View the site at food.marcos.ac!

The context

So, I have a condition called eosinophilic esophagitis (EoE) that basically causes my throat to close up over time (in the absence of medication) due to a chronic allergic reaction. It’s a pretty rare disease, but every once in a while I’ll learn of somebody else who has it, like my dentist or John Green. But unlike other allergic conditions where consuming the allergen produces a near-instant reaction, I wouldn’t know anything was wrong until about a month of continuous exposure to the allergen, at which point I start to have difficulty swallowing. This makes finding the culprit(s) much less straightforward.

When I was younger, I tried a few elimination diets. First, we eliminated milk from my diet for two months, but saw no improvement in the endoscopy. Next, we removed milk, wheat, soy, and eggs for another two months. No dice. Undeterred, I went two more months without eating those four ingredients plus corn, potatoes, and rice. Still no good. Finally, I tried a last-ditch attempt which also removed chicken and beef and similarly did not work. At that point, I was a high schooler who just wanted to eat. So, I started taking prescription drugs to treat my condition and decided to worry about it later.

Then, later came. I had graduated college and didn’t want to take expensive¹ weekly injections for the rest of my life if I could help it. I was determined to figure out once and for all what I was allergic to, so I turned to the “Plan Z” in these situations: the elemental diet.

The elemental diet

I think of the elemental diet as the complement of the elimination diet: instead of eating everything except a small list of ingredients, you eat nothing except a small list of ingredients. The first step was to live solely off of these insurance-covered² amino acid shakes for two months. Assuming that worked, we would slowly re-introduce foods into my diet in small batches every 1-2 months. At that pace, it could easily take two years to build up a normal-looking diet that doesn’t trigger my condition. For reference, this is what a 30-day supply looks like:

Those first two months were pretty great, actually. Just kidding, obviously it sucked. I remember the first day especially well. The night before, I’d enjoyed my “last meal”; a philly cheese steak from the corner deli. Waking up and remembering that I had two foodless months ahead of me, followed by a much longer stretch of extremely limited and philly-cheese-steak-less diets was instantly demoralizing. The temptation to give up before I’d even started was overwhelming.

At first, I planned on breaking up my amino acid shake consumption into meals: I would drink two boxes for breakfast, three for lunch, another three for dinner, and another two spread out throughout the day as snacks. That morning, I downed my first two boxes, practically forcing the second one down because the taste was so strange and artificial. I somehow felt hungrier than before. This doesn’t bode well, I thought. The whole day, I felt hungry and tired.

The next day was a workday that I definitely should have taken off. I decided that I needed to make this diet productive somehow, somehow gamify the experience so that I could externalize my progress and see it at a glance. Vaguely inspired by Nolen’s One Million Checkboxes, I got to work.

My countdown to real food

The tracker was made more or less in one sitting. I used a Firestore database which contained a list of status updates and a single number, representing the total number of “juiceboxes” drank so far³. I had a number (originally 300, then later 600) representing the number of boxes I’d need to drink before getting a taste of “real” food again.

Unlike a few of my previous web dev projects, I focused on designing the site for both mobile and desktop instead of leaving the mobile experience as an afterthought. One styling trick I learned this time around was the use of auto-fill in the grid-template-columns CSS property, allowing the number of columns in the juicebox grid to adapt responsively to the available width. I like using Firebase for these one-off projects because (a) it’s free, and (b) react-firebase-hooks provide super convenient hooks for optimistic updating and component-level reloading when certain collections change.

I had a shortcut on my phone to increment the count and add updates whenever I felt like it. I can honestly say it made a surprisingly big difference to watch that number go up each time I downed a box or two, and to log my journey without food, even if my updates went straight into the void.

Well, not completely into the void. There were a few people that checked in with my website regularly. To those people: thank you so much. I appreciate you.

As time went by, the diet got easier and easier. I suppose my brain learned to associate the amino acid shakes with nutrition, because they started to taste better, and I started to feel full after drinking them. Anticlimactically, the days kept passing, and eventually, I had made it through two months, and an endoscopy proved that my condition had gone away in the absence of food. We take our wins where we can get them.

The journey continues

This story probably won’t have a conclusion for a long time, but I have gotten to eat some food again. Recently, I’ve been eating pork and quinoa, which isn’t half bad if you salt it right. I still go through 8-10 amino acid shakes a day to get my caloric intake. Pork and quinoa is definitely a step up, but I’m still counting the days until I can eat a slice of pizza again, or have some tiramisu.

I decided to stop updating the site, since I no longer need it to motivate me, nor do I have the reflex to increment the counter each time I drink a box. But I’m still very happy with it, since it goes to show that one day of effort can have a positive impact for months to come, if not longer.

Addendum

Don’t take thirty amino acid juiceboxes through TSA.

Also, a lovely wood cake my family made me to celebrate my foodless birthday:

Footnotes

1. Without insurance, my injections would run me $3,993.36 per carton, which would last me two weeks. That or take steriods, which are cheaper, but have well-studied negative side effects. ↩︎

2. I’m lucky that I was able to get my insurance to cover there. Health insurance companies are notoriously stingy on covering medical food on the basis that they’re not a “medical necessity”. ↩︎

3. In retrospect, it would have been smarter to represent this as a list of timestamped increments, so that way I could plot a nice time graph and still get the total by summing the number of rows. Oh, well. ↩︎