Cochrane review finds liraglutide likely aids 5% weight loss, but other evidence is uncertain

A Cochrane review funded by the World Health Organization concludes that liraglutide, a GLP-1 receptor agonist, likely helps more adults with obesity lose at least 5% of their body weight than placebo does, while rating the evidence on several other outcomes as limited, uncertain or showing little meaningful difference ^[1].

The review pooled 24 randomized controlled trials with 9,937 participants, ages 31.3 to 64.7, mostly from middle- and high-income countries. Searches ran through Dec. 17, 2024, and trials had to follow participants for at least six months ^[1]. Seventeen of the 24 studies focused on specific subgroups, including people with diabetes, non-alcoholic fatty liver disease, polycystic ovary syndrome and obstructive sleep apnea. Most studies compared liraglutide with placebo; two included semaglutide as an active comparator and one included orlistat, and two compared liraglutide with structured lifestyle-modification programs ^[1].

What the numbers showed

Over medium-term follow-up of 26 to 68 weeks, liraglutide roughly doubled the share of people reaching at least 5% weight loss compared with placebo (risk ratio 2.10, 95% confidence interval 1.80 to 2.45; 18 studies, 6,651 participants). The review rated that finding as moderate-certainty evidence ^[1]. Over longer follow-up of 104 to 162 weeks, the effect on that same 5% threshold was smaller but still present (risk ratio 1.78, 95% CI 1.51 to 2.09; two studies, 1,262 participants), also rated moderate certainty ^[1].

Results were less firm for average percentage weight change. Over the medium term, the mean difference from baseline was -4.72 percentage points (95% CI -5.32 to -4.12; 16 studies, 6,050 participants), but the review called that evidence very uncertain, largely because study results varied widely ^[1].

On safety, the review found low-certainty evidence that liraglutide may increase any adverse events (risk ratio 1.07, 95% CI 1.04 to 1.11; 16 studies, 8,147 participants) and serious adverse events (risk ratio 1.20, 95% CI 1.00 to 1.43; 20 studies, 8,487 participants) ^[1]. Evidence on withdrawals because of adverse events was very uncertain but pointed upward (risk ratio 1.98, 95% CI 1.30 to 3.02; 19 studies, 8,628 participants), with a similar signal in the long-term data ^[1].

For major adverse cardiovascular events, the review reported moderate-certainty evidence of little to no clinically meaningful difference versus placebo (risk ratio 0.86, 95% CI 0.66 to 1.10; six studies, 5,762 participants) ^[1]. Quality of life, measured by a physical-function scale, also showed little difference at medium term (mean difference 2.90, 95% CI -0.08 to 5.89; six studies, 3,733 participants), rated moderate certainty ^[1]. The evidence on mortality was very uncertain (risk ratio 0.44, 95% CI 0.08 to 2.25; 18 studies, 8,224 participants) ^[1].

The authors noted that 22 of the 24 trials were funded by the drug's manufacturer, with major manufacturer involvement in design, conduct, analysis or reporting in 13 of them, and said this raises concerns about potential conflicts of interest. They called for further independent, long-term studies ^[1].

Why it matters for patients

Liraglutide is an older GLP-1 medicine, and this review is one of three the Cochrane group is conducting on GLP-1 receptor agonists for adults with obesity, so it speaks to how much is actually known about a drug class many people are now considering ^[1]. The clearest signal is a threshold measure: whether someone loses at least 5% of body weight. For average weight change, heart events, quality of life and death, the review either found little meaningful difference or could not reach a confident conclusion ^[1].

The safety findings matter for day-to-day decisions. The review found low- or very-low-certainty evidence of more overall adverse events, more serious adverse events and roughly twice the rate of stopping treatment because of side effects, though confidence in those estimates is limited ^[1].

One limit readers should weigh: because 17 of the 24 trials enrolled specific subgroups, such as people with diabetes or sleep apnea, the average results may not describe every patient, and the review did not report results by age, sex or other personal factors ^[1].

Whether liraglutide's benefits hold beyond the study periods examined has not been established. The review's own conclusion is that the medium- or long-term impact on percentage weight change, cardiovascular events, quality of life and mortality may be limited or uncertain ^[1].

It also is not yet known how these findings compare with newer GLP-1 medicines. Only two of the included trials used semaglutide as an active comparator, and the review reported no head-to-head conclusions from them ^[1].

The review was published by the Cochrane Collaboration with funding from the World Health Organization, under original protocol DOI 10.1002/14651858.CD015092 and updated protocol PROSPERO CRD420250654193 ^[1].