Ice cream confounders

And a fine example of how hard it is to identify symptom causes/triggers

My mind is badly cluttered with messy hunches about why my symptoms surge when they do. I am always hoping to find modifiable risk factors — like the silly ice cream thing I brought up a couple weeks ago. I have been wondering if my summer rough patch is related higher consumption of cheap ice cream.

I have a “result” from that experiment today. But not really, because it was muddled by other issues… predictably. Science is hard and messy. But tasty, in this case!

The power of data

I now have about five months of detailed logging of symptoms and potentially relevant factors. That spreadsheet is a beast now. And I am starting to see the potential to begin to upgrade (or downgrade) some of my many hunches with something resembling actual data. Maybe I can start actually answering questions about these things. Questions like:

• Does heat really give me trouble? Kinda seems like it.

• Does ibuprofen blunt my crazy soreness after exercise?  I do have that impression.

• Does my symptom severity correlate with cumulative sleep deprivation strongly, or only loosely?  I’m thinking "strongly," but I’m not really sure.

• Do I feel better when I eliminate alcohol? Really does not seem like it, but have I ever carefully tracked my consumption and symptoms to actually confirm it?  I have not.

• Is exercise the one trigger to rule them all, the only thing that fairly consistently predicts rough patches?  There's a good chance the data will show that pattern, despite some exceptions… I reckon.  And it will remain a reckon without data.

And the (somewhat silly) question of the hour, introduced a couple posts back…

Is daily consumption of cheap ice cream the reason I’ve been in a rough patch all summer? 

Data says nope. After a surge in consumption for several days, I et up the last of the ice cream, then cut it off sharply.  If the ice cream was trouble, then I should now be getting some relief.

But I am not. 😒  There have been no changes.  This rough patch just keeps going.

But the data is soooo messy

There are at least four things potentially obscuring the results of my ice cream test, good or bad, four perfect examples of “confounding factors”:

1. Insomnia. A multi-day episode of extreme insomnia left me severely sleep deprived… peaking right at the same time I was hoping I might see some improvement from backing off on the ice cream.

2. Exercise. There’s another plausible source of my summer rough patch, starting at about the same time I got into eating extra ice cream: I have been testing myself with stronger workouts than I have dared to attempt in years.  My goal is to establish a baseline, to find out just how much

3. Exercise… in the other direction. Meanwhile, I actually removed another source of exercise intolerance recently — a long story, to be told soon in another post. Any improvement in the last week could also be better explained by that.  Not that there was any improvement.

4. Heat. And then there’s the summer heat, which got going around the same time I started eating way more ice cream than usual — not a coincidence — and yet it is also a potential source of symptom aggravation.

“Science” is hard

I repeat: I don’t take this ice cream experiment seriously at all.

But the question is legitimate, and I have some data, and it’s a good model for more serious questions. It’s practice! Just forcing me to think about confounders is useful, because they are why it’s so tricky to figure out symptom causes and triggers. The only defense is more/better data, and learning to control for confounders as well as I can.

And so, as silly as it is, I actually will do three more well-defined periods of high ice cream consumption — each one at least a week long, separated by ice cream droughts. And I will try to schedule around the confounders.

So, more tests. Which means more ice cream. 😋