Am I crazy because I eat too much sugar? ~A CGM experiment ~
When I saw this skit on SNL, I felt relieved. Because this is what I want do at every meal but too ashamed to admit it. I love you Tina Fey.

I always loved my carbs, and I started wondering if my mood fluctuations are results of riding the glycemic roller coaster, after reading this post.
I am not diabetic.
I got myself a sensor to see if my diet is the cause of my bad moods. I got some control equipment to see how different the CGM (continuous glucose monitor) and capillary blood reading was. The CGM manufacturer said its readings are delayed since it reads extracellular glucose (interstitial) rather than actual blood glucose. CGMs became big about 5 years ago, when the FDA approved these devices for diagnosis and monitoring. Apparently my coworker’s tiny diabetic yorkie wears one. Abbott sells the cheapest one compared to its competitors, where Medtronic is on the high end. Yes Abbott as in the maker of Similac.


Now on to stapling this baby onto my arm! The applicator is well-designed and I didn’t feel a thing! I got over that self-harm hesitation by counting to five. The needle punches through to place the sensor and then it retracts. The needle itself does not stay in your arm. Just the flexible sensor. (See end of post when I remove it).



I mostly went about my business as usual wearing it. I tried to record my fasting glucose using a blood sample every morning and test glucose whenever I’m feeling a hangry monster coming on. I also did some ‘loosely controlled’ experiments.
The sensor itself can store the entirety of the 14 days’ data. To check your live reading and download a chunk of data, hold your phone within 4cm of the sensor and scan yourself like produce. (Okay I know it’s not a barcode. It uses RFID/NFC)


Now of course we should start with a glucose tolerance test. For diabetes diagnosis, 75g glucose is used. I did 50g since it’s the common gestational diabetes test (which I have failed before).


I ate 12.5 glucose tablets in 5 minutes. Not the best way to start a morning. But FOR SCIENCE!
My glucose level rocketed up to 200! At 1 hr it was barely under the 140 threshold. By 2 hours in, it was back to baseline.
You can also see that the blue line for the CGM reading lags the blood readings by about 20–30min. Which would explain some of the differences between capillary and interstitial readings. The large fluctuations also made the CGM think it was going crazy and prompted blood sample testing.
Below is what the tolerance curve looks like for diabetics and criteria for diagnosing diabetes.


I have done a 100g glucose test while I was pregnant since I failed the 1hr one, I felt fine during the 3 hour test, but boy did I feel crappy late afternoon! I think it’s from reactive hypoglycemia, where your body dumps out so much insulin to take care of the glucose that your blood glucose actually undershoots — otherwise known as a sugar crash.

Now let’s look at some more normal scenarios. (aka. a deeper look at my crappy eating habits.)




I have a pretty sedentary lifestyle. My glucose level went up in response to exercise, which makes sense that I needed energy to avoid the falling children. Supposedly athletes’ bodies can keep their glucose levels pretty even despite the physical load.
I started this experiment wanting to know if sugar levels caused my hangry and food coma episodes. So I logged these events in the Abbott app and back traced a couple hours to try to figure out what led up to them.


I thought I would get food coma at the peak of a glucose spike, but it was actually on the downward slope. Hangriness was harder to predict since I had less data. But I am definitely prone to this if I had a small carby meal too long ago. I didn’t get hangry when I had lots of protein, or had big meals even if they were carb-heavy (it probably lasted me longer).

It took a bit of prying and lots of cringing from my family to take off this thing. The sensor is a flexible wire with a contact on it. There are videos on the web for tear-downs for anyone interested in what’s onboard.
Conclusions:
I think I get hangry because my blood glucose is at fasting levels for too long. Unless I have a soda for breakfast, the fluctuations from eating food are not big enough to trigger a reactive hypoglycemia episode. So don’t have a soda for breakfast guys! (And plan your meals reasonably.)
Storing massive amounts of sugar into glucagon makes me sleepy. If I have mainly protein, I can basically stay mentally sharp all day and barely think about food. It’s kind of joyless. (Says the sugar addict.)
Food for thought:
Does eating fiber and fat with sugar really decrease the glycemic response (putting a coat on naked carbs)? Maybe next time I’ll take Metamucil or bacon grease with my glucose tablets.
I don’t really know what value there is for testing gestational diabetes at 50g. Why use a different dose? The results should also be compared to your non-pregnant self since there’s loads of interperson variability in glucose sensitivity.
I’m excited for this technology to get integrated in the future in personalized health products like Fitbits and Apple watch! Blood pressure and oximetry monitoring are also elements I’d be interested in studying. I think it’s great to see real-time data from your body so you are more aware of the impact of your lifestyle.
Disclaimer: I’m not affiliated or receiving compensation from any product or person mentioned here.