Scott Hanselman

My buddy Garrick just got a LifeScan UltraSmart meter and here's his detailed review:

I finally an UltraSmart meter (just before they started to give them away). I intended to get one sooner, but they were not available in mmol/L format, which is what I use here in the great white north.  I started using it August 1, and there is now about 230 data points in it now.

Most of what there is to mention about it has already been said, so I apologize if I repeat much.   I do have an AccuSoft Advantage Complete (also called a Manager) from several years ago, that does much of the same sort of graphing, though far less sophisticatedly. I had high hopes for the Complete initially, but I ceased using it because of the lack of precision and accuracy of the tests.  The Ultrasmart uses the same sensing system as the Ultra, so I assume it will have the same precision,  but also with an 8% low bias in the normal to high range.  [This helps to show better data than we really get, which makes us feel good.  I always add 8% to any reading I get that is not too low, and to my averages.] Unfortunate, considering it doesn't need to be inaccurate.

The Ultrasmart is larger than the regular Ultra, as is the carrying case, therefore it is more cumbersome to carry around, so I likely won't use it as a portable meter.  It does use easily available triple A batteries, not as cheap as double A, but far cheaper than button cells would be in a meter like this.

It looks nice, and can be easily confused for a PDA by strangers... as long as they don't see the blood spilling (:c>).

Every 25 tests, it asks for confirmation of the test strip code.  This is inconvenient and a possible waste of a strip if you apply the blood sample without noticing that it hasn't become ready yet.  I have wasted 3 so far. Lifescan has made a couple of extra bucks for J & J.

3000 test memory.  This is very practical for myself.  I don't download my regular Ultra meters, which have only a 150 test memory, to the Onetouch computer program regularly enough, so I have several large gaps in the stats.  Since this will be my most commonly used meter, there should be less gaps.  (I have 8 or 9 Ultras, altogether)

To turn the light on or off, hold the OK button down for at least 1.3 seconds.   Note that you can turn the meter on with the light, and also go straight to the Fastfacts menu, by holding the OK button.

The Ultrasmart is as easy to use as it possibly can be.  Very intuitive, even without the instruction book.  The Fastfacts button is my favourite.  It accesses graphs and averages, amongst other ways to observe the data.  The Graph of all results (push Fastfacts, select Glucose Analysis, push OK, push OK for Graph of All Results, push OK one last time or wait until the meter goes there on it's own) can only cover 3 days on the display screen, so it can be moved using the arrow keys, one data point at a time.  By holding an arrow key for long enough, you can get a 'flowing graph' that moves at about 8 data points per second.  Also, each individual point can be selected (push OK for the one that is flashing...hard to see especially without the light on) to view the 'comments' that were added to it... not something I will be doing.  The Graph by Time of Day works similarly, but can only include data points from one of the seven time zones (before and after each of three meals and one for night time).  You choose which one.  After you have seen one, you can use the back button to go back and choose another.

Average of All Results is also easily available.  (push Fastfacts, select Glucose Analysis, push OK, select Average of All Results, push OK)  The display shows the averages for the last 7, 14,30, 60, and 90 days all on one screen.  Quite useful for quickly spotting trends over the last 3 months, though not as inclusive as the 'flowing graph'.  Average by time of Day gives the you the averages for the seven time zones on one screen.  First, you have to choose how many days you want the averages to cover.  You have a choice of 7, 14, 30, 60, or 90 days.

One thing it lacks is the standard deviation (SD) of the data.  I consider the two most important numbers to be the BG average, but also the SD.  The SD tells me how tightly controlled I have been.  [It is available on the computer program though, however less convenient.]  SD isn't perfect of course.  The main problem would be having a lot of hypoglycemia.  Since hypoglycemia is not far from perfect, (compared to hyperglycemia),  having a lot of it would still show the SD to be quite low, and the average BG would also look very good, but I certainly wouldn't be feeling my best.

It does have hypo information, which is useful, especially considering that hypo level is customisable, as is the 'acceptable' bg range, both before and after meals.  For the post meal data to be reasonably accurate, you would have to eat at regular times.  I just use the same range for all times anyway so that is not a problem for me.  [My target is 4.7 (85), or my target range being 4 to 7 (72 to 125), which is the range for which if all is stable, I generally take no corrective action]

As usual with all things techy, there is more here than I care to use.  While I do agree that the exercise, health, meds/insulin, and food data buttons are useful for many diabetics, I find I can remember these things easily enough now (except for my medical records which are stored on good old paper), that I haven't any real use for  their accompanying programs.  As well, the calculations and protocols I've developed for myself are much more sophisticated than the Ultrasmart allows for.  [For instance, my insulin protocols for backpacking took me 3 days of hiking at various distances and elevation gains.  The result is a schedule that reduces my basal and bolus dosages, and also the lead time on the boluses, based on distance traveled and how many thousands of feet of elevation gain I have done and/or will do  ...BTW, it seems incredible to consume 700 grams of carbs in a day, and only need a total of 15 units of insulin, including the basal.]

The food data inputs capability may be suitable for type 2s, but are far too primitive for accurate enough calculating for tightly controlled type 1s, not to mention cumbersome.  I would however, highly recommend the food and exercise functions to new type 1s or type 1s just starting to use a basal/bolus routine.... partly because the meter provides some degree of fun.

One thing I consider lacking, is the choice of viewing the averages back farther than 90 days. Though 90 days nicely covers  HbA1c, with 3000 tests stored, it easily could go back 8 months or even years, depending on the frequency of one's tests.  From my own bg computer programs, I do know that December, July and August are my worst months. However, making use of that much data would presume this meter survives as my favourite for long enough.  In the present climate of ongoing improvements that are happening to meters recently, it may not. 

Garrick Neal

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By