I've been reading a lot of the threads here and see a lot about a low carb diet. However, I have a question or series of questions. First let me start out on this premise: Let's say I am a type 1 diabetic (how about that, I am!) and I know my I:C ratio is 1:20 as an example. As long as I give myself proper insulin for the carbs I don't have a problem with my BG being out of control after meals regardless of whether I have 200 carbs a meal or 60. In this type of situation, what is the benefit to me as a diabetic, not just general health benefits anyone could get, from being on a low-carb diet? In essence, I'm doing what my pancreas would normally do and dose for the meal (although my pancreas was much more efficient and precise of course).
If this is a hot-button issue here then forgive me, I don't intend to open up a can of worms. My books from Dr. Bernstein and 'Think Like a Pancreas' are still on their way, so maybe they will answer this question. I've just seen a few people here who seem to feel very guilty that they were still "living like they didn't have diabetes and dosing for the meal." If you're BG's are in check and your otherwise healthy and doing okay, why is this a problem? Again, we're discounting the benefits anyone could get, and focusing on why it's beneficial specifically for diabetics.
NOTE: I DO count my carbs religiously (I don't see how that could be optional as a T1) and try, in general, to eat fewer than I did before I was diagnosed. Also, the ratios and scenario above do not fit me, they are just setting the stage for the discussion.
Zoe, once you've downloaded your ping data, go to the 'compilation' tab. The SD is in the left hand box. If you need more help, please ask! Mine as of the last couple of weeks is 25. Gerri is right, finding the real sd is a long pita, but I'm only interested in my deviation from the mean.
Thanks for the info, jrtpup. This thread is going off in all kinds of directions! I believe you're talking about the EzManager software when you say "downloaded the ping data" not anything on the pump itself or the remote. I actually never downloaded it. I glanced at it and it seemed like a pita and I haven't gone back...lol. I keep old school records and it works for me.
So I guess the range I described earlier of how much is above, how much in range and how much below can't produce the SD? That's ok, I was just curious how mine compared to other people's but the ranges are useful info as well. Thanks anyway!
I do kind of like AR's approach below of just looking day by day and seeing what to change. I've had my pride at my 5.7 A1C eroded by struggling numbers in the last month so easy come easy go..lol
But when you calculate your sd, all it is doing is estimating exactly that, what percentage of your BGs fell within a certain range. For SD, the range is 1 and 2 standard deviations away from the mean, with 68% of your readings expected to fall within one standard deviation of the average, and 95% of your data falling within 2 standard deviations from the average.
You could just decide what is a useful range for you, and determine what the percentages actually are just like Zoe has done.
Back in the 80's I got fructosamines done regularly - then it became 'old fahioned' and the A1c took over.... I was too young (10yrs) and the internet wasn't on anyones radar so I have no idea of what the reasoning was!
Sorry, I don't know about extracting SD info from your Ping. I'm Ping-less & pump-less:) You can use two week's readings to get a pretty good idea. Of course, the more data, the better if you've got 30 days. I only did SD a couple of times because I don't log consistently (hate logging). Once it was 18 & the other was 27. I was curious because I have gastroparesis.
Oh yeah, I know Gerri. Jrtpup told me I need to download software to do it. But I don't even know what it means, so it doesn't seem worth the trouble. You said you got 18 and 27. Are those good numbers?
That percentage within range, percentage over range and percentage under range for before and after meals that I have on my pump (without the need for either software or math) seem like enough useful information enough for me.
18 is good, 27 isn't. Having the precentages is very useful. I'm not a math person, so I've only done SD twice. Probably won't ever do it again.
Without getting into a YMMV discussion, an SD of 20 or less is considered to be good because it indicates a small variance.
Zoe asked if what I listed was good.
You can't really begin to say if 18 or 27, or any other standard deviation is good or bad just by looking at that one number. The calculation may be horrendous, but it's meaning really isn't all that complicated and is pretty straightforward.
If my diabetes caregiver told me that my SD was important but couldn't tell me what it means, I'd back away slowly and find a new caregiver.
Basically, assuming all your BGs fall under a bell shaped curve, you expect around 68% of your BGs to be within one standard deviation of your average and aound 95% of your BGs to fall within 2 standard deviations of your average.
So, if my average BG is 100, then a standard deviation of 27 means that a I would expect to find 68% of my BGs to fall between 127 and 73, and 95% to fall between 154 and 48. If my average is 100 and my standard deviation is 18, I'd expect 68% of my BGs to fall between 118 and 82, and 95% of my BGs to fall within 136 and 64.
In this case, is 18 better than 27? Sure. But is 27 bad? If so, based on what?
Wanna be proactive in bringing down your standard deviation? You don't have to make any changes at all. Just take more samples and, assuming nothing else changes, your standard deviation will go down simply because of the way standard deviation is calculated.
If you look at the formula, the "variation" is in the numerator and the number of samples (how many times you take your BG) is in the denominator. Mathematically, all you have to do is take more samples without ever changing how much your BG varies at all, and, voila, lower standard deviation.
I know this might not be statistically correct (please don't flame me)... standard deviation for me is how much you vary from your median. So if your median BG is 100, and your SD is 30, you've been between 70 and 130. The smaller the better.
Gerri, I don't think 27 is 'not good', though I do aim lower.