If you’ve been offered an antidepressant, this is probably the question you actually want answered. Not “do they work” — you can find a thousand articles on that. You want to know which antidepressant with the least side effects? The ones ithat are least likely to make you feel strange, wreck your stomach, or leave you flat.
Here’s the honest answer, and then the reasoning behind it.
The short version
For most adults starting treatment for the first time, Sertraline has the best balance of tolerability and effectiveness. In fifteen years of prescribing it in general practice, I’ve had very few patients tell me they hated it. That’s not a claim I’d make about every drug on the list.
That said, “fewest side effects” is the wrong question if it’s the only one you ask. The better question is: which side effects am I least willing to live with? Because the drug with the mildest overall profile isn’t automatically the right one for the person who can’t sleep, or the person who’s already tried two tablets that did nothing.
What the research actually says
Two large studies dominate this area, and it’s worth understanding what each one measured — because they don’t measure what most people assume.
The 2018 Lancet study (Cipriani and colleagues) pooled 522 trials and nearly 117,000 patients across 21 antidepressants. On tolerability, agomelatine, citalopram, escitalopram, fluoxetine, sertraline and vortioxetine came out best, while amitriptyline, clomipramine, duloxetine, fluvoxamine, reboxetine, trazodone and venlafaxine had the highest dropout rates. Cipriani and Collegues
Now the important caveat, which you’ll rarely see mentioned. The study’s tolerability outcome was all-cause discontinuation at eight weeks — in other words, how many people stopped taking the drug for any reason at all, used as a stand-in for how well it was tolerated. That’s a reasonable proxy. But it isn’t a measurement of side effects. Someone who quits because the tablets weren’t working counts the same as someone who quits because of nausea.
The 2025 Lancet study (Pillinger and colleagues) filled part of that gap. It combined 151 clinical trials and 17 FDA reports covering more than 58,000 participants, looking at what 30 different antidepressants do to the body over roughly eight weeks. The differences were larger than expected — up to a 4 kg gap in average weight change between drugs, from around 2.5 kg of weight loss on agomelatine to about 2 kg of gain on maprotiline. The authors concluded that treatment guidelines should be updated to reflect these differences in physical risk, while stressing that the choice still has to be made individually.
So: one study tells us which antidepressants people stay on. The other tells us what those drugs do to weight, blood pressure and heart rate. Between them, they still leave out three of the things patients complain to me about most — sexual side effects, emotional blunting, and what happens when you try to stop. Sexual dysfunction was explicitly excluded from the 2025 analysis, and it’s unclear whether the physical changes it did find persist beyond the short trial period.
That’s not a criticism of the research. It’s a reason not to treat a league table as a prescription.
Personal experience. What people actually report to me in clinic
When a patient comes back at two to four weeks having stopped their tablets, this is roughly the order in which the reasons come out:
- Diarrhoea or an unsettled stomach. By far the most common early complaint with SSRIs.
- Dizziness or light-headedness, especially standing up quickly in the first week.
- “It makes me feel weird.” Hard to pin down, and patients often apologise for not being able to describe it better. Don’t apologise — it’s real and it’s common.
- “It messes with my head.” A sense of being not-quite-yourself, foggy, or oddly detached.
- Headaches. SSRIs change serotonin levels in your brain and body, which can trigger mild or temporary headaches as your nervous system adjusts
Notice that none of these are the things the newspapers write about. They’re mundane, they’re mostly early, and — this is the part worth knowing — the first three usually settle within one to two weeks if you can ride them out. Most people who stop at day four or five stop just before the point at which it would have got easier.
The four I prescribe most, and why
Sertraline. Often my default first choice. The side-effect profile is genuinely mild, it’s safe in most people with heart conditions, and it’s cheap. The trade-off is diarrhoea in the first fortnight for a fair number of people.
Escitalopram. In my experience it tends to be the more potent of the two, so I reach for it more often when the depression is severe rather than mild-to-moderate. It also performed well on tolerability in the 2018 analysis. The catch is that it needs a bit more care at higher doses and in older patients, because of its effect on heart rhythm.
Mirtazapine. This is the one I choose when insomnia is a dominant part of the picture — and I use the sedation and appetite effects deliberately, not as an unfortunate by-product. If you’re not sleeping and you’re not eating, a drug that makes you sleepy and hungry is doing two useful jobs at once.
There’s a quirk here that surprises most patients: Mirtazapine is generally most sedating at the lower dose, around 15 mg. Go higher and the sedation often eases off, while the antidepressant effect strengthens. That’s the opposite of what everyone expects from a tablet. The pharmacological explanation is that at low doses the antihistamine effect dominates; as the dose rises, other actions come to the fore and partly offset the drowsiness. If your sleep is the problem, going up may not be the answer. If your mood is the problem, it may be. That’s a conversation to have with your prescriber rather than a decision to make yourself.
Vortioxetine. Good tolerability data, but I generally hold it back until two SSRIs have failed. In private practice there’s an additional factor nobody writes about honestly: it’s considerably more expensive than the alternatives, and for a patient paying out of pocket that matters. A drug you can’t afford to stay on for six months isn’t the best drug for you.
Which ones give people more trouble?
Not “bad drugs” — drugs with a heavier profile, which sometimes buys you more effect. Venlafaxine and duloxetine had among the highest dropout rates in the 2018 analysis. Venlafaxine can nudge blood pressure up. Paroxetine has a reputation for difficult withdrawal. Fluoxetine is well tolerated but was among the less effective in head-to-head comparisons.
Any of these can be exactly right for the right person. The point is that they warrant a specific reason, not a default.
“Most people who stop at day four or five stop just before the point at which it would have got easier.“
Five things worth knowing before you start
Give it two weeks before judging. The stomach upset and the dizziness tend to fade. The benefit tends to arrive later — usually between two and six weeks. The early experience is a bad predictor of the eventual one. This is the most important thing I tell my patients. The initial side effects almsot always disappear by two weeks.
Take it with food, and don’t start on a Monday. Both reduce the odds of the first few days derailing you.
Weight is a fair thing to raise. The 2025 data confirms the differences between drugs are real and measurable, so if weight matters to you, say so at the outset rather than six months in.
Ask directly about sexual side effects. They’re common with SSRIs, they’re barely captured in the headline research, and most consultations skirt around them. A prescriber who can’t discuss it comfortably isn’t giving you a full picture.
Never stop suddenly. Stopping abruptly can cause dizziness, electric-shock sensations, irritability and flu-like symptoms. It doesn’t mean you were addicted, and it’s largely avoidable by reducing gradually with your GP’s guidance.
The honest conclusion
If someone forced me to pick one answer for a first-time patient with straightforward depression: sertraline. The evidence supports it and everyday experience supports it.
But “fewest side effects” is a starting point, not a destination. The right antidepressant is the one whose particular trade-offs fit your particular life — your sleep, your weight, your other conditions, your budget, and what you’re least willing to tolerate. That’s a sixty-minute conversation, in of itself, not a search result. We also need to consider how severe your depression is and the potency of the specific SSRI. An impotent SSRI for severe depression is oftent not a good starting point, no matter how tolerable the side effects are.
Choosing between antidepressants is easier when the diagnosis is clear. If yours hasn’t been, a comprehensive psychiatric assessment is usually the more useful place to start.
This article is general information and not a substitute for individual medical advice. If you’re struggling with your mental health, speak to your GP. If you’re in crisis, contact your GP urgently, call 111, or call Samaritans free on 116 123 at any time.
