The NHS has started a trial-run of a weight loss programme replacing food with 800-calorie shakes and soups for 3 months, offered to people with newly diagnosed type 2 diabetes (T2D) to lose weight and put their diabetes into remission. Some healthcare professionals and charities are sceptical about the programme's effect on people's mental health. They fear it may trigger people to have a negative relationship with food (disordered eating). Some studies show indirectly that these programmes are somewhat safe; however it is not known for sure if it could affect people's relationship with food for the worse. Investigators will invite 56 people with T2D and disordered eating (picked up by questionnaires they will fill in) to participate in a trial. Of these participants, 28 will get TDR and the rest will get their standard care. Investigators will then measure how their scores of disordered eating change at 1, 3, 4, 6, 12 and 24 months. Investigators also plan to analyse the recorded sessions to better understand participants' experiences using TDR and their thoughts about eating and body image. This study will help shed light on how safe this type of diet is for people with disordered eating. It may lead to screening for eating disorders if TDR becomes standard care. If concerns are unfounded, it can reassure people with type 2 diabetes and healthcare professionals.

Inclusion Criteria:

• Willing and able to give informed consent, can communicate in English
• Age between 18 and 65 years inclusive
• Live in England
• BMI ≥27 kg/m2 or higher (adjusted to ≥25 kg/m2 for people of Black, Asian and minority ethnic origin)
• Diagnosed with T2D within 6 years prior to the day of screening
• Global EDE-Q scores ≥2.67
• To have an HbA1c taken in the last 12 months
• Latest HbA1c ≥43 mmol/mol (6.1%) and ≤87 mmol/mol (10%) if on diabetes medication OR latest hbA1c ≥ 48 mmol/mol (6.5%) and ≤87 mmol/mol (10%), if not on diabetes medication.
• If diagnosed with type 2 diabetes more than 1 year since day of screening, the participant must have attended their GP surgery for monitoring/ diabetes review when last offered
• Commit to continue annual reviews with their GP, even if T2D remission is achieved

Exclusion Criteria:

The participant must not enter the study if ANY of the following apply:

• Current or previous clinical diagnosis of an eating disorder
• Combination of EDE-Q ≥4 AND a CIA score ≥16 at screening
• Currently participating in a structured weight loss programme or self-reporting that they have lost >10% of their body weight in the last 3 months
• Insulin use
• Known kidney disease of stage 3/4/5 or eGFR <60 mls/min/1.73 m2 within the last 12 months
• Active substance use disorder
• Active cancer other than skin cancer
• Known proliferative retinopathy that has not been treated
• Porphyria
• Undergone or is awaiting bariatric surgery
• Myocardial infarction or stroke within previous 6 months
• Severe heart failure defined as equivalent to the New York Heart Association (NYHA) grade 3 or 4
• Active liver disease (not including non-alcoholic fatty liver disease)
• Pregnant, breastfeeding, or planning to become pregnant during the course of the study
• Soy or milk or fish allergy, lactose intolerance, or following a vegan diet
• People currently participating in another study or clinical trial of a CTIMP/non-CTIMP

An independent member of the research team, who will not be delivering the intervention or conducting baseline or follow-up assessments, will do the randomisation by inputting the data onto the software. The research team will then notify the participant of their allocation group and further steps.

Due to the nature of the intervention and the control group, blinding of the participants and the researchers to the intervention is not possible. However, the assessment of the primary outcome will be blinded, as questionnaires will be self-administered online. The PIS will present the study to participants in a more general context, focusing on the impact of total diet replacement on well-being, without mentioning disordered eating, as this could predispose participants in their questionnaire answers and heavily affect the outcomes of this study, introducing measurement bias and weakening the conclusions drawn.