Table 3.

Summary of findings of type 2 diabetes remission at 1 year after diet intervention compared with baseline with GRADE certainty of a body of evidence

Diet	Conclusion statement	No. of participants (no. of studies)	Certainty in the evidencea	Comments
TDR	TDR leads to a large increase in T2D remission by a median of 54% from baseline (range 46–61%), when compared with standard care (4–12%).	445 (2 RCTs)	⊕⊕⊕⊕ HIGH	Low-risk-of-bias RCTs, pre-specified outcomes with power calculation
Meal replacement	Meal replacement likely leads to T2D remission by 11% from baseline, when compared with standard care plus diabetes education (2%).	4503 (1 RCT)	⊕⊕◯ MODERATE Due to possible publication bias	Ancillary observational analysis of RCT
Mediterranean diet	Mediterranean diet may lead to T2D remission by 15% from baseline, when compared with LFD (4%).	215 (1 RCT)	⊕⊕◯◯ LOW Due to imprecisionb and possible publication bias	Small sample size, and ancillary observational analysis of RCT
Very low carbohydrate ketogenic diet	The evidence is very uncertain about the effect of ketogenic diet on T2D remission due to serious risk of bias of the study methods and imprecision, although one non-RCT reported a remission rate of 20%, compared with no remission in usual care with diabetes education.	349 (1 non-RCT)	⊕◯◯◯ VERY LOW Due to serious risk of bias (rated down 2 levels) and imprecisionb	Lack of randomisation, uncontrolled confounding, selection bias, incomplete outcome data, possible selective reporting, imprecision and imbalance between groups
VLED (food based)	The evidence is very uncertain about the effect of food-based VLED on T2D remission, although one small uncontrolled intervention study reported a remission rate of 22%.	9 (1 single group uncontrolled intervention)	⊕◯◯◯ VERY LOW Due to critical risk of bias (rated down 3 levels), imprecision and potential publication bias	Lack of randomisation, uncontrolled confounding, selection bias and selective reporting of result. Only one positive, small study

Remission is defined as either HbA_1c < 48 mmol/mol (<6.5%) or fasting plasma glucose <7 mmol/l and no glucose-lowering medication

^aGRADE level for certainty of evidence: ‘high’ indicates that we are very confident that the true effect lies close to that of the estimate of the effect; ‘moderate’ indicates that we are moderately confident in the effect estimate (the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different); ‘low’ indicates that our confidence in the effect estimate is limited (the true effect may be substantially different from the estimate of the effect); and ‘very low’ indicates that we have very little confidence in the effect estimate (the true effect is likely to be substantially different from the estimate of effect)

^bRated down one level due to imprecision, as the sample size is less than an optimal information size of 400

T2D, type 2 diabetes; TDR, total diet replacement