Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm T1D (insulin-dependent). I used a Dexcom for a couple years a while ago, back around the G3 through G5 era or so. I ended up stopping using it because my diabetes is very well managed, and I found it didn't make a huge difference to my management. With my insurance, each new sensor every week cost about $50, and would fail in various annoying ways: sometimes it would just stop working in the middle of the week, or not work at all from the start, or disconnect randomly for a few hours. Each sensor also used a ludicrous amount of plastic for its applicator, like a tennis-ball-sized hunk of hard plastic, that you just chucked in the trash, every week. Maybe things have improved, but it was a bad combo of unreliable and expensive so I just quit using it and went back to finger pokes.

If it were like $5 per week, I might put up with it. Maybe this over-the-counter model will be affordable. We'll see, I guess.



Dexcom and Freestyle both have options (G7, FSL2) that have a much narrower form factor than the old sensors. They end up being far more reliable, because the sensor doesn't dislodge. The new Dexcom sensors are more accurate. Less drift and no calibration.

Out of pocket cost is obviously your own situation but I think anyone with T1 should look at them. For me they are life changing, even moreso if you have a pump that can deliver based on CGM readings.

Plastic waste situation is still bad.


> life changing

Yeah, seeing ~300 datapoints a day instead of 4 while pricking your fingers is amazing QoL improvement. Dead serious.


Glad to hear they're improving, though the waste situation is bizarre. I will probably go back on one at some point, I didn't hate it, just didn't feel it was worth the cost/benefit ratio. It didn't impact my A1c, I've always been in the low 6s, and I've only had two lows that I would consider significant in my 25 years with the disease. Benefit just wasn't there for the hassle and price.


That is impressive control, especially with no CGM!


> The new Dexcom sensors are more accurate. Less drift and no calibration.

I found the Dexcom G7 to be inaccurate by 10-20 mg/dL out of the box. I have LADA, and G7 initially was giving me a very rosy view of my blood glucose levels, until I started pessimistically calibrating it with a finger prick glucose monitor.


> The new Dexcom sensors are more accurate. Less drift and no calibration.

As someone using a freestyle libre who's never calibrated, how often do you need to calibrate it? Reason for asking is that I'm likely to switch to the g6 if I go with the TSlim X2 for looping. TIA!


No calibration is required for Dex G6/G7. I think you can optionally calibrate the G6 but I never have nor have I ever had to.

I am going to be biased: I both work on and use the t:slim X2. It is well worth it, especially if covered by your insurance. CIQ took me from ~70 to 85%, sometimes 90% time in range; my understanding is that it's the best closed loop algorithm out there currently.

Regardless of which pump you end up on, having a closed loop system will significantly improve your control and helps you get that extra little push from the 7's into the 6's a1c range, which I'm sure you know is tricky to do manually without lows.


Thanks! Hearing (I'm assuming) a fellow T1 using and working on a TSlim is quite a good recommendation haha.

It's good to hear that calibration isn't really required.

My "concern" (or rather, apprehension) with the TSlim/G6 setup (which fortunately should be fully covered by Dutch insurance) is that I'm aware I can do a DIY loop using the omnipod dash and with a libre/g6 (using a 3rd party app like diabox & nightscout if necessary). From what I've heard TSlim does seem to be the best "official"/FDA approved loop; however AndroidAPS looks very powerful and probably has much finer tunings possible.

However my diabetic nurse is strongly against the Dash/DIY setup, which is understandable (she isn't familiar with it and can't guide me, I'm also close to the 150u/3day limit, and it's my first time with a pump.)

For what it's worth an A1C of even 8 would unfortunately be an improvement where I am :') so I'd rather be sure of reaching say 7 rather than try aggressively for something like 6.3 (which I had achieved in a distant era...). I think I'll try the 1 year trial option if I can, the thought of locking in a 4 year period feels quite long.

If you're familiar with them, what are your opinions on diy loop setups? Do you have any other suggestions for someone in my situation? Thanks a lot!


I actually have tried the DIY loop with the older insecure Medtronic pumps that could be controlled over radio. It was openAPS. I'm not sure if there are others or how they work.

It ended up being fairly inconvenient. My setup was a pi zero with a radio bonnet running on my desk; it would connect to the internet to pull CGM data, and deliver boluses accordingly.

I think the main issue was that I wasn't babysitting it enough to know if it was working or not. It was hard to see/understand bolus decisions it was making unless I checked iob, and it absolutely destroyed the battery life of my pump; I used a AAA once every couple days when I was near my desk, nevermind the battery on the rig (I had it plugged into the wall).

Overall, the system that you actually use the most, that is the most reliable, is the most effective system, even if it's a little bit suboptimal. In my case, CIQ is probably a bit less aggressive than the openAPS system; but importantly, it is truly "set and forget." I never have to worry if it's enabled or when, or how, it's just always working, 24/7. With a condition like ours, consistency is the best predictor of performance. That makes it the clear winner imho, unless you really like to tinker with things and you want to be very active with your loop.

The other thing is that the software (openAPS) was Javascript and Bash on a Linux system. Those architecture choices just make me nervous. Too many bad things that could happen. I'm not saying it's impossible for open source software to be reliable, certainly not, but the complexity of their system was just too large of a surface area to fully test. There are extremely significant cybersecurity risks using those older pumps also. It's basically root access to the bolus button.


Thanks, those are good points. I think with an omnipod5/libre/androidAPS setup it doesn't need more than a single android phone (as the omnipod is controlled through the phone) but what you mentioned about risks is definitely true. I remember someone talking about bluetooth vulnerabilities on the flipper zero thread possibly affecting/hacking pumps.


> Each sensor also used a ludicrous amount of plastic for its applicator, like a tennis-ball-sized hunk of hard plastic, that you just chucked in the trash, every week.

Erm, let's be a little defensible here.

These sensors have to be stored in sterile packaging--that means thick enough to be a barrier. That means it also has to hold the sensor in such a way that a palette can be gamma irradiated. It also needs to be resistant to drops, crushes, and other accidents.

The sensor has to not be harmful with organisms irrespective of the understanding of the person using it. Any change to that packaging has to be re-evaluated for effects on that sterilization.

And while I understand people having concerns, if the only reason we needed to use plastic was to treat folks with diabetes--I think the world would somehow manage.


I work in the medical device space (mostly software as a medical device). I attended the AAMI Nexus conference two weeks ago and attended a talk about making devices more eco friendly. The tension between safety and sterilization and packaging was definitely brought up. It’s a hard balance to make. One example I thought was interesting was that the NIH did a study to see if it was necessary to swab people’s skin and use gloves before a shot. It turned out it wasn’t and it didn’t make a difference, so they stopped requiring the providers to swab people’s skin or wear gloves, thus saving an immense amount of waste when taken over all vaccine shots.


Eh, maybe, but the bulk of the applicator is just empty plastic molded to fit an adult's hand. I have to believe they could figure something less wasteful out (reusable applicator?) if they wanted or needed to. Regardless, it was just another downside for me weighed against the relatively few upsides that made me quit using it. It felt terrible to chuck all that in the trash, especially when I'd have to do it two days in a row because a sensor failed.

[1] https://fm.cnbc.com/applications/cnbc.com/resources/img/edit...


T1D currently using G6 / Omnipod Dash / open source Loop app.

The worst is every 90 days when I get a new G6 transmitter. It comes in a box the size of Two video tapes stacked on top of each other. I've tried reaching out to my mail-order pharmacy about it since I know there are smaller versions of it (or used to be), but nothing yet.

From what I've seen, the packaging for Dexcom G7 and it's applicator seems much smaller than G6's.

Side note: That applicator for G4 instilled fear every time I had to use it and I was already 20 years into T1D by then. The spring loaded applicators in G6/G7 are a big improvement as far as end user experience IMO :-)


Hahaha, the huge syringe thing? That was pretty wild! The spring ones felt like putting a needle to a balloon though, always made me tense when I was putting one on.

Yeah the Dexcom stuff always felt like they were going for super-premium Packaging Experience. But after the first couple weeks that wears thin, and now you've got all this crap to throw away. Like, guys, my pen needles come in a dense little cardboard box and it's totally fine. Chill out with the packaging.


"Let me just hyperventilate really quick before I literally plunger this thing into my body" was always how it felt lol.

Agreed that it's gotten old, especially if I'm traveling somewhere. Needing to cart what you needed (plus backups of course) ends up taking up a surprising amount of space.

I will say that making the switch from a tubed pump (t:slim) to a tubeless (omnipod), it's drastically cut down on the waste for pump stuff at least.


The g7 is definitely a bit better on all counts except maybe bluetooth reliability. I use them with [open source closed loop](https://loopkit.github.io/loopdocs/) insulin pump software and it is pretty great.

Still a lot of plastic but the sensor comes in a smaller jar/applicator in a cardboard box and it pretty reliably lasts 10 days cutting down waste.

It sounds like this new sensor is just a g7 but with less stringent accuracy standards for non insulin users allowing it to last 15 days.


I just started with the G6 (still using it), and when I started I'd have that sometimes. Whatever they changed in the past year or so, it has gotten waaaaay better. (They would always replace them if they failed early, but was still a huge pain.) I'm exited to move to the G7 (mainly because of the shorter warmup time).

Also, pairing it with the Omnipod 5 is just next level. Closed loop. Never having to inject in public. No tubes.


I have the G7 and it's incredible. On a medtronic pump... wanting to switch asap


The omnipod isn't setup with the g6 yet. Can't wait.


I think you mean G7.


Yup. Those numbers are, like, right next to each other on the keyboard :)


fwiw, my wife uses Freestyle Libre 3 and the with-insurance cost every two weeks is about $40. Whenever she's had one fail (which seems to be about 1:10), Abbott has replaced it free of charge and sent a pre-paid label so she can return the failed device for QA review.


Yeah, I feel like prices are on a downward trend the last couple years, which is great. For me personally, I try to spend absolutely as little time and effort as possible thinking about my diabetes. So having to call the manufacturer or fill out forms and receive a package (more waste) and ship a thing back (even more waste) is just too much time and effort for me to want to bother. I hope they keep improving, it's cool tech and I'd like to like it. Just not there yet.


Worked at a PBM. Interesting factoid.

The FreeStyle Libre 3 has a lower ongoing out of pocket cost for people who just need diagnotic monitoring but no integration with a pump. Dexcom , however, offered rebates that made it cheaper in initial outlay by the PBM to get people using, then left them paying the higher upkeep costs for functionality they may never need.

Something to keep in mind. Your insurer doesn't optimize for you. They optimize for them.


The plastic waste issue is one of the first things that usually comes up around this topic regardless of who you ask. It's such a shame because it's not like you have many options unless you're comfortable going without sensors.


All my diabetes stuff is free, in my country. But it must be shipped to my house in Quarterly deliveries.

Shipped in several boxes, each filled and wrapped. then the GCMs, strips, refills, pump connectors come in their own boxes. Some per ten, some single. Each sub-box has several booklets with instructions and safety guidelines. Each unit is individually wrapped in plastic, paper. All of them are single use. Even connectors and rods.

Our combined household has less waste than my insulin therapy produces. One single insulin therapy produces more plastic, paper and wrapping weight than what my wife and I produce together on everything else.

It's utterly insane.


I'm really intrigued by your statement that your diabetes is well managed, if I wake up in the wrong side of the bed, my insulin resistance change!

Wondering how you can maintain that, do you have a very rigid routine and meal plan?

I'm using Libre2 as part of my prescription and it has been working wonders by alerting early on before it is too low or too high to have enough time for a correction dose, or a small bite. The fact that you can see the trajectory of Glucose going up or down is by itself a reason to get.


Hm, I'm not sure. I don't talk with other diabetics much so I guess I don't know what other's experiences are like. I go to sleep and wake up on the same schedule every day including weekends. My eating is not rigid, but does tend to be consistent just because my daily routine is consistent. I'm not "low carb" or anything, but I do follow a healthy-ish diet and avoid hyper sugary foods (soda, coffee milkshakes, sugary cereals) like the plague. Diabetes or not, no one should be eating that stuff :)

I do miss having the moment-to-moment trajectory information. Knowing whether I'm still going up 90 minutes after a meal was valuable. But I find I'm able to feel when I'm above 200 mg/dL, so can just correct by feel most of the time.


I think your experience was unusual in that you had so many failures (hopefully you at least got your broken sensors replaced for free; both Dexcom and Libre do this) and that you are good at managing your diabetes without many measurements.

The out-of-pocket expense can certainly be rough; I use Libre right now because my Dexcom out-of-pocket would be similar to yours and Libre is a third.

I lend my CGM to people to who are possibly prediabetic. I'm very glad I can point them to an OTC solution now.


You went back to finger sticks? I am floored. Do you follow a strict low carb diet? How often are you exercising?


Definitely not low carb, no. I have a bowl of oatmeal for breakfast 4-6 times a week :) My diet is pretty consistent though. I know how that oatmeal will respond to insulin 90% of the time, and the remaining 10% random quirks I can just deal with by feel & a finger prick to confirm. The thing I'm most strict about is I only allow myself to have about one "bad diabetes meal" per week. So that's like any restaurant food, or doughnuts, or pizza, or whatever. I mostly only eat home made meals, my wife & I both love to cook. But those aren't low carb, just whatever we feel like cooking that week.

I walk about a mile a day, to the bus & back for work and a bit for leisure and to get up from my desk. Otherwise I don't exercise.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: