Keto for type 2 diabetes: what the studies show

Cut the carbs low enough and blood sugar falls — that part isn't controversial. The harder question is whether a ketogenic diet can push type 2 diabetes into remission, and whether that lasts. The trials give a split answer: real improvement in the first six months, and a much smaller edge by a year.

What remission looks like at six months

Blood-sugar control improves fast on carbohydrate restriction. A 2021 meta-analysis of 23 randomized trials and 1,357 people found that at six months, low-carb diets reached diabetes remission — defined as an HbA1c below 6.5% — in 57% of participants versus 31% on control diets. The size of that gap fits the mechanism: eating fewer carbohydrates lowers the glucose load the body has to clear, so blood sugar and insulin demand drop.

See the type 2 diabetes research for each trial behind those figures.

Why the advantage fades

The six-month numbers are the high-water mark. In that same 2021 meta-analysis, the improvements in weight, triglycerides, and insulin sensitivity had diminished by twelve months, and the remission data at a year were sparse. A separate 2017 meta-analysis found low-carb lowered HbA1c about 0.34% more than a higher-carb diet in the first year, then found the two diets similar after a year. A two-year randomized trial saw a clear HbA1c advantage at six months that was no longer statistically significant at twelve.

The studies with the biggest numbers carry the biggest caveats

The most striking remission figures come from a continuous-care program that put people into nutritional ketosis with heavy coaching. At one year it reported HbA1c falling from 7.6% to 6.3% and insulin reduced or eliminated in 94% of the people using it; at two years it reported diabetes reversed in about half the group. Those results are real, but the study was non-randomized and funded by the company selling the program, so participants chose their own group and the comparison isn't a clean test.

Low-carb versus low-fat, head to head

When calories are matched and both groups are supervised, the gap narrows. A 52-week randomized trial found low- and high-carb diets produced the same HbA1c drop and the same weight loss. The low-carb group's edge was using less medication and having steadier blood sugar, not a lower HbA1c.

What this means if you're managing diabetes

The practical read: carbohydrate restriction is a legitimate tool for blood-sugar control, strongest in the first several months, and it tends to reduce how much medication people need — which is exactly why it can't be a solo project. Lowering carbs while on insulin or a sulfonylurea can drop blood sugar too far, so any change belongs in a conversation with the clinician managing your medication. If you do track it, Copper Keto Companion logs meals by voice and keeps a running carb picture, so you and your doctor can see how a week of eating actually lands. Talk to your doctor before changing a diabetes diet or your medication.

FAQ

Can keto reverse type 2 diabetes? Some trials report remission — normal blood sugar without medication — at six months, including 57% versus 31% on control diets in one meta-analysis (Goldenberg, 2021). The effect often fades by twelve months, and remission is not the same as a cure. This is education, not medical advice.

How much does low-carb lower HbA1c? About 0.34% more than a higher-carb diet in the first year, per a 2017 meta-analysis (Snorgaard), with the difference gone by a year. A 52-week trial found the same HbA1c drop on low- and high-carb diets (Tay, 2015).

Is it safe to do keto on diabetes medication? Only with medical supervision. Carbohydrate restriction can lower blood sugar enough to require reducing insulin or sulfonylureas, and that change should be managed by your doctor to avoid hypoglycemia.

Why do the big "reversal" studies get questioned? The largest results come from a non-randomized, industry-funded coaching program (Hallberg, 2018; Athinarayanan, 2019). The findings are real, but the design makes them hard to compare against usual care.

Where can I read the actual studies? The type 2 diabetes research index lists each trial with its numbers and limitations, and the blood sugar research covers the insulin-sensitivity side. For a general primer, see keto and diabetes.