2 ms·
There's a lot to dig into here that's wrong: One fundamental problem with this is the lack of baseline determination of glucose sensitivity. People with diabet
by ineedasername 6y ago
There's a lot to dig into here that's wrong:
One fundamental problem with this is the lack of baseline determination of glucose sensitivity. People with diabetes, especially type 2, have different sensitivities & levels of insulin production. A group with greater or lesser averages on those factors would strongly influence results. They don't even appear to have accounted for medication dosage, merely whether or not the person was on biguanide. Or even mention weight, another significant factor for insulin sensitivity.
Second is the glucose increase: Glucose tests have margins of error, an issue not addressed here, and the difference in glucose levels this study measure were small considering that factor, about 20 points. Specifically: Even a blood-drawn lab test has a total error of about 8%. If that's tge type of test they did, then an 8% swing could reduce the results to insignificance. If they used a rapid test, even the sort used in a doctor's office, it's even worse: those are considered accurate if it produces results that are within 15% of what a lab test would produce, meaning +/- 15% of a value that already has a total error of 8%.
There's just too much wrong with a study like this that's poorly controlled with ~15 people in each group. And unfortunately there's a decent chance we'll see headlines soon along the lines of "Diabetes: Think sugar matters? think again ::lol lol we made a pun::"