How I Slept After Quitting Caffeine

Five months ago, I started to reduce the number of caffeinated drinks I consumed each day. Three months ago, I completely quit caffeine to see how my sleep and daily life would be affected.

Being mindful of your caffeine intake is one of the most commonly listed self-help tips for better sleep. Show me a sleep hygiene article that doesn’t mention it, and I’ll show you a pillow stuffed with golden unicorn feathers.

As a sleep blogger who also recommends thinking about caffeine consumption, I wanted to put my money where my mouth is and see if giving up my daily buckets of coffee would really improve my sleep.

The short answer is that I feel like I’ve been sleeping better, and my sleep tracker has also shown some modest improvement to the total sleep time and sleep score. I also feel better during the daytime, with less anxiety and an overall improvement in my mood.

However, I don’t think quitting caffeine is the last step I’ll need to take if I want to consistently sleep better, and I’ll probably need to consider other factors as well.

Recap: Nine months of experimenting with my sleep

This past year has felt at times like an epic journey towards an increasingly extreme form of abstinence. No drugs, no alcohol, no nicotine, no caffeine, no added sugar or artificial sweeteners.

Nine months after I began the first part of my personal experiment, I finally know what it feels like to live a straight edge lifestyle, and importantly for this blog: how I sleep without eating or drinking all the fun stuff.

In this article, I’ll be focussing on how quitting caffeine affected my sleep tracker data, and my observations of how I’ve been sleeping. I’ll also compare the time without caffeine with the data I have from the preceding year and a half.

In some ways, this is the third in a series, but it’s a standalone piece that doesn’t require a reading of the previous two to understand.

If you’re interested though, you might like to read part one in which I gave up alcohol and sugar for a month. And then there’s part two, in which I discuss four months without consuming them.

Just my experience

My intention is to discuss my personal experience of quitting caffeine and other substances, and is not intended to be medical or dietary advice. I’m not a doctor or nutritionist, so if you have any concerns about your diet or health, please discuss them with your doctor or a qualified nutritionist.

I’ll also do my best to resist the siren’s call of self-righteousness. I think it can be helpful to hear about other people’s experiences if you’re considering giving something up but aren’t quite ready yet. It’s perhaps less helpful when someone overzealously tries to convince you that “This way is better!”

And honestly, I’m still not entirely convinced my black or white abstinence is ‘better’ anyway.

Why I gave up caffeine and what I hoped to gain

Giving up alcohol and sugar nine months ago now has definitely been good for me, and in many ways. However, my sleep tracker data hasn’t shown the kind of vast improvement to my sleep that I’d expected.

My personal feeling is that I sleep better without alcohol or an uncomfortable belly after eating too much chocolate late in the evening. But my wearable sleep tracker data just hasn’t backed that up in a way that was clear (I use the Fitbit Versa 3).

So I decided to search elsewhere for sleep-stealing culprits. And caffeine was the obvious target, especially considering how much I typically consumed.

I averaged two to four cups of strong coffee most mornings, maybe another one in the afternoon, and probably a couple of black teas as well during the afternoon. Add my love of chocolate to that, and my daily caffeine consumption was quite high.

Caffeine is a stimulant, and research has shown that it can have important disruptive effects on sleep. For example, a study in 2006 found that high caffeine intake led to worse sleep in adolescents, and a study in 2013 found that caffeine intake in the six hours before going to bed led to more disrupted sleep in adults.

Interestingly, I’ve long wondered whether I’ve been stuck in a vicious circle of needing caffeine to wake me up because I’m sleep deprived, when one of the things that caused the sleep deprivation in the first place may have been caffeine the day before!

There was a fascinating review of the link between caffeine, sleep quality and daytime functioning published by Australian researchers in 2018 that’s worth reading if that concept interests you.

To find out if my coffee habit was robbing me of sleep, I decided to cut down slowly and then switch to decaf. If nothing else, it was going to be interesting to see what I’m like as a person with no caffeine in my system. After all, by my calculations, I’ve been drinking it daily in one form or another for 27 years!

Step one: No caffeine after midday

On August 9, 2021, I started the reduction in my daily caffeine consumption. My aim was to drink no coffee or tea after midday, which naturally resulted in me cutting down to between one and three cups most mornings as I just didn’t feel like drinking more than that.

By October, I was down to just one cup most mornings as that seemed enough to wake me up and get on with the day. After that, I just drank decaf and was perfectly happy with it.

Let’s take a look at my Fitbit Versa 3 data for the period of moderating only.

1. Total sleep time

The table below shows the average sleep I got per night from August 9 – October 26.

DatesAverage sleep
per night
Aug 9 – Aug 156 hrs 47 mins
Aug 16 – Aug 226 hrs 47 mins
Aug 23 – Aug 29 7 hrs 13 mins
Aug 30 – Sep 56 hrs 54 mins
Sep 6 – Sep 126 hrs 56 mins
Sep 13 – Sep 186 hrs 43 mins
Sep 20 – Sep 266 hrs 55mins
Sep 27 – Oct 36 hrs 27 mins
Oct 4 – Oct 106 hrs 58 mins
Oct 11 – Oct 177 hrs 2 mins
Oct 18 – Oct 246 hrs 48 mins
Average6 hrs 52 mins

So the average amount of sleep I got during the two and a half months moderating phase was 6 hours and 52 minutes. This is a reasonable amount for me, but how does it compare to the previous phases of my experiment?

  • The year before quitting anything: 6 hrs 52 mins
  • Month 1 with no sugar or alcohol: 6 hrs 44 mins
  • Months 2-4 with no sugar or alcohol: 6 hrs 53 mins
  • Months 5-6 with no caffeine after midday: 6 hrs 52 mins

You can see that the total sleep time shows no improvement at all, and is the same as it was during the year before I started my experiment.

I’m starting to wonder whether I’ll always average just under seven hours of sleep, no matter what I do. Perhaps that’s all the sleep I need and I shouldn’t be concerned?

2. Fitbit sleep score

In the table below, you can see the Fitbit sleep score during the moderating phase.

DatesFitbit Sleep Score
Aug 9 – Aug 1581
Aug 16 – Aug 2282
Aug 23 – Aug 29 80
Aug 30 – Sep 581
Sep 6 – Sep 1282
Sep 13 – Sep 1882
Sep 20 – Sep 2681
Sep 27 – Oct 377
Oct 4 – Oct 1082
Oct 11 – Oct 1780
Oct 18 – Oct 2482
Average81

And once again, let’s compare that to the previous stages of my experiment:

  • The year before quitting anything: 81
  • Month 1 with no sugar or alcohol: 82
  • Months 2-4 with no sugar or alcohol: 81
  • Months 5-6 with no caffeine after midday: 81

Just like with the total sleep time, there’s no change. The sleep score is actually remarkably stable for the entire year and a half.

3. Restless time

I decided not to look at the sleep stage data in depth after the last article because, on reflection, I just don’t know if I can trust it enough to warrant including it. At least with the total sleep time, and the score, I can kind of compare it with my manual sleep diary and feeling of how well I slept.

One other measuring point which I have managed to ‘roughly’ double-check based on how soundly I feel I sleep is total restless time.

I’m not including a table this time, as the Fitbit app doesn’t give you weekly averages. Instead, I manually calculated the average. Here it is compared to the preceding time periods. I didn’t include the entire year before, because it takes too long to manually extract the data from the app.

  • 2 weeks before quitting sugar and alcohol: 7.9%
  • Month 1 with no sugar or alcohol: 7.1%
  • Months 2-4 with no sugar or alcohol: 8.4%
  • Months 5-6 with no caffeine after midday: 9.2%

Seeing my total restless time going up to 9.2% surprised me. I’m not sure what the explanation could be. So far, my best guesses are a change in the firmware of the Fitbit, or something else making me more restless that has nothing to do with my diet. There was a month of more stress than usual in there, so perhaps that had an impact.

4. Resting heart rate

The final point I’ve been keeping track of throughout the experiment is my resting heart rate. Again, it’s too time consuming to pull a year’s data out of the app, so I’ve only included the two weeks before I started giving things up, and was eating and drinking whatever I liked.

  • 2 weeks before quitting sugar and alcohol: 60
  • Month 1 with no sugar or alcohol: 57.8
  • Months 2-4 with no sugar or alcohol: 55.8
  • Months 5-6 with no caffeine after midday: 56

There’s nothing in this result that suggests moderating caffeine had much effect on my resting heart rate beyond the improvement that already came from quitting alcohol and sugar.

It’s good to see it’s still lower than the year before, so I imagine that’s the reduction in alcohol, healthier diet, and increase in exercise continuing to benefit me.

5. How I felt while moderating caffeine

I don’t feel like keeping the coffee to the morning had much impact on my sleep, so the sleep tracker data seems reasonable – even if I secretly wish there was an improvement to my sleep!

I felt better during the daytime, with slightly less nervous energy in the afternoon. But nothing of note where my sleep is concerned. Perhaps I was still managing to drink enough in the morning that the benefits of reducing it weren’t noticeable. The next section will provide some more insight on that front.

Step 2: Zero caffeine

This step was hard! Even though the cut was only from one coffee a day to none, I still had headaches, lethargy, and low mood for several days.

I think either I’m particularly sensitive to caffeine, or I’d found the secret to making the world’s strongest coffee! Maybe when they say steep for 2 minutes, adding an extra 0 isn’t such a good idea after all…

Let’s see what happened to my sleep with no caffeine at all in my system.

1. Total sleep time

In the table below, you can see my average sleep time from October 26, 2021 to January 8, 2022:

Dates Average sleep
per night
Oct 26 – Oct 307 hrs 0 mins
Nov 1 – Nov 76 hrs 59 mins
Nov 8 – Nov 14 6 hrs 58 mins
Nov 15 – Nov 21 6 hrs 53 mins
Nov 22 – Nov 28 7 hrs 18 mins
Nov 30 – Dec 56 hrs 47 mins
Dec 6 – Dec 126 hrs 35 mins
Dec 13 – Dec 18 6 hrs 51 mins
Dec 20 – Dec 267 hrs 15 mins
Dec 27 – Jan 26 hrs 58 mins
Jan 3 – Jan 86 hrs 52 mins
Average6 hrs 57 mins

On average then, I got 6 hours and 57 minutes of nightly sleep with no caffeine. Let’s compare it to the previous time periods again:

  • The year before quitting anything: 6 hrs 52 mins
  • Month 1 with no sugar or alcohol: 6 hrs 44 mins
  • Months 2-4 with no sugar or alcohol: 6 hrs 53 mins
  • Months 5-6 with no caffeine after midday: 6 hrs 52 mins
  • Months 7-8 with zero caffeine: 6 hrs 57 minutes

It’s the most average sleep so far since I started the experiment, but still only five minutes more than the year when I ate and drank whatever I felt like.

I was surprised it wasn’t more time, as I do feel like I’ve been sleeping better with no caffeine in my system on the whole. More on that later.

2. Fitbit sleep score

Below are my Fitbit sleep scores for the time with zero caffeine:

Dates Fitbit Sleep Score
Oct 26 – Oct 3085
Nov 1 – Nov 781
Nov 8 – Nov 14 84
Nov 15 – Nov 21 84
Nov 22 – Nov 28 84
Nov 30 – Dec 583
Dec 6 – Dec 1284
Dec 13 – Dec 18 82
Dec 20 – Dec 2682
Dec 27 – Jan 282
Jan 3 – Jan 882
Average83

Finally, there’s a result of note. The average of 83 is the highest so far, and even though it’s not a huge leap, it’s still an improvement. Below you can see the sleep scores for the whole time period again:

  • The year before quitting anything: 81
  • Month 1 with no sugar or alcohol: 82
  • Months 2-4 with no sugar or alcohol: 81
  • Months 5-6 with no caffeine after midday: 81
  • Months 7-8 with zero caffeine: 83

One interesting point is that this is the first time period since I began the experiment in which there was no week with a score below 80.

So even though the overall average might not seem significantly higher than before, the consistency of those 80+ scores is satisfying to see. Perhaps that helps explain why I feel like I’ve been sleeping well this past couple of months.

Nights with 90 sleep score

In addition to the consistent sleep scores of 80+, there were three days in the zero caffeine period when the sleep scores were 90, 90, and 91.

Considering this was just a 2.5 month period, and there were only five 90+ days in the whole year before quitting alcohol, sugar, or caffeine, it gives me room for optimism that my sleep is heading in the right direction, albeit slower than I had hoped for.

3. Restless time

There’s something happening with the restless time that I can’t explain. 8.8% on average with no caffeine is not what I expected. Below is the comparison again:

  • The year before quitting anything: 7.9%
  • Month 1 with no sugar or alcohol: 7.1%
  • Months 2-4 with no sugar or alcohol: 8.4%
  • Months 5-6 with no caffeine after midday: 9.2%
  • Months 7-8 with zero caffeine: 8.8%

Although it’s a bit lower than the months with caffeine in moderation, it’s still higher than all the previous time periods. Honestly, I have no explanation for this increase. I plan to continue without caffeine, so this is one sleep measure I’ll be keeping a keen eye on.

4. Resting heart rate

In the previous article, I talked about how pleased I was to see a large improvement in my resting heart rate. The drop was quite noticeable when I gave up alcohol and sugar. Since then, it seems to have leveled off though. Below you can see the change over time:

  • The two weeks before quitting anything: 60
  • Month 1 with no sugar or alcohol:  57.8
  • Months 2-4 with no sugar or alcohol: 55.8
  • Months 5-6 with no caffeine after midday: 56
  • Months 7-8 with zero caffeine: 55.9

My resting heart rate on average was the same with caffeine in moderation (56) and zero caffeine (55.9). And it’s not really any different from the previous two months when I was still drinking caffeine (55.8).

Interestingly, I took two weeks off work at Christmas, and did a huge amount of exercise on most days. The average for that two weeks was 54, which is lower than all the previous time periods.

So I wonder if my resting heart rate can only really be improved further with exercise? Perhaps my diet is less of a factor simply because it’s so healthy already, there’s not much more that will happen through tweaking the diet alone. That’s a theory I’d have to put to the test in the future I think.

It’s also possible that being so healthy has opened up space in my life for more exercise, and more enjoyment of exercise. I’m currently doing considerably more exercise than I used to, so perhaps I need to view measures like the resting heart rate from multiple angles and not get tunnel vision because of my current interest in diet.

5. How I felt after quitting caffeine

Giving up caffeine seems to have objectively improved my total sleep time and Fitbit sleep score, but not by a huge amount. And at the same time, I appear to be more restless at night, which is somewhat contradictory.

So once again, it’s perhaps more important to consider how I personally feel my sleep was before and after quitting caffeine.

I’ll start with an additional benefit I’ve felt because I think it’s important and relevant: less anxiety. Cutting down caffeine helped a bit, but I noticed a big improvement to my anxiety levels and general stability of mood when I cut it out. And on that note, less trembly hands too when I accidentally overdosed on coffee!

For me, and many people I suspect, anxiety and sleep are not friends. Having an overall reduction in daily anxiety and stress levels has meant going to bed at night with less on my mind, which can only be good. In the last month, I can count on one hand the number of times I’ve been awake in bed late at night ruminating.

Whether the sleep tracker shows it or not, I do feel that quitting caffeine completely had a positive impact on my sleep. I feel like I more regularly get a good night’s sleep without being awake for ages in the night. And there have been many nights recently when I’ve slept very deeply – at least that’s how I feel when I wake up.

Future plans

Continuing without caffeine

I previously wrote about how I’ve struggled at times with abstaining from alcohol and sugar, especially at parties and special events.

With caffeine, it’s the complete opposite; I’m happy not to be drinking it and have zero desire to go back to the dark days of panicking in the morning when you discover there’s no coffee in the cupboard.

This clear and unwavering feeling that life is better without caffeine is very welcome news, as it reduces the amount of mental work I have to put into resisting temptation on a daily basis.

Interestingly, when I gave up alcohol and sugar nine months ago, I felt many positive benefits. With caffeine, it’s been more a case of seeing negatives disappearing instead of benefits appearing. That may seem like two side of the same coin, but I think it’s an important distinction to make.

To give some clear examples, I had brighter eyes, more energy, better skin, better digestion, and more energy when I quit alcohol and sugar. I felt like these were positive additions that I could really see growing over time.

With caffeine, I have less grumpy mornings until I’ve had a coffee, less trembling hands when a coffee is too strong, less yucky feeling in the stomach from drinking too many coffees back to back, less toilet breaks during the day, and less washing up. These feel more like the removal of negatives rather than the addition of positives.

Anyway, no matter which way you choose to look at it, giving up caffeine has been nothing but good for me, and surprisingly easy to keep up once you push past the first few days of withdrawal.

The late exercise, late stress, and late snacks need to calm down

I’ve noticed in the past that the average sleep times and sleep scores can be heavily affected by just one or two very bad night’s sleep. So, during the last five months, I’ve kept a diary of what I was doing in the evenings before I slept particularly badly.

As well as trying to improve my sleep in general, I’ve also been thinking about what I can do to reduce the worst outliers: the nights when the sleep score is around 70; the nights when I only get five or six hours’ sleep.

Below you can see a screenshot of my Fitbit app sleep data. Notice how the sleep score was high for six days, with three in the high 80s, but the weekly average was pulled down by the worst night of 77:

screenshot of the fitbit app showing a week's sleep data

One pattern I’ve seen is that late-night events, like intense exercise, stressful conversations, or snacks while watching a movie often precede a bad night’s sleep.

So going forward, I’m going to try to reduce how energetic my late evenings are whenever possible, both physically and emotionally.

Obviously, I can’t control everything, and have no plans not to have any fun in the evening because it might affect my sleep score! But if there are some decisions I can take that don’t have an impact on my enjoyment of life, I’ll be more mindful.

A good example is not eating a giant bowl of popcorn just before going to bed! Another would be having a lighter workout if it’s getting late rather than going nuts and then just hoping I’ll be able to sleep.

Continuing without alcohol or sugar

As I type this, it’s been nine months and twelve days since I had a drop of alcohol. And even though sugar is harder to keep track of, I think I’ve done a good job of cutting it out massively.

I continue to feel better in myself living life this way, so will keep it up for longer. I’d like to reach a year, at which point I might reassess the situation. Right now though, I can see myself continuing this way for a considerable length of time.

In the photos below, you can see me two weeks after I gave up alcohol and sugar on the left, and nine months later on the right. It’s not an exact copy of the same pose, but I personally think it shows how I look healthier nowadays.

two photos of ethan green. One just after quitting alcohol and sugar, and one nine months later.

What happened to the no phone and stable sleep schedule?

At the end of the last article, I came up some more plans to improve my sleep in addition to giving up caffeine. They were:

  • Go for a 20 minute relaxing walk every day, even if I do other exercise that day.
  • Turn the light off to go to sleep between 10:30 p.m. and 11 p.m. as often as possible.
  • Set an alarm for 7 a.m. every morning.
  • Don’t use my phone in bed. Books only.

I have to admit, I haven’t done so well on these points!

I have managed to go for a walk pretty much every day, so I can tick that one off the list.

My bedtime hasn’t been that stable though. It’s not helped by the fact that I don’t get home until after 10 p.m. on a couple of nights of the week at the moment, as I have some fun activities in my life right now. I could do better on the days that I am home though.

As for the phone, I’m getting there. Slowly. I’d say I turn it off before getting into bed on maybe two nights of the week on average, and only use it for 30 minutes on another two or three. But there are still nights when I get sucked into the screen for longer than I should.

So, I’m going to rethink the sleep schedule and see if I can find a compromise that works.

And here it is: my promise to myself for the next three months is that I will not use my phone in bed, but only read.

Wish me luck on that one; it could end up being the hardest challenge so far!

Your views

Have you tried giving up caffeine before? How did it affect your sleep and well-being? If you haven’t, but would like to, what is it that’s stopping you?

I’d love to hear your story in the comments below, or any thoughts you have about the experiment I’ve been doing with changing my lifestyle to change my sleep.

8 Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

  1. I am on the 6th week of zero caffeine. My sleep has been very irregular since I quit, but I am gradually falling asleep and staying asleep between 10pm-4am. I have found it really helpful to go to bed right around when I get tired which is between 9:30/10pm. The other noticeable change is, like you experienced, a reduction in anxiety. I feel so much more calm and level-headed throughout the day. I won’t be going back to caffeine primarily because I don’t want to be regularly on edge again.

    • Hi Courtney
      Thanks for sharing your experience. It was interesting to read about your reduction in anxiety. I am absolutely certain now that it’s made a huge difference for me. As an experiment, a little while ago I spent a few weeks back drinking caffeine (I’ve gone back to zero now). I quickly started to notice anxiety about things that I didn’t have before, and was more sensitive to life’s daily problems and challenges. As you said, I feel much calmer and level-headed without any caffeine.
      As for sleep, I also believe it’s better not to drink caffeine, at least for me. The number of nights with disrupted sleep seemed to increase when I was back drinking it, and importantly, I think my nighttime worries were more present, which also has an impact on my sleep.
      So yes, as far as I’m concerned, caffeine is going to be a big no-no going forward again!
      Regards
      Ethan

  2. Hey Ethan,

    Thanks for all the work you’ve done in putting this whole article together! I love when people take the time to assess such things upon their own bodies, I find it fascinating.

    I personally used to be extremely addicted to caffeine and would consume anywhere between 200mg – 450mg/day for about 3 years back-to-back.

    In 2021, I decided that after quitting sugar for 90 days and seeing its results, it was time for me to try and quit caffeine. I am not a balance type of guy, so I went cold turkey. It was INSANELY hard.. for about 2.5 weeks I couldn’t recognise myself anymore:

    – I was sleeping more and waking up relatively tired
    – I wouldn’t be able to do any productive work after 5-6 pm most days
    – I lost my drive and focus for at least 2.5 weeks if not more.

    After month 1 however, things started to shift. I successfully did a full reset of 5 months, which might have been the biggest challenge I did up to date.

    Now, about 14 months later, I find myself falling into old habits again where I tend to have a caffeinated beverage every day. It’s not a lot, but I decided to do a 30-day fast this time.

    What I would suggest you though for improving sleep is to use a more accurate sleep tracker such as WHOOP or Oura ring. Fitbits are great for counting calories and for running, but they’re definitely not spectacular when it comes to keeping track of REM or deep sleep which is probably what you should be looking at in your analysis.

    Again, super happy to have found this article! Do you have a newsletter by any chance?

    Pascu

    • Hi Pascu

      Thanks for sharing your experience – it was very interesting to read!

      I can totally relate to the period of time you describe in which things didn’t seem quite right. I definitely went through a similar transition phase when I gave up sugar and alcohol, and also caffeiene. I think it takes time for the body to adapt, and it’s particularly pleasant until it does. But it’s well worth sticking it out though as the benefits, when they come, far outweigh the initial discomfort and loss of energy/motivation etc.

      As for the sleep tracking, thanks for the suggestion. As a general rule of thumb, I try not to put too much stock in the sleep stage tracking. I’m just not convinced how accurate any of them really are, and can’t confirm it myself. At least with total sleep time, I am able to confirm more or less that I was likely to have been asleep when it said I was. For example, if it says I was asleep when I was just reading in bed quietly, I know there is a big problem. I can’t do that with REM and deep sleep though.

      I don’t currently have a newsletter, but keep meaning to get one up and running again! Have you seen my youtube channel? I mostly publish reviews there, but at least you can keep an eye on what I’m up to.

      Regards
      Ethan

  3. Thanks for your insightful, personal story. I was just looking online to see told other’s experiences. I’ve given up sugar for 1 month. It’s surprisingly easy to live without!!! I don’t drink or use caffeine. I may have a black tea, but I don’t feel the effects, and never have it daily. Anyhow, thanks again for sharing.

    • Hi Amanda
      You’re welcome – I’m glad you liked the article. Well done on giving up sugar! It isn’t as hard as it might seem once you get used to it. I have to admit, I still get cravings for something sweet though, which is usually satisfied with some plain yoghurt, fruit, and crushed nuts. That seems to do the trick!
      Regards
      Ethan

  4. Great post! Though I’ve never had the courage to quit caffeine completely, I’ve definitely learned about its effects on sleep. Best wishes!

    • Thanks Jim, I’m glad you liked the article. If you decide to give it up one day, I hope it goes well! I don’t think it takes much courage – just some acceptance that there will be a couple of days of discomfort potentially. And after that it’s much easier than you might imagine, in my experience.
      Regards
      Ethan