Each study, what it measured, what it found, and where to read it. Every entry is marked with who paid for it.
A network meta-analysis of liraglutide, semaglutide and tirzepatide at approved weight-management doses, run outside the manufacturers.
Adults with a BMI of 30 or above, or 27 with a weight-related condition and no diabetes, randomised to weekly semaglutide or placebo alongside lifestyle counselling.
Adults with obesity, or overweight with a related condition, randomised to 5mg, 10mg or 15mg weekly tirzepatide or placebo.
A network meta-analysis comparing tirzepatide and semaglutide across the trial literature rather than relying on the single head-to-head study.
A systematic review restricted to studies comparing tirzepatide and semaglutide head to head, covering five observational studies and two randomised trials, followed for six to twelve months.
The first direct comparison: adults with obesity and no type 2 diabetes, on the maximum tolerated dose of either tirzepatide (10 or 15mg) or semaglutide (1.7 or 2.4mg).
The registration trials measured people while they were taking the medicine. This is what follows, and the trial evidence and the real-world evidence do not fully agree.
An Oxford team systematically reviewed every study tracking adults who stopped a weight-management medicine, from older drugs like orlistat through to semaglutide and tirzepatide. Average time on treatment was 39 weeks, with 32 weeks of follow-up after stopping.
A meta-regression modelling the shape of regain rather than a monthly average, assuming rapid early regain that progressively slows.
Pooled weight change after stopping, separated by which medicine people had been on.
A Cleveland Clinic study of people who started semaglutide or tirzepatide and stopped within three to twelve months, following what they did next. Trials measure controlled withdrawal; this captures people restarting, switching, or changing approach.
An analysis of electronic health records covering people who took a GLP-1 for at least 90 days, lost at least five pounds, then stopped. Published by Epic Research, a health-data organisation analysing its own dataset, and not peer reviewed.
Adults 45 and over with existing cardiovascular disease and a BMI of 27 or above, but no diabetes. The question was whether semaglutide prevents cardiovascular events, not whether it reduces weight.
The prediabetic participants from SURMOUNT-1 continued on treatment for a further two years, testing whether weight loss held and whether diabetes was delayed.
Healthy female mice aged 20 months, roughly equivalent to a 60-year-old human, were given semaglutide until the end of life and compared with untreated mice and with a group fed 24% less to match the appetite suppression.
People taking these medicines often report that other cravings quieten too. The mechanism is plausible and well studied in animals, but human evidence is early: two randomised trials on alcohol, and observational data everywhere else. Nothing is approved for any of this.
A phase 2 randomised, double-blind trial at an academic medical centre, using low doses of 0.25mg and 0.5mg weekly. Participants had alcohol use disorder but were not seeking treatment for it. The primary measure was how much they drank in a laboratory setting before and after treatment.
A Danish trial of once-weekly semaglutide in treatment-seeking patients who had both alcohol use disorder and obesity, addressing the main limitation of the earlier trial.
A systematic review of preclinical and early clinical work on GLP-1 agonists across alcohol, nicotine, cocaine and opioid use disorders.
A survey of where the field stands, canvassing addiction specialists rather than reporting a single result.
Medsafe holds the approved data sheets and consumer information for every product sold here and reviews emerging safety signals. Pharmac decides what is publicly funded.