This transcript has been edited for clarity.
Stephen M. Strakowski, MD: Hello and welcome to all of you viewing this video and reading the transcript. I’m thrilled to have two illustrious guests here to raise the quality of today’s conversation, which is about recent trends surrounding the decreasing use of lithium in the treatment of bipolar disorder.
Our first guest is Dr Kay Jamison, a professor of psychiatry at Johns Hopkins School of Medicine and a well-known author of many books about bipolar disorder. Kay’s groundbreaking work has helped redefine how we think about this condition so that everyone, not just physicians, can better understand it.
Our second guest is Dr Michael Ostacher, a professor of psychiatry and behavioral sciences at Stanford, and the director of the Bipolar and Depression Research Program at the VA Palo Alto Health Care System. Michael has the unfortunate personal experience of having known me since medical school, so I apologize for that.
Welcome to both of you, and thank you for talking today about this important topic.
In a 2020 publication in the American Journal of Psychiatry, Rhee and colleagues reported something that several others have as well in the past 20 years or so, which is that the use of lithium for the treatment of bipolar disorder has dramatically declined. Unfortunately, even back in 1997, the period when Rhee and colleagues began their analysis, lithium wasn’t as commonly used as one might imagine. But since then its usage has gone from approximately 30% to < 20% of individuals being treated for bipolar disorder. In contrast, there has been a dramatic upswing in the use of so-called second-generation antipsychotics, from 10% in 1997 to > 50% in 2016. That’s what we want to talk about today.
Dr Jamison, what do you think about that? Is it okay that we apparently decided to abandon lithium in the care of people with bipolar disorder?
Kay Redfield Jamison, PhD: Not only is it not okay, it’s deeply concerning. First and foremost, bipolar disorder is a potentially lethal illness. One of the things that we know about lithium from more than 100 studies by now, and many meta-analyses, is that lithium is the only drug that we have that really acts against suicide. So that’s concerning.
Also, when lithium works well, it works extremely well. I think it’s gotten a very bad reputation as being, in other people’s words, “toxic,” “old fashioned,” “cumbersome,” and so forth. I think none of those things is warranted.
And lithium has nobody to promote its use. It’s not a drug that makes anybody any money, but it’s an effective drug and it has neuroprotective effects. I think it’s really bad that it’s not more widely used.
Strakowski: Michael, do you have anything to add to that?
Michael J. Ostacher, MD, MPH: I think Kay has put her finger right on many of the problems that have led to the decrease in lithium prescribing.
One is this narrative, which I think is probably a false one, about lithium being a poorly tolerated drug with terrible side effects. I think that is a result of a form of anti-drug marketing that’s taken place.
It’s also true that there’s nobody to sell lithium, but there are companies that have a great interest in marketing their bipolar disorder treatments in a manner that pushes lithium out of the way. It’s not that these companies need to continue to market their drugs, many of which are generic now. But the narrative has been put out to the community, providers, and patients, and that simply doesn’t go away when the marketing does.
I’m on board with Kay about my concerns that lithium prescribing has decreased and that it’s dangerous for patients.
Jamison: I don’t think we know that lithium has a worse side-effect profile; it has different side effects. But all these medications, and certainly the antipsychotic medications, have very potentially problematic side effects. I think it’s just hard to change a narrative once it’s out there, but it’s really important that it be done.
What’s Behind the Decline?
Strakowski: Michael, why do you think this phenomenon is occurring? What’s causing such a dramatic decline in the use of a drug that — as I’m in agreement with Kay about — is uniquely useful in bipolar disorder?
Ostacher: It’s really hard to comprehend how the narrative around lithium has become established among the public and providers.
This idea that somehow lithium is worse because of its side effects comes from a lot of different places. Lithium has been compared to other drugs in a number of studies. Well, of course, those studies are funded by companies that have an interest in their product looking better. One of the ways they can do that is by making sure that the comparator (eg, lithium) is dosed in a higher range than necessary for patients to respond to it. And so, people will get tired, because lithium certainly has side effects, and it needs to be dosed very carefully.
I also think we’re in a risk-averse environment. Whether it’s true or not, if people hear that a given drug will be less likely to cause problems, and that the other drug will lead them to have bad patient outcomes and being potentially sued over that, they may choose to use drugs that they perceive to be safer in some way and avoid lithium.
The issue of suicide is a very important one. Kay is referring to meta-analyses looking at outcomes for death by suicide and by all causes in studies of people on lithium compared with placebo and other drugs. Those show that the overall death rates are lower for the people who are on lithium. Yet there’s still this perception that lithium is so toxic and dangerous to use, and in fact that you can’t give it to patients who are suicidal because they might harm themselves by overdosing with the very drug that they would benefit from.
It’s so complex to understand how patterns of prescribing have changed. But I think the point about the money involved is really important. Also, somehow this narrative that’s arisen in the context of — I don’t know if “litigious” is the right word — but in a setting in which people are afraid to practice good medicine takes place.
Jamison: I think an aspect of litigiousness is that if a patient who had bipolar illness commits suicide and it’s not in their chart that lithium was considered as a treatment, the potential litigation for that is considerable. People have to at least consider that as a possibility, particularly in somebody who has a history of suicide.
I also think the point about dosing is really important. The reputation that lithium got for being “toxic,” “mind numbing,” and so forth, certainly those things have an element of truth to them. But they come from earlier studies where people were kept at a much higher doses and blood levels of lithium. Those blood levels have been drawn back now, and people have far fewer side effects, far fewer cognitive effects, than they did when they were on higher doses of lithium.
The Downsides of Risk-Averse Clinical Decisions
Strakowski: One of the concerns is that potential renal issues follow this drug wherever it goes. What’s the state of the art when it comes to our understanding of that aspect now?
Jamison: People are monitored much more carefully now, and they’re more aware that that’s a possibility. The evidence that it’s a really serious concern is not very strong, compared with the concerns about not being on lithium.
Ostacher: I think the population-level data suggest that there’s not a great risk of lithium in terms of causing chronic renal disease. There are studies conducted in Europe that that bear this out.
It’s certainly the case that acute lithium toxicity can lead to acute renal failure, and I think that’s an important thing to consider. It ties into what Kay was talking about, which is this older concept that lithium levels needed to be run at a pretty high range in order to be effective. That puts the patients at risk for related issues that can elevate toxicity, such as taking a diuretic or nonsteroidal anti-inflammatory drugs, which can reduce people’s lithium levels. If people are running their lithium levels in a pretty high range, they’ll be at a higher risk for acute problems. But the evidence for chronic renal failure from lithium remains uncertain, even at population levels.
At the same time, people are much more frightened of negative outcomes than they are drawn by positive outcomes. People are really risk-averse in their decision-making. We know this from behavioral economics. I think the narrative that there could be some harm leads people to avoid using lithium when they might otherwise be inclined to do so.
Jamison: And I also think we overemphasize the possible negative effects of lithium but don’t mention the possible positive effects, namely, staying alive, staying well, and the possible neuroprotective effects. We don’t mention those to patients in the discussion of potential risks and benefits.
That’s really going to have to change at some point, as we’re learning more positive things about the effect of lithium on the brain. These effects are simply just not mentioned. Well, I think it should be mentioned, because it’s the brain that’s most involved.
Strakowski: The perspective from my own practice and research is that when lithium works and people are responsive, the response is qualitatively different from other drugs. People will truly be well in a way that they’re not when they’re on other antipsychotics, divalproex, or something.
Does that make sense or am I just imagining that because I’m old enough to think it’s a good treatment?
Ostacher: Well, there is the clinician’s bias: The patients that do well come back and the patients that don’t do well don’t want to see us again. There’s always that problem.
But I think your comments represent the experience of most people who work in the bipolar specialty, which is that lithium is remarkable for those people who respond to it.
This probably has something to do with the data from the study you first described, which indicates that the diagnosis of bipolar disorder has probably gotten somewhat broader. When this study was initiated years ago, bipolar II disorder — characterized by episodes of hypomania without mania and depression — wasn’t yet a diagnosis. It’s much more commonly diagnosed now, but we have limited findings to guide the treatment of those patients. Certainly, lithium has not been adequately studied in that group of people to determine whether or not it’s helpful for them.
But people who have bipolar I disorder — characterized by episodes of clear mania — really deserve to have lithium as their first-line treatment and have it prescribed in a way that allows them to tolerate the treatment not only acutely, but primarily for maintenance treatment. That’s where the benefit of lithium is greatest: in preventing episodes of future mood problems, whether mania or depression, and as Kay pointed out, in its potential anti-suicide effects.
Jamison: I think one of the things everybody is hoping for is to have more specific ways of predicting those people who are going to be good responders to the lithium vs those who probably are not going to be. We already know some of the things associated with good lithium response, but as time goes by, we’ll know more and more.
That’s also important, because not everybody is going to tolerate or respond well to lithium, but many people will. And it’s those people you want to reach and be as precise as possible about trying to predict who’s going to do what.
Strakowski: And that same comment can be said about every treatment for bipolar disorder.
How to Increase Lithium Prescribing
Strakowski: This video arose out of conversations we were having in the Bipolar Task Force and the National Network of Depression Centers, which is working on how to improve the care of people who struggle with bipolar illness. One thing would be to resurrect the use of lithium.
What are your thoughts about how we might reinvigorate the use of lithium among clinicians who clearly have become overly cautious, unfamiliar, or unwilling to prescribe it?
Jamison: First, we have to learn who’s more likely to respond and who’s not, and what the advantages are. It’s important to keep educating people about the advantages of lithium and the problems, but to get the knowledge out there, because I think it’s just been smothered by other medications. There’s no evidence that they work better. Many of them certainly have bad side effects.
I think doctors also need to be persuaded before they can persuade patients. That’s very simple. If you aren’t knowledgeable about lithium, or you’re concerned about it or about getting sued because you aren’t using it, then you’re not going to have any kind of ability to talk to your patients into considering it.
Strakowski: Whenever we take a useful option off anybody’s treatment menu, we’re doing them a disservice. If we’ve put this aside as something we’re not going to use, then we’re not providing all the options possible for a given person.
I do a lot of consulting to people with bipolar disorder, and I wish I had a rubber stamp that said “try lithium,” because I’ll look at histories of dozens of medication trials, and never a one included a trial of lithium.
Ostacher: It’s important that we provide lithium education to both providers, who are the ones who really control what medications are prescribed, but also the people with lived experience of having bipolar disorder. They need to learn how helpful it can be to their treatment and their lives.
From the provider side, we’ve made it very difficult to treat people with lithium because of the aggressiveness with which people use high doses of lithium. I don’t think it’s necessary at all. We should frame lithium as primarily a maintenance treatment for a chronic illness. People can be very gentle and careful with their prescribing. The range of lithium levels that will lead to benefits in maintenance treatment are probably quite low.
In Bob Prien’s original lithium maintenance studies from decades ago, the minimum lithium level in the study that they tried to shoot for was 0.6 mEq/L, which is lower than I think many people use. In the BALANCE trial that I was a part of, which John Geddes at Oxford ran, people were on a combination of lithium/valproate or lithium or valproate monotherapy. The minimum lithium level that was accepted in that trial was 0.4 mmol/L. That ended up being an effective dosing strategy, with people receiving both the combination and lithium monotherapy doing better than those who got valproate alone. Therefore, I think we can frame this as a drug that can be dosed at lower levels.
From my experience with patients, lithium’s most problematic side effects are the ones that are visible to others — namely, tremor. People who have a tremor often have to explain themselves in public. It’s often very embarrassing for people to have one. They pick up their cup of coffee, it spills, and somebody gives them a stare and asks them what’s going on. It’s quite possible to dose lithium in a way that minimizes that external side effect. It can’t necessarily be eliminated. But that side effect should be a focus of treatment if it needs to be so, for example with beta-blockers like propranolol, which work pretty well for suppressing that tremor.
I have very few patients who were addressed in this way, with an understanding of what the stigma and bias against the drug might be and an effort to try to minimize side effects that were problematic for them. And my patients have done very well in terms of not having mood episodes. So, I think that education on both sides would be useful.
Jamison: I also think it’s important to emphasize to patients and their families the very compelling data about reducing suicide, and to present it as an interesting drug — which it is — instead of just being presented in this dreary manner. It’s a drug that has interesting characteristics and is the sort of thing that you can work with. You’re not trapped into a high dose or low dose. You can work with your doctor to try to figure out what’s best for you.
Also, lithium provides ongoing care of your brain. In this wellness-oriented universe of ours, we often ask, “Okay, you’ve got your brain now, what do you do with it?” One aspect of this is preventing the damage from future manic or depressive episodes. It’s still somewhat preliminary, but it looks like there are these advantages to using lithium. There are lots of things going on in the brain — not just prevention of mania and depression, but possibly helping with degenerative diseases or whatever. I think it’s important to not oversell that, but also to want to be aware of it.
I’m somebody who’s been on lithium for a long time. It was a really good day when I started reading literature about the neuroprotective effects of lithium, and I think that’s true for a lot of people, as long as it’s put in the correct context. Again, the science is limited, it’s new, but it all seems to be going in one direction.
Strakowski: Thank you both for educating me and others about lithium. I think the three of us are not-so-secretly conspiring that this will be the first step in leading to increased use of lithium in the United States and elsewhere. We encourage everyone to look at that manuscript by Rhee and colleagues that we’re referring to, and also then expand your education about lithium. The National Network of Depression Centers continues to develop its work in this area, in order to help people become more informed.
Thank you for tuning in today and for listening to us. We hope that it was useful for you.
Stephen Strakowski, MD, is the founding chair and professor of psychiatry at Dell Medical School, University of Texas. His research focuses on the brain changes that occur at the onset of bipolar disorder.