Suicide Risk and Smoking

[Dragonfly]

Population-Attributable Fractions of Axis I and Axis II Mental Disorders for Suicide Attempts: Findings From a Representative Sample of the Adult, Noninstitutionalized US Population -- Bolton and Robinson 100 (12): 2473 -- American Journal of Public

___________________________________________________________________

So, 8% of the suicide attempts could be attributed to smoking cigarettes. The article goes on to hypothesize that this may be that nicotine might reduce serotonin levels in the brain. And it is the lowered serotonin levels that are associated with suicide attempts. If this is found to be true in other studies, this is amazing!

 

[David Baxter PhD]

From the abstract:

Results. Of the 25 disorders we examined in the model, 4 disorders had notably high PAF values: major depressive disorder (PAF = 26.6%; 95% confidence interval [CI] = 20.1, 33.2), borderline personality disorder (PAF = 18.1%; 95% CI = 13.4, 23.5), nicotine dependence (PAF = 8.4%; 95% CI = 3.4, 13.7), and posttraumatic stress disorder (PAF = 6.3%; 95% CI = 3.2, 10.0).

That's quite remarkable: nicotine dependence is a higher risk than PTSD.

I haven't read the full article but off the top I wonder if that isn't the result of a covariate. For example, schizophrenic patients have a higher rate of smoking than the general populations, leading to hypotheses that the nicotine is somehow self-medicating. Could it be that it's the presence of certain other disorders associated with nicotine dependence rather than the nicotine dependence itself? Or is it truly the nicotine dependence itself, possibly because of its association with reduced serotonin levels?

 

[Dragonfly]

From the abstract:

That's quite remarkable: nicotine dependence is a higher risk than PTSD.

Yes. The authors comment on this remarkable finding, as well.

..... I wonder if that isn't the result of a covariate. For example, schizophrenic patients have a higher rate of smoking than the general populations, leading to hypotheses that the nicotine is somehow self-medicating. Could it be that it's the presence of certain other disorders associated with nicotine dependence rather than the nicotine dependence itself? Or is it truly the nicotine dependence itself, possibly because of its association with reduced serotonin levels?

Yes. I wondered about this as well - that nicotine use could actually be a surrogate marker for a major mental illness - such as schizophrenia, as you noted above. It is interesting. I went back and reread the article carefully .... the short answer is that I can't tell because of how the analysis were done. They used this relatively new statistical technique called PAF - population attributable fraction - which essentially looks at sequential attributable risk for an outcome. But the method, as with all stats methods, is not without its shortcomings. Here is an article that specifically addresses some of the shortcomings of this type of analysis:

Epidemiologic Perspectives & Innovations | Full text | Partitioning the population attributable fraction for a sequential chain of effects

In the article on risk for suicide attempts, indeed they used only single variable analysis - so there is no way to tell from the article if controlling for a psychotic disorder would decrease the relative risk of nicotine with respect to suicide attempts. Having said that, the authors present a relatively convincing argument that smoking cessation in the US would account for one million fewer suicide attempts per year. Imagine ....

 

[David Baxter PhD]

Thanks, Dragonfly. I have no idea what PAF is either. I'm not sure I want to... I started getting confused about statristics when meta analyses became the next big thing.

 

[Dragonfly]

..... I have no idea what PAF is either. I'm not sure I want to...

So .... here's where I confess that I used to read stats books. ..... Yup. Literally read them. From one cover to the other for enjoyment and relaxation .... And I confess that I still have several newer ones on my e-book reader. But I have been able to let that habit go Well. Mostly. But I do admit that every once in awhile, I find myself re-immersed in, and fascinated by, the world of numbers, equations, approximations, hypothesis etc. I am so glad that there is room for all types of beings on this planet of ours ....

 

[David Baxter PhD]

Yes, indeed.

One of my former graduate students loves statistics and still teaches it. I swear at her thesis defense, she was the only one who knew what she was talking about when she got to the advanced analysis part.

My graduate stats professor used to answer questions with an expression of puzzlement and the words, "But it's intuitively obvious!". I always wanted to smack him for that.

 

[Dragonfly]

I always wanted to smack him for that.

:rofl:

 

[CarlaMarie]

I'd rather smoke than read statistics.:rofl:
So the point is....