This is just a follow-up of a study where the same result has already been published. It is remarkable how the lack of evidence of any beneficial effect whatsoever from their pills gets spun into a promotional talk for the pills. It gives the impression that their pet theory for why they failed to find the result that they wanted was because the patients did not adhere to the pill regimen. Adherence can be physiologically monitored, but it requires blood samples - a lot mor ecomplicated than just asking the patients, and looking at their pill bottles to see how many of their 14 pills remain after two weeks.

The discussion of the inflammatory markers is shalow, but the topic has great potential for explaining the marked survival difference between depressed and non-depressed people post-ACS. The topic to search is "pro-inflammatory cytokines sickness behavior." However, no one is really saying that a treatment for depressogenic cytokines is an SSRI. SSRIs do have a role in interfering with healing-type inflammatory repsonses -specifically, making blood less "sticky." but I believe the science has not yet caught up to the enthusiasm of the SADHART ppl.

Psychotherapy for post-ACS has been tried, in a similar trial, similar timeline, also, called "ENRICHD." The study found no survival benefit for post-ACS ppl receving talk therapy. However, this trial had a handful of "fatal flaws," including the most striking flaw: the "depression" group was 20% depressed ppl, 20% ppl with minor depression, and 25% of ppl with low social support. So, because ppl have noted that depressed ppl post-ACS die quicker, they decided to test a psychotherapy intrevention with ppl who have poor social support? Pharma has likley taken the failure of a talk therapy intervention as promising for pills, but has so far failed to find a pill effect. But they will keep on trying.

I think there are two ways to look at it. Perhaps Pfizer is hoping to get headlines: "Depression linked to post-heart attack mortality." Not reading the study, people think, "Depression is bad; antidepressants treat depression; therefore, I should take an antidepressant, like Zoloft, after my heart attack."

On the other hand, it could be an honest study that will lead to other studies, with a psychotherapy arm and with other drugs. The claim that treating depression will help isn't proven by the data, but it's not disproven either (except for Zoloft). You can expect more studies -- probably a few dirty ones -- trying to capitalize on this new market, though.

Philip, I think the reason you "don't know what to do with" this study is simply that it's an insignificant piece of research. SADHART was a major clinical trial (yes, funded partially by Pfizer) and this is just a follow-up looking at what "features" of depression were associated with death within the next 7 years. Basic epidemiologic, number-crunching stuff, really (but publishable... these guys have to make tenure somehow!!).

As David accurately pointed out, the study did find that more severe depression was correlated with increased mortality, but there are numerous possible explanations for this (and just as many ways to treat that depression-- not just Zoloft). The problem is that only Pfizer has the $$ to do the study and get it published, so it comes off as a not-so-subtle advertisement for Zoloft.