Does anyone know of a way to test blood glucose without a pinprick? I'm open to commercial products (perhaps optical?) or ways to estimate it based on other observable factors. I'm not diabetic, so I don't need the accuracy that a normal glucose meter offers.

Some background to put this question in the context of fitness: Blood glucose, while usually considered in the context of diabetes, is directly correlated to insulin efficiency - or in too many cases, insulin resistance (inefficiency). Because insulin is the primary mechanism by which fat is stored, I don't believe it is possible to design an effective long-term weight management strategy without accounting for your own body's utilization of insulin. This is the hormonal basis of low-carb diets, slow-carb diets, etc.

I believe that, combined with detailed logs of sleep, diet, and exercise, measuring blood glucose on a frequent basis can provide information that will help to understand the impact on my personal physiology of these activities as they pertain to weight management; it's another tool to be used in conjunction with the scale and the measuring tape.

2 Answers 2


As a diabetic, I can very easily tell you that the answer is that there isn't a way. Trust me, if there were, we wouldn't be using blood monitors. All the expletives in the English language cannot describe how much they suck. Anyone who claims the spring loaded lancets are painless is trying to sell you one. Period.

  • DexCom and Minimed both offer user managed continuous glucose monitoring now: insert a single site the lasts several days to a week (on-label, and more off label). Alas, they (1) are still pretty expensive and (2) have a floating absolute calibration so they must be cross-checked with standard prick based tests several times a day at a minimum. Jun 9, 2011 at 2:31
  • Continuous glucose monitoring still requires a blood sample, and from personal experience I can tell you that they are every bit as uncomfortable as pinpricks, though you do not get things like swollen knuckles from overusing a finger.
    – tmesser
    Jun 9, 2011 at 2:53
  • Absolutely - I do not want to use an invasive tool like this, especially since I don't have to (knock on wood). Because this is more of a self-measurement experiment than medical monitoring, I would be willing to sacrifice some accuracy for making the measurements easy and comfortable.
    – G__
    Jun 10, 2011 at 4:32
  • BTW, it seems as though there are a number of non-invasive solutions that are in prototype phase now, such as orsense.com/Glucose and groveinstruments.com/glucose_monitor.html. Hopefully these will prove out and hit the market soon!
    – G__
    Jun 10, 2011 at 4:38

One method seems to be urine test strips. Although the strips are inexpensive and completely non-invasive, there are a couple drawbacks:

  • There is usually not glucose in the urine until the blood glucose exceeds the renal threshold, so this technique can't work for lower glucose levels.
  • The readings actually reflect the level of blood glucose from a few hours earlier.

Typically, the strips seem to simply measure an under/over threshold. For instance, these have a sensitivity of 100mg/dl (this seems to be a standard threshold, although others do seem to exist).

A urine-based system might involve obtaining test strips with different sensitivities. It wouldn't be possible to get an exact reading this way, but as long as blood sugar fluctuated above the renal threshold then it might be possible to chart it in a range bounded by two consecutive strips.

  • Urine strips aren't used by diabetics any more, because they are wildly inaccurate. If you aren't diabetic, you will practically never have glucose in your urine, because your blood sugar will never be high enough for glucose to spill over into your urine.
    – nhinkle
    Jul 20, 2011 at 5:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.