When it comes to blood glucose testing, with only a little simplification there are three types of people in the world – those who use insulin, those who control their diabetes without insulin, and those who don’t yet have diabetes.

People with type 1 diabetes and pregnant women who take insulin need to test at least three times a day, according to The American Diabetes Association’s Position Statement on Standards of Medical Care in Diabetes. But the recommendations are murky for people with type 2 diabetes who use insulin and anyone changing therapy and are especially murky for people who control their diabetes on diet alone. The statement doesn’t even consider people who control their diabetes with pills or with exercise.

Technically, self-monitoring of blood glucose (SMBG) is what we call blood glucose. The position statement says, “The role of SMBG in stable diet-treated patients with type 2 diabetes is not known.” Until now I never could understanding why they said that.

Now, a huge debate has broken out in the ADA’s professional journal, Diabetes Care. This debate explains why the professionals don’t know how often type 2s who don’t use insulin should test.

For people who use insulin the benefits of testing are obvious and well documented. If your levels are too low or too high, you can correct them immediately.

But what about those of us who don’t use insulin? Of course, we can exercise or not eat if our level is too high, or we can take a glucose tab if it’s too low.

Mayer Davidson, an M.D. who is a past president of the ADA, shocked me with his editorial that it is “a waste of money” for type 2s not using insulin to test. He does believe that if they test the best time is before and one or two hours after a meal.

Rather than checking fasting levels, in the last five years or so testing after meals has become the preferred time for those who have good blood glucose control. More on this later in a separate article.

Why does the ADA say the benefits of testing for non-insulin users is unknown? Because until quite recently there have been few well-designed studies. Few studies – but not no studies.

This year a team of Dutch doctors systematically reviewed these studies and found that six of them were randomized controlled trials. Their article, Self-Monitoring of Blood Glucose in Patients With Type 2 Diabetes Who Are Not Using Insulin, is available in full online. Those who tested were able to decrease their A1C level by 0.4 percent – statistically significant and clinically relevant. A decrease of that magnitude reduces the risk of eye, kidney, and nerve disease by 14 percent, according to the huge U.K. Prospective Diabetes Study.

Two more studies are so new that they haven’t even been published yet, but were presented at the ADA’s Scientific Sessions in June and are abstracted online.

A group of doctors in Atlanta presented their study of 552 people not on insulin. Those who tested at least once a day had an average A1C of 6.9 percent, but those who never tested had an average of 7.8 percent. But the A1C hardly changed if they tested more than once.

Doctors in Germany and Switzerland reported on their 10-year study of 3,268 people with type 2. Those who tested their blood glucose levels – whether or not they were taking insulin – were 33 percent less likely to develop complications of diabetes or to die than those who didn’t test.

The evidence is clear. All of us who have diabetes need to test our blood glucose levels. Others can wait.