Treatments for generalized anxiety, medications and non-medication options, as ranked by the hosts. From the Psychofarm podcast, July 2026: 27 ranked, each with the reason and the moment it was said.
A real treatment that brings overall anxiety under much better control when done right, ideally with a professional guiding the person.
“Mindfulness Mindfulness and Essence. S tier. This is an actual treatment. If you do this right, and don't get me wrong, there are a lot of ways to do this wrong, which is why often it helps to have a professional guide you through it”
Self-directed CBT books that work much like mindfulness practice and are a great way for most people to improve their anxiety.
Placed together. One host finds Feeling Great more like homework and The Happiness Trap lighter.
“This is essentially It's very similar in my - my mind to mindfulness practice because it's essentially CBT, self-directed CBT. This is also In a great way. Way for most people to improve their anxiety problem”
They work and relieve anxiety, but mostly trade one problem for another through dependence and side effects, and they weaken the learning that therapy depends on.
One host wanted B, the other F; they settled on C for now. Both agree they work well in the right short-term patient.
“you're almost inexorably going to have either benzodiazepine dependence and side effects in exchange for the anxiety, or you will develop almost a psychological dependence on the benzodiazepines or both. So it's just a matter of harm. It's high efficacy”
Not a medication, but medication works poorly at best without an accurate diagnosis, so it comes first.
“To be honest, this is, uh... the most important thing. Okay, not a medication. Medication, but you can't give a medication unless you have an accurate diagnosis”
Not found very efficacious in one host's experience; reserved mainly for comorbid migraine, and the other host reaches for it with pain or fibromyalgia.
Moved during the episode: first talked about as a B or B plus, ended on the board in C. The other host would have kept it in B.
“This is a personal bias. I've used it a lot, and I just don't find it to be all that helpful. The one group where I'll use it is comorbid migraine disorder with depression or anxiety diagnosable”
Seems to work quite well, but it has a very severe withdrawal syndrome, which keeps it out of a higher tier.
“Venlafaxine and paroxetine are very efficacious in my experience. Not saying that there's any clear evidence that's better or worse than the other ones in the same class, but the fact is they seem to work quite well”
Anticholinergic effects, weight gain and the worst pregnancy profile among SSRIs make it worth avoiding for the average prescriber.
Sedating with some sleep support, and seen as efficacious, but it is the least prescribed of the group.
“So I think I don't like the anticholinergic effects. I don't like, um, there's weight gain. Uh, it's not good. It's got the worst, I think it's the worst SSRI for pregnancy”
Seen as underrated: as good as a benzodiazepine for anxiety in the studies, low risk, no sexual side effects, and few long-term side effects; held to S minus because of twice-daily dosing.
One host used to think it was a scam medication and now is among its biggest believers; the other has since come around.
“So basically low risk, a little, I put it at S minus because you have to do it twice a day and people don't like to take medications twice a day a lot of the time, but don't forget buspirone”
Stimulants cause and worsen physiological anxiety, so reducing them is worth considering, with care when they treat a true neurodevelopmental condition.
Caffeine was included in the discussion.
“So it's a very delicate balancing act. So it's not an S tier, but you should definitely think about it, especially in today's world where stimulants in the United States, at least, are probably highly overprescribed”
Good for physical and mental health and a helpful adjunct, though not a primary treatment for true generalized anxiety.
Hosts disagreed: one host wanted B, the other S; A was the landing spot.
“Stop putting an S here. I'll stipulate to the A, okay? But it's just not a primary treatment for anxiety problem in the group that has such high physiological anxiety that they genuinely meet what I would diagnose as a generalized anxiety disorder”
Effective and well tolerated with no sexual side effects and good for sleep disruption, but weight gain holds it back; weaker evidence for anxiety than for depression.
“No sexual side effects but weight gain, weight gain and what are you going to do? If you need someone that wants to gain weight or could use some weight or doesn't mind some weight gain and they have some sleep problems, mirtazapine, it's a good choice”
Low risk and may help with allergies, itching or sleep, but it makes people too sleepy and is not a good PRN.
Hosts disagreed: one host wanted B, calling it the best PRN.
“It is a C tier. First of all, you know I don't believe in PRNs. Right. So that's why I'm not going to pick it. And second of all, I think it's actually a terrible PRN, though that might be because my patient population”
Probably works about as well as buspirone or worse, and is too expensive while brand new, so wait for it to go off patent; its once-daily dosing is a plus.
“I suspect it works about as well as Buspirone or perhaps worse because it's probably less antipsychotic. Than buspirone And it's way too expensive because it's brand new. So there's no good reason to use it”
Looks risky and acts on many things in the brain, with possible habit formation and liver toxicity raised.
One host first said D, then F, since the board has no D tier; the other host had little to say about it.
“it's very much a kind of this looks kind of risky and it does a lot of different things in the brain. I'd personally put it on D, but that means I would probably put it F tier to be honest”
Low risk with minimal side effects and little dependency risk, though the data is small and benefit may fade after months.
Hosts disagreed: one host wanted A, the other B.
“We've got this medication that's got very minimal side effects, minimal dependency risk. If you need something that someone's already loaded up and they're begging you for something else, it's a great option”
The traditional research-based last resort for refractory generalized anxiety, with a real evidence base, held back by many side effects.
Went back and forth between A minus and B plus or B; nortriptyline would be the first TCA to try.
“This is the traditional research-based method of last resort for generalized anxiety disorder, refractory to first-line treatments. Um, It probably belongs in A. Really, it should just be TCAs”
These rankings are the hosts’ and guests’ opinions from one episode, for clinicians and the curious. They are education, not medical advice, and not a guide to choosing a treatment for any one person.
Educational content only. These resources are for education. They are not medical advice and not a substitute for clinical judgment, current guidelines, or individualized patient care.