one side effect of SSRI's seems to be an increase in suicidal ideation / urges. given that these are often precisely the reason people are prescribed the effect seems somewhat... paradoxical, shall we say?

another side effect of SSRI's seems to be all those problems people typically report while they are on the medication and also while they are trying to stop taking it. sexual problems, headaches, nausea, weight alterations, brain zaps and so on.

i'm fairly sure that the 'harm' to children was due to an increase in suicide for those taking the medication compared to those taking placebo (where both groups were indeed matched for severity).

now people are looking at the increase in suicidal urges / ideation for those adolescents taking the medication compared to those taking placebo (where both groups were indeed matched for severity).

surely the other side effects also count as likely harms that should be weighed against likely benefits for taking them.

and when calculating the benefits don't forget to calculate the differential between non-active placebo and active and also (even more importantly) non-SSRI active placebo and active. the last ensures the double blind isn't broken when people experience unpleasant side effects, you see...

the trouble comes from... firstly, the sheer amount of money that the drug companies make off medication sales. they are of course reluctant to include warnings on their product.

the drug companies put a lot of money into sponsoring research studies. if you are a researcher there is a huge difference in how much a university will pay you for your studies and how much a drug company will pay you for your studies. if you want more money (which is surely fairly rational) then you would be wise to conduct studies which (are likely to) support the efficacy of the product being marketed by the drug company.

drug companies are required to disclose information to the effect that their medication is harmful. one can of course get around this by the way one interprets the statistics. for example, one can report 'this study found that overall participants reported less symptoms of depression when taking xxx compared to non-active control' even when (at the very same time) one could also report 'this study found that in the 5-15 age group participants were twice as likely to commit suicide when taking xxx compared to non-active control'. the statistics are the same. it is the INTERPRETATION of the statistics that can be problematic. or perhaps more accurately, the fact that we do need to be selective in what conclusions we draw from a study (when there are an indefinate number of conclusions that could be drawn). people do of course draw the conclusions that they most want to / that seem most surprising.

drug companies are not required to disclose findings that show their medication to be ineffective. if a company replicates one study 50 times and finds that on 49 trials medication xxx is no more effective than placebo they will not report the findings of those 49 trials (they are not required to report the findings of those 49 trials), instead they will report that they have shown their medication to be more effective than placebo (which they have indeed shown on that one trial).

20% of scientific studies (across the board) are rejected due to people pleading 'methodological flaws' or 'contaminated samples' or some random %#@&#! up. those findings are not reported to the scientific community. what is scary is that when less than 20% of the drug company sponsored studies report that the medication is more effective than placebo this is the finding that is reported to the scientific community, the doctors, the public. when this is the situation... it makes it more likely that the studies showing medication to be effective are the mess ups and the 80% showing no effect (or fairly clearly showing harm but being interpreted as showing no effect) are the most replicable.

the stuff that is coming out now... isn't due to any new trials that have been conducted. it is due to people doing a literature review of the studies that have been conducted thus far (studies sponsored by drug companies in fact) and basically... reinterpreting the stats. how is it that we never interpreted the stats this way in the first place?

there are none so blind as those who do not want to see...

and of course even if the medication is 40% effective and placebo is 100% effective it is considered UNETHICAL TO PRESCRIBE A MEDICATION THAT IS IN FACT A PLACEBO / INERT. now... is that due to good medical practice or due to politics / profits do you think? you can't patent a placebo... can you?