In the same JCP issue in which atrocities were committed, McGirr and friends looked at 351 Canadian  consecutive suicides, and then performed a psychological autopsy to find out what had been wrong with them.

Almost all of the women killed themselves with two methods: overdose (46%); surprisingly,  hanging (38%).  For men, it was hanging (53%) or firearm a distant second (16%).  I don't know what's going on in Montreal, but it's different than LA.  (Less guns?  More trees?)

In comparison to men, women were more likely to be college grads and have jobs, to have a lifetime history of depression or anxiety, but less likely to have ever abused alcohol (26% females vs 44% males).

In the six months prior to the suicide,

Depression:   males 52%; females 56%

Anxiety: males 10%; females 15%

Alcohol: males 31%; females 18%. 

So there it is, more than half of suicides were depressed at the time of the hanging/shooting/OD.

Which is fine, but there is one statistic the authors neglected to report:

Number of patients who had been in psychiatric treatment at the time of death: 10. 

The number 10 doesn't appear in the study, and repeated attempts to get the actual number from the authors were failures: "we don't have systematic data."  Ok: the same group put out another study: out of 422 suicides, 28% had been to psychiatry in the past year.  Let me translate: 70% had not. 

If a tree falls in the forest, and no one hears it fall, shouldn't we get some guys out to the forest? 

---

Addendum: in the Oct 2006 Am J Pub Health, the authors find that suicide rates have been decreasing-- dramatically-- especially for the elderly since 1985 (from 21/100k to 16/100k) and youth since 1995 (14/100k to 10/100k).  But it's worth repeating that the number of actual suicides is still very small.