3 Years with the Dex...my experiences to date (UPDATED below at 3.75 years)

Hard to believe, but I am now starting my 4th year with the Dex. I have been a T1 for 31 years, and after some missteps in the first year (back then it was still urine testing and pork-based insulins...enough said!) I have been in very tight control and have no complications to date. I do not use a pump, but I was very diligent about fingersticking...probably 12-15x a day. But three years ago I had a low blood sugar episode that prompted my endocrinologist to suggest a CGM system, and with that I started on the original Dexcom. Here are some thoughts after three years on the system:

Much more awareness of my BG trends following different activity levels, meals or insulin dosages. That was an eye-opener my first few months on the Dex.

Activities that were problematic for fingersticking (skiing, golf - since you grip the club blood would continue to gush out of a fingerstick for several holes due to the grip pressure, gym workouts, business meetings, movies or shows) were now significantly better. The example I give is standing at the top of a ski run, bundled up, and wanting to make sure my BG was ok before going down. That was impossible to do with a One Touch because the outside temp was too low, not to mention the fact that you'd have to take off gloves and get all the "stuff" to do the test. With the Dex it is simply a matter of reaching into a pocket.

No more "overreacting" to high blood sugars...this is related to my first point above. I could now see where higher blood sugars were peaking and starting to settle out. With only fingersticking, sometimes I would over-medicate to get high blood sugars heading downward.

Lower stress levels related to blood sugars. Just being able to look at the Dex and see that sugars are in a normal range with a flat arrow is relaxing...or rather it at least reduces the stress levels about maintaining healthy blood sugar levels.

The Dex has made me aware the fingersticking is not fool-proof. Often times if I get a fingerstick that is materially different from the Dex reading, I always fingerstick again to check. The One Touch is not perfect! If your fingertips are not clean then there is the possibility that there is some sugar residue there that can make the fingerstick inaccurate. My guideline is that if fingersticking is materially lower than the Dex, the fingerstick is probably accurate and the Dex is slow to catch a decline in BG. I don't bother to recheck the fingerstick...I treat the low blood sugar. If the fingerstick is materially above the Dex, then I don't do anything until I fingerstick again to see if the value is closer to what I'd expect given the Dex's reading. I NEVER GIVE MYSELF INSULIN BASED ON A SINGLE FINGERSTICK READING...that is dangerous because even with normal blood sugars it is possible to get an inaccurate high reading on a fingerstick.

For long drives, monitoring blood sugars are easier. I always fingerstick before getting behind the wheel, but for monitoring trends (i.e. watching for down arrows) while on the road it is useful to have the Dex.

Customer service has been excellent with almost no exception. Sensors have been replaced without question when I have called with problems.

Net net, my highs have been lower, and my lows have been higher than before, and my A1C's have continued to be in the same ranges that I had when I fingersticked frequently.

I have had every possible frustrating experience with the Dex over the past 3 years:

The sensor that just goes bad for no reason...instant "failed sensor"

The "???" episodes that are much less frequent with the Dex Seven Plus than they were with the original, but still happen.

The inaccurate low readings at night that keep me and my wife awake. Sometimes the only way to get some sleep is to turn off the sensor. When these first happened I would over-correct the lows by eating or drinking juice just to get the low alarm to stop. Then later that night I'd have high blood sugars from the over-correction.

The static electricity reset....all of a sudden the sensor goes dark.

The tape that starts coming loose after 4 or 5 days.

The sensor that just never seems to be accurate, or even close to accurate. Usually I give a sensor two days before killing it and calling for a replacement. Sometimes you never know...I've had sensors that barely worked the first day and they somehow become perfect for the remainder of the week.

Related to the "low alarm at night" comment above, I've had periods where the sensor flat lines at a high or low level for an extended period of time. Frustrating when it is either low or high. Usually a flat lining sensor is on its last legs and I kill it.

In general, the sensors that make me think the system should be called the Dexcom Five or the Dexcom Six...sensors that start slow and end prematurely. On occasion I have had sensors that perform really well and the adhesive also works well, and I have used 9, 10 or even 11 days. But those are pretty rare...usually performance for me starts to degrade and I've never experienced the great performance in the "extension period" that others here have seen.

The system is expensive, though worth it to me. Blue Cross has covered it for me since day 1, which helps, but it still is a lot of $$$ out of pocket.


Please make the receiver smaller. Holy Toledo, three years now and there has been no improvement in the form factor. Make it an iPhone app or at least make the receiver smaller. My original One Touch from the 80's was gigantic, and so was my first cell phone, but these days to have a form factor that stays unchanged for 4 or 5 years is terrible.

Have a Mac version of the software. I don't use the software much at all, but when I need to (when working with Customer Service to diagnose a problem with a sensor) I need to borrow a PC to upload. I don't have the know-how to put the software on my Mac using parallels or whatever.

Use a more universal charging system. One time on a business trip my receiver lost power and I couldn't restart it using the reset button (stick a pencil or paperclip in the hole in the back). Customer service said I had to stick the charger in it...problem was I didn't have the charger with me so I had no Dexcom for a couple of days.

On the receiver, have a countdown reading so I know how long a sensor has until expiration. I sometimes forget what day or what time I started a sensor. Would be great to be able to find out how much time a sensor has left.

When you re-start the receiver after shutting it down, please eliminate the "test" of the alert function that is loud enough to startle someone in another zip code. I once restarted my receiver in a quiet room (I can't remember why I shut it down) and the noise that the receiver made when it restarted had heads all turned my way.

Did I say make the brick receiver smaller?


Hope this summary was of interest. On the whole, I would say I am glad I have the Dexcom, though I wish it were better for all the reasons above. If I were to give the Dexcom a rating, it would be something in the range of 7.0-7.5 out of 10.0. At times it is frustrating, but on the whole it beats not having it. When there is a time that I don't have a sensor attached (like when I kill a sensor and want to wait until the next morning to start a new one) I really miss having the sensor for what it provides, even though I like not having the bulky receiver in my pocket. I will keep going now into my fourth year (this is my second receiver that I got about a year ago...my first receiver started losing charges quickly and the receiver was getting weaker) but I hope some of my suggestions above will be incorporated in whatever new version comes out (whenever that date may be).

It is wonderful reading your peice. Thanks

Thanks for the review. I expected some inconsistent performance so i am reasonably satisfied with it. Cost is a big issue and I wish I could get 14 days per sensor but my body usually allows 10, sometimes 12. I relate to getting tired of it but then missing it.

I am disappointed with the inherent inaccuracy once ISF gets stale during inactive sleep hours but i see no reason to fault Dexcom for an issue that is unavoidable (except of course, all the PR never mentions this). I get around it by placing sensors closer to the midline of my stomach to stay away from ISF pooling as I sleep on my sides and i set my low alert lower at night.

Gen 4's receiver will be smaller and in color, and you raise a good point, I hope it charges by USB. Ultimately, I'd like readings sent to my cell phone (vs pump) which I pull out frequently anyway.

First off, stomach mount has been both poor accuracy and wildly off while arm mount in my case has corrected that well as others have reported on many other sites.

Overnight perfomrance on arm with exception of faulty sensors has been fine.

Accuracy of sensor has been good and as long as blood BG is slowly moving, both finger stick and Dexcom can agree within 2 to 10 units.

When there is fast changes in Blood BG at finger tips, Dexcom sensor is slow tracking and catching up at interstitial tissue. My impression today was that it was not inaccurate.

My read is that sensor is accurate at its mounting point and only suffers from interstitial blood distribution delays.

Biggest complaint that I have requested many times to fix is when receiver overheats. Readings stall at 164 and stays there till chilled down. Second unit they sent did not work any better. FDA asked me to request 3rd unit from Dexcom.
I did over 6 months ago and never replaced.

My work around is to use chilled sealed gel pak in a small food container that enables getting 24/7 performance
since March 2011 without stalls a slong as cold gel pak behind receiver.

I the low limit alarm blocked to 100 max is highly inconvenient and would prefer being able to set to 120 to 130.
I have need to block BG from going sub 100. Once it hits 100; it drops fast and below 80 extremely fast. 10 minute and interstial delay make that worse.

Interference with other RF sources in a condo environment can cause receiver to stall out without warning and display last reading it could obtain prior to interference with no warning of alarm.

As for customer service, I am glad you obtained full and complete polite servive without discrimination.

My experience has been spotty and mostly unhelpful on serious matters. This is not some mass produced fully canned product like a wrist watch sold in volumne.

Single finger sticks are untrustworthy and I have seen on finger bee at least 20 units off. I always test each hand and as long within 10 points, I average out. If not I retest another finger on hand with highest reading and usually see that second reading many times come in unde 5 points difference. Ditto for doing insulin shots - absolutely critical.

Thank you for sharing your experience.

From the Children with Diabetes Web site I found following from a senior member:

After trying almost every spot execpt leg , we are now arm people.

Belly gives us the worst results, I just won't waste a sensor on the belly. The numbers will be 30 + points off for us!

Butt and lower back are fine execpt at night, which is really annoying. So we now use arms almost exclusively , unless he requests another area. We are on day 9 with this arm sensor and are very happy with the numbers .

I know, where is the iPhone app already!?

I posted the following into the Mac compatibility thread a bit earlier today. I'm reposting in here because I think it's kind of important, and it might get lost in that other thread with some many pages to go through. It addresses some of your concerns, TT, as well as those of the OP. Hope you don't mind the reposting. /\/\


I wrote my Dexcom rep today. I referenced this 2/23 article from BusinessWire, and asked her about what the future holds insofar as new ways for us to reap our data (especially us Mac users). This is what she had to say:

"Hi Michael,

Thanks for staying on top of recent developments. The article you attached regarding the acquisition of Sweetspot is exactly what is new with Dexcom regarding web applications. As you read, we just acquired the company so it will take time to develop the applications for approval through the FDA and FCC because not only do we want to bypass the need for a Mac app, but we also want to bypass the need for a receiver—so that we can send the signal to any technical piece of equipment, i.e., cell phone/smart phone. I do know that we have discussed the concepts with the FDA so they can be more in tune with what we want to do as a company moving forward. But it will take time and like Jay says, the timeline is yet to be determined.

If it would help, I am happy to meet with you from time to time, maybe once a quarter to download and review the data from your receiver with you. Just let me know. RM"

The "Jay" is the sales rep I originally wrote to who responded and forwarded my note on to Rose Mary. It's nice that she is willing to meet up with me and look at my numbers. I'm not really all that concerned as my REAL use is the day-to-day trending (which has been awesome!!!), but it would be nice to meet up with her once or twice maybe, just to see what's up. Of course, seeing the numbers might lead me to want to see them more often (the first one is always free...), so you just never know.


Excellent article.
I never, however, heard of anyone sticking themselves 15 times per day. Seems a bit extreme.
Now that your using a CGM how many times do you stick?
I know Dex. reccommends twice daily.

UPDATE almost 10 months later:

I just re-read my 3 year report above, and most of what I wrote above still holds true today. A fair amount of "good" but still a lot of frustration with the Dex that leaves me taking my overall grade down from 7.0-7.5 to something like 6.5.

Quite frankly, over the last few months my sensor reliability has been terrible far too often. I just killed a sensor that I had started 5 days ago. For the entire 5 days it was very unreliable and inaccurate. The tell tale sign was the fact that often there was no trend arrow present...the receiver unable to track the trend in my BG levels even though the levels were not changing rapidly. And I got the fake "low" alerts in the middle of the night that were false. This morning I finally gave up on the sensor and tossed it in the garbage. Its replacement has been active now for four hours and just alerted me to a "45" reading, while two fingersticks (and my feeling) verified was a false low...both fingersticks from clean hands, two different fingers showed mid-90's. So same old song and dance with the new sensor...at least so far.

Which leads me to wish that the next generation Dexcom was available NOW. I'm coming up on almost 4 years using the same system and the honeymoon is long over. I still prefer the Dex to wearing nothing, don't get me wrong, but for the length of time that the system has been in the market it has stayed at the same level of performance (or mediocre performance) for far too long.

If there was a more up to date alternative, I would look in that direction without question. As it is, if someone were to ask me if I recommended Dexcom, my recommendation would be a very qualified/weak one...or at least I would set expectations for reliability at a realistic level.


I am curious where you live and run this unit - rural/dense residential - main city and in free standing house - versus condo. Any comments?


We have exchanged messages over the last year on RF disturbances. Here is my new experience:

I have been in West Los Angeles for over 2 years. Dense city, 70s most of the time, not too humid, but I'm an avid surfer and get a lot of aerobic exercise.

I now work in the country 60 miles north of Atlanta, very hot, very humid, not as much physical activity for me, VERY RURAL. I'm on a plane round trip coast to coast on the weekends every week.

Result...no noticeable difference. I have been using Dex for over 18 months, never had to get a replacement sensor sent, generally within 5-10 bg with my OneTouch glucometer. On average 10-14 days per sensor with the help of Flexifix tape.

My conclusion:

As an educated patient (which I'd say we all are)I'd say user error has nothing to do with it. As a chemical engineer in the biochemical medical industry, I'd say manufacturing variability has nothing to do with it (exactly why FDA approval for new generations, pump integration, etc. takes so long these days). I'd agree there is slight difference lot to lot, but I truly believe a more contributing factor is just how the probe ends up in us sensor to sensor (regardless of insertion technique). Process of elimination leaves....variation in body chemistry.

I have never used a site other than my belly, and have a device that gives me what I feel is an accurate bg every 5 minutes. Yes, sometimes I get an out of range when it is a foot from the sensor, but it comes back (there is a workaround that has been previously discussed that works for me). I bolus my MDIs every day off my Dex. I feel lucky because my body chemistry works with the technology for whatever reason. I am also optimistic, because the technology is only going to get better.

Hang in there everybody, and remain STRONG ON INSULIN!


I live in a semi-rural area on a one-acre lot. I work in an urban office environment. But performance is the same everywhere, including golf courses in the middle of nowhere, a water ski lake in an even more remote area, and driving in a car. For the entire 3.75 years I have rotated the insertion point around the abdomen, and haven't changed my insertion method. Variability week to week cannot be tied to anything obvious.

Truetheory and KevinMR:

Thank you for responding. Your comments are on track for the way it should be - all things being equal.

Thank you for responding. Appreciate greatly your response.

jim snell

Isn't T1 amazing that we put up with expensive, unreliable tech? My hope is that some of the variable performance from sensor to sensor is due to the hand-made process and that the machine-produced G4 will be better. I wish they had skipped straight to a smartphone implementation but I imagine endless FDA delays.

I feel compelled to write that the sensor I am currently using has been as accurate as an atomic clock, despite its slow start. So, I am back to liking my Dex.

The Dex is definitely a love/hate relationship. Better than nothing, for sure, but comes with a lot of drama and volatility.

Good summary recap!

Some things have changed in 2015, but many of the things you point out are good reality checks for anyone in this diabetes world.

The stress level is key. Just being able to look down and check almost feels like you have and assistant following you around ready to give you your BS numbers. If you want a lower A1c and don't want to look like a dog who has be hit too much, waiting for another reaction, a CGM is key.

I also use a pump. I have dropped the Medtronic and went with the Tandem T-Slim. It's display is crisp and easy to read , even when you never bring it up from the waistline. Check you BS and give yourself a dose and no one needs to know.