Diabetes Drug Class May Affect COVID-19 Outcomes

Nurse uses glaucometer to check a female patient’s blood sugar leve.
Researchers sought to identify which medications for type 2 diabetes may be associated with severity of COVID-19.

Premorbid use of glucagon-like peptide-1 receptor agonist (GLP1-RA) or sodium-glucose cotransporter-2 inhibitor (SGLT2i) therapy was found to be associated with decreased odds of mortality and adverse outcomes following a COVID-19 diagnosis in people with type 2 diabetes, according to research results published in Diabetes Care.

Using data from a cohort study of COVID-19 data from the National COVID Cohort Collaborative (N3C), researchers sought to determine how the use of new antihyperglycemic medications might be associated with COVID-19 severity. In particular, the investigators set out to evaluate the association of premorbid use of GLP1-RA and SGLT2i with COVID-19 outcomes compared with a comparator group of patients using dipeptidyl peptidase-4 inhibitors (DPP4i).

Patients in the N3C cohort had any encounter after January 2020 and 1 or more SARS-CoV-2 laboratory tests. The N3C data set included electronic health record (EHR) data beginning on January 1, 2018. Participants were adults aged 18 and older in 2020 with a positive SARS-CoV-2 polymerase chain reaction (PCR) test and at least 1 ambulatory prescription of a GLP1-RA, SLGT2i, or DPP4i in the 24 months preceding the positive SARS-CoV-2 PCR test.

The primary study outcome was mortality within 60 days of any positive SARS-CoV-2 PCR test. Secondary study outcomes included mortality during any time after the index date (total mortality) and emergency department visits, hospitalization, and mechanical ventilation (such as intubation or ventilation) within 14 days of any positive SARS-CoV-2 PCR test.

By February 23, 2021, data from 3,453,824 adults from 42 contributing sites were available in the N3C database; of these, 629,242 had COVID-19. A total of 12,446 individuals (62.5% White; 53.4% women; mean age, 58.6±13.1 years) from 35 contributing sites were eligible for study inclusion.

Patients in the DPP4i arm were generally older and had a lower body mass index (BMI) compared with those in the GLP1-RA and SGLT2i subgroups (64 vs 56 and 58 years; BMI, 33 kg/m2 vs 37 kg/m2 and 35 kg/m2). More patients in the DPP4i group vs the other groups had chronic kidney or end-stage renal disease, myocardial infarction, congestive heart failure, cancer, dementia, or stroke. The researchers reported a slightly lower incidence of insulin use in the DPP4i group.

The crude 60-day mortality rate in the total study population was 3.11%, and this differed by class of premorbid medication use: 2.06% and 2.32% used GLP1-RA and SGLT2i, respectively, vs 5.69% who used DPP4i. The total mortality rate over the entire observation period was 2.29%, 2.48%, and 6.18% for each medication class, respectively.

After propensity score weighting, 60-day and total mortality in the GLP1-RA group was 2.31% and 2.58% vs 4.86% and 5.33% in the DPP4i group; similarly, these values were 2.70% and 2.87% for the SGLT2i group vs 4.74% and 5.18% for those taking DPP4i.

Results of a targeted maximum likelihood estimation analysis showed that GLP1-RA users had lower odds of 60-day mortality vs DPP4i users (odds ratio [OR], 0.54; 95% CI, 0.37-0.80) with an estimated risk difference of -0.020 (95% CI, -0.035 to -0.0044), reflecting a total of 2 fewer deaths per 100 COVID-19 cases.

The use of GLP1-RA medications was also associated with lower odds of total mortality vs DPP4i use (OR, 0.56; 95% CI, 0.39-0.92) and emergency department visits, hospitalization, and mechanical ventilation (ORs, 0.81, 0.73, and 0.73, respectively) within 14 days of a COVID-19 diagnosis.

Use of SGLT2i also showed lower odds of 60-day and total mortality (ORs, 0.66 and 0.63; 95% CI, 0.50-0.86 and 0.49-0.82), for a total of 1.6 fewer deaths per 100 COVID-19 cases. Estimated risk difference in 60-day mortality was -0.016 (95% CI, -0.026 to -0.0057). Lower odds were also noted for emergency department visits and hospitalization within 14 days of a COVID-19 diagnosis (OR, 0.90 and 0.91), but the odds of mechanical ventilation were not significantly different.

The main limitation of the study, according to the researchers, involved comparison by prevalent drug prescribing rather than by drug initiation. Other limitations included the large differences in characteristics across all medication groups, potential residual confounding due to severity of comorbidities, and bias stemming from factors that were, according to the researchers, difficult to measure. The researchers noted in particular that GLP1-RA are more expensive than either of the other medications studied, and data on socioeconomic status were not available, which the researchers acknowledged was a significant limitation of the data set.

“This study provides evidence for antihyperglycemic medication class-based differences in COVID-19 outcomes, where premorbid GLP1-RA or SGLT2i prescribing is associated with lower mortality and other adverse clinical outcomes in the setting of a COVID-19 diagnosis as compared with DPP4i prescribing,” the researchers concluded.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Kahkoska AR, Abrahamsen TJ, Alexander GC, et al; for the N3C Consortium. Association between glucagon-like peptide 1 receptor agonist and sodium-glucose cotransporter 2 inhibitor use and COVID-19 outcomes. Diabetes Care. 2021;44(7):1564-1572. doi:10.2337/dc21-0065