> By LOTS of anecdotal reports - what do you mean?

What I mean is that if you find ONE anecdotal report that 'I tried to stop taking Effexor and I had severe nausea' then even if the nausea was due to the Effexor (rather than eating something bad, say) then one is probably unlikely to experience that side-effect oneself. If only one person in every hundred thousand people experience a side-effect you probably don't need to worry about it too much.

If there are lots and lots and lots of anecdotal reports then it becomes less likely that they are ALL due to other things (eating something bad, having a bad day etc) and it becomes more likely that they are indeed effects of people coming off the drug.

Of course what we would most like is follow up studies for every single person who has ever taken Effexor. We would like to know:
- How much were they taking?
- How long were they taking it for?
- How much did they reduce the medication for and over what amount of time?
- What experiences did they have that they think was related to the reduction in medication?

Do doctors compile these statistics? No, they do not.
Do medication companies compile these statistics? No, they do not.
Do researchers compile these statistics? No, they do not.

Do they do the studies on a smaller scale? Not really. Why? Because most of the funding for studies comes from the drug companies. If you want some research funding to do a project then you won't get funding to do your research if you don't design a study that the funding companies like. This is what some academics refer to as 'selling out'. It is tempting for scientists (chemists and nuclear physicists and defence researchers in particular) to 'sell out' aka to make one hell of a lot more money producing findings (or applications) that people are prepared to sponsor...

(Philosophers can sell out by writing dence and incomprehensible rationales for stupid policys and decisions on where to spend the funding and so on and so forth)

Even if there is a lack of scientific research to show that people have severe withdrawal from Effexor that doesn't mean that people don't have severe withdrawal from Effexor (it is possible the studies have not been done).

By analogy: Even if there is a lack of scientific research to show that cellphones are harmful that doesn't mean that cellphones are not harmful (it is possible the studies have not been done).

By analogy: Even if there is a lack of scientific research (as there surely used to be) that smoking is harmful that doesn't mean that smoking is not harmful (it is possible that the studies haven't been done and it is also possible that the political clout of the tobacco industry (or drug companies) mean that 'suspicions' that it is harmful simply aren't followed up on with respect to doing the studies.

Drug companies have to disclose findings that support the hypothesis that their drug is HARMFUL. This can be avoided by manipulating statistics, however.

There was one anti-depressant that was shown to be better than placebo if you compared people aged 14-45 on the drug to people 14-45 off the drug (the precise numbers have been changed because I don't remember but the take home message is the same).

Trouble was that if you homed in on the 14-19 age group... More people committed suicide on the drug than off the drug.

They only homed in on the statistics (examined them more thoroughly) after several parents sued the company after their kids committed suicide on the drug. That was part of the reason behind the proposed 'warning this antidepressant may cause suicide' label. It was also part of the reason behind the proposal that drug companies should have to disclose ALL of their findings and not just the ones that were favourable.

See how manipulation of statistics alters the 'our drug harms' vs 'our drug helps' conclusion?

I don't have a beef with science...

But it is surely wise to think critically about it.

Reading numerous anecdotal reports is one way around this. I've heard a number of people say that when they told their doctors that they were having nausea etc from reducing Effexor their doctors told them 'rubbish - no side effects have been reported from Effexor withdrawal'. That is changing now. Doctors are starting to take Effexor withdrawal seriously (and actually report them to their peers etc). But for a while there... Doctors didn't report their patients withdrawal symptoms because they REFUSED to believe there were any. That is called a confirmation bias and it is BAD science (there is a lot of bad science out there especially when there is a lot of money to be had in the vicinity)

Wise not to over-estimate science. Especially when there is such a political agenda behind it.

Don't get me wrong - medication does indeed help a lot of people out there and I think that most people do go on it in something of a last resort kind of way. Still, I think people should be informed about what it is they are getting themselves into. Anecdotal reports are fine. What you want, however, is LOTS of them. And... Try and think critically about them (e.g., people who are disgruntled are more likely to be vocal than happy customers... Perhaps?)