’s a scam, and GSK knows it. But it allows them to make this point in their letter to Senator Chuck Grassley, their biggest Congressional critic:

    “GSK’s 1989 and 1991 analyses of the clinical trial data for Paxil, properly analyzed, revealed no increase in the risk of suicides or suicide attempts for adult patients on Paxil.”

    “Properly analyzed,” indeed.

Leave a Reply