Antidepressants That Aren’t “Antidepressants”

(Written 10/2005; updated 6/2013)

Not Pills “Natural” pills Psychiatrist’s pills
Exercise omega-3 fatty acids(fish oil) lithium
Cognitive-Behavioral Therapy n-acetylcysteine(NAC) lamotrigine
Light manipulations optimize thyroid quetiapine

More on this table below.


Although this page will repeat some ideas you’ll find elsewhere on this website, they seemed worth presenting together. After all, these may be the most important concepts on this website. Consider the following logic:

  1. Depression is the most common symptom in bipolar disorder.
  2. Antidepressant medications (Prozac, Paxil, Celexa, Effexor, etc.) can be very effective against depression.
  3. Result: antidepressants are commonly used in bipolar disorder.

That’s where we are at the present time, in practice. But here is more logic to consider.

  1. Antidepressants can make bipolar disorder worse in several ways (some of these are controversial; here are the data on that).
    • They can cause hypomania where there was none.
    • They can induce cycling, or make it worse.
    • They may keep a person from becoming truly stable.
    • And they might, just might, cause some long-term harm, perhaps even irreversible harm.
  2. Therefore: antidepressants should be avoided, as much as possible, in bipolar disorder treatment.(This statement in particular is controversial).

Now, let’s combine these two bits of logic:

  1. Some antidepressant tool is commonly necessary in bipolar disorder.
  2. Yet true “antidepressants” should be avoided, if possible, in bipolar treatment.
  3. Therefore, one should maximize use of all other antidepressant approaches before using typical “antidepressants”.(Again, this is my personal conclusion, not an expert opinion or mandate).

Fortunately, there are at least nine such approaches to consider. These approaches are not associated with increasing cycling, long-term destabilizing, or the concerns about possible long-term harm which have been raised about typical antidepressants. They all have other risks, but so do any treatments you might consider. For any
approach you take, you will always be balancing risks versus possible benefits.

Before we turn to that list of nine options, however, there is an even more important principle to consider — because it might make those eight options unnecessary.

Central principle: first, stop the cycling

For many people, their depressions are not constant. Instead, they “cycle” into depression, and out again — although the respite is often brief. However, if it is clear that your moods do indeed cycle, then the first thing to do is stop the cycling. A lot of people will find that their mood smoothes out at an acceptable level.

I hope you see what this means: if you stop the cycling, you might stop having depressions entirely — so you wouldn’t need an antidepressant, of any kind. And this makes bipolar treatment a lot simpler, because you can stop the cycling with “any old mood stabilizer”, not just the ones (shown below) with antidepressant effects. This allows you to choose from the entire menu. That’s good, because sometimes you’ll want that entire menu available, to pick the one with the least troublesome side effects for you.

After that step is complete, when you are no longer having cycling, then, if you are still depressed, you should consider options from the list below. Then, when you have considered these options and perhaps used one or several, then you may be one of those people who really need to add an antidepressant, with caution, to your mood stabilizers — in my opinion. (There are exceptions, including especially people who have been on an antidepressant for years and doing well, with no cycling. These people may do best to stay on their antidepressant, although even that is controversial: see Controversy 3.)

But before turning to antidepressants, there are at least 9 alternatives. Remember them as “3 columns of 3″.  Well, at least 3 in each; the lists are growing, which of course is good.

Nine antidepressants that aren’t “antidepressants”

(Each of the “pill” options in this table has its own page of additional information; see Mood Stabilizers table of all the options.)

Not Pills “Natural” pills Psychiatrist’s pills
Exercise omega-3 fatty acids(fish oil) lithium
bipolar-specific psychotherapies n-acetylcysteine(NAC) lamotrigine
Light manipulations(dark therapydawn simulator, chronotherapy)(some wish to include ECT here as well) optimize thyroid quetiapine (lurasidone too, soon)

Okay, I admit, dark therapy doesn’t quite belong on this list, as it does not have direct antidepressant effects, only mood stabilizing effects, but it’s such a cool idea, I think everybody needs to know about it — because some limited version of it is within reach for almost everyone. If you’re interested in how mood is affected by light, see an essay on Bipolar Disorder: Light and Darkness. You’ll see how the biological clock is affected very directly by both (and find interesting links about how the clock works and how lithium affects it; and about why blue light appears to be more important than any other wavelength when it comes to mood effects).

Similarly, optimizing thyroid — getting near the hyperthyroid side of normal, or at least making sure you’re not near the hypothyroid side of normal — is not directly antidepressant but may allow whatever antidepressant things you’re doing to work better. It too is on the list because I think it is so nice to be able to offer options which carry zero, or near zero risk (as long as you don’t end up becoming hyperthyroid, which should be relatively easy to avoid). For more, see Thyroid and Bipolar disorders, including the references on which this approach is based and a reminder that this is not even an experimental approach, just a working guess.

Finally, lithium could be listed in the “natural” column. Okay, that’s a stretch, perhaps, as it is made by pharmaceutical companies and must be handled just like other medications, with caution. But it does come from the earth, after all.

A brief word on some medications that other psychiatrists might have included in this table.  I have not included risperidone and valproate in the table above because:

  • I find risperidone actually has too much antidepressant-like action and may induce a subtle agitation/insomnia/cycling that is hard to manage, because one will always be wondering if it is coming from the patient or from the risperidone;
  • although divalproex/Depakote has a recent pilot studyDavis demonstrating an antidepressant effect, as well as a small study of less direct designWinsberg, it has not impressed me with its antidepressant effects compared to those three listed.)


I hope this essay may have led you to wonder: with all these options to manage depression in bipolar disorder, and with the risks antidepressants pose (or may pose, depending on how you interpret the current controversies), why does anyone persist in using antidepressants in bipolar patients — at all?

Obviously I wonder that myself. As I listen to doctors talk about this, it seems there are several factors:

  • Habit: we’ve been using them for a long time, and they do help quite a few people (even many people with bipolar disorder — initially).
  • A short-term view of treatment, focusing on the symptom du jour, instead of the long-term goal of stopping cycling as a means of addressing repeated episodes of depression. It’s easy to get drawn into treating today’s symptom, when you’re sitting with a patient who is really depressed, who wants an antidepressant, and who has not learned all you have learned about antidepressants’ possible risks.
  • A different interpretation of the available data on risk. You might hope this would be the main explanation. However, I fear that many doctors are unaware of the data we have on the risk controversies.

(Many people are already on an antidepressant. Before you think about going off of it on your own, you might want to read some thoughts from a colleague about how to do that — not on your own please).