After coffee, I take on the morning’s math problem: figuring out the insulin needed for my kid’s breakfast. So much for total carbs, minus amount spilled and not eaten, plus replacement carbs eaten, minus some for gym class in the morning…
Life is all about numbers when you’re living with Type 1 diabetes (T1D). Type 1 is an autoimmune disease that destroys the body’s ability to produce insulin or regulate blood sugar. You then own figuring out how much insulin the body needs – and that can vary minute by minute, depending on over 42 factors. Too little, and blood sugar goes high, causing complications. Too much, and blood sugar can go dangerously low. You’re constantly trying to hit a moving target.
“Just adjust based on experience,” is the standard, keep-calm-and-carry-on advice.
To which the appropriate response is:
Trying to capture that “experience” is incredibly challenging. It means logging blood sugar and insulin doses 12+ times a day (ideally with notes), and then analyzing patterns after a week or so, usually in Excel. Why Excel? Because the various devices used to manage diabetes (blood sugar meters, insulin pumps, continuous glucose monitors) all use proprietary systems, so there’s no connecting their data. No wonder less than 5% of people with T1D look at their numbers!
Enter a frustrated, tech-savvy parent who saw the need. Howard Look had worked at TiVo, Pixar, and Amazon, and when his daughter was diagnosed with Type 1, he was appalled at the how outdated management tools were. He kicked off the #wearenotwaiting movement and developed Tidepool, an open source platform that sucks in data from a range of devices and displays them in a clean, interactive, user-centered dashboard (“Blip”).
As an end user, I love this. I’m no longer staring at a cloud of numbers or trying to wrangle disparate columns in Excel. Now I’m looking at graphs that highlight blood sugar patterns I need to fix – and I can see updates in minutes.
But I might not think to look for a week or two, so I love that another app we use, Dexcom’s “Clarity,” forces me to look by sending weekly alerts and reports that highlight key measurements (time in range, severe events, standard deviation from target blood sugar, etc.). The reports even suggest causes when they’re apparent (i.e. rebound highs caused by over-treating low blood sugars) – things I might easily miss.
The next step is turning algorithms and machine learning loose on the numbers to do more complex analysis and eventually putting them in the driver’s seat instead of a beleaguered human making decisions between meetings, at a birthday party, or in the middle of the night. It’s starting, and I can’t wait to hand over the keys!