Semaglutide vs dulaglutide in UK care
Why this study matters
Randomised trials tell clinics whether a medicine can work under controlled conditions. They do not always show how it performs among patients encountered in routine care. A 2026 study in The Lancet Regional Health – Europe examined that gap by comparing once-weekly injectable semaglutide and dulaglutide in UK primary care.
The researchers used IQVIA Medical Research Data incorporating THIN records. They included adults with type 2 diabetes who started either medicine between January 2019 and December 2022 and followed records through June 2023. The study was an active-comparator, new-user cohort rather than a randomised trial. Statistical weighting was used to account for differences between treatment groups.
What the researchers reported
The cohort contained 6,616 new users. The main per-protocol analysis included 1,901 semaglutide users and 2,735 dulaglutide users who remained on treatment; 1,980 people, or 29.9% of the overall cohort, discontinued early.
At one year, the study estimated that semaglutide users had a 0.22 percentage-point greater reduction in HbA1c than dulaglutide users (95% confidence interval 0.15 to 0.30 percentage points). The estimated difference in bodyweight reduction was 1.92 kg in favour of semaglutide (95% CI 0.93 to 2.91 kg).
The authors also tested how closely routine-care patients resembled participants eligible for the earlier SUSTAIN-7 trial. They reported that 87.7% of patients in the per-protocol cohort would not have met that trial’s eligibility criteria. Semaglutide’s comparative effects were maintained in this larger, non-eligible group: the estimated differences were 0.23 percentage points for HbA1c and 2.01 kg for bodyweight.
Overall safety was reported as comparable between the medicines. However, people who discontinued early had gastrointestinal-event rates of 23.5–26.1 per 100 person-years, compared with 10.6–13.8 among those remaining on treatment. The early-discontinuation group also had smaller treatment effects.
How clinics should interpret it
This is useful UK real-world evidence, but it does not establish that semaglutide caused every observed difference. Treatment selection, adherence and unmeasured patient characteristics can affect observational comparisons even after statistical adjustment. The study also assessed adults treated for type 2 diabetes. Its results should not be presented as direct evidence for private weight-management patients without diabetes or for different semaglutide doses.
For clinic governance and compliance teams, the most relevant signal is not a treatment recommendation. It is the reported relationship between persistence, gastrointestinal events and outcomes. Records and patient communications should distinguish expected adverse effects, reasons for stopping and clinical outcomes rather than treating discontinuation as a single administrative measure.
The findings do not change MHRA product information or professional prescribing duties. The MHRA's current GLP-1 guidance stresses use for authorised indications under clinical supervision, while NICE guideline NG28 sets the wider treatment framework for adults with type 2 diabetes. This observational comparison can inform evidence reviews, but it does not override either source or establish that one medicine is universally “better” or safer.
Key takeaway
In this UK primary-care cohort, the researchers reported larger one-year HbA1c and bodyweight reductions with semaglutide than with dulaglutide and comparable safety overall. The high proportion of patients outside the original trial criteria strengthens the study’s relevance to routine diabetes care, while its observational design and diabetes-only population limit wider conclusions.