Ambien, Melatonin, and More: What We Know — and Don’t — About Sleep Medications and Dementia Risk

Older adults often turn to prescription and over-the-counter sleep aids to help them get a restful night’s sleep, but what do we really know about their long-term effects on brain health? 

A 15-year UCSF study, published in the Journal of Alzheimer’s Disease, examined the association between sleep medication use and risk of dementia in older adults. Researchers studied 3,068 older adults without dementia and found that frequent use of sleep medications (at least five times per month) was linked to increased risk of developing dementia among white participants but not among Black participants. Researchers said more work is needed to understand why the risk differed by race and what mechanisms may be involved.

Dr. Yue Leng, PhD, an epidemiologist at UC San Francisco whose work explores how sleep, circadian rhythms, and napping relate to neurodegeneration and cognitive decline in older adults, was one of the researchers on the study.

In this conversation with Being Patient’s founder Deborah Kan, Leng discusses the study’s findings and explains why epidemiological research can show associations but cannot prove cause and effect. She also cautions not to over-interpret sleep-stage estimates from wearable devices, which can increase anxiety and even worsen sleep. She discusses what scientists know about deep sleep and REM sleep, why melatonin is often misunderstood as a treatment for insomnia, and why safer first steps for chronic sleep problems typically include behavioral approaches and screening for sleep apnea before relying on medication.

Being Patient: Let’s talk about the study. Epidemiological, for our audience who doesn’t know the difference, is really looking across existing data to determine possible theories in research. Is that correct?

Dr. Yue Leng: Yes. I’m really glad that you mentioned this because I think one of the key points about epidemiological studies that we should let people know up front is this is not a designed trial study, which’s best to tell people which is the cause and which is the effect.

One thing that we should keep in mind as we interpret findings from these studies is that this shows more of an association — not exactly that one is causing the other. So I think it’s important to have that in mind.

Being Patient: How much data did you look at out there, and what did it pertain to? Did it pertain to specific types of sleeping medication? Because, as we all know, there’s everything from melatonin to Ambien, which are completely different. What did you look at, and how much data was out there?

Leng: We looked at over 3,000 participants from the U.S. This is existing data. This is actually coming from a study that was conducted many years ago. The study started in around 1997 to ’98. 

This is a community-based study, which means we included older adults living in the community, not people recruited from sleep clinics. Importantly, we excluded people who already had dementia at baseline so that we were able to follow them forward in time to see who later developed dementia. 

This is what we call a longitudinal study. We can really see whether the sleep medications people were taking are related to their future risk of developing dementia in many years. In the case of this study, the follow-up length is 15 years. It’s a really long study.

We did look at sleep medications very broadly. As I mentioned, this is existing data, so this is more like real world medications that people are actually taking. We are not giving people extra sleeping pills. People basically reported what medications they were using, how much, how frequently they were using. And we looked across both prescription and over-the-counter (OTC) sleep medications. These are the kinds of things that people commonly use in real life — things like Ambien, melatonin, benzodiazepine, antihistamine — so both OTC and prescription medications.

I think the key thing we were interested in was the frequency of their use, given that there are so many different varieties of sleep pills that people were taking. We were really more interested in how frequently they were using these meds and whether more frequent use will be associated with higher risk of dementia in the long run.

Being Patient: What did you find in terms of elevated risk of dementia with different medications?

Leng: Coming back a little to what you were just saying, this is actually one of the main reasons we set out to conduct this study. Part of my lab’s work is really focusing on how people’s sleep problems affect their brain health.

As we know, as people get older, especially, sleep problems are just so common, and they are exhausting. So people do want something that works. People usually turn to sleep pills, which is very understandable and normal.

One thing is, from our work and others, we have seen very frequently that sleep problems themselves may be associated with higher risk of dementia. The sleep problems themselves may be a modifiable risk factor for dementia.

At the same time, you expect [sleeping pills] to improve sleep. So there is really confusion there: Do sleep pills help protect the brain by improving sleep, or could long-term use of sleep pills themselves affect brain health in other ways? 

Being Patient: There’s REM sleep, there’s deep sleep. My understanding is deep sleep is the most important for memory and detoxifying your brain, because that’s when something called the microglia come out and they purge the toxins. It’s like the cleaning cycle. They’re called the janitors of our brains. They really only come out, as I understand it, in deep sleep. Is that correct?

Leng: Generally, people think deep sleep — what we also call the slow-wave sleep — is the brainwave that’s most important, especially for our glymphatic — our brain to clear out the bad proteins.

There is a ton of work from animal studies that shows that, for mice, their brain is actually very active during sleep. This is especially happening during what we call deep sleep. And this is where they clear out all these bad proteins, including what we call amyloid and tau, which is related to Alzheimer’s disease. 

However, I also want to say for the past few years, our work and others have increasingly also seen a very important role of REM sleep — rapid eye movement sleep that you just mentioned. This is the stage where dreams usually happen. This stage also is known to help with memory consolidation, and especially emotional memory.

So I would say it’s one stage of sleep that’s important. You’re supposed to get a more balanced, good distribution of different sleep stages. But in terms of protein clearance, the brain dynamics, metabolism, deep sleep is really important. That’s right.

“Do sleep pills help protect the brain by improving sleep, or could long-term use of sleep pills themselves affect brain health in other ways?”

Being Patient: Do we know what proportion of each stage of sleep we should actually be getting — REM versus deep — in terms of brain health?

Leng: Yeah. Very, very good question. I’m actually wearing a number of devices myself, and I probably wear five, six, and some of them are in bed, so I don’t need to wear all of them to sleep, because I do want to know how my sleep is.

It’s also important to clarify what these watches can and cannot do, because I think this is sometimes causing a lot of anxiety when someone sees from their watches, “I’m not getting enough deep sleep,” and, “I’m only getting a sleep score of 50.” I think that the anxiety itself may be causing a lot of sleep problems. We actually have a term called orthosomnia, and this is where this anxiety is termed.

For most of these wearable watches, they really just track activity. They are not really tracking your sleep, which means they can tell for some things they’re pretty good, like total sleep time, how regular your sleep patterns are — especially night by night — what’s the longitudinal pattern of that.

However, they are not great for telling you how much time you spend in deep sleep versus what we call the dream sleep, the REM sleep. So the first thing to keep in mind is: Don’t rely on those numbers, especially for the sleep stages.

In terms of the actual composition, usually you would expect your deep sleep to be around — I think 20 percent to 25 percent is pretty good for a typical night — and then about the same for REM sleep. It also kind of shifts across the night. In the first half of the night, this is where you get most of the deep sleep. And then as the night goes on, you get more REM sleep in the later half of the night.

I also want to say this composition really changes across your life course. When you are younger, you get more deep sleep. This is also the stage where growth hormones and a lot of things happen. So when you are a kid, you get a lot of deep sleep. That number will gradually decrease as you get older.

So in older adults, generally you get less deep sleep. Your total sleep time might decrease and you have more fragmented sleep, and you have less REM sleep. In general, it’s very common for older adults to note that they wake up more frequently during the night and they don’t get as solid sleep as they were younger. That’s absolutely normal.

Being Patient: What did you determine by looking at over 3,000 people who were taking sleep medications? Could you isolate the findings by specific medications that people were on and attribute it to risk?

Leng: Coming back to the study, I want to start with the overall finding: We did find an association between frequent use of sleep medications and long-term risk of dementia.

We also split the participants up by race — white and Black participants. Maybe I forgot to mention at the beginning: I think this is also a very unique part of this study. We were able to look at the two different race groups, which is very rarely done in previous studies.

What we found was: Among white participants, those who reported frequent use of sleep medications — by frequent use we meant reporting using sleep medications often or almost always — and that’s defined as at least five times a month, compared to those who never or rarely used them. These frequent users had nearly an 80 percent increased risk of developing dementia over this period.

Being Patient: So that’s not a lot at all. I thought you were going to say, five times per week, but five times per month is not that much, actually. 

Leng: Yeah. Again, this is an epidemiological study, so we can’t really count on the dose too much. This is also by people’s self-report. So there’s also a potential issue of reporting bias — that people might not be remembering that accurately.

In this study, we asked people to report their frequency in five categories. The top two categories: One is “often,” and the most frequent one is “almost always.” “Often” is at least five times — I believe it’s five to 15 times a month. And then “almost always” is above more than 15 times a month, if I remember correctly. But you can see how big of a range that is for each category. It might just be arbitrary cutoffs.

“These frequent users had nearly an 80 percent increased risk of developing dementia over this period.”

Being Patient: What about the different types of medications? Did you just group it all into one? Because there’s obviously the NyQuils versus Ambien versus the melatonin.

Leng: Unfortunately, as I mentioned, we only have 3,000 people, and not a lot of them reported taking sleep meds. So we were not able to examine the specific medication types for the statistical analysis.

However, I do want to say descriptively, we saw that benzodiazepines and antihistamines were among the most commonly used sleep medications in this population. Although we can’t really say which medication is linked to the highest risk, this is really more about the frequency. You can see the general pattern of what medications people are taking in this population.

Being Patient: And don’t we know that those drugs already pose higher risk?

Leng: Yeah. I think benzos are probably the one drug with the most evidence because it’s a drug that has been on the market for such a long time. There has been a lot of evidence in terms of falls and short-term memory impairment that people have reported when taking these drugs. But I think it’s still very commonly used.

Also, remember this study was done back in 1997 to ’98. At that time, probably the type of drugs that people were taking were also very different from what you would use these days.

For benzos, we do have more evidence in terms of their effects. But for the others, especially the OTC ones, like antihistamines and melatonin, the evidence is pretty much still mixed and lacking. Part of the challenge is it is very hard to do controlled trials for these OTC drugs, because people are just taking them so much..

Being Patient: What do we know about the link between melatonin and dementia? Melatonin we produce on our own, right? Is a decreased production of melatonin — does that increase risk of dementia?

Leng: This is a really common point of confusion, so I’m really glad that someone asked this. Primary care physicians do tell people just to take melatonin whenever they have sleep problems.

First to clarify, as you said, melatonin is a hormone that everyone produces, and it does decrease with age. That said, melatonin is not really a sleeping pill. It’s not a treatment for insomnia. So if you suffer from insomnia — problems especially with staying asleep — melatonin is not something that would help.

What melatonin does is really to help shift your body clock. That’s why it’s commonly used when you’re trying to deal with jet lag, shift work, or times when your sleep timing is off.

I think it’s true that for older adults, because natural melatonin tends to decline, low-dose melatonin can sometimes be helpful. However, the danger is taking high doses and relying on it. So it’s always important to remember that an occasional low dose is the way to go.

And then the second is coming back to this personalization point. Unless you know that your sleep problems are absolutely caused by declining melatonin, it’s not really safe to keep taking melatonin just to assume that this is caused by melatonin.

Being Patient: I have to ask this to a sleep researcher. Would you ever take — or do you take any — to help you sleep?

Leng: Great question. First of all, I don’t have trouble falling asleep, but my problem is I do wake up sometimes during the night.

Yes, sometimes when I’m traveling, when I have jet lag, I do take melatonin. I’ve tried other, more supplement-type of medications, just for curiosity. But I wouldn’t take a prescription sleep drug because I think there is a lot of next-day impact. Also, I think we just don’t know enough about what this is doing to our body and our brain. So I’d be very cautious about taking sleep meds.

Being Patient: What’s the next stage of this study? Will this translate into further research?

Leng: We are really trying to move toward more personalized research because one size doesn’t fit all. We are using sleep technologies to track people’s sleep at home for long periods so we can better phenotype — divide people into subgroups — and then identify who may be more vulnerable, who may actually benefit from certain treatments. The goal is really to match the right approach to the right person.

The other direction is intervention studies. Because all these studies that we have done, and others have done, are observational, we’re also trying to design new sleep intervention studies to compare these drugs and also behavioral treatment. We didn’t really touch on this today, but I want to say two things that are important.

There are a lot of behavioral strategies out there that people should really go to first before turning to drugs. This is actually considered first-line treatment for insomnia. This is what we call CBT-I, cognitive behavioral therapy for insomnia, and they do help.

Being Patient: That’s like going to bed at the same time, making sure you sleep in a dark room, getting up with the sun, right? 

Leng: That’s a good summary. [It’s about addressing] your behaviors and your thoughts about sleep.

The other thing is sleep apnea, which is a huge problem, very common. It’s a breathing disorder that, if you don’t know it, a major sign is snoring, and you repeatedly stop breathing frequently during the night. It’s important to rule out other sleep problems such as sleep apnea. 

So if you have sleep apnea and you keep taking sleep medications thinking you have insomnia, you’re actually not dealing with the right problem, and that can make things worse. For that, you really should go to the sleep lab, talk to the doctor, get a sleep test done if you suspect that you may have these problems, before taking any drugs.