Effectiveness and care continuity in an app‐based, glucagon‐like peptide‐1 receptor agonist‐supported weight‐loss service for women with overweight and obesity in the UK : A real‐world retrospective cohort analysis

In highly adherent patients, real-world weight reduction outcomes approach those seen in randomized controlled trials.

“The observed weight reduction in clinical practice overall tends to be lower than in randomized controlled trials; however, outcomes approach those seen in trials when focusing on highly adherent patients.”

The weight-loss effectiveness of GLP-1RA was similar to what has been observed in trials when focusing on the subgroup of adherent patients who consistently took their prescribed GLP-1RA.

“Importantly, the weight‐loss effectiveness of GLP‐1RA was similar to what has been observed in trials when focusing on the subgroup of adherent patients who consistently took their prescribed GLP‐1RA, for example, at 12 months follow‐up, 50 , 53 , 55 , 56 , 57 , 58 , 61 whereas studies reporting results in an ‘as‐started’ approach found considerably lower effectiveness. 52 , 59 , 63 , 66”

Selection bias is a frequent concern in real-world studies of GLP-1RA that only include 'drug survivors' (those retaining treatment over a prolonged period) or that restrict analyses to high-dose users.

“Areas of particular focus are how to avoid selection bias, which is a frequent concern in studies that only include ‘drug survivors’, 53 , 55 , 58 , 60 , 61 , 62 , 65 , 67 , 70 , 71 that is, those retaining treatment over a prolonged period, or those restricting their analyses to high‐dose users. 57”