Not outperforming a placebo means they don't actually do anything.
SSRIs are potent pharmaceutical drugs that have multiple, very strong effects.
For example, Setraline (SSRI) is one of the most effective drugs to treat premature ejaculation. It used to be prescribed for exactly this purpose (and maybe still is). If anything, this is one of the most pronounced and easily measurable effects of SSRIs.
If I took 25mg of setraline per day (extremely small dose) for a week, it would become impossible to ejaculate.
It also has a whole multitude of other effects that are more dependent on individual (age, gender, bodyweight, etc), some other effects for males are -> crushing testosterone levels, weakness.
For me personally, it turns me extremely, extremely emotional, makes me able to sleep all the time, and mentally turns you into a zombie where your lifes dreams and aspirations just disappear.
Does it help with depression? I would say no. That however doesn't mean it doesn't do anyting, because they do all sorts of things. It is a very potent and poorly understood pharmaceutical.
For example, cherry pick people who have the flu. Some people in the placebo group curing their flu in the study period shouldn't be surprising. That happens by default.
If they are comparing the effect of placebo with no treatment at all, then no, the measured effect of placebo actually can't be that. The reversion to the mean effect would appear for the no treatment at all people too.
So no one really knows, in a scientific, quantifiable sense, how people who suffer from depression evolve without treatment.
One problem with that is that SSRIs are much more accessible than therapy. In my country, they are fully covered by social security and procuring them is relatively easy, while therapy is too expensive for lots of people.
> If you read further in your [1] you'll see that the rate of side effects is not "in excess of 70%" but lower, and it depends on both the medication and the dose. Switching medications and changing doses is often sufficient to ameliorate some or all of these effects.
In my case, and according to studies I'm not the only one, the side effects can be persistent. Neither the doctors nor the medication's notice warned me of this. If I knew that when I was young, I would have made different choices.
SSRIs did help me, but it cost me a lot, and I still cry about it. Of course, there is no way to know if this is a better outcome for me than what would have happened if I managed to refuse to take them.
"Not much better than placebo" is burying the lede.
The real problem is that placebo performs very well in depression studies. It's a well studied phenomenon.
Effective antidepressants are marginally better than placebo in the studies because the placebo group improves so much, not because the antidepressants don't do anything.
> and are shown to be about as effective as therapy — which is actually durable.
False dichotomy. The recommendation is for people on SSRIs to also do therapy.
You don't have to choose one or the other.
> There’s also rates of sexual side effects in excess of 70% [1]
If you read further in your [1] you'll see that the rate of side effects is not "in excess of 70%" but lower, and it depends on both the medication and the dose. Switching medications and changing doses is often sufficient to ameliorate some or all of these effects.
That paper also mentions newer alternatives such as Vilazodone (SSRI plus 5-HT1A action) which are shown to have lower incidence of these side effects.
> and they cause weight gain which is separately associated with depression.
SSRIs aren't really associated with weight gain once you exclude the older ones like Paroxetine which have anticholinergic effects. A lot of studies find statistically insignificant weight loss or slight gain.
Weight gain is really a negligible decision factor in modern SSRI treatment: https://www.ccjm.org/content/ccjom/70/7/314.full.pdf
There's a lot of misinformed fear mongering in your comment.