When I started following my standard deviation several years ago, I targeted <= 1.7 mmol/L (30 mg/dL). I read one study that concluded that the 14-day report was a good period to use. It stated that 14 days provided good data sufficiency while still sensitive enough to reveal the parts of the day that need work. This was the study that coined the term, “ambulatory glucose profile.”
I believe that glucose variability is a key metric for monitoring glucose exposure. When it is low enough, it permits targeting a lower blood glucose average safely. One mistake people make is trying to lower their glucose average without keeping an eye on glucose variability as indicated by the standard deviation.
Aiming for a lower average without lowering the standard deviation sets one up for undue hypoglycemia risk. This is what most doctors worry about. They seem to assume that all their patients experience high variability and so jump to a hasty conclusion that a low glucose average equals a high hypo risk.
Using a low carb way of eating (< 30 grams/day) and an automated insulin dosing system, I’m able to keep my standard deviation in the range of 1.0-1.4 mmol/L (18-25 mg/dL). For the last two weeks, however, I’ve relaxed my discipline since I’m on a European holiday. My standard deviation has slipped to 1.8 mmol/L or 33 mg/dL. I’m flying home today and I’m confident I’ll be able to restore my control to former levels.
I measure my time in range performance using 3.6-6.7 mmol/L or 65-120 mg/dL and usually can meet that 80-92% of the time. The clinical standard for time in range is usually set to 3.9-10.0 mmol/L or 70-180 mg/dL. While this standard is suited for a broad range of the patient population, I use the tighter range because I can often meet it and I believe that extended time in the 7.8-10.0 mmol/L (140-180 mg/dL) range is less healthy.