This is an update of an existing systematic review, now covering 38 randomised trials and 25,816 adults with overweight or obesity but without diabetes, treated for at least 16 weeks. It adds 14 new trials and 11,000 participants to the previous version, which makes it the most current single answer to the question every clinician is now asked in every clinic.
Placebo-subtracted weight loss, at the top of each drug's range: liraglutide 5.8% (95% CI 3.6% to 8.0%), oral semaglutide 14.3% (11.4% to 17.2%), subcutaneous semaglutide 14.8% (13.4% to 16.2%), orforglipron 12.4% (9.7% to 15.1%), tirzepatide 19.0% (16.4% to 21.6%). Among agents not yet available, retatrutide reached 22.1% (19.3% to 24.9%) and amycretin 23.9% (18.5% to 29.3%). Head-to-head data favoured semaglutide over liraglutide, and tirzepatide and cagrilintide-semaglutide over semaglutide.
The safety picture is the part worth memorising, because it is what patients actually experience. Gastrointestinal adverse events occurred in 76.0% on active drug against 40.1% on placebo — that is most people, not a minority. Discontinuation because of adverse events was nonetheless low at 10.7% against 3.4%, and numerically higher with some oral agents. Serious adverse events were 6.5% against 5.2% and deaths were rare in both arms, with no new safety signals.
Two things follow for general practice. First, the differences between agents are now large enough to matter: liraglutide and tirzepatide are not interchangeable, and the gap between them is roughly threefold. Second, the arrival of effective oral agents — oral semaglutide at 14.3% and orforglipron at 12.4% — changes the access calculation in settings where cold chain, injection technique and needle cost are genuine barriers, which describes a great deal of Indian practice.
- Warn every patient that gastrointestinal side effects are the norm, not the exception — three in four.
- Most people who get side effects still continue; only about one in ten stops because of them.
- Choose the agent on expected effect size, not habit: the range across drugs is roughly threefold.
- Oral agents now achieve figures close to injectables and remove the cold-chain and injection barriers.
- The review found no new safety signals, but trials of this length cannot settle long-term safety.
The statistics, in plain English
Placebo-subtracted means the weight lost on placebo has already been taken off, so these are the drug's own contribution rather than the total a patient sees on the scale. The reviewers deliberately did not pool the trials into a single number, because the studies differed too much in population, dose and duration — so these are the best figures from individual trials rather than a combined estimate, and comparisons between drugs that were never tested against each other are indirect. The head-to-head results, where they exist, carry more weight than the ranking implied by the list.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for top clinical updates, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free