Refine a Trial

Study Outcomes 

Patient Reported Outcomes

  • Sense of disease burden
  • Energy level
  • Cognitive functioning
  • Depression
  • Anger/Frustration
  • Stress
  • Hunger/satiety
  • Impact on planned activities



  • BP
  • Weight
  • Heart rate
  • Heart rhythm

Prospective Measures of Glycemic Variability

  • Time in Range
  • Incidence of hypoglycemia
  • Incidence of hyperglycemia
  • “Length of Line”
  • Number of hyperglycemia measures and hypoglycemia measures per day
Help Improve This
View Discussion
melissablee's picture
I have two thoughts on this one. I would also like to see an outcome measured that looks at the difference between variability in and out of range. A person can have very little variability but a higher A1c because they trend toward 200 with a tight standard deviation. Does this affect the person's QOL? Do they feel better without the fluctuation? Also, would study participants all be asked to define the same target range or would it be adjustable per user?
Anna McCollister-Slipp's picture
These are great suggestions and questions! What do you think? Should we have one "standard" range or let people choose their own? For data analysis methods, we would likely need to have one standard range for everybody.
dougkanter's picture
what does "length of line" refer to?
Anna McCollister-Slipp's picture
Some people have suggested that measuring the "length" of the line from a CGM - i.e. if you stretched out the data trend into one long line - it would be a straightforward measure of the amount of fluctuations and the extremes of fluctuations. The more extreme or frequent your ups and downs, the longer the line. It's a proposed "digital biomarker" for glycemic excursions/variability.
dougkanter's picture
Beyond measuring the number of high and low blood sugars, it would be good to have some sort of general metric for big/fast changes in blood sugars. i feel like glycemic variability most often leads to not feeling well when my day is like a rollercoaster, going super high, then an overcorrection that results in a low, etc.
dgreenwood's picture
Consider trying to use data to develop predictive algorithms.