The eHealth Diabetes Remission Trial (eDIT)

Overweight patients with type 2 diabetes are offered a total diet replacement with the goal of weight loss and diabetes remission. Study participants are randomised to eHealth follow-up or face-to-face follow-up, but the dietary advice is the same in both groups. A healthy control group with normal glucose tolerance is examined once but is not randomised and does not receive any intervention.

One hundred and six overweight patients with type 2 diabetes replace usual foods with total diet replacement (850 kcal/day) for 3 months with the goal of 15 kg weight loss and diabetes remission (HbA1c < 48 mmol/mol without diabetes medication). After 3 months of total diet replacement, food is reintroduced stepwise, and an individually tailored energy prescription is used to prevent weight regain. If study participants do not reach the weight loss goal at 3 months, they are recommended two additional months of total diet replacement. If weight regain occurs during the weight maintenance phase, a rescue plan with total diet replacement will be recommended. To increase the chances to maintain their lower body weight, participants will use a program specifically design for that purpose based on cognitive behavioural therapy. Total study duration is two years. The one hundred and six participants are randomised to either the eHealth follow-up or the face-to-face follow-up. The dietary advice to achieve and maintain weight loss is the same in both groups and delivered by the same dietician, physician, and nurse but the method of follow-up differs between groups (eHealth vs face-to-face). eHealth group: All study information and the cognitive behavioural therapy program is given by an eHealth application in Stöd och behandling which is part of 1177.se. Participants register body weight, fasting blood glucose, and blood pressure every morning at 1177.se. Regularly a measurement of HbA1c is taken at home. Participants will have scheduled video appointments with the study dietician, study nurse or study physician. Face-to-face group: Participants have appointments in the medical office with the study physician/nurse/dietician. At the appointments, body weight, blood pressure and capillary blood glucose is measured. HbA1c is measured during the appointments at 0, 6, 12 and 24 months. The cognitive behavioral therapy program for the control group is identical to the program of the intervention group, but the program is delivered during the face-to-face appointments. Healthy control group: Fifteen healthy participants with normal glucose tolerance, stable body weight for at least one year and matched to the study participants for sex, age, and weight after one year study duration, will be examined once. These participants are not randomised and do not receive any intervention. Outcomes: Primary and secondary outcome measures are compared between 1) the experimental eHealth group and 2) the experimental face-to-face group. For the other pre-specified outcome measures the experimental eHealth group and the experimental face-to-face group are taken together as one group, examined twice (baseline and 12 months) and compared to the healthy control group, that is examined only once. Long-term follow-up: Because long term data on diabetes remission by total diet replacement are lacking, study participants will be followed after they have finished the study. Data of the national diabetes registry from participants of this study will be compared to other patients in the national diabetes registry not participating in this study but matched for age, sex and diabetes duration to the study participants.

HbA1c without diabetes medication as a continuous outcome (HbA1c correction according to Tsapas et al 2020 if on diabetes medication). The outcome will be tested for non-inferiority first, with a non-inferiority margin of 8 mmol/mol; followed by a test for superiority.

HbA1c without diabetes medication as a continuous outcome (HbA1c correction according to Tsapas et al 2020 if on diabetes medication). The outcome will be tested for non-inferiority first, with a non-inferiority margin of 8 mmol/mol; followed by a test for superiority.

Number of participants with HbA1c < 48 mmol/mol without diabetes medication. The outcome will be tested for non-inferiority first with a non-inferiority margin of 10 percentage points; followed by a test for superiority.

Partial diabetes remission (<7.0 mol/l), complete diabetes remission (<6.1 mol/l) without diabetes medication.

Karolinska Exhaustion Disorder Scale. Minimum value 0. Maximum value 54. A higher score means a worse outcome.

The scale estimates how satisfied the participants is with different areas of life, as well as how important each area of life is. Minimum scale 0. Maximum scale 48. A higher score means a better outcome.

The questionnaire investigates the dimensions mobility, self-care, usual activities, pain and anxiety/depression. Minimum value 1. Maximum value 5. A higher score means a worse outcome. In addition, a general health score is assessed. Minimum value 0. Maximum value 100. A higher score means a better outcome. All dimension will be analysed separately, but EQ-5D-5L will also be combined to one score.

Inclusion Criteria: Type 2 diabetes with duration 0-6 years BMI 27 kg/m2 and higher HbA1c 43 mmol/mol or higher (48 or higher if without diabetes medication) Exclusion Criteria: Insulin treatment Weight loss more than 5 kg during the past 6 months Diagnosed eating disorder eGFR < 30 ml/min/1,73m2 Myocardial infarction last six months Severe heart failure (NYHA class III) Ongoing cancer Pregnancy Treatment with antipsychotic drugs

Tsapas A, Avgerinos I, Karagiannis T, Malandris K, Manolopoulos A, Andreadis P, Liakos A, Matthews DR, Bekiari E. Comparative Effectiveness of Glucose-Lowering Drugs for Type 2 Diabetes: A Systematic Review and Network Meta-analysis. Ann Intern Med. 2020 Aug 18;173(4):278-286. doi: 10.7326/M20-0864. Epub 2020 Jun 30.