SGLT2i treatment is associated with weight and Hba1c reduction. Previous work has demonstrated clinical predictors of HbA1c and weight reduction with canagliflozin. We aim to analyse predictors of HbA1c and weight change across the class.

Methods: People with type 2 diabetes treated with any SGLT2i in the ABCD nationwide audit and available baseline and follow-up data were included. Multiple linear regression model was used to assess independent predictors of Hba1c and weight change. Analysis was performed using Stata 16.

Results: Data for 275individuals were included (n=13925 Empagliflozin, n=106Dapagliflozin, n=2976 Canagliflozin) . Baseline (mean±SD) were: age 59.8±10.5 years, baseline Hba1c 9.1±1.6%, weight 97.7±22.2kg, BMI 33.8±6.9kg/m2 and eGFR 81.8±13.6. Median (IQR) diabetes duration was 8.3 years (4.4-12.7) and follow-up time 1.5 years (0.7-2.7) . 61.3% were men. HbA1c fell by 0.87% (95%CI 0.85-0.89; p<0.01) . Factors associated with greater Hba1c reductions include: older age (ß=0.01, p<0.01) ; shorter diabetes duration (ß=-0.01, p<0.01) ; higher HbA1c (ß=0.65, p<0.01) ; greater weight loss (ß=0.02, p<0.01) ; higher alanine aminotransferase (ALT) (ß=0.005, p<0.01) ; higher baseline eGFR (ß=0.002, p<0.01) and canagliflozin or empagliflozin use (vs. dapagliflozin; p<0.05) . Weight fell by 3.0kg (95%CI 2.2-3.8; p<0.01) . Predictors of greater weight reductions include: older age (ß=0.05, p<0.01) ; lower HbA1c (ß=-0.29, p<0.01) ; higher weight (ß=0.03, p<0.01) or BMI (ß=0.17, p<0.01) and larger HbA1c reductions (ß=0.41, p<0.01) . Empagliflozin was superior to dapagliflozin (p=0.01) - no other differences between drugs were noted.

Conclusion: Our models demonstrate that baseline characteristics can help predict clinical outcomes with SGLT2i. Individual patient levels factors appear to provide only weak predictive value but may still be useful in assisting therapeutic choices.


T.S.J. Crabtree: Other Relationship; Abbott Diabetes, Novo Nordisk, Sanofi. A. Gallagher: Other Relationship; Dexcom, Inc. S. Sivappriyan: None. K. Dhatariya: Advisory Panel; AstraZeneca, Boehringer Ingelheim International GmbH. R.P. Raghavan: Speaker's Bureau; Novo Nordisk. Other Relationship; Lilly Diabetes, Novo Nordisk. A. Bickerton: None. J. Elliott: Advisory Panel; Abbott. Speaker's Bureau; Abbott, Dexcom, Inc., Insulet Corporation, Novo Nordisk. G. Rayman: None. I.W. Gallen: None. I.R. Idris: None. R.E. Ryder: None.


The ABCD SGLT2 audits are funded by unrestricted grants from Boehringer Ingelheim, AstraZeneca and Napp Pharmaceuticals

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at