Slicing the Coffee Data

Ever since I started collecting data to track my headaches, I have not seen a connection between coffee intake and headaches. Seven months of data and no relationship. Until now. I think I see a pattern and it isn’t what I expected.

CoffeeFrequencyAverage Headache Intensity (0-5)

The table above shows that low coffee intake actually results in a slightly higher headache intensity. Note that these headaches are in the middle of the night. They feel nothing like caffeine withdrawal headaches, yet caffeine intake makes them go away quicker.

Now for the pattern I spotted. Over the past 7 months, I collected data for 102 days where I track the time I drank my last coffee for the day.

  • 81 days my last coffee was after Noon with a Headache Intensity of 1.1.
  • 21 days my last coffee was before Noon with a Headache Intensity of 2.2.

Based off this sample, I can expect to reduce my headache intensity by 50% if I have an espresso after Noon. Of the 5 killer headaches I had during this period (5/5), only one of those dates did I have an afternoon espresso.

Since I also believe that high caffeine levels result in reduced sleep quality and may be contributing at least partially to the core headache problems, I think the best strategy for me is to reduce my morning coffee intake and delay my last espresso into mid-afternoon.

Of course I could be wrong on all this. Additional testing should answer this question.


Add yours

  1. Just quit coffee. I am trying to at the moment. You really need to get a handle on this headache issue.

    Actually went to a local coffeeshop this morning, got in line, thought about, thought about how a woman told me my teeth were stained and gross, got out the line, went back home.

  2. @thomas – It is a delicate balance. If my data analysis is correct then cutting out coffee would make my sleep WORSE. My plan of action is to push my last espresso to later in the day. Then reduce the morning coffee levels. Establish a new normal and reassess.

  3. It is an odd proposition to consider that coffee could be a medicine. If so, what is it curing? Or, rather, what ailment is it masking?

    Recently, someone told me the same thing about tobacco. It’s the fucked up thing about drugs. Nothing is what it seems. I don’t doubt your data but I doubt addiction.

  4. @Txomin – I’ve really taken your comments regarding addiction seriously. I am addicted to caffeine. Many people are, so I am not alone. The poison is the cure.

    Yesterday I had an espresso at 4 PM and then proceeded to have the most restful 9 hour sleep I’ve had in months.

  5. Have you tried Milstead and Co in Fremont yet? It just opened last month.

  6. @Becka – Yes. I’ve been there 3 times. I organized an event with the Coffee Club of Seattle, where we hiked around lake Union and stopped into Milstead for coffee.

  7. there is caffeine in migraine medication.

    my friend’s son has gotten horrible migraines since he was a small child (he’s now 12). usually in the middle of the night. hydration and a can of coke did the trick getting him back to sleep.

    she didn’t realize it was the caffeine that was doing it until one night he woke up with a migraine, and she didn’t have a coke handy. she grabbed a ginger ale. he couldn’t get any relief, and they were nearly up all night. the next day when she told me, i told her it was the caffeine that did it – she finally made the connection.

  8. @Aviva – I have a friend that has migraine issues. She uses coffee to cope. I tried a low-caffeine cold tea with poor results. Now I’m thinking I should up the caffeine until I find that sweet spot that tames the headache, but doesn’t keep me awake.


  9. @Aviva – Last night I brewed a full strength tea and sat it by my bed. At 3 AM, I woke up with a headache. I sipped it between 3 AM – 5 AM. By the time I woke up at 6 AM, I had no headache. Although this won’t prevent headaches, it may be a good management tool. I’ll keep testing!

  10. I recently used 23andMe’s service for genetic analysis and learned that I was a “fast caffeine metabolizer” … I had wondered if coffee drinking was interrupting my sleep but it seems less likely given that information. I wrote about the experience of using the 23andMe (service) if you’re interested:

    I have some privacy concerns about the service, but it was fascinating to get some detailed data in regards to some questions I had about my own genotype (MTHFR variation, fast vs. slow twitch muscle composition, etc.)

  11. @JD – That is pretty cool information. Thanks for sharing.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.