Physiology Friday #241: How Caffeine Dose and Timing Affect Our Sleep Physiology
That afternoon coffee is subtly changing your brain overnight—even if you can’t detect it.
Greetings!
Welcome to the Physiology Friday newsletter.
Details about the sponsors of this newsletter including FSTFUEL electrolytes, Examine.com, and my book “VO2 Max Essentials” can be found at the end of the post!
We all know someone who can drink an after-dinner espresso and still fall asleep effortlessly an hour or so later. Admittedly, I’m one of these people.
“Caffeine at night just doesn’t seem to affect me,” they claim.
But here’s the problem—how we feel we slept is not always an accurate indicator of how well we actually slept. There’s a difference between objective and subjective sleep quality that’s not only important but physiologically relevant.
Ask someone how they slept and you’ll probably get a decent answer about their duration, deepness, and quality of sleep. People aren’t terrible at making this assessment.
But if you really want to know how someone slept, hook them up to brain electrodes and monitor their neural activity. This is a technique known as polysomnography, and it gives insights into the time someone spends in the various stages of sleep throughout the night—something called sleep architecture—or objective measures of sleep quality.
Back to caffeine. Sleep experts argue that, while one might assert to be unbothered by some late-night caffeine, what’s really going on under the hood paints a different picture.
Caffeine may have undetectable (by us) effects on our sleep quality that inevitably impact our next-day cognitive and physical function. Given the increasingly well-established role of poor sleep quality in promoting cardiovascular disease and cognitive decline, it’s certainly worth knowing if any of our habits around caffeine should be modified to promote the best sleep possible.
It’s not that all caffeine is bad of course. The dose (and timing) make the poison. A few sips of coffee in the late afternoon might be fine, but a higher dose close to bedtime isn’t. The only way to truly know is to study it, and that’s what was recently done by a group of researchers from Australia.1
Their findings provide us with a framework and practical recommendations for when (and when not) to consume caffeine if we want to sleep well.
Participants in the study completed a total of 7 separate experimental trials in which they received one of two doses of caffeine in pill form (100 mg or 400 mg) 12 hours, 8 hours, and 4 hours before their bedtime. For example, on one day, the participants consumed 100 mg of caffeine 12 hours before bed and had no caffeine for the rest of the day. On another day they consumed 100 mg of caffeine 8 hours before bed. So on and so forth for the other time points and caffeine doses
The participants also consumed their habitual morning caffeine in pill form on all of the experimental days and throughout the study. In a control condition, they received a placebo at each time point. A period of 48 hours separated each of these conditions.
Each night, the researchers measured the participants’ sleep using a portable sleep-monitoring device (a partial polysomnograph). At night and in the morning, the participants were asked how good they perceived their sleep to be.
Let’s start with how the different doses and timing of caffeine affected levels of caffeine in the participants’ saliva. Not surprisingly, the 400 mg dose led to higher levels of caffeine in the body than the 100 mg dose at all time points, and in all cases, caffeine rapidly rose after consumption and gradually declined.
But as you can see in the (blurry) figure below, all caffeine doses at all time points led to caffeine concentrations above zero right when participants were getting into bed and even persisted until they woke up the next morning. Even when you consume 400 mg of caffeine 12 hours before bedtime, there’s still some caffeine left in your system the next morning when you wake up. That’s all thanks to the half-life of caffeine which is 3–6 hours—after this period, only half of the initially consumed caffeine remains in the system. After another 3-6 hours, the amount would be halved again, and so on.
Despite the 100 mg dose leading to detectable levels of caffeine in the body, consuming it 12 hours, 8 hours, and even 4 hours before bed did not impact the participants’ total sleep duration, sleep efficiency, sleep latency (the time it takes to fall asleep), wakefulness during the night, the number of nighttime awakenings, or their sleep architecture (time spent in N1, N2, and N3 sleep stages).
The 100 mg dose also did not affect subjective sleep—the participants perceived their sleep duration and quality to be no different whether they received the placebo or consumed caffeine at any of the three time points. They also didn’t perceive differences in how long it took them to fall asleep, how much they woke up during the night, or their sleepiness and alertness levels before bed.
This would indicate that it’s safe to consume 100 mg of caffeine as little as 4 hours before bedtime without it negatively impacting your sleep—perceived or otherwise.
This wasn’t true for the 400 mg dose…
When consumed 12 hours before bedtime, the 400 mg dose of caffeine had very few effects: light sleep increased by around 6% and the duration of deep sleep decreased by 20 minutes, but there were no effects of this dose and timing on total sleep time, objective measures of sleep quality, or perceived sleep quality.
Consuming 400 mg of caffeine 8 hours before bed was a bit more impactful: sleep efficiency declined by 7% and deep sleep duration decreased by 15 minutes. The participants also reported feeling like they slept about 45 minutes less and spent more time awake during the night—about 30 minutes longer on average. However, this dose and timing did not affect total sleep duration or most other measures of sleep quality.
The largest effects were observed with the 400 mg dose of caffeine consumed 4 hours before bed: total sleep duration was 50 minutes less, sleep efficiency dropped by nearly 10%, it took the participants 14 minutes longer to fall asleep, and the participants spent 26 more minutes awake during the night and woke up an average of 1.4 times more per hour.
Sleep architecture also changed with this dose and timing regimen: light sleep increased by 5.6% and deep sleep decreased by around 30 minutes. The participants’ perception of sleep quality was also much worse—they felt like they slept 1.3 hours less and reported higher pre-sleep alertness and less sleepiness before bed. They also felt like it took them longer to fall asleep (by about 30 minutes) and reported a reduction in their sleep quality.
This means that the only safe time to consume 400 mg of caffeine is 12 hours before sleep. When consumed 8 and 4 hours before sleep, this dose of caffeine negatively impacts objective and subjective sleep and this gets worse the closer one consumes caffeine in relation to bed.
A significant implication here is the discrepancy between objective and subjective measures of sleep quality following caffeine consumption.
While participants could often perceive negative effects when a high dose was consumed close to bedtime, their ability to detect disruptions diminished as consumption occurred earlier in the day.
Notably, the participants often failed to perceive caffeine-induced sleep fragmentation, even when it was objectively measured. Those of us who claim to be immune to late-day caffeine might be underestimating the true extent of caffeine's impact on our sleep. Food (or drink) for thought.
A few other interesting takeaways from this study are that habitual caffeine consumption and genetics don’t appear to play a huge role in how caffeine affects our sleep.
The participants were habitual caffeine consumers—they reported consuming 70–200 mg per day on average, with some participants consuming up to 300 mg per day. This means that even regular coffee drinkers (or however you prefer to take your drug of choice) experience caffeine’s effects on sleep in a dose- and time-dependent manner and never habituate against them.
The study investigated two specific caffeine genotypes: the CYP1A2 variant rs762551 (this gene is involved in the metabolism of caffeine in the liver and variations in this gene can affect how quickly an individual metabolizes caffeine, resulting in "fast" or "slow" metabolizers) and the ADORA2A variant rs5751876 (this gene codes for the adenosine A2A receptor, which caffeine binds to. Variations in this gene can influence an individual's sensitivity to the effects of caffeine).
Genetics didn’t appear to be a factor in caffeine’s effect on sleep—there were no differences attributable to the participants’ CYP1A2 or ADORA2A genotype, however, it’s worth noting that the study probably didn’t have a large enough sample size (it was underpowered) to detect these differences.
The takeaway from this study is plain and simple: Avoid any dose of caffeine larger than 100 mg in the ~12 hours before your plan to go to sleep. The closer you get to bedtime and the more caffeine you consume, the greater the chance your sleep quality will suffer.
I’m in bed by 9 or 10 on most nights—this means I need to cut my caffeine off around 4 p.m., which I pretty much always do. However, I will start to be a bit more conscious about consuming larger doses of caffeine early in the day, since the effects might carry over into my sleep that night. From now on, I’ll probably avoid 200–300 mg of caffeine after 9 a.m.
There’s a famous quote by the renowned physicist Richard Feynman:
“The first principle is that you must not fool yourself and you are the easiest person to fool.”
We might think our sleep is robust to the mid-afternoon cup (or two) of coffee, but this study suggests otherwise. What we perceive about ourselves (and our sleep) isn’t always true—a more objective analysis sometimes paints a different picture.
Will you put yourself at risk for early cognitive decline and other sleep issues by having caffeine too close to sleep on occasion? Certainly not. But these findings are a friendly reminder that even the habits we don’t think twice about—that after-dinner espresso—may be having subtle effects on our health.
Thanks for reading. See you next Friday.
~Brady~
The VO2 Max Essentials eBook is your comprehensive guide to aerobic fitness, how to improve it, and its importance for health, performance, and longevity. Get your copy today and use code SUBSTACK20 at checkout for a 20% discount. You can also grab the Kindle eBook, paperback, or hardcover version on Amazon.
Examine.com: Examine is the largest database of nutrition and supplement information on the internet.
FSTFUEL combines electrolytes with amino acids to help your body maintain hydration and optimal functioning during exercise or intermittent fasting, so you don't have to choose between fasting and fitness. If you want to try some, the guys at FSTFUEL have agreed to give my audience a 30% discount on their orders. Just use the coupon code BRADY30 at checkout.
This makes me wonder about the long term effects of the changes in sleep large doses of caffeine make. My dad used to drink coffee all day long. Usually a whole pot a day and then a cup at night after dinner, every single day. He died in his 80s with type 2 diabetes, and loss of vision in one eye due to several strokes. Granted his diet wasn't great, but I suspect things may have been different if he hadn't drank so much caffeine (not to mention the sugar he put in his coffee...😂).
This study was really eye-opening (no pun intended) for me. I am a heavy coffee drinker consuming up to 400-450 mg a day on most days. I often stop consuming caffeine by 6 pm and am in bed by 2:30 am and I can feel the sleep disruption on some days. For me, stopping around 2 pm would be ideal.