username2 comments on Open thread, Jul. 18 - Jul. 24, 2016 - Less Wrong

3 Post author: MrMind 18 July 2016 07:17AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (123)

You are viewing a single comment's thread. Show more comments above.

Comment author: Soothsilver 20 July 2016 08:54:07PM 2 points [-]

How likely am I to die from taking SSRI's?

This review says:

In conclusion, this review suggests the need for caution in the use of SSRI therapy, particularly among patients with other risk factors for bleeding and those taking NSAIDs. The estimated rate of hospitalization for upper GI bleeding in the US during the 1990s was ∼ 155 per 100,000 population per year [46]. The data examined suggests that the risk of GI bleeding may be increased 2 – 4-fold in patients taking SSRIs and 3 – 12-fold when combined with NSAIDs. However, one must recognise that the incidence of upper GI bleeding increases with age. As such, the risk/benefit profile of these compounds could be influenced by the patient’s age and other co-morbid conditions.

on which I performed calculations like this:

  1. It's a 4-fold increase with 155 per 100.000 population getting hospitalized for bleeding by any cause
  2. Bleeding has a 30% mortality rate
  3. The increase of chance in death is +0.001395 per year
  4. Which is 1395 micromorts per year

Which seems like a lot, for just one sideeffect of SSRI's. Would somebody please check my calculation to see whether I haven't made a major error?

One thing I've neglected is that GI bleeding appears to occur more with the elderly and its mortality rate rises sharply with age, but I haven't found any good data on that. Does anybody have some experience with this?

I'm suffering from a recurring depression (in my 4th depressive period now) which apparently often responds well to SSRI's, but I'm still not sure how much of a risk it is.

IANAL, but I'm not requesting medical advice, just help with math.

Comment author: username2 25 July 2016 11:29:25PM *  0 points [-]

In direct contradiction to your request, I am not helping with your maths but providing general (not specific) advice.

Bleeding has a 30% mortality rate

This depends very much on the population you are looking at (age, source of bleeding, severity of bleeding, comorbidities at time of bleeding - both acute and chronic). I would suggest that if you are not fulfilling some of these criteria your bleeding mortality is much lower. It is certainly possible that an SSRI could have a statistically significant effect in cases of severe bleeding like major trauma, but you'd have to estimate the risk that you'll experience major bleeding.

As Cariyaga has said, acetaminophen/paracetamol is an alternative, which doesn't have the same interaction with SSRIs. Individual doses of NSAIDs may be ok, but I don't know if there's any good data to support or deny this. Part of the risk with NSAIDs is that they affect coagulation AND directly cause bleeding (gastric ulceration). Past history or family history of clotting/bleeding disorders, gastric ulcers or significant gastric reflux increase the risk of this.

Another alternative is non-serotonergic antidepressants. There are many classes of antidepressant and some may not feature the same level of evidence of this increased bleeding risk (although all feature potential side-effect profiles longer than my arms). To confuse matters, some are classified structurally (tricyclic antidepressants TCAs, tetracyclic antidepressants) and some functionally (SSRIs, SNRIs, NaSSAs, RIMAs, MAOIs). TCAs may behave functionally like SSRIs, and may confer a bleeding risk as well.

I think the solution will depend on some research by yourself and some discussions with your doctor. Certainly avoiding NSAIDs is an easy first step. It is easy to get paralysed by the ocean of studies, but doing nothing may not be the right decision for your quality of life.

Comment author: Soothsilver 26 July 2016 05:53:53AM 0 points [-]

Thank you very much.