What endocrinologists and diabetes teams should know about exocrine pancreatic insufficiency (EPI)

by Dana Lewis

Recently I was asked to give a talk to a group of endocrinologists and other diabetes providers about exocrine pancreatic insufficiency, and while building my presentation I was thinking through everything I wish endocrinologists knew about EPI. (We call it EPI in the US, but if you are elsewhere in the world you may see it referred to as PEI, for pancreatic exocrine insufficiency. This is the same condition.) 

This is targeted toward endocrinologists and other diabetes care providers and heavily focused on people with diabetes but this is also relevant to gastroenterologists and primary care providers as well, and includes info on prescribing enzymes that is relevant to anyone with EPI regardless of any co-condition.

1. Exocrine pancreatic insufficiency occurs – and is common – in people with diabetes. All types. 

I was diagnosed with EPI back at the end of 2023 by pure luck, because I saw a post about a dog with GI symptoms that were solved by dosing enzymes with their food. I was curious and looked up if this condition occurred in humans, too – and it does. It’s called exocrine pancreatic insufficiency (EPI or PEI) and happens when the pancreas no longer produces enough digestive enzymes to help digest the fat, protein, and carbohydrates that you eat. I had a textbook case of symptoms, yet despite ongoing investigations with a gastroenterologist for over a year, it was never brought up on the list of possibilities as a condition we should screen for. The more I read, particularly seeing that yes, it occurs in people with diabetes, and studies show around 30% prevalence in people with type 1 diabetes and people with type 2 diabetes, I was flabbergasted that my gastroenterologist never considered it.

And why had I never heard of it? I had heard of the connection between T1D and celiac, which is why I ended up getting screened and diagnosed with celiac very quickly during college when I developed GI symptoms. I hear of gastroparesis quite frequently as well in people with diabetes. But EPI is likely more common than celiac and gastroparesis combined and I never hear it talked about by endocrinologists, researchers, or people with diabetes.

I could have been saved from over a year of suffering if my gastroenterologist was more aware of this or if my endocrinologist was more aware of the prevalence of EPI in diabetes. It’s now my goal to make sure we cut this window of suffering for other people with diabetes. So the number one takeaway for endocrinologists is that EPI occurs in people with diabetes, it’s more common than celiac and gastroparesis, and even if you don’t screen for EPI first, if GI symptoms are occurring and you’ve ruled out celiac and gastroparesis, screen for EPI next! (If not first, depending on symptoms).

  1. How do you screen for EPI in people with diabetes?

Endocrinologists can absolutely order the screening test for people with diabetes.

(Yes, you can refer to a gastroenterologist…but…sadly…many gastroenterologists are not necessarily more advanced in their care of EPI, as my experience has illustrated. Gastroenterologists also underscreen, underdiagnose, and undertreat EPI across a wide variety of conditions, so unless you have a top-notch GI you know personally who is familiar and treats EPI with the actual standard of care, more on that below, you should consider ordering the test yourself and managing the EPI process for your patient with diabetes.)

The screening test most commonly used is not perfect, admittedly, or fun. The fecal elastase test requires a stool sample and getting it back to the lab in a certain time frame. But, it can be illustrative of EPI or likely EPI. The typical guidelines say <=200 ug/g indicates EPI. Historically there have been cutoffs for <100 and <=200 for different categories, but it doesn’t really matter because you don’t prescribe based on severity level or treat any differently. The 100/200 cutoffs just give you data from past research on the accuracy (sensitivity and specificity) of the elastase test.

But human elastase fluctuates, so someone could have a test with a result of 205 ug/g, for example, at the time they are tested and other times be more obviously under the 200 cutoff, e.g. 155 ug/g or some other random number.  My plea for endocrinologists is to be aware of the limitations of the elastase test in both directions, and contextualize the fact that an elastase >500 ug/g likely means not EPI, but something in the 200-500 ug/g range, though, is worth repeating a test and/or considering a trial of enzymes. (There’s an interesting study showing people with 200-500 elastase benefit from enzymes, so keep that in mind). I wrote more about elastase testing and the sensitivity and specificity here.

The TLDR I wish people knew is that 200 is not a precise, exact cutoff. I think it is incorrect to treat 201 ug/g as “you don’t have EPI!”. This is where clinical judgement comes into play, depending on if you’ve already ruled out obvious other issues and symptoms continue, and you have an elastase near or below the 200 ug/g ‘cutoff’, then you may consider enzyme trials with your patients.

Another factor to take into account is EPI-specific symptom frequency and severity. See the EPI/PEI-SS section 5 below.

And yes, EPI occurs in people who are overweight and it’s possible to have EPI and not (yet) have vitamin deficiencies. It doesn’t require unexplained weight loss or pain/pancreatitis symptoms or vitamin deficiencies in order to diagnose EPI.

It was also interesting, during my presentation, to hear from an endocrinologist, the wonderful Dr. Marina Basina from Stanford, who was sharing how she is also screening some people with diabetes who are “asymptomatic” (GI-wise) but have unexplained extreme spikes and drops on their CGM. Once they are on enzymes, they’ve been telling her how they feel like different people and have so much energy compared to before. This “asymptomatic” presentation – with some possible ways to detect based on CGM data and other atypical symptoms – is something I hope we research more in the future.

  1. Endocrinologists can prescribe pancreatic enzyme replacement therapy (PERT) and should do so according to guidelines.

Endos absolutely can and should be willing to prescribe enzymes (aka PERT) to their patients with diabetes, if they’re able and willing to help manage the prescription. But there’s a few things to know.

The first is that most guidelines recommend 40-50,000 units of lipase per meal. ( EPI in diabetes doesn’t have its own specific ‘guidelines’: because EPI is pretty similar across conditions, there’s no reason not to use guidelines developed for EPI in other conditions as a starting point for dosing. Here’s a systematic review of all dosing guidelines across conditions including the evidence and best practices for diabetes and EPI.)

Coming from units of insulin, this sounds like a lot, so you (and patients) have to wrap your brain around the dosing sizes, first. But it’s actually not that much. Pills come in a variety of sizes depending on which brand, ranging from 3,000 units of lipase per pill up to 60,000 units of lipase per pill. The pills actually contain THREE types of enzymes: lipase, for fat digestion; protease, for protein digestion; and amylase, for carbohydrate digestion. They all have different amounts per pill but fat is the most commonly (95+%) researched factor and so the pills are described solely by fat/lipase amounts. I’ll stick to that in this section, but read more about protein below.

Please, oh please, don’t prescribe 3,000 or 10,000 to “start small”. The dosing guidance is 50,000 (ish) per meal as a starting dose. Most people actually need more than that, depending on what they are eating (fat, protein, etc). Enzymes are like insulin in that underdosing can feel like a drop in the bucket. But, ‘too much’ is not like insulin. ‘Too much’ usually requires a lot more than the equivalent insulin dose to cause issues. Most people are massively underdosed and if you start with a 10,000 dose, for example, it’s common to then infer that “it’s not working” and it must not be EPI, and people stop taking it (or the prescription is ended). In reality, ‘it’s not working’ is usually “this is not enough” for the person’s individualized needs and what food they are consuming. Again, the starting dose is around 50,000 units of lipase per meal, with it being common to increase as high as 150-200,000 per meal, depending on diet. There are some studies that have some correlation to ‘severity’ of EPI, but it’s also very individual on the person’s body AND what they are eating. Someone who eats a higher fat diet – such as people with diabetes who might choose a lower carb diet with more fat and protein – will need more enzymes than someone else who eats a low fat, high carb diet. (And individual to individual it varies).

So how much should you prescribe? And how?

4. How do you prescribe enzymes to someone with EPI? 

Given the above, you can Rx a starting dose of around 40,000-50,000 units of lipase a few ways. One brand, Zenpep, has pills of 25,000 units of lipase. Let’s do the math: this means 2 pills of Zenpep 25,000 per meal. Another brand, Creon, has pills with 36,000 units of lipase. You could choose to start with 1 pill per meal (36,000 units of lipase) or two (72,000 units of lipase). Either would be suitable, BUT regardless of which brand and which dose you choose to START with, this has to come with a conversation and education to the patient about what they are taking, WHY, and WHEN.

I found that both fat and protein have to be titrated. You can give someone a low fat meal with enzymes that work and then give someone the same amount of fat with protein and get massive symptoms. Some people conclude that they can’t eat that food whereas in reality, it’s the protein that needs additional enzymes. (I have a lot of specific examples in this blog post showing how common low-fat high protein meals like chicken and veggies may need as much enzyme dosing as higher fat meals, because of the protein content.)

This means you need to teach people that they can expect to develop a ratio of lipase:fat and protease:protein similar to how they have an insulin:carb ratio. This is something they can self-titrate and figure out about how many grams of fat or protein one pill “covers”, keeping in mind that they can have different pill sizes.

You should write the Rx to be taken with food. Do NOT write this as “take three times a day” because you assume people eat three meals a day. This is faulty because 1) people don’t then understand it is WITH FOOD and if they’re not eating they shouldn’t take it and 2) some people eat snacks and they need to dose for snacks, too, plus 3) some people (especially those who also have gastroparesis or other situations) eat several small meals a day.

I would explicitly explain to people that they need to take this with food and drinks with nutrients (so water and diet soda do not need enzymes) and that you are writing the prescription for X pills a day to be taken with food. And, tell them that once they figure out what dose is working for them and how many pills a day they need to call/message in and that you will update the prescription for them. That’s what happened for me: my Rx was originally written for 2 pills per meal, so I got 6 pills per day. I ended up calling and asking them to change my Rx to 8 pills per day* so that I could have snacks some times, or to otherwise to support the flexibility of sometimes having a larger meal that requires more than 2 pills.

(* Remember the dose matters; I am not specifying my dose/pill size/brand here, but this is not to say that 8 pills is the magic number for everyone.)

The benefit of educating people this way is it helps anchor that 1) they need to figure out what dose works for them and 2) the starting dose may be not what they need.

I am over-emphasizing this because in online communities I see a huge number of people who are not told that this Rx is for consuming with food (or drinks with nutrients, e.g. milky drinks) so they take it on an empty stomach because “they’re supposed to take it”, or they don’t realize they need to adjust it in proportion to their food; or they run out and can’t get a refill because of how the original prescription was written and they for whatever reason can’t get in contact with a provider to increase the prescription. So yes, this type of education when you provide the prescription matters a lot to set people up for success! 

You can also, eventually, prescribe multiple ‘sizes’ (doses), e.g. a Creon 24,000 and Creon 36,000 or a Zenpep 25,000 and a Zenpep 60,000 based on what works for people. If someone can take 1 Zenpep 60,000 for most meals and is satisfied with that, great; they may not need that dose for small snacks though and would benefit from the additional size (e.g. Zenpep 25,000) to have on hand for snacks. Or to ‘top off’ a bigger meal without doubling up on the 60,000. (2x 60k is 120,000 whereas a 60k and a 25k becomes 85,000 units of lipase). That’s likely a follow up conversation after they’ve done some logging and titration work to figure out what works for them and what they typically eat, plus outlier meals.

Here are some common scenarios of when most people should take enzymes and some “it depends but most people don’t need enzymes”:

  • One bite of something (e.g. tasting as you are cooking dinner) – because the amount of fat, protein, and carbs in a lick or taste is pretty minimal, people could maybe get away with no enzymes.
  • Black coffee – probably doesn’t need enzymes, unless:
  • Black coffee with milk/cream/etc added – probably needs enzymes depending on the quantity added.
  • Grazing as you are at a buffet with a nibble here, a nibble there – probably needs enzymes as the grams of fat and protein go up.
  • A snack that has more than a few grams of fat/protein – probably needs enzymes
  • A piece of fruit – varies person to person. Some people can eat fruit standalone with no enzymes, others seem to need the enzymes. Why? Because of the carbohydrate digestion, even though there is little to zero fat and protein involved.
  • A salad that is low fat – probably needs enzymes especially if it has protein, even though it doesn’t have “much fat”. (Also some fat likely still needs enzymes, in and of itself.)
  • A high fat meal – needs enzymes and likely more enzymes than typical if it’s a bigger meal or a meal with a higher quantity of fat than typical. E.g. if a typical meal is around 45g of fat and someone takes two pills (of whatever size), that means their pill:fat ratio is 1 pill per ~22g of fat. If they were to then sit down to Thanksgiving dinner and have 60g of fat, they probably need at least 3 pills because the additional fat beyond 45g of fat needs to be dosed for. More examples here about figuring this out.

How many grams of fat and protein require enzymes? The great “it depends”. For me, I don’t have an absolute number but if it’s greater than 2-3 grams of fat or I am uncertain about the quantity I am eating, I take enzymes.

I do not take enzymes for “low” treats that have no fat and protein. So this means no enzymes (for me) for:

  • Juice
  • Soda
  • Candy without significant fat/protein (airheads, gummies, fruit snacks, etc.)
  • Glucose tabs

(It likely depends on the amount – probably a larger amount of carbs-only would still require enzymes. But for me since I do small or medium amounts of carbs for lows, I haven’t found a threshold where this seems to make a difference.)

I also don’t need enzymes normally if I was drinking water or diet soda.

I WOULD take enzymes for lows if I was consuming:

  • Peanut butter crackers (fat and protein need enzymes)
  • Hot chocolate made with milk (milk has fat and protein that need enzymes)
  • A snickers bar (fat and protein, needs enzymes) or other candy like this that has fat and protein alongside the carbs.

There’s probably also a threshold for which you need enzymes for an amount of carbs, but both because enzymes have a rather large dose of amylase in them in every pill; there is little (zero?) research on carb digestion needs with amylase among EPI research; and I have little anecdata to give you on exactly where this falls, I think it is less typical to run into this. Probably most of the time people get away with dosing for mostly fat and protein and carbs are mostly covered by that; but it’s also common for people to dose for fat and underdose for protein and get symptoms as a result.

  1. It gets better (EPI can be successfully treated) and we can track it with the EPI/PEI-SS

It does get better. This is an important message for people with EPI. It can be more challenging when there are multiple conditions going on at the same time, but EPI in and of itself can be managed with dosing enzymes.

And, we can track it getting better. I made the Exocrine Pancreatic Insufficiency Symptom Score (the EPI/PEI-SS) because there wasn’t a good symptom screening or tracking score for people with ALL types of EPI. (The existing ‘scores’ were developed only from people with EPI and chronic pancreatitis or cystic fibrosis and I found that the symptom list was too narrow and did not account for frequency and severity). It uses 15 symptoms and a self-rating of both frequency (0-5) and severity (0-3), and the resulting score can range from 0 to 225.

Like elastase, the ‘cutoff’ is not precise – based on an early real-world study of n=324 people, we found around 55 to be a cutoff indicating a greater likelihood of EPI. Can someone with a score of 45 have EPI? Yes. Can someone with a score of 62 not have EPI? Yes. It’s not that precise, but that doesn’t mean it’s not useful.

This symptom score can be used in a variety of ways:

  1. To increase the pre-test probability for subsequent elastase screening
  2. To help physicians and patients understand exactly what frequency and severity of symptoms the patient is dealing with.
  3. To be able to see over time with enzyme dosing how the symptom experience is changing.

For example, someone who was diagnosed with EPI had a symptom score of 72. After they went on enzymes and titrated their dosing, their score is now 7. That is a huge improvement and puts their symptom experience on par with people who don’t have EPI (or any other GI condition).

Another scenario may be someone who had more extreme symptoms – imagine an EPI//PEI-SS score of 155. They went on enzymes, and a few months later their score was 96. That’s an improvement, yay! But, it shows that they still likely have room for improvement and their enzyme dosing likely needs ongoing review.

(Note: the EPI/PEI-SS was evaluated in an online, real-world study of n=324 people, including a large number of people with diabetes. The diabetes sub-analysis is here, showing that having either type 1 or type 2 diabetes did not significantly influence the EPI/PEI-SS scores. We have an ongoing clinical study right now based at University of Florida in partnership with Dr. Brittany Bruggeman evaluating the use of the EPI/PEI-SS with elastase screening for people with diabetes. We also have an EPI/PEI-SS study running in New Zealand and Australia, with Dr. Amanda Landers, for people with pancreatic cancer. All this is to say that EPI/PEI-SS is being evaluated further for screening for EPI but we have not yet kicked off studies for using it for ongoing titration of enzymes or for capturing what ‘asymptomatic’ presentations may look like. This is an area I’d love to study, though, so if you are interested in studying its use alongside titration protocols for enzymes or asymptomatic presentations – please reach out to collaborate!)

—

A blue square with white text that says "What endocrinologists (and diabetes teams) should know about exocrine pancreatic insufficiency (EPI/PEI), a blog post by Dana M. Lewis from DIYPS.orgIf you read any or all of this – thank you for your interest in supporting people with exocrine pancreatic insufficiency. We need more of you! If I can help your work in any way – please reach out to me any time (Dana@OpenAPS.org).

TLDR summary:

  1. EPI occurs in people with diabetes and is more common than gastroparesis and celiac. Screen for EPI in people with diabetes with GI symptoms! Some patterns of unexplained low energy/fatigue and extreme glucose spikes and drops on CGM also might warrant EPI screening in PWD.
  2. Elastase is not perfect but what we can most easily access. Endocrinologists can test for it. A result around 200-ish or lower warrants an enzyme trial. Endos and diabetes teams should care about EPI because it impacts glucose levels!
  3. Give a reasonable dose of enzymes to start. Reasonable means 40-50,000 units of lipase or more. Endocrinologists could manage the enzyme prescription.
  4. Write an Rx for enzymes to take with food. Teach people to adjust their dose in relation to what they eat; the fact that it will take some time to figure out a dose; and encourage them to call back and ask for an adjusted Rx based on what they need.
  5. EPI sucks when untreated, but it does get better and we should continue pursuing better and adjusting enzyme doses until symptoms are improved.

How I work with AI (September 2026) and how I do more types of things with AI than before

by Dana Lewis

I have previously written about how I work with AI and the kinds of things I do with it. Even since I last wrote about that in February, things have changed. A lot. In February, I described creating a hacked-together remote LLM prompting tool to enable me to trigger ongoing use of my LLM while I was away. I didn’t use it much, just occasionally, because it was hacky and required me to set it up in each project I wanted to use it for (by design – for security reasons). But, then remote use came out for most AI agents, and my hacked-together-solution was no longer needed. My use of AI for “laptop tasks” went up dramatically, while my time spent at my laptop physically went down. This was a huge win for me because for various reasons sitting at my laptop is not always feasible. Now, I could be out walking or hiking and keep projects running. I could also be sitting in a different room and also keep projects running, even with my laptop not in reach. This was a big and delightful change for me.

The development of coding ‘agents’ that can go off and do a task as directed, more reliably than before, has also been a huge change and benefit in the last six months, too. (And now, sub-agents! One agent can direct others to run in parallel.)

If you haven’t given it a try, I highly recommend you try some of the coding agents and use them remotely from your phone to see what they can do and what situations you might benefit from using them in.

But it’s not just “work” that I’m doing more of or differently. I’m branching out to solve more and more of my software-shaped-feeling problems. I continue to build custom apps (like Scheduler Pilot and Protocol Pilot – the latter of which is available if anyone else needs/wants an app to help with tracking and titrating medications) at the drop of the hat. Collectively, I have over half a dozen custom apps I use *daily* which really surprises me (that I have and use so many). Some I only need to use once a week or so, but several I use daily.

I’ve also been able to drastically improve other existing custom apps I have. One big example is Baseline Pilot, which I built last year to enable Scott and me to share our biometric data every morning. For the last ~10 months or so, that has involved us manually opening the app and hitting a button to export the data, and the other person clicking it and opening the app to load it in. It was revolutionary at the time, but it required manual effort on both sides. Recently, though, I had to deploy a server for a work project (which, hello imposter syndrome, I’ve never done before) and do all kinds of technical stuff that I am hugely intimidated by (or – was!) and was still able to do. Because of that, I lowered the friction / barrier to entry for other projects to use a server for data storage and data transfer. As a result, I ended up building a family-focused 3D printing management app so that multiple people in the family could submit files to the person in the family who has a 3D-printer (and so they could access the files in one place and put them in queue etc). That was intimidating, but I had figured out how to do that, so the next thing on my mind after work use and the 3D printing management app use was to then think about if this would solve my manual-click burden for Baseline Pilot. And the answer was yes. Using the same infrastructure that I had set up for the other projects, I was able to quickly (less than an hour) get Baseline Pilot set up to load data into the database in the server and pull down to my and Scott’s apps – automatically – without having to click to share data every day (and click to import the other person’s data every day). It…just works! I open Baseline Pilot and see Scott’s data, without him ever clicking and exporting the data. Woohoo! It thrills me every day for the past several days to open the app and see the data. (And yes, I have it set up where it will send me a push notification for any data above the cutoff I specified, eg >3 standard deviations from normal on key metrics).

I also built a new app called RoutePilot, which allows someone to pull in completed workouts from Apple Health and see if any run intervals were detected. This way, someone learning to run and doing run/walk intervals can track how much running they did without having to hit things on their watch or phone to manually track it while they’re doing it. Because I built this and was showing it to Scott, he realized some of my visuals gave him ideas for how he might use some of this to create cool displays to show pace versus distance PRs of different segments, both for retrospective analysis (my app’s original intent) but also live during a run. This then evolved as we realized this was useful for our serial hiking when we repeat hike routes and find it motivating (sometimes) to know if we are near or above our previous best for a route, so we added in some neat hiking features too. Just yesterday while out hiking, we were doing a “new” route, where we did a hike we’ve done before as a loop but in the reverse direction. Scott wasn’t aware on the way down/back which connecting trails we were taking. It wasn’t as easy as he wanted to be able to adapt the route we had plugged into the app. I suggested being able to load the trail maps and tapping the segments to quickly build a route, that way we could see it’s distance and elevation. He set the agents off and by the time we were home, it was built. Today, I picked up his code and worked on it some more, solving some bugs to allow me to connect segments in different ways, improving the look of it, and adding additional features to this. This type of project used to take us months. Now it’s minutes/hours/days instead of days/weeks/months.

It’s very cool to use AI to help me learn to do new things and then be able to take those skills or awareness of what is possible and apply it to more and more projects. Plus, because I am getting more done on ‘work’ and ‘personal projects’ (using quotes because as an independent researcher there’s a lot of overlap, some unfunded projects are also work), I have more momentum and energy for other things in my life, too.

One surprising example is that I am cooking a lot more. But not just because I have more time. It’s mostly because I have figured out what kind of feedback and input I can get from an LLM to make it easier for me to cook. This actually started by getting Scott to help me cook more. I would figure out what we wanted to make; plan and buy the ingredients (e.g. put them in our grocery delivery); decide what day we wanted to make it; put time or a request for time on the calendar to do it; then tell Scott what I wanted him to prep or get out to help me. Sometimes I’d then do the ‘cooking’ or baking, and other times I’d have him do it all. He gave me feedback on what methods were most helpful – which is a preferred style of listing ingredients in the recipe and certain steps he wanted spelled out. We began using a shared project folder in our LLM of choice where he could see recipes that I customized in the format of his preference. He also liked being able to ask it questions as he went, whether it was converting units on the fly or what cooking instrument to use (e.g. spoon versus whisk, etc). Then at the end, we’d update the recipe for next time to include all the decisions we made and tweaks to the methods or ingredients, so we had a better working artifact (err, recipe) for next time if we made it again. This iterative feedback process made him more willing to do this and it made it easier for me to start exploring other recipes we could make this way, too. I began to be more ambitious with finding varieties of the things we already liked. I even was ambitious and after eating something at a restaurant I really, really liked, had my LLM look up and find the original recipe and then help me adapt it.

(We do a lot of adapting of recipes. First, because I have celiac and I’m gluten free. Second, because we are fairly minimalist and are happy to use frozen pre-chopped or canned vegetables or other ingredients like that when possible. The lovely thing about talking to an LLM to convert a recipe is to say “we have X, Y, Z – use that” even if the original recipe called for something else, or to ask if it B works as a substitute for A or not, and what the tradeoffs are if we used it anyway. This is helping me overcome some of the stigma I faced earlier in life when talking to other people about the hardships I found cooking – some of which was because cooking for 1 person was hard, some of which was because there were far fewer gluten free options or substitutes a decade ago, some of which is because modern technology is awesome and now there’s fantastic options like more canned and frozen pre-chopped vegetables (frozen pre-chopped onions! Frozen cubes of garlic or ginger!) that didn’t exist back when I first started. I have had family members and friends be fairly dismissive when I talked about how I found it challenging to cook the way they did. Which was frustrating because it always felt like a moral failing on my part when in reality I had more barriers and fewer resources accessible to me. Now, the resources are all here – more GF options, more general resources like frozen pre-cut veggies, and LLMs to help me work through recipes and oops I did X instead of Y how do I fix or address that or does it matter? Some of this is AI and some of this is just nice growth in food resources coming together at the same time.)

I am also using AI to create art. Except, put down your torches and pitchforks, please. Not in the way that statement might bring to mind (the way that people have knee-jerk reactions to). I am painting with acrylics on canvas, by hand. But, I use AI quite a bit in the process. I usually find a painting that I like and ask AI to help me adapt a reference painting that is in my skill level (e.g. beginner or advanced beginner), so I have something to work from. Or, I’ll give it a photo from a hike that we did and ask for it to turn it into a reference painting for me. Then, I paint. I take pictures of my canvas as I go and sometimes upload them and ask it for feedback on a specific area or overall, or ask it to give me suggestions on what order to create my painting in. That’s actually how I started, because as someone with no exposure to painting before, I had no idea where to start! I would ask it to give me a step by step high level approach to the painting, and would follow that. (Obvious things like working back to front, starting with the sky etc, in a landscape are not obvious when you are a 100% beginner). Eventually I learned additional ways to get help with my painting. I sometimes ask it to give me advice on fixing a color, or how to mix a color, given the paint options that I have. Other times I’ll ignore the original reference and just tell it to focus on my work in progress picture and give me high level feedback. It’s been incredibly helpful to me and has facilitated me painting >30 paintings in the last year or so since I started painting. It’s really useful for customizing feedback to where I am, skill-wise. I also ask it sometimes to assess my portfolio and give suggestions for what skills to work on or document what progress I’ve made on existing skills I’ve been working on. That’s also helpful for seeing where you do have actual skills already (or strong skills compared to others), so you don’t just feel overwhelmed and inept all of the time. You can drill in or scale out, feedback wise, and you can ask it over and over again if you are stuck on painting something and can’t get it quite right. It doesn’t get bored, it doesn’t (openly) judge you – unless you ask it to, and it’s nice to be able to have something to give feedback and input on the specific things I want input for. It’s better than the next available human, because I don’t have a next available human who is interested in my hobby project. (Scott, who occasionally gives surprisingly prescient feedback especially about color values in a way that my eyes/brain don’t intuit the same way, isn’t always available or interested to the degree that I am interested in this – so LLMs often win for iterative feedback in the absence of an available, interested human).

How I use AI in September 2026 to do more types of things than ever before. A blog post by Dana M. Lewis on DIYPS.orgBoth cooking and art are examples of things where AI facilitates me to do a lot more than ever before. It’s less surprising that AI has helped me build all kinds of software projects from apps to drug titration tools and simulators. But being able to use AI to interact more with the real world is a use case that I don’t know that I would have predicted, and it’s exciting to see how else it might enable me in the future. There’s probably half a dozen other physical world interactions it’s facilitating for me, too, that I don’t think of as obvious examples. This includes things like giving me feedback on my strength training work to tweaking and improving my spreadsheet where I log and automate filling in my strength training workouts into my other all-activity tracking spreadsheet…it’s a nice mix of physical world and software interactions that all are making my life easier and better than before.

If you have cool, surprising uses of AI helping you in the physical world, I’d love to hear about them! Send me (or comment) your ideas and experiences.

Meta-thoughts about feedback loops in the era of AI

by Dana Lewis

One of the critiques I hear from others about AI is lack of credit for work. I have a different motivation for a lot of my work which is maximum impact and scale, such that for a lot of things it’s less important to designate “credit” as it is for the work to scale and reach as many people as possible. Thus, it’s a benefit for my work (whether it’s my book – which is open source, or OpenAPS or any other projects) to be training material for LLMs. And it’s a benefit for it to be accessible and referenced and used, because it will reach more people. 

I was thinking the other day about some of the smaller scale projects that are most meaningful to me. Particularly, things like writing the post after I broke my ankle in 2019, which for years picked up steam and reached other people (via search) who also couldn’t find other resources about returning to weight-bearing after an ankle fracture. I’ve written a variety of other posts knowing they would not be high-volume traffic and would instead sit ‘on page 17 of Google’, or the like. Interestingly, with AI, I think some of that content has the ability to be surfaced up more easily into AI – which is awesome! Usually it links to posts so theoretically someone could click the link and read it in its full form, rather than the paraphrase or summarized content in the AI they are using.

I don’t mind the summary or paraphrasing in their AI, actually. Again, my goal is to get information to people who need it – so this AI use case does serve my goal.

However, one thing I realized the other day is that this breaks a feedback loop. I maybe get 1 comment on a piece of content (e.g. a blog post) for every ~500 reads of it (order of magnitude estimate). That feedback via the comment is a huge signal and part of my process, because it indicates there is value in continuing to write and provide this niche content (whether it’s about broken ankles or book recommendations or anything else). But, that ratio is changing – meaning I get fewer comments over time – as people ingest the content elsewhere, not realizing that there is a place to give feedback or acknowledge that it was useful. That then influences me, because it makes me question in some cases whether there is still value in creating (writing, etc) things. 

Meta thoughts about feedback loops in the AI era. A blog post by Dana M. Lewis on DIYPS.orgI’m not sure what to do about it. However, it has given me a renewed motivation to start to give other people feedback and acknowledgement when I see something of theirs that is useful. I’m trying to build this muscle and habit to go look for a way to give someone feedback. (PS – if you now or in the future read any of my posts and find it useful – I would benefit from you leaving a comment or shooting me a private note or emailing me to let me know!)

Protocol Pilot – an app to track medications

by Dana Lewis

I originally built Protocol Pilot for myself, to keep track of a complicated medication trial that involved slowly ramping up my dosing and keeping track of side effects & my resulting experience (hopefully, improvements). It was really helpful, and as I used it I added in a bunch of other features to make it easier to track things. I also realized that sometimes I wouldn’t want to have a day that followed the next level of the plan, or was in a different order, etc, so I ended up making the app more flexible to allow me to log what I did however I did it.

I mentioned it to a friend who was curious if it would work for her purpose (trying a taper – the opposite of what I was doing – of a different medication), so I added a few more features and put it into TestFlight. (If you want to try it, you can! Send me an email to Dana+ProtocolPilot@OpenAPS.org or message me wherever you usually talk to me, and I can send you the link to download it.)

I’ve been surprised at how useful Protocol Pilot has been, for several different use cases I’ve had over the years, and maybe some of these match your experiences, too:

  • Have you ever taken an over-the-counter medication and then, a few hours later, tried to remember exactly when you took it?This may be because you have a new injury and you’re trying to stay on top of pain meds or because sometimes you take things regularly (like during your menstrual cycle) but the rest of the time you don’t have to remember to take it at all, because you don’t need it.
  • You need to start a new medication that is known for gastrointestinal (or other) side effects and want to slowly increase your dosing over time to minimize risk of side effects.
  • You are ready to stop taking a medication but you can’t just suddenly stop it, you need to slowly decrease the dose over time.

With Protocol Pilot, you can do all of those things. You can add a medication like Tylenol (acetominophen) or Ibuprofen and log when you take it, when you need it. This way, you can keep track of when you took your last dose. You can also set up a plan, whether that’s increasing or decreasing a med, to help you track as you progress through those different dose changes.

You can log side effects or notes related to your dosing, or none at all. You can also schedule reminders to check in and log how you’re doing overall, however you want to rate that.

Protocol Pilot: tracking medications (sometimes), a blog by Dana M. Lewis on DIYPS.orgThis is all useful because “how has it been going?” and “is this working?” can be hard to answer against everything else you’ve got going on in life, if you don’t have a convenient way that works for you to track it. Protocol Pilot has become the way that works for me – and a few others – to track things. If you want to try it yourself, email me (Dana+ProtocolPilot@OpenAPS.org) and I can give you the (iOS) TestFlight link to try it. And if you try it – I’d love feedback and also feature requests for what would make this helpful for YOUR use case, if it doesn’t hit the spot in the way it’s currently designed.

A portfolio dashboard to be able to see everything I’m building, maintaining, and growing

by Dana Lewis

That time I built a project… to keep track of my projects.

Two weeks ago, I was thinking through my current work and how I keep tabs on everything. Some of my project work is grant-funded, some of it is personal projects, some of it is early research that later transitions over to grant-funded and/or collaborative projects, etc. This means some projects have set up pathways for reporting and tracking progress, but others don’t. I was noodling through what I might do to better track all the things I was working on, because of the lack of visibility for a lot of the maintenance and tiny but meaningful tweaks across a lot of projects. Actually, I didn’t know exactly how many projects I’m working on and juggling.

Coincidentally, I saw a re-post of an interview Simon Willison did about technical blogging from earlier this year, which was interesting in of itself. He also linked to a blog about a talk he gave about ‘coping strategies for the serial project hoarder’. I figured there would be some useful ideas in there given I, too, have a serial project habit (although not at the same scale Simon does).

That triggered my brain to realize that because I don’t have all of my projects on Github (nor do they need to be), I can’t use the same style of issue-driven development. Some of my projects are code and do have git repositories, some do not. I have a mix of manually tracked projects and git repo-tracked projects locally. What if I could combine those, though? I’d like to be able to keep track of all types of projects, ranging from public-facing iOS apps like PERT Pilot (plus the Android version of PERT Pilot) and Carb Pilot; n=1 iOS apps like my macro-while-running/hiking or medicine trial/protocol app I recently built; non-iOS apps and public-facing tools like an infusion dosing simulator or BookPilot; plus various funded and un-funded research projects that don’t have a concrete finished artifact. Some are done, some are being maintained, some I’m working on, some I lost interest in but might come back to.

What might work for managing all of this? And what would give me visibility into the history and progress of some of those git-tracked repos where I have a lot of progress but it’s not visible anywhere, leaving me feeling like I’m not doing ‘enough’ even when I am doing quite a bit of work (it’s just spread out across a lot of projects)?

How meta to realize I needed to build a thing to track the things I’m building (and have built already).

A portfolio dashboard

I ended up building a simple dashboard to track all of my projects; automatically update tied to git repos across my computer; and also paired with manual to-do and task tracking.

The really nice thing is that it now automatically logs completed tasks based on local git commits, which is something I didn’t know I needed. It ties in with the manual to-do list. It also helps me see alllllll the projects I have active and their various states plus a high level ‘last touched’. If I’m bored, I can glance and see which project might deserve some attention and what state I left it in last, without having to go and open it and spelunk to figure out what state I left it in. When I’m busy and overwhelmed, I can see which projects based on those statuses are fresh enough to be left alone and deal with more urgent tasks (like specific project deadlines or things that are broken).

Plus I built in rollups so I can see what I’ve done today, this week, this month. I also added a mini dashboard section so I can see weekly, monthly, quarterly, or this year what I’ve worked on and what projects have gotten my attention the most at various times, as well as show me velocity over weeks. I don’t want to beat myself up on days when I’m not feeling well and am not productive, but I also want to motivate myself to do a little more when I am capable of doing more. This serves both purposes with flexibility, giving me grace – not every day is going to be the most productive ever, and some weeks because of other projects or life priorities I will do less. At the same time, I can see when I start trending toward less productivity and if there’s not a valid reason (e.g. competing life priorities like several weeks this summer where we prioritized time off with both sides of our families), that helps motivate me to correct the trend.

The combination of being able to log to-do’s and see that they are done, plus the automated history, means I can track things that aren’t digital projects (like paintings, which is a hobby project I have spent substantial time on in the last year) as well as digital projects.

And, I didn’t want to spin up complicated architecture, so this dashboard is simple. Local json data that is served up in an HTML page. I keep a server running on my laptop to show the page as a tab on my computer. I can also see a snapshot of this HTML on my phone (although I can’t edit data because I don’t want to spin up more complicated architecture, I keep it as view only for now). I can then while on the go review things and think about what I might want to work on next, etc.

Already, this has been super helpful the first week that I’ve been using it. It has motivated me to chip away at a big, hairy, and scary project that landed on my lap last week (and succeeded with, despite my fear that I would fail at this hard, technical project). I’ve been able to see that actually, even when I’m feeling non-productive, because of the huge portfolio (dozens) of projects of various sizes, even light maintenance work builds up to be large productive volumes of work. It also helps me realize when apps or projects are getting a little outdated and while not urgent, deserve to be refreshed. And I can already tell it’s reducing the context-switching “tax” of how hard it is to dig back into a project where it’s been a while since I touched it. (A good example of this is a bug report someone sent me for the Android version of PERT Pilot, which I had not touched in months. The bug was something not impacting all users and was intermittent, so I had to figure out how to debug it and reproduce the intermittent issue, then fix it, then re-figure out how to push the fixed version to the Play Store, which felt hard because I hadn’t done so in months. It was a little bit easier to do the ‘figure out what I did last’ part, which made the rest of the harder parts a little easier, too, and I didn’t procrastinate it quite as long as I might have otherwise.)

A project...to track projects? A blog by Dana M. Lewis on DIYPS.orgI thought about whether this amount of effort to build a custom setup was worth it, in contrast with trying to shoehorn my processes and tracking into other existing project-tracking systems. But, the amount of effort to shoe-horn my stuff into an existing system or to get everything into Github would take just as much time and not have all of the features I want. (I’ve tried various versions of that in the past, and always dropped them). This project may not last forever and I may eventually drop it, too, but for now it seems to work and like many things these days, the little bit of investment in setting myself up now to track and automate tracking of what can be automated goes a long way.

Custom app for tracking medication trials and complicated medicine protocols (Protocol Pilot)

by Dana Lewis

It’s awesome and terrible at the same time in this day and age to be handed a medicine with a complicated protocol and given no ideal way to visualize or track the schedule and the feedback from the medicine. This can lead to “non-adherence” and “non-compliance” in clinical-speak, or “I don’t know if it’s working” or “I don’t know how and when I should take this” in patient-speak. For medication trials, it risks missing the signal of determining if it works. The awesome part is that now with AI, it’s easier than ever for someone to use AI to help them to think through what they need to track and customize it to fit their lifestyle and preferences to actually track what they are doing. But it’s terrible that we have to do that, and I recognize that not everyone will think to do this; have the time to do this, etc. 

When one of my doctors and I decided to try a new medication for me, the conversation circled around the fact that there is no data for my use case. (A recurring theme). So, using their best knowledge of basic protocols for this medication used for other conditions, they outlined the approach they recommended. Take dose A. Then take dose A twice a day. Then take dose A once and dose B the same day, separated by at least X hours. Then take dose B twice a day. You can take dose B eventually up to Z times per day. Let me know if it works for you. 

(Determining if it works for me is up to me, because it’s subjective.)

Even with all of my experience and resources, I came home feeling overwhelmed about implementing this medicine trial. Not in the least because the effect of this medicine is very time-specific, and also because the effect isn’t “do I feel better” but do certain things get better when using the medicine. I need to track those specific things. There’s the variable of taking a small dose to acclimate and minimize side effects and then ramping up; then there’s the assessing efficacy; then there’s trying to determine (if it’s working at all) what’s optimal.

It’s overwhelming to think about working my way through that. I need a spreadsheet for tracking, of course, except it’s complex enough that I wasn’t sure how to set up a spreadsheet for tracking this. There are too many variables to track (including non-med ones that influence the interpretation of efficacy)! Plus I need reminders to track outcomes at very specific time intervals. Like I built Scheduler Pilot when I needed the ability to move and reschedule treatments based on the relationship to each other, I need something more flexible than a spreadsheet, or manual phone alarms I have to set, that can track according to the relation to the time of the medication and tests I am doing.

An app it is, then. Luckily, I have enough experience recently with app development and various technical projects and applying this to niche, boutique or “custom” and “n=1” software that I know I can do this. I spent an hour with an LLM discussing all of the app specifications and building out a spec, then set the AI agent to work implementing this in an app and pushing a build to my phone for me to test and validate. It worked. I maybe spent an hour discussing; it spent 20 minutes building the first MVP; I spent 5 minutes using it and creating a list of feedback and tweaks; and the cycle continued. I now have a fully functioning medication trial app for complex medication protocols called Protocol Pilot, built and on my phone ready to go before the medication arrived at my door step. I can now track where I am progressing through the testing protocol, record time-specific feedback around different aspects of my unique scenario as well as medication side effects, and have all this data stored and reviewable not only in the app but the ability to export as json or a spreadsheet for review and analysis elsewhere. And, most importantly, there is a robust notification setup in this app. When I say I’m starting a task to test, it prompts me at the right intervals (there are multiple) to reflect and record any side effects of the med and whether the task is different with the medication peaking, plus any other time-anchored review periods I want to do. Instead of having to remember to track, it tracks for me when it’s time to track something. This is a huge benefit of doing this as an app rather than a spreadsheet.

Why “all this work” for a tracking app? Part of the reason it’s worth the “effort” is because I have learned that putting in the work up front to have a good system that works for me to track means I’ll actually track and record what matters. I dislike looking back and not being able to use tracked data because I didn’t track the right things, because it was burdensome. The whole point is building it into UIs that work for me.

The nice thing is, I don’t necessarily have to design it in a way that works for everyone else or indeed for anyone else. It lowers the threshold of action to build what works for me. Later, if someone else wants it, I can certainly share it and scale it and modify interfaces to support other use cases – but I don’t have to slow down a working prototype that works for me, when I need it, because I need to design for everyone else at the same time. And part of this is because the barriers to starting are so low, thanks to AI. (Another plug for you to try again with AI if you haven’t tried in a while, especially using agents and sub-agents where you can set goals and let it keep running and testing it’s work before it bugs you to review, so you can provide higher level direction without having to review line by line of code or fix trivial bugs.) 

Protocol-Pilot-tacking-medication-trials-and-complex-protocols-DanaMLewisThat being said, I do have in mind the ability for this app to scale and be used by other people, for example actually baking in an LLM to take an unstructured protocol (e.g. “take dose A then B then a plus b at XYZ time frames over ABC days, record JKL symptoms and watch for MNO side effects and record if those happen, when) and transform the app interface to reflect that protocol and when to record things. If this type of app sounds like something you would use for your use case – let me know and I’d love to collaborate with you to build in that feature and have you test it for your use case (whatever it is, you don’t have to tell me specifics). Email me any time – Dana+ProtocolPilot@OpenAPS.org – and let me know.

—
Update: a week later and I have already built in the ability to support different users (aka, anyone can use this: it’s not hardcoded to my use case) and a friend has already asked to try it. Plus, it supports multiple protocols/tests at once! Reach out if it’s something you’d like to see if it works for your use case.

Why I built another titration tool, this time for titrating subcutaneous IG (ScIG Pilot)

by Dana Lewis

Why is everything a titration problem; why does the traditional healthcare / research system not solve titration for so many therapies; and why do I always find myself having to do it?

Good questions that I won’t answer (because I can’t), but it is a good introduction into what I have been working on. I built a titration simulation tool for clinicians and people who want to explore different adjustments to dosing immunoglobulin, especially through subcutaneous methods. AKA, subcutaneous IG, AKA ScIG (rhymes with “rig”).

What exactly is the problem?

ScIG is studied in very few conditions – the approved FDA indications are generally for PI (primary immunodeficiency) and CIDP (chronic inflammatory demyelinating polyneuropathy). But, ScIG (and IVIG) are used to treat a lot of autoimmune conditions (often used off-label, because the few other conditions studied have a much smaller population studied and lower-quality evidence bases). Thus, there are no approval studies nor good data for these other condition use cases. Instead, there is a smattering of literature showing that occasionally it works here and there. The dosing regimens are based on the PI/CIDP studies and there are very few of those studies beyond product-specific studies. ScIG has mostly been tested at a 7 day interval (meaning once per week) and there’s one study checking to see if you do it every 14 days (every two weeks) but double the dose to maintain the total dose per month, you get about the same outcome…that is, in people with PI, which means they were measuring infections. That doesn’t answer the “does it work” question for most other use cases for ScIG, which is a variety of other autoimmune diseases. It also doesn’t appear that they tested lower doses (?), so we can’t tell if the same dose over 14 days was effective because they doubled the weekly dose for the 14 day regimen they studied.

But does it work well if you do the same dose every 9-10 days? No one can tell you that, it hasn’t been studied, so if a patient is struggling with every 7 days scheduling, they may benefit from a schedule change, but there’s no data or guidance from research for the clinician to help estimate whether that may be ok, or a problem. Or, what happens if you slightly drop your dose by 10mL? Or what happens if you split your dose across 14 days? What happens if your dosing is evenly split over time or very differently split based on the pre-filled cartridges you have available?

Clinicians don’t often seem to prescribe ScIG to very many patients, so they also don’t have a strong feedback loop of data on alternative dosing regimens. And most indications are off-label and diverse, so that further dilutes that data stream: what works for someone could be unique to their disease rather than useful to extrapolate to everyone on ScIG.

Why does this matter?

ScIG is dosed based on body size. If you have a larger body, you have a larger dose. You are limited to how much you can infuse per site which influences how many needles you use. Needles have a risk of causing bruises and lumps (resolving to small clots of tissue that can sometimes take weeks to months to go away), so in some cases fewer needles is ideal, but this then drives up the size of the fluid depot which can be painful due to skin stretching and the amount of fluid that has to be cleared. It is a tradeoff between all of these factors. And, over time, people may have less skin real estate for various reasons. But, there’s no tool that helps patients make this kind of tradeoff. Current dosing calculators can help with dose and volume and figuring out infusion times, and tracking apps can record what happened, but they usually do not show what different schedules might mean over time. For example, whether doing smaller amounts less often changes the overall amount dosed, whether this creates bigger highs and lows in IgG levels, or simply reduces the number of sites. There’s also the question of how much time it takes for these varied dose schedules given the tradeoffs where fewer needles means larger depots and slower infusion times.

I wanted a tool to make those trade-offs easier to see. I wanted to be able to compare possible schedules side by side, showing the expected pattern over time, the amount per week, the number of infusion sites, the volume per site, and also model what the impacts are for the IG itself.

ScIG Pilot

ScIG dosing decisions often involve trade-offs among dose interval, total dose intensity, infusion volume, number of sites, per-site volume, and patient tolerability, but existing product calculators don’t cover this. They generally are designed for clinicians or pharmacists and focus on narrow tasks such as converting weight-based dose to volume, estimating sites or infusion time, or supporting labeled weekly/biweekly schedules. Published PK models describe ScIG absorption, peak/trough behavior, and exposure under standard regimens, but these are not available as a practical tool for comparing based on real-world constraints.

ScIG Pilot is a literature-informed simulator that compares candidate ScIG schedules against a reference regimen (aka a standard 7-day dosing schedule), showing normalized exposure curves, predicted peak/trough patterns, weekly-equivalent dose intensity, site burden, dose volume per site, and infusion frequency. (It is also open source.)

It can help someone think through a scenario and ask these types of questions of a proposed change:

  1. If I change the schedule, how does this change the volume I am getting over time? How does this compare to the existing schedule?
  2. When might this dose change impact the body?
  3. What is the burden trade-off: sites, dose per site, infusion days, time per infusion, longest gap? How does this stack up over a month or a year?

Then, based on outcomes (does it work) and various preferences (less time; fewer sites; smaller fluid depots; etc), someone can decide what works for their scenario, or at least have visuals to help them think through a decision and understand various tradeoffs. This is also helpful when trying to estimate out to longer time frames, for example: not just what this regimen does in a month in terms of number of sites, number of infusion days etc; but also how does this translate to a year of this therapy in terms of number of infusions and sites, etc?

–

ScIG Pilot: a subcutaneous immunoglobulin titration tool, a blog post by Dana M. Lewis on DIYPS.orgIt can be really hard to titrate these types of therapies and infusions, because there is no good data. I’d love to see studies in future collect this type of data or have repositories (like I described here) that are not limited to one disease indication, so a patient or clinician could go and research across populations to better get some data to triangulate from to better inform a subsequent decision to go on a therapy or change a treatment regimen. Right now, there are no visuals, no data, and no tools. But now, for this, there is at least ScIG Pilot to help visualize and triangulate across what little data we do have.

(And it’s open source so if you have ideas to improve ScIG Pilot or suggestions to make the models more useful, please feel free to open issues or PRs or email me to discuss – Dana+ScIGPilot@OpenAPS.org)

An update of what I think about my exoskeleton after ~1000 miles of wear and tear

by Dana Lewis

Reader, I bought longer shorts.

Chafing was my main complaint about wearing an exoskeleton. (Here’s what I got, why I got one, which one I chose, etc. You can also read a second post I wrote about a second version I tried, as well). I got my exoskeleton in July 2025, and now in August 2026, I’ve put close to 1,000 miles (!) on my legs with an exoskeleton, everything from walking, hiking, cross country skiing (classic), and a tiny bit of running.

Because of my situation (which is not something that translates to anyone else), I wear my exoskeleton a lot. I’m also an active person, although my activity level is down from before, it has gone up again (not all the way to my old baseline, but well out of the lowest levels I was before getting an exoskeleton). Thus, a thousand or so miles using the exoskeleton in 11-12 months of using it. Because I wear it for just about everything, and for long hours at a time, and started using it last summer, I would get chafing from the ongoing use and heat and humidity of many-hour sessions with it. (For example, 5-6 hour hikes). I wrote about it in my first post, but I had tried everything from wrapping the straps on the exoskeleton to pre-taping my legs so the straps and pads didn’t rub. I eventually got all the chafing to heal and rigorous pre-taping at the first signs of rub helped minimize it, so that’s what I planned to do this year as well once it got warm enough to switch from walking and hiking in pants to shorts. But…this year, I had other reasons to change things (including the fact that I now do subcutaneous infusions of large volumes in my legs which coincidentally makes my legs very sensitive to additional tape), and the pre-tape method was no longer a viable option. Instead, I sucked it up and bought a few pairs of longer shorts that came down to my knees. I had avoided doing this because I thought it would be hot, but I picked a few to try including some that are looser around my upper leg (more like jogger pants through the thigh and hip) and they are not too much warmer compared to bike-style shorts with a 7” hem that I used to prefer for walking, running, and hiking. And…they work. The pad and straps rub mostly against the shorts and don’t cause the cuts and abrasions from chafing on my legs. Woohoo. So, TLDR I bought long shorts (11” works for my leg length), and that’s what I’ll be using during shorts season moving forward. (During winter, there was no problem over pants for walking and classic cross country skiing.)

Activity changes with an exoskeleton

I originally thought I would only wear my exoskeleton for hiking (to help me with stepping up, rocky terrain, etc.). In my unique situation, though, my muscles overcompensate for muscles that are not working (firing/getting the signal/reasons unknown) and cause a really extreme sensitizing feeling that can last for days and make it really uncomfortable to walk. I did enough A/B testing to figure out it was better to do all of my long (more than a mile) activities using the exoskeleton. I was able to bring my activity levels back up, not quite to baseline (I was an ultramarathon runner before), but to a reasonable level. This meant more miles of walking on my beloved paved trail most days of the week; more hiking (with more back to back days) with longer trail choices and also no longer being intimidated by the elevation gain profile any more.

A big change for hiking was not just more miles, more elevation gain, and more days of hiking…but also how I’m hiking. I had started to use poles to compensate both for lung/chest muscle weakness and balance/stability on rocky terrain (which I didn’t know at the time was likely from that missing muscle signaling/firing so I was compensating with poles and arms for lower body changes). But now, even coming downhill on rocky terrain or big rocky steps, I rarely find myself pulling out my collapsible poles to use. Rarely meaning I think I’ve done so three times so far on hikes out of ~30 or so this year.

I have more confidence stepping, whether that’s stepping across rocky areas; crossing a rocky stream; stepping down a step; or taking big single leg step-ups (like stepping up onto a boulder). I even think, as a result, that I’ve managed to build some strength back up in both legs to facilitate stronger single leg step-ups. This is likely because the exoskeleton helps me swing my leg up farther than I could swing it and ‘reach’ for a step, but now I have more power to finish the step, driving the force through my leg to come up and through the step. (The exoskeleton does a tiny bit of force generation down, but I still do 90+% of that effort). That’s awesome both because it’s cool to see myself gaining strength as a general concept, but it also shows me that not all my muscles are impacted by whatever I’m dealing with, and it is still possible to gain strength. (Given my thing is technically undiagnosed, that’s possibly a very powerful signal we can use to test future hypotheses with.)

It’s also nice to not have to use poles because then I don’t use up my hand muscle power gripping my poles (since my hands are also impacted somewhat), so while the exoskeleton only works on my legs, it results in decreasing the physical impact to other areas of my body as well. Again, a lot of this is unique to my likely rare situation, but I have been pleasantly surprised to think about how improving the ability to stride or step up overall helps with stability on even surfaces, and that’s something that other people likely would experience too even in different body situations.

I also tried and loved using it for classic cross country skiing over the winter. “Classic” cross country skiing refers to where there are two tracks in the snow and your skis glide along in these tracks (or path). I wasn’t sure if the exoskeleton would work for the motion, but because you lift your leg up and forward like a step, the exoskeleton does help with that leg lift motion from the hip, just like walking. And like walking and hiking, I was able to do more miles of cross country skiing than I was last season while causing fewer issues from my muscles not firing correctly. I still was limited in some cases by my lung/chest muscles, but my legs were usually not the limitation in the muscle-disease-sense. I didn’t quite get back up to my mega ultra mileage season from 2 years ago (where I was doing regularly 15-16 miles with one big 28 mile day); but I also wasn’t at my low from last year without the exoskeleton where my ski days were more like 6-8 miles on the best days but also had several 3-4 mile days with disproportionately more effort. Instead, I was able to clock several 10-12 mile days as my “low” end mileage and had many 13-15 mile days throughout the season. Woohoo!

My specific experiences with Hypershell as a brand

As I wrote before, I bought my exoskeleton for $1,199. I try to make that very clear because I was then chosen to try the Ultra version for free (which I wrote about here). They didn’t try to influence what I said about it, and you’ll see from my writeup that I actually think the Pro model I originally got was the best value, and still prefer that over the Ultra. The Ultra is supposed to be reengineered such that it is more efficient, but from my experience, it feels like it has just minimized power output in order to get a longer ‘range’ from the same amount of battery. There’s a chance I haven’t put it through its paces enough, and this may be unique to my muscle situation (I tend to run mine on higher settings than most people?), so of course this is based only on my experience, but between a $1200 price point and a $2000 price point, I doubt most people would get $800 benefit more from the Ultra over a Pro (ditto the Carbon, as I wrote about originally). That being said, in May 2026 they came out with several new versions, confusingly overlapping in names – there’s another Pro (now called the Pro S?) and other models with S in the name. I haven’t done a lot of reading into them and comparing specs, especially the new ones against the old models, but I wouldn’t be surprised if there are similar tradeoffs of power output versus range across different models again.

Anyway, the point of this section is to talk about my lived experience with it. Mine broke in November 2025 (I bought it in July 2025). Ack. The back bar – which you adjust the ‘hip width’ setting with – broke suddenly and unexpectedly while I was out on my (thankfully near to home) flat paved walking trail. I was trying a run interval, which I had done before, and I was hearing an odd noise. I stopped after 15 seconds and turned it off and back on, inspected the motors on both sides, but didn’t see anything wrong with it. When I started my next run interval a minute or two later, the bar broke in the back and it was clear the right side leg wasn’t working right. But because I had it on hyper (red) mode at high power, and it didn’t detect that it was broken since it was still strapped to my leg, the motor kept moving and trying to drive the leg. It was essentially loose and flailing around on my right side so it was hard to grab and press the button to power it down. I had to unstrap it from my leg and avoid getting whacked with it flailing around until after a second or two, the “it’s not strapped” auto-stop detection worked, and it stopped. Phew.

Hypershell as a company was great about it, though. I emailed the customer service account and included pictures and they said “yup, it’s broken, and we’ll replace it”. (The warranty is for a year, and this was ~4 months into my using it, probably around 250 miles of usage at that point). They asked a couple of questions about the box I would use to ship it back, the weight, etc, then sent me a free shipping label. We packed it up, dropped it off, and sent it back. Once they received it and verified it was broken like they said, they sent me a replacement. This did take about 3 weeks in total, in part because I was traveling for a week by the time they got me the label and I had to wait to ship it, plus it was over a holiday. Luckily, I still had the Ultra model and was able to use that in the meantime, but if I hadn’t had another one, I would have been without an exoskeleton for ~3 weeks. (For me, that’s meaningful time without, given how I use it to facilitate my preferred activity levels). The new one arrived and I started using it again, and it worked without issue from there.

I have seen some complaints online about slow customer service responses, and for other communication processes have also seen intermittent speed, but for straightforward “hey this is broken, how do I get a replacement?” it was on par with decent brand customer service and I have no complaints in that regard. If you were on the fence with that as a reason not to consider getting one – I wouldn’t let that stop you from deciding to get one. They do hold to their warranty.

(I will say, it did make me a bit nervous that mine broke (and in such a weird way at that back bar point) so ‘soon’ in – after only 250 miles – because they had somewhere online an estimate of standard lifetime being well over 5x that point. I had asked them if this was a common failure point or if I was doing anything wrong that I should do differently with my next one, but I didn’t get any good feedback on that. I do wonder if, because I use a wider hip setting, it may generate more force and torque on the bar, plus my higher mileage and higher power settings, than their testers have used in the past. But I don’t know if it is that or just a fluke.)

I will clarify that I’ve put over 100 miles on the Ultra, and it didn’t break. I passed the same mileage point with my replacement unit, and it also hasn’t broken. So I don’t think these units just break around 250 miles for no reason; that contextual data would indicate my broken one was likely a fluke.

Range anxiety no more (return of the mileage, thanks to additional batteries)

My original Hypershell x Pro model that I bought came with 2 batteries. I started using both batteries for hiking last year and managed to run through both batteries on a hike or two. It did cause some ‘range anxiety’ and I would limit how much power I ran it on so that I wouldn’t run out before the end of the hike. You use less power going downhill, but because I prefer mine on ‘walking’ mode for downhills so I still get assistance lifting my leg up as I step, I wanted enough power to get me all the way back to the car. Once or twice on bigger hikes, I managed to run out of battery over half a mile from the car, so I began to pay much closer attention to how much power I used on the outbound and uphill part of the hike to try to better manage having enough battery to finish the hike with at least some low assist.

When I was given access to the Ultra, it came with 2 batteries also, and the batteries are also interchangeable. This meant that I had 4 batteries I could use if I wanted to carry the 3 spares. I did immediately start carrying 2 spares on every hike or ski excursion because I had previously fully used 2 several times, so it was nice to have 3 total (one on device plus two spares), and I regularly would switch and be using the third one as we completed the activity. I had actually been on the fence about buying more batteries anyway, to give me more range (and more miles), when I ended up with more as a result of the Ultra. Also, when mine broke and I returned it, they didn’t have me send back the batteries (which are subject to air shipment restrictions), but the new one they shipped did come with 2 more batteries. So now I had…6. Wow. Even with my mileage tendencies, I have never managed an activity to use more than 4 batteries, even running the exoskeleton on higher power intensities. I have been on the 4th one for some 10-12 mile hikes that had 3-4,000 feet of elevation gain by the time we finished (because I also use it on a low ‘walk’ setting for downhill so I still get assist lifting my leg with every step), but haven’t even managed to fully go through all 4 and need a 5th.

The tradeoff is that you have to carry these batteries – I usually carry 2 spares in my pack and if I want a 4th, have Scott carry the third spare in his backpack. But given how the exoskeleton assists me (and prevents multi-day recovery issues as a result of using it with the power settings I need), it’s worth carrying that extra weight and rotating the batteries. When it dies or is about to run down, it’s very easy to stop, have Scott do the swap (battery is on the backside under that adjustable bar, so it’s faster for him to unclip a battery, grab and swap the other one from my pack, then I turn it back on while he zips up my pack), and carry on. It takes 30-60 seconds to do a full swap. Whereas when I do the swap, I have to stop, put the device on standby, fully take it off and set it on the ground, take my pack off and get the battery out, pick up the exoskeleton and swap the battery, put the battery back in my pack, the pack back on, and the exoskeleton back on. I’ve done it plenty of times – especially when I’m off cross country skiing by myself – but it’s a lot faster when hiking with someone to get them to do it. But for me and how I use the exoskeleton and why, it’s worth the extra weight of additional batteries. People who don’t use as high of power settings will not need as much battery, and this may be where the ‘efficiency’ of the Ultra and possibly the new models just released (the x S series) might benefit people who need lower power output.

What’s next? What are my future exoskeleton choices?

My warranty is now over, so I expect at some point in the next 6 months my device is likely to break (because of my high usage patterns) and because it’s out of warranty, I’ll need to decide what to get next. I am slightly disappointed that these don’t hold up to the full predicted lifetime mileage…but…a little less so now that I see how much value I get out of it and again, the fact that they do stick to their full year of warranty if it does break. A solid year of my high-mileage use under the warranty is really decent value for me. I also now have plenty of my data to understand the power settings vs battery vs time expectations and have a little better way to contextualize against the “max range” and “max battery time” metrics the company provides for how to translate that to my experience (with likely higher power settings).

I actually don’t know yet what I would buy next, though. If there were no other options, I would definitely get another Hypershell brand exoskeleton, because I am very happy with my Pro (and how they handle the warranty). I’m just not sure which one because I haven’t studied and compared the new models and the price points to the existing line. But…there are other brands coming to market. There is one coming via Kickstarter (to start) that has a hip + knee option that is interesting. I don’t know that I “need” a knee-based motor, but I also don’t know without trying it that it wouldn’t be beneficial. In fact, I have a weak hypothesis that it would help me overall get more mileage out of lower power settings, because it would better help compensate across all my muscle motions (not just the hip), so I maybe could go farther with less battery requirement. Maybe. I won’t know unless I get to try it, and I don’t know if I would commit the money to buying it at full price without knowing if it would be more beneficial for my situation. (It’s unclear after the Kickstarter when it would be available for general purchase). They also have the option to buy it without the knee motor, but again, I don’t know if it would work better than Hypershell’s version, in terms of the physical structure, battery life, power output, ability to get data out of the app to help guide my usage, etc. I’d be very interested in testing it (and A/B testing with and without the knee component!) if that was an option, but without that ability to test, I don’t have a good prediction right now for what I’ll choose for my next one.

A powered exoskeleton is an instrument of freedom: thoughts after 1,000 miles. A blog post by Dana M. Lewis on DIYPS.orgAnd…there are likely more coming. I’ve seen two or three others advertising online. It’s an exciting time for those of us who want or need leg-based or lower-body exoskeletons. If you haven’t read my other post, do read it – it seems like exoskeletons are benefitting a lot of people for a lot of reasons and I get a lot of feedback that it’s helpful to hear about my experiences so people can consider it for their own context (even though my situation is very unique).

A heuristic approach to deciding what is safe to eat with celiac disease

by Dana Lewis

When you’re diagnosed with celiac disease, it is nice in the sense that you have a known issue (consuming gluten causes an autoimmune reaction and damage to your body) and that the solution is known: eating 100% gluten free. The challenge, though, is that “being gluten free” is not easy. Every food, every drink, every morsel – and even some non-food items – can risk cross-contamination, even if something itself is gluten free. There are countless decisions a day about what is safe to eat. And what many people don’t realize (and may find confusing from the outside) is that this decision process can change, over time or even on a day to day or situational basis.

I had been thinking about writing a post about my approach to celiac and how I decide what is safe for me to eat (and why), and how that may differ from another person’s perspective. Even person to person with celiac, we may make different decisions! But more importantly, I myself may make different decisions on different days for the same food or situation. (Plus, people who are gluten free for non-celiac reasons may have their own, differing decision making process about gluten.) There was also recently a really interesting scientific study published that I think can help illustrate why decisions vary at different points in time for the same person, so I wanted to talk about the high level approach and then what this study shows us about why this type of approach is helpful for people with celiac. I have had celiac for 18+ years, so I have a lot of experience evaluating food options, food labels, and deciding what to do. I rarely get glutened – maybe once a year, if that – and this is even with eating out and getting takeout from a lot of restaurants. I am also very symptomatic (even tiny amounts of gluten are noticeable to me), thus I have some confidence in categorizing myself as someone who does not get glutened often.  Disclaimer is probably needed though: I’m not a medical professional and I’m not telling you what to do. Instead, I’m trying to show you how I think and evaluate things:

  1. Eating certified gluten-free is best, but not always possible

It’s helpful to realize that there are certifications for gluten free. In the US (where I live), there are a couple of different labels to understand the difference for. A food labeled “gluten free” means it has been tested to the FDA standards of <20ppm. However, that doesn’t mean it’s certified. There are several GF certifications beyond this standard, most of which are to the standard of <10ppm, and usually brands that bother to get certified also have good processes for vetting ingredients (inputs) as well as batches (outputs) of products. Different certifications have different requirements for these things so they slightly vary, but usually it means a higher standard – meaning a lower amount of gluten tolerance – than a standard word of “GF” on the label.

That being said, it’s a lot better than NO indication of gluten free on a label!

My rough hierarchy is:

  1. Certified gluten free (usually <10ppm)
  2. Gluten free (FDA standards, so <20ppm)
  3. No indication of ‘gluten free’, but no gluten-containing ingredients and no cross contamination warning such as “made on shared equipment”. This means it could be <20ppm or could be above, we don’t know.
  4. No indication of ‘gluten free’ and a yellow flag – some cross contamination risk such as “made on shared equipment”; or a red flag “may contain wheat”. Again, we don’t know if it’s <20ppm or above.
  5. Red flag, will not eat for any reason – contains wheat or a gluten-containing ingredient (such as malt or malt flavoring, rye, barley, etc.)

There are a lot of foods that I eat that fall into that third category though – say, picking up a banana from the produce section at the grocery store. It isn’t labeled as gluten free – is it safe to eat? I say yes and choose to eat it, even though it’s not labeled, because it doesn’t make sense to have a gluten free label on a banana when it’s a self-contained item and not processed with other foods that are likely to contain gluten.

Same for more processed items, such as other fruits and vegetables, or a container of pre-washed and pre-cut fruits or vegetables, or even a bag of frozen vegetables. As long as it doesn’t have a “may contain wheat” warning on it, I am more likely to be ok consuming it if it doesn’t have gluten-containing ingredients.

But there are other foods that ARE an issue in this category, which I have learned from experience of reading a lot of food labels. Trail mix, processed fruit (like yogurt-covered and other dried fruit products), etc because they are often mixed on shared production lines with trail mix items that do contain wheat. For these, I am more likely to be cautious and avoid items that use shared equipment because it feels like the risk of cross contamination across product lines or batches is higher. (It’s a bummer, because it rules out a lot of dried fruit, trail mixes and snack mixes, nuts, etc. that otherwise should be gluten free, but for me the uncertainty is too high). Oat related products, which I’ll talk about more below, also fall into this category.

Generally, though, some brands or sellers of products aren’t incentivized to test and label their products GF, even if they ARE gf (bananas and other single-item fruits and veggies are a great example). But, given they themselves have no gluten ingredients and are minimally processed, they’re usually a reasonable choice (e.g. a banana). GF testing costs a lot of money, which is why brands may not do it, and why we as people with celiac have to decide how to deal with the fact that there are a lot of foods in this category that may be safe to eat. Ruling them out because they don’t meet the costly most stringent standards would drastically limit our options and mess with quality of life. If you’re newly diagnosed and trying to recover from months or years of immune system damage, it may make sense to be more strict at first, and then expand back out (or not). But know that some of us with celiac ARE choosing to eat foods that are not definitely labeled gluten free (such as a banana), and that does not necessarily mean we are getting glutened.

2. The amount of food matters

Because I am someone who gets symptoms from tiny amounts of gluten and know how much gluten hurts me, I have a clear red line: I don’t eat gluten. I will not knowingly eat “may contain wheat” products: for me, it’s not worth it, even if the risk of a single bite is low.

One example is a few months ago, when I was online grocery shopping and found a new product, a chocolate-covered frozen strawberry that was new to me. It didn’t list gluten ingredients on the label, but they didn’t have any cross contamination warnings in the product listing, on the picture of the item online, nor anything on the manufacturer’s website. So I bought it. When it arrived, I went to try some, and glanced at the back of the bag and saw a ‘may contain wheat’ warning. ARGH. I was so mad: I had actually cross-checked the manufacturer’s website before I bought it, in case the listing was wrong! I was mad enough to email the brand and tell them they should be listing this on their product site, plus the grocery store listings (even though they don’t always have control over that). But they DO control their site and they should fix that. Super annoying. Anyway, the point is, I didn’t eat these and gave them to my husband, Scott, because they weren’t worth a possible glutening. I could’ve tried one just to taste it, but it wasn’t worth that possible risk. Sloppy labeling and communication about labeling like that makes me not trust a brand.

But, as above, there are times when there are products with NO label / cross contamination warning…and I do try it. However, I recognize the risk could be higher in this product versus the example above (chocolate covered strawberries that may contain wheat) – and there’s no way to tell. The difference, though, often also has to do with whether the risk is known/not and the quantity. If I decided to try the chocolate covered strawberries, I’d likely mentally categorize them as safe and something I would keep eating, so I would consume more quantity of them, driving up the risk. Whereas if I “noped” them and put them in Scott’s gluten section of the freezer, I wouldn’t be at risk of mindlessly eating them at a later time in larger quantities, increasing the risk. For foods that are not labeled and don’t have a cross contamination warning, I usually am cautious about the quantity, because as the quantity consumed increases, the risk of consuming gluten also goes up.

And this is where that new scientific study comes in. They did a gluten challenge test via pills with celiac volunteers in Australia (shout out to these brave folks who participated in the study), and recorded symptoms *and* did a test for a proxy of immune activity to test response to multiple levels of gluten, from <1mg to 2mg, 3mg, and much higher doses. (It was randomized, blinded, and they also used a placebo for control!) And what they found was that doses <=2mg didn’t trigger an immune response…but 3mg did (in a sizable minority of participants). (Not necessarily symptoms though, even in those with an immune response). Note a HUGE caveat of this study is that immune response does not mean proven damage, but it does give us a threshold to understand at what point the body might start reacting and how to think about cumulative exposure over the course of day…for which we previously didn’t have the data to have a mental framework.

Thinking about 3mg is super duper helpful because we can now use that in context of the thresholds we are most familiar with: <10ppm (certified GF usually) and <20ppm (FDA standard for GF). If you eat 30g (a small snack) of a food that is <10ppm (parts per million, so < 10/1,000,000th of it is gluten), you get a possible 0.3mg gluten dose – which you wouldn’t expect an immune reaction to. Same for doubling that, a larger snack – that would be 0.6mg. But if you 10x that (think meal-size amount), that puts you at 3mg. Whoa. That’s the level of an immune response (possibly), if gluten was present at that top end (10ppm) for that product. Now, does that mean symptoms and damage long-term? Maybe not – but this starts to show us how the amount of food matters IN ADDITION to the labeling.

The FDA standard, <20ppm, shows this even more. A 30g food amount at the top end of the 20ppm standard would be a 0.6mg dose; a larger 60g snack would be 1.2mg…whereas a 10x meal size amount (eg 300g) would be 6mg, which is definitely in the category where the immune system is more likely to respond (according to that study).

Now imagine a food in my third category, unknown ppm. That means it could be even below certified standard (eg <10) or FDA standard (<20ppm), but it could also be more – we just don’t know. But in any case, use the same logic: 30g of 40ppm (double the FDA standard, for easy math) would mean 1.2mg. Maybe fine. But as the amount of food increases, you’ll see that possible gluten amount increase: 60g of food at 40ppm would be 2.4mg, and well before you get to a meal size (300g), you would go well beyond the point of gluten exposure (>=3mg) where more people experienced an immune response and – as the amount got bigger – also increasingly saw symptoms.

This is how I think about the label/not uncertainty and the amount of uncertainty, now: in context of the quantity I might want to eat. I had previously intuited “this is more risky” when the label was uncertain and the food quantity was bigger, but this helps me put a numerical framework on the risk. 

Some loose categories to think about now with this framework:

  • For naturally GF / simple or few-ingredient foods, like fruit, vegetables, plain rice, potatoes, beans, plain dairy, or rice pasta, “no gluten ingredients” may be enough for many routine decisions, especially if the product is simple and not from a high-risk category and does not have any cross-contamination label warnings.
  • For complex processed foods, “no gluten ingredients” is weaker. A 30 g snack may still be a reasonable pragmatic choice for some people, but a 300 g meal is a different decision.
  • For oat-containing foods, “no gluten ingredients” without GF labeling or process information is a much weaker signal. Oats deserve their own caution category because of cross-contact risk in the supply chain.

3. Oats are evil* 

* Evil in the sense that they require their own mental category of gymnastics and special handling, above and beyond what we have discussed. Some people with celiac perceive that they are sensitive to oats themselves, but not everyone is – and I am not. I do eat oats, see below. But, oats are OFTEN cross-contaminated at harvesting, processing, etc because of the nature of being grown in fields near gluten-containing grains, and thus anything with oats is a hot mental mess of risk. (They’re not really evil, they just require a lot of thought, see below). 

Oats are themselves gluten-free, but they have higher cross-contact risk from growing, transport, processing, and manufacturing. Whole or rolled oats prominent in the ingredient list should bump risk assessment up; a controlled gluten-free oat process with certification, purity protocol, dedicated facility, and batch testing should bump risk assessment down.

But sadly, some of this is really hard to figure out. It used to be that there were “purity protocol” oats, where oats were grown in fields far away from gluten, harvested separately, processed separately, transported separately, etc. And with more testing on arrival, etc. But this was very expensive, and it’s rare now. And…that process was still not flawless; sometimes cross contamination of oats still happened. (The site “Gluten Free Watchdog”, for example, which does lab testing on consumer GF products used to recommend certain brands of purity protocol oats, and as of 2026, no longer does so.)

So what is a person with celiac supposed to do? Avoid all oats?

You could, but you don’t have to. You could avoid oat-related products for any of the following reasons:

  1. Oats bother you for other reasons
  2. You are too tired to go check the brand and their product testing protocols for a particular product
  3. There’s no indication of testing for GF or special handling consideration even though it uses oats

You could also decide to consume oat-containing products for any of the following reasons:

  1. The product is certified gluten free, which means there are requirements around testing the ingredients and the batched products, which reduces the risk
  2. The product is one where oat is used for oat flour (meaning anything is ground up and distributed across the flour and then across products, so the risk of one bar/cookie/etc itself is lower than a whole oat product), lowering the risk of a particular bite having >3g of gluten on its own
  3. The product is not certified gluten free but the brand has publicly shared their process for testing batches of products and preferably how they test the oats themselves before using them inside products.
  4. YOLO and you want to eat whatever the product is anyway, regardless of the above

What do I do? It depends.

Sometimes, depending on the situation, I fall into scenario A (avoid) based on reason number 2: I am too darn tired to go research the provenance and figure out if 3 applies. It’s safer to avoid as a rule, and eat something else (or not eat, and eat later something safe).

Other times I will do the research and if it falls into reasons 1-3 of scenario B, I will choose to consume it.

Here are some practical examples, which ended up glutening me, and what I changed as a result, and some maybe surprising categories of things that I *do* eat as a person with celiac:

A. Remember Nature Valley Oat & Honey bars? I looooved those as a kid. So when KIND came out with oat&honey bars that were marked gluten free, I was excited. I avoided them for a while, but I heard other people with celiac were eating them, and I decided to try them. I did ok, but I did eat one once (when I didn’t have a lot of food options while traveling), and BOOM I was glutened. It was definitively that which glutened me and it was brutal. Those went back on the NOPE list for me, along with all other KIND products containing oats. Yes, they are marked ‘gluten free’ which means that they batch tested at FDA standard (<20ppm), but cross contamination can still happen, and it was such an extreme level that I permanently decided this was not a risk worth taking for me. (The exception might be if I was stranded on a desert island and that was the only food item; but even when very very hungry on a travel day I will choose staying hungry over trying one of those again, or anything else that is a similar whole-oat type product rather than something where the oat has been processed and spread as a flour. The latter reduces the risk more than whole-oat products)

B. Speaking of Nature Valley brand, recently I noticed a product listing for new chewy protein bars that were supposedly gluten free. I assumed the listing was wrong, because in my brain ‘nature valley’ is strongly associated with gluten. But, I double checked, and it’s actually certified gluten free. More importantly, I looked at the ingredients – these are specifically the chewy nut protein bars, and there are no oats, they are nut (and soy protein) and chocolate based). Ok. So I can apply my rubric – not the oat one – and because they’re certified gluten free, try them. I did (and actually like them), so these have entered my ‘snack bar’ rotation as a reasonable option. I wouldn’t eat a lot of them in one day – because the risk grows with quantity, but because they are certified gluten free, that is good enough for me at single-serving quantity. And so far, I have not gotten glutened from these.

C. But what about other oat-containing products? Because of my past experiences, I am VERY skeptical. A cousin-in-law shared a new product she thought I might like that she had gotten for her toddler, because it was marked GF. It was an oat-based PBJ-stuffed oat bar. I scoffed and dismissed it, because in the past I had checked out this brand and they didn’t say anything about GF processes and testing, so I let my oat rubric (NOPE) override the choice. However, when I had more energy (and was bored), I went and looked them up online. Nowadays, this brand DOES list their processes very clearly. They say they are GFCO-certified gluten-free, its oats are certified gluten-free, its bakery is a gluten-free facility, and it does daily testing on every batch of oat products. Ok, now we are talking! (They also talk about purity protocol oats, which as I said, is a great detail but does not make it infallible). But because they used gf-tested oats before they even start cooking AND do daily testing of batches, that means there is a greater chance of catching issues. This is important because one bar is a 60g serving, which is double a serving of my previous example…but it’s the certified (<10ppm) standard, so that means 60g of snack, even if it was 10ppm, means a likely 0.6mg of gluten exposure risk. That’s still well below the ~3mg immune response threshold, so that makes me more comfortable with the risk of this product. Plus, like I said, the in/out testing. Would I eat 5 of these in a day (300g of it)? No, the risk definitely goes up with more. But, I not only eat these, I eat these on a periodic basis (a few times a week), because of the testing and certification. Even though they are oat-based.

(What is this product/brand? Bobo’s PB&J rounds. I haven’t tried many other Bobo’s products but the above detail is supposed to apply to their other products, too – check their website and on-product labeling for the latest details in case this has changed.)

D. Another example might be Honey Nut Cheerios or Lucky Charms, which are now all GF by default. But they’re made with oat flour. General Mills has pretty good testing processes for batches, but they also had some kerfluffles when these products were first marketed as GF with feedback from the community that it was causing issues. Now that I have the above scientific study to help frame QUANTITY with risk, it makes me wonder if it was the quantity that was causing the issues, because cereal is hard to do a single-serving of, and as I’ve shown in the data above, as you eat more of something the risk goes up, EVEN IF IT IS CERTIFIED GF. A large bowl (or two) of cereal would be more likely to get increasing exposure, even if the per-serving amount was under the FDA standard or certification threshold of acceptable ppm. I personally also have a hard time limiting my quantities of cereal, so based on this I don’t eat oat-based cereal like this…but recently I decided I could get single-serving cups of Lucky Charms occasionally, because it’s a fixed quantity (30g); labeled as GF; and it’s an oat flour product rather than whole-oats. Plus General Mills has made it clear they test after sifting their oats; after the oats become flour; ship oat flour separately; then they also do batch testing at the end. (Same for other General Mills GF oat flour products). Like Bobo’s, that’s as decent of a process that I think we can probably get. So, like Bobo’s, I am now willing to eat these products. But, again, for cereal, only as a single-snack-serving product, because of the risk quantity calculation and my own experience with resisting the temptation of a larger bowl of cereal.

E. I even eat oatmeal. Quaker Oats, like General Mills, has a process where they evaluate oats in a separate facility for gluten as well as evaluate their batches of finished products. Note that this does not mean the general Quaker Oats products: they only have a few GF products. One is a big can of quick oats (like you might use in recipes for say, oatmeal chocolate chip cookies), and another is prepared packets that you mix with water or milk to make a bowl of oatmeal. These GF versions with prepared packets come in plain or maple & brown sugar flavors, and I eat the latter periodically and have never gotten glutened from these. Like the cereal, I make this decision based on the brand’s processes and testing of input ingredients as well as batch testing the products, plus the awareness that there is still a risk but that I am further mitigating this risk by consuming single-serving quantities of it.

—

All of the above is a lot of words but this chart may help, with colors to show how as you go down the levels of labeling certainty, the type of product (the three columns: simple; complex processed; oat containing) also should factor in, AND so should the quantity because the risk goes up the more you eat something.

Example chart showing the 5 row categories from certified GF to 'may contain wheat', plus columns for simple food that is naturally gf to more complex processed food, to an oat column broken into two categories for oats with good testing processes and those without that info. As you go up in row category, or to the right toward complex food or oat-containing foods, the amount of food consumed increasing also drives up risk. The chart shows ok-ish decisions versus those that start to transition to caution or to outright avoidance categories.

It’s absolutely ok to make different decisions at different points in time, based on the situation. If you are struggling with recovering after diagnosis, you may want or need to be more careful and limited in what you eat. If you have been getting glutened and can’t figure out where, this may help you think through types of food or quantities where the risk may be in the quantity. (P.S. – people with celiac can also have exocrine pancreatic insufficiency (known as EPI or PEI), so be aware if you have eliminated all sources of gluten and still have GI symptoms, you may want to take the EPI/PEI-SS and ask your gastroenterologist and discuss elastase testing to rule out EPI).

And sometimes you’re just too tired to think through the risk and just stick with the safest possible option – meaning only choosing the things labeled definitively GF. That’s fine, too. You can also choose differently when you have more energy to do research and find out if you should rule something permanently out or rule it back in if the brand has changed their labeling / processes /etc.

Personally, if I’m put on the spot (meaning someone offers me food or shows me a label and asks me if I want any), if there is any uncertainty I’m prone to saying no thanks, it’s not worth it (right then). Sometimes I’ll go back later when I’m not on the spot and look it up and look for more information about testing/etc from the brand. That’s how I ended up adding the PB&J rounds into my snack rotation, after first excluding them because they failed my oat rubric test based on the on-product label.

Another big caveat is to remember to check products often, even if it’s something you eat often. Brands could add more testing and validation to their products; they can also change their products and add new cross contamination warnings to their products that weren’t there before. Anything that is processed is definitely worth reading the label for: I have found that cheap store-brand bags of frozen vegetables usually have “may contain wheat” warnings on them (argh!) on the product itself, even though the online pictures of it and the label don’t show them online. So keep in mind online listing and online pictures and actual on-product labels can vary. (Name brand frozen veggies tend to not have this warning on them, so in that scenario I would pay more (sigh) for the name brand without the warning).

What do I eat as a person with celiac? And how do I decide what is safe to eat? A blog post by Dana M. Lewis on DIYPS.orgAnd again, I want to emphasize that you can make difference decisions for different situations. How I decide – and my rubric above – may not match how you handle things. Everyone is different. If you’ve been glutened lately and want to be more strict/careful in your decision-making, you can. If you are traveling and don’t want to risk feeling bad, you can be more stringent in your criteria for what you do/don’t risk. These are just a few examples of where I make varying decisions at any point in time.

—

Disclaimer is probably worth repeating: I’m not a medical professional, I’m just a person with celiac for 18+ years, I am not telling you what to do, but this reflects how I handle uncertainty and decision making and how/why it changes over time, especially as someone who eats a lot of packaged and prepared foods.

Want signal? We need more noise (more examples to address the quiet bottleneck)

by Dana Lewis

Note: I’m assuming if you are reading this post that you’ve first read this post, summarizing the concept. The below is a companion piece with some expanded details about the concepts and more examples of how I think addressing this bottleneck will help make a difference.

—

You might hold the perspective that in the growing era of AI, there’s too much noise already. Slopocalypse. (Side note: this entire post, and my other post, was written by me, a human.) That’s not the biggest problem in healthcare, though: in both research and clinical care, so much critical data is simply not collected. We are missing sooooo much important data. Some of this is an artifact of past clinical trial design and how it was hard to collect or analyze or store the data; there were less established norms around data re-use; and some of it is a “collect the bare minimum to study the endpoints because that’s all we are supposed to do” phenomenon.

Nowadays, we should be thinking about how to make data and insights from clinical work and research available to AI, too, because AI will increasingly be used by humans to sort through what is known and where the opportunities are, plus make cross-domain connections that humans have been missing. And we need to be thinking about whether the incentives are set up correctly (spoiler: they’re not) to make sure all available data is able to be collected and shared or at least stored for humans and AI to have access to for future insight. What is studied is also ‘what is able to be funded’, which is disproportionately things that come to the commercial market eventually after regulatory approval.)

“But what’s the point?” you ask. “If no one is looking at the data from this study, why collect it?” Because the way we design studies – to answer a very limited-scope question, i.e. safety and efficacy for labeling claims and regulatory approval – is very different than the studies we need around optimizing and personalizing treatments. If you look at really large populations of people accessing a treatment –  think GLP-1 RA injectables, for example – you eventually start to see follow-on studies around optimization and different titration recommendations. But most diseases and most treatments aren’t for a population even 1/10th the size of people accessing those meds. So those studies don’t get done. Doctors don’t necessarily pay attention to this; patients may or may not report relevant data about this back to the doctor (and again even if they do: doctors don’t mentally or physically store this data, often); and the signal is missed because we don’t capture this ‘noise’.

This means people with rare diseases, undiagnosed diseases, atypical or seronegative diseases, unusual responses to treatments, multiple conditions (comorbidities), or any other scenario that results in them being part of a small population and where customizing and individualizing their care is VERY important…they have no evidence-based guidance. Clinicians do the best they can to interpret in the absence of evidence or guidelines, but no wonder patients often turn to LLMs (AI) for additional input and contextualization and discussion of the tradeoffs and pros and cons of different approaches.

We have to: the best available human (our clinicians) may not have the relevant expertise; or they may not be available at all; or they may be biased (even subconsciously) or forgetful of critical details or not up to date on the latest evidence; or they may be faced with a truly novel situation and they don’t have the skills to address it because they’re used to cookie-cutter standardized cases. This is not a dig at clinicians, but I recognize that we now have tools that can address some of the existing flaws in our human-based healthcare system. I keep talking about how we need to recognize that evaluations of AI in healthcare shouldn’t treat the status quo as the baseline to defend, because the status quo itself has problems, as I just described.

And AI has some of the same problems. (Except for the ‘not available’ part, unless you consider the lower-utility free access models to mean that the more advanced, thinking-based models are ‘not available’ because of cost access barriers). An AI may not have any training data on a rare disease… because nothing exists. It may drop information out of the context window, and we may not realize that this has happened (i.e., it ‘forgets’ something). Usually these are critiques of AI, juxtaposed against the implication that humans are better. But notice that these critiques are the same of humans, too! This happens all the time with human clinicians in healthcare. A human can’t make decisions on data that doesn’t exist in the world, either!

So…how do we “fix” AI? Or, how do we fix human healthcare? We should be asking BOTH questions. Maybe the answer is the same: increase the noise so we increase the signal. Argue about the ratio later (of signal:noise), but increase the amount of everything first. That is most important.

How might we do this? I have been thinking about this a lot, and Astera recently posted an essay contest asking for ideas that don’t fit the current infrastructure. (That’s why I’m finally writing this up, not because I think I’ll win the essay contest, but mostly because i’s an opportunity for people to consider whether/not this type of solution is a good idea as a separate analysis from “well, who is going to fund THAT?”. Let’s discuss and evaluate the idea, or riff on it, without being bogged down by the ‘how exactly it gets funded and managed over time’.)

I think we should create some kind of digitally-managed platform/ecosystem to do the following:

  1. Incentivize and facilitate written (AI-assisted allowed) output of everything. Case reports and scenarios of people and what they’re facing. All the background information that might have contributed. All the data we have access to that is related to the case, plus other passive, easy-to-collect data (that may already be collected, e.g., wearables and phone accelerometer data) even if it does not appear to be related. Patient-perspective narratives, clinical interpretations, clinical data, wearable data – all of this.

    A.) And, be willing to take variable types of data even from the same study population. For example, as a person with type 1 diabetes, I have 10+ years of CGM data. For a future study on exocrine pancreatic insufficiency, for example, not everyone will have CGM data, but because CGM data is increasingly common in people with T2D (a much larger population) and in the general population, a fraction of people will have CGM data and a fraction of people are willing to share it. We should enable this, even if it’s not for a primary endpoint analysis on the EPI study and even if only a fraction of people choose to share it – it might still be useful for subgroup analysis, determining power for a future study where it is part of the protocol, and identifying new research directions!

  2. Host storage somewhere. There can be some automated checking against identifiable information and sanity checking what’s submitted into the repository, plus consent (either self-consent or signed consent forms for this purpose.)
  3. Provide ‘rewards’ as incentives for inputting cases and data.

    A) Rewards might differ for patients self-submitting data and researchers and clinicians inputting data and cases. Maybe it’s one type of reward for patients and a batch reward of a free consult or service of some kind for researchers/clinicians (e.g. every 3 case reports submitted = 1 credit, credits can be used toward a new AI tool or token budget for LLMs or human expertise consult around study design, or all kinds of things.)

    B) Host data challenges where different types of funders incentivize different disease groups or families of conditions to be added, to round out what data is available in different areas.

  4. Enable AI and human access (including to citizen scientists/independent researchers who don’t have institutions) to these datasets, after self-credentialing and providing a documented use case for why the data is being accessed.

    A) Require anyone accessing the dataset to make their analysis code open source, or otherwise openly available for others to use, and to make the results available as well.

    B) Tag anytime an individual dataset is used by a project, that way individuals with self-donated data (or clinicians submitting a case) can revisit and see if there are any research insights or data analysis that applies to their case.

    C) Provide starter projects to show how the data can be used for novel insight generation. For example, develop sandboxes for different types of datasets with existing ‘lab notebooks’, so to speak, to onboard people to different datasets/groups of data, different types of analyses they might do, and to cut down on the environment setup barriers to getting started to analyzing this data.

    D) Facilitate outreach to institutions, disease-area patient nonprofits and advocacy groups both to solicit data inputs AND to use the data for generating outputs.

Why should we do this?

  • Clinical trials do not gather all of the possible useful data, because it’s not for their primary/secondary endpoints. Plus, they would have to clean it upon collection. Plus storage cost. Plus management decisions. Etc. There’s a lot of real barriers there, but the result is that clinical trials do not capture enough or the most possible relevant data. So we need a different path.
  • Clinicians/clinical pathways don’t have ways to review data, so they avoid doing it. There’s no path to “submit this data to my record but I don’t expect you to look at it” in medical charts (but there should be). So we need a different path.
  • Disease groups/organizations sometimes host and capture datasets, but like clinical trials, this is limited data; it often comes through clinical pathways (further limiting participation and diversity of the data); it doesn’t include all the additional data that might be useful. So we need a different path.
  • Some patients do publish or co-author in traditional journals, but it’s a limited, self-selected group. Then there are the usual journal hurdles that filter this down further. Not everyone knows about pre-prints. And, not everyone knows how useful their data can be – either as an n=1 alone or as an n=1*many for it to be worth sharing. A lot of people’s data may be in video/social media formats inaccessible to LLMs (right now) or locked in private Facebook groups or other proprietary platforms. So we need a different path.

Thus, the answer to ‘why we should do this’ is recognizing that: AI does not create magic out of nowhere. It relies on training data and extrapolating from that, plus web search. If it’s not searchable or findable, and it’s not in the training data, it’s not there. Yes, even the newer models and prompting it to extrapolate doesn’t solve all of these problems. What AI can do is shaped disproportionately by formal literature, institutional documents, and whatever scraps of publicly available content remain. If we want to shape what’s possible in the future, we need to start now by shaping and collecting the inputs to make it happen. We can’t fix the past trial designs but we can start to fill in the gaps in the data!

Here is an example of how this might work and why it matters to build and incentivize this type of data sharing.

  1. If we build and incentivize data sharing, the following individuals might self-donate their data and write-ups, or a clinician might submit them. Later, someone might analyze the data and put the pieces together.
  • Someone with type 1 diabetes (an autoimmune condition) and this new-onset muscle-related issue. Glucose levels are well-managed, and they don’t have neuropathy/sensory issues (which are common complications of decades of living with type 1 diabetes), and muscle damage and inflammation markers are normal. MRIs are normal. The person submits their data which includes their lab tests, clinical chart notes (scrubbed for anonymity), a patient write-up of what their symptoms are like, and exported data from their phone with years of motion/activity data.
  • A different person with Sjogren’s disease (also an autoimmune condition) and a similar new-onset muscle-related issue. There are known neurological manifestations or associations with Sjogren’s, but more typically those are small-fiber neuropathy or similar. The symptoms here are not sensory or neuropathy. MRIs don’t show inflammation or muscle atrophy. Their clinician is stumped, but knows they don’t see a lot of patients with Sjogren’s and wonders if there is a cohort of people with Sjogren’s facing this. The clinician asks the patient and gains consent; scrubs the chart note of identifying details, and submits the chart notes, labs, and a clinician summary describing the situation.
  • Later, a researcher (either a traditional institutional-affiliated researcher OR a citizen scientist, such as a third person with a new-onset muscle-related issue) decides to investigate a new-onset muscle-related issue. They register their hypothesis: that there may be a novel autoimmune condition that results in this unique muscle-related issue as a neuromuscular disease (that’s not myasthenia gravis or another common NMJ disease) that shows up in people with polyautoimmunity (multiple autoimmune conditions), but we don’t know which antibodies are likely correlated with it. The research question is to find people in the platform with similar muscle-related conditions and explore the available lab data to help find what might be the connecting situation and classify this disease or better understand the mechanism.
  • They are granted access to the platform, start analyzing, and come across an interesting correlation with antibody X, which is considered a standard autoimmune marker but one that doesn’t differentiate by disease, and which does highly correlate with this possibly novel muscle condition when it exists and is elevated in people with at least one existing autoimmune condition and these muscle-symptoms that are seronegative for every other autoimmune and neurologic condition. Further, by reading the clinical summary related to patient B and the self-written narrative from patient A, it becomes clear that this is likely neuromuscular – stemming from transmission failure along the nerves – and that the muscles are the ‘symptom’ but not the root cause of disease. This provides an avenue for a future research protocol to 1) allow these types of patients to be characterized into a cohort so this can be determined whether it is a novel disease or a subgroup of an adjacent condition (e.g., a seronegative subgroup);  2) track whether the antibody levels are treatment-sensitive or not or stay elevated always; and 3) cohorts of treatments that can be trialed off-label because they work in similar NMJ diseases even though the mechanism isn’t identical.
  • The mechanism for this novel disease isn’t proven, yet, but in the face of all the previous negative lab data and neurological testing patient A and patient B have experienced, it narrows it down from “muscle or neuromuscular” which is a significant improvement from their previous situations. Plus, this provides pathways for additional characterization; research; and eventual treatment options to explore versus the current dead ends both patients (and their clinical teams) are stuck in. And because there are no good clinical pathways for these types of undiagnosed cases, this type of insight development across multiple cases would not have occurred at all without this database of existing data.

2. The above is a small-n example, but consider a large dataset where there are hundreds or thousands of people with CGM data submitted. Plus meal tracking data, because people can export and provide that from whatever meal-logging apps some of them happen to use.

  • By analyzing this big dataset, an individual researcher could hypothesize that they could build a predictor to identity the onset of exocrine pancreatic insufficiency, which can occur in up to 10% of the general population (more frequently in older adults, people with any type of diabetes, as well as other pancreas-related conditions like pancreatitis, different types of cancer, etc), by comparing increases in glucose variability that correlate with a change in dietary consumption patterns, notably around decreasing meal size and eventually lowering the quantity of fat/protein consumed. (These are natural shifts people make when they notice they don’t feel good because their body is not effectively digesting what they are eating). They analyze and exclude the effect of GLP1-RA’s and other medications in this class: the effect size persists outside of medication usage patterns. This can later be validated and tested in a prospective clinical trial, but this dataset can be used to identify what level of correlation between meal consumption change and glucose variability change happen over what period of time in order to power a high-quality subsequent clinical trial. This may lead to an eventual non-invasive method to diagnose exocrine pancreatic insufficiency through wearable and meal-tracking data. (None exists today: only a messy stool test that no one wants to do and is hampered by other issues for accuracy.)

–

These are two examples and show how even small-n data or a large dataset where there are additional subgroups with additional datasets can be useful. In the past, it’s been a “we need X, Y, Z data from everyone. A, B, C would be nice to have, but it’s hard and not everyone will share it (or is willing to collect it due to burden), so we won’t enable it to be shared for the 30% of people who are willing”. Thus, we lose the gift and contributions from the people who are able and willing to share that data. Sometimes that 30% is a small n but that small n is >0 and may be the ONLY data that will eventually answer an important future research question.

signal-noise-more-examples-DanaMLewisWe are missing so much, because we don’t collect it. So, we should collect it. We need a platform to do this outside of any single disease group patient registry; we need to support clinician and patient entries into this platform; we need to support intake of a variety of types of data; and we need to have low (but sensible) barriers to access so individuals (citizen scientists, patients themselves) can leverage this data alongside traditional researchers. We need all hands on deck, and we need more data collected.