Wing 1994.

Methods	Trial design: RCT Randomisation procedure: "...randomly assigned.." No further detail given. Allocation concealment: Unclear Blinding: Recipients: No Providers of Care: No Outcome Assessors: Unclear Quality Assessment Category: C (one point awarded) ‐ high risk of bias.
Participants	Country: USA Setting: Out‐patient clinic Number: 93 participants in total were randomised (48 to IG and 45 to CG) at the start of the trial. At six months: No data reported. At one year: 38 participants in IG and 41 participants in CG were followed up. At two years: 36 participants in IG and 37 participants in CG were followed up. Age: At baseline: IG: 52.3 ± 10.7 years CG: 51.3 ± 8.7 years Sex: At baseline: CG: 38% male; 62% female. IG: 33% male; 67% female. Diagnostic criteria: 1. All participants must meet NDDG criteria for type 2 diabetes mellitus. Inclusion criteria: 1. More than 30% or 18kg above their ideal body weight (based on Metropolitan Life Insurance norms). 2. Aged 30 to 70 years. 3. No health problems that would preclude use of a VCLD.
Interventions	Trial intervention(s): Low calorie diet (LCD) with two x twelve week periods of very low calorie diet (VCLD). Participants were prescribed a diet of 400 to 500 kcalories per day for two periods of twelve weeks (weeks 1 to 12 and weeks 24 to 36) interspersed with a low calorie diet (LCD: 1000 to 1200 kcalories per day as in the control diet). During the VLCD period of the study, participants could either consume up to 500 kcalories per day in liquid formula or from lean meat or fish. The LCD was slowly reintroduced after the VCLD period had finished, slowly increasing the calories over four weeks. Comparison intervention(s): Low calorie diet of 1000 to 1200 kcalories per day. Participants were able to select the food items they wished but they were encouraged to spread their calories over the day and to limit dietary fat to less than 30% of calories.
Outcomes	The low calorie group had a baseline mean weight of 104.5 ± 21.5kg and a follow‐up mean weight of 97.7 ± 17.4kg twelve months later. The very low calorie diet group had a baseline mean weight of 102.1 ± 11.7kg and twelve months later had a mean weight of 93.5 ± 10.4kg. The mean blood concentration of glycated haemoglobin in the low calorie group was 10.4 ± 2.0% at baseline, and 11.8 ± 2.7% at follow‐up twelve months later. The mean glycated haemoglobin concentration in the participants of the very low calorie group was 10.4 ± 2.2% at baseline and 9.2 ± 2.1% at the twelve month follow‐up. 1. Weight 2. Glycaemic control 3. Blood pressure 4. Lipids were all measured at six, twelve and twenty‐four months.
Notes	Source of funding of trial: None mentioned A deposit of US$150 per participant was required at the start of the study, but was refunded in full for reaching behavioural goals and attending assessment sessions. Drop‐outs: At one year, ten participants from the IG and four from the CG had dropped out. At two years, twelve participants from the IG and eight from the CG had dropped out.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear

HbA1c = Glycated Haemoglobin 
 BMI = Body Mass Index 
 CG = Control Group 
 IG = Intervention Group 
 NDDG = National Diabetes Data Group 
 WHO = World Health Organisation 
 Unless specified, ## ± ## is mean ± SD