A study published on 16 October 2026 in Nature Medicine reports outcomes for 8,127 women in the UK who started GLP-1 treatment, almost all of them on tirzepatide, through a structured online programme. Average weight loss at six months was 16.1 kg, or 15.5% of starting weight. It is one of the largest prospective looks yet at obesity drugs delivered by a telehealth service. It also has real limits: the programme was co-run by two companies that employ two of the authors, there was no comparison group, and everyone paid privately.
What the study measured
The 8,127 women started treatment through a digital programme that the authors describe as including mandatory blood-test monitoring, weekly video-supported follow-up and clinician-guided dose adjustment; 99.8% were prescribed tirzepatide. The study followed them for six months. Mean weight loss was 16.1 kg, with a 95% confidence interval of 15.9 to 16.2 kg. An interval that narrow reflects the large sample: it says the average is pinned down precisely, not that most women lost close to 16 kg. Dividing 16.1 kg by 15.5% implies an average starting weight of roughly 104 kg.
The spread of results was wide. In the study, 83.1% of women lost at least 10% of their starting weight and 22.4% lost at least 20%, and the authors report similar responses across age, starting weight and socioeconomic groups. Among women with blood results at both time points, HbA1c fell by an average of 3.3 mmol/mol, about 0.3 percentage points on the scale used in the US. It returned to normal in 84.5% of those who started in the diabetes range and 96.3% of those who started in the prediabetes range. Total cholesterol fell by 0.77 mmol/L (about 30 mg/dL), LDL cholesterol by 0.61 mmol/L (about 24 mg/dL) and triglycerides by 0.47 mmol/L (about 42 mg/dL). Of the women who started, 91.4% were still in the programme at six months, and the authors report no serious treatment-related adverse events.
How it fits what came before
Most of what is known about tirzepatide’s effect on weight comes from randomised trials, where participants are screened, followed by study staff and measured at 72 weeks. Those trials are the reason the drug is approved, but they say little about whether the same results hold when the drug is prescribed through an app and a video call. That question matters because a large share of private GLP-1 prescribing now runs through online services of exactly this kind. Our provider reviews look at how those services differ.
What this study adds is evidence that one programme with specific safeguards, routine blood tests and weekly contact, kept more than nine in ten women engaged for six months and produced substantial average weight loss. It does not show that online prescribing in general produces these results. Services that skip blood tests or follow-up were not studied, and the paper does not test which of the programme’s features mattered. The six-month result also cannot be set directly against the 72-week trial figures, because weight loss on these drugs usually continues beyond six months. Our page on tirzepatide side effects covers what the trials recorded on safety.
What is not yet known
This was an observational study with no comparison group, so it cannot separate the drug’s effect from the programme’s, or either from the kind of person who signs up and pays. Everyone in it was a woman, in the UK, paying out of pocket, a group likely to be more motivated and better resourced than the wider population with obesity. Follow-up was six months, so weight regain and longer-term side effects are outside its scope. About 700 women (8.6%) had left by six months, and blood-marker results come only from those with paired tests. The programme was co-operated by eMed Healthcare UK and SheMed, two authors are their employees, and several authors report fees from GLP-1 manufacturers. The authors themselves flag the lack of a comparator, the self-pay population and the short follow-up. We read the published abstract; the full paper was not accessible to us, so details such as how adverse events were collected are not reported here.
Sources: Deanfield J, Wong JLC, Katsoulis M, et al. “Real-world, digital-first delivery of GLP-1 receptor agonist therapy in women.” Nature Medicine, published online 16 October 2026; PubMed record 42888249.
