If psychiatric hospitalization is so effective, why is the suicide rate highest immediately after release from the hospital?
A massive study (1,185,727 patient-years) published in January ("Higher-risk periods for suicide among VA patients receiving depression treatment: Prioritizing suicide prevention efforts," M. Valenstein et al., Journal of Affective Disorders, Volume 112, Issues 1-3, January 2009, Pages 50-58) on a military veteran population being treated for depression found that suicide rates were dramatically elevated immediately following a psychiatric hospitalization.
The base suicide rate of the population was found to be 114 per 100,000 person-years. Clearly, the depressed VA patients are at a much higher suicide risk than the general population of the United States, for which the suicide rate is estimated at only 16.7 per 100,000 person-years. But the suicide rate for these depressed veterans shot up to 568 per 100,000 person-years during the 12 weeks following a psychiatric hospitalization - five times the already high base rate for the non-hospitalized depressed veterans, and 34 times that of the overall American suicide rate.
But if hospitalization were actually effective in "treating" suicidality, wouldn't we expect the suicide rate to be quite low after a hospitalization?
One problem with this line of thinking is that we might expect only the most seriously suicidal patients to be hospitalized at all. One hypothesis is that hospitalization is effective in reducing suicidality, and that the suicide rate of hospitalized depressed veterans would have been much higher if they hadn't been hospitalized. No study I am aware of attempts to compare the suicide rates of patients who meet criteria for hospitalization, but who are randomly assigned to be or not to be hospitalized. Given the widespread faith in hospitalization as a suicide treatment mechanism, to conduct such a study would probably be considered a breach of professional ethics toward the non-hospitalized group.
However, in the absence of data from such a study, I think the alternative hypothesis needs to be considered: that hospitalization is so horrible, demeaning, and above all ineffective, that it does nothing to prevent suicides and may actually increase one's resolve in that direction. Far from showing caring and compassion, forced psychiatric hospitalization demonstrates to the patient that he is a prisoner. For a prisoner, there is a clear method of escape. Yes, there are people who claim to have benefited from involuntary psychiatric hospitalization, just as there are people who claim to have benefited from childhood beatings and from those wilderness camps they send bad kids to. But there are also people who have suffered involuntary hospitalization and found it to be a life-changing, demeaning experience. In fact, I think we must be suspect of the "glad it happened" group. The psychological defense mechanism of denial, the discomfort of cognitive dissonance, contribute to people interpreting past events with unwarranted optimism. As long as you can convince yourself that the involuntary hospitalization was good for you, you don't have to admit to yourself what an insult to your dignity was done to you.
Sadly, the authors of the study are using the results to recommend yet more coercive practices. What is really needed is more intensive "treatment" following a hospitalization, they say - or a "firm connection to outpatient services," in the Orwellian words of the study's authors.
For those whose link to the study is gated, here's a Washington Post summary of the study: "With Depression, Vets Face Higher Suicide Risk."
Update: Commenter Jessa continues the discussion on coercion at her site, Made with Awesome.
Update: Zarathustra responds to my arguments here.
Update: Jim adds to the discussion here and here.