BMJ study compares kidney safety of GLP-1 and SGLT-2 drugs in type 2 diabetes: what Canadian patients on weight-loss medications should know

This article is for informational purposes only and is not medical advice. Consult your Canadian healthcare provider about your situation.

A September 2026 BMJ study of 75,455 adults with type 2 diabetes found that GLP-1 receptor agonists (such as semaglutide, sold in Canada as Ozempic and Wegovy by Novo Nordisk, and dulaglutide, sold as Trulicity) and SGLT-2 inhibitors (such as empagliflozin, sold as Jardiance, and dapagliflozin, sold as Forxiga) reduced the five-year risk of serious kidney deterioration by roughly 40% compared with DPP-4 inhibitors — but only among patients who had albuminuria at the start of treatment. Among the much larger group without albuminuria (about 82% of the cohort), no statistically meaningful kidney benefit appeared over the study period. For the roughly 3.7 million Canadians living with type 2 diabetes, this distinction matters: a urine albumin test that many patients may never have had could determine whether their second-line diabetes drug is actually protecting their kidneys.

Albuminuria is the presence of excess albumin protein in the urine, a sign that the kidney's filtering system is under stress. It is measured using an albumin-to-creatinine ratio (ACR); a ratio at or above 30 mg/g is the threshold the study used to define albuminuria. Diabetes is the leading cause of chronic kidney disease in Canada, and kidney failure is one of the most serious long-term complications of poorly managed type 2 diabetes. The question of which second-line drug best protects the kidneys — after metformin, the standard first-line treatment — has significant consequences for patients and for provincial drug formularies.

What this means in Canada

All of the GLP-1 receptor agonists and SGLT-2 inhibitors studied are approved by Health Canada and are available by prescription across the country. Semaglutide (Ozempic for diabetes, Wegovy for weight management), dulaglutide (Trulicity), empagliflozin (Jardiance), and dapagliflozin (Forxiga) are all on the Canadian market. Coverage varies by province. In Ontario, OHIP+ does not cover these drugs for adults, but the Ontario Drug Benefit (ODB) formulary lists empagliflozin and dapagliflozin for patients with established cardiovascular disease or chronic kidney disease meeting specific criteria. Quebec's RAMQ covers some SGLT-2 inhibitors under similar conditions. British Columbia's PharmaCare and Alberta's AHCIP have their own listing criteria, generally tied to cardiovascular or renal risk. GLP-1 receptor agonists face more restricted formulary access across most provinces, often requiring prior authorization and documented cardiovascular risk.

Health Canada has not issued specific guidance in response to this study, which is expected given it was published September 16, 2026. The SOGC (Society of Obstetricians and Gynaecologists of Canada) does not have a direct mandate over diabetes drug selection, though it does address metabolic health in reproductive-age women. Diabetes Canada's 2023 clinical practice guidelines already recommend GLP-1 receptor agonists and SGLT-2 inhibitors preferentially for patients with established cardiovascular disease, heart failure, or chronic kidney disease — a position this new study reinforces and refines by specifying that albuminuria status, not just kidney disease diagnosis, may be the more precise predictor of who benefits.

Canadian telehealth platforms such as Felix, a Canadian online prescribing service, and Science & Humans (scienceandhumans.com), a Canadian men's and metabolic health platform, prescribe GLP-1 receptor agonists for eligible patients. Cleo, a Canadian women's health platform, and Maple, a Canadian virtual care service, also connect patients with physicians who can prescribe these drugs. None of these platforms replace the specialist nephrology or endocrinology assessment that patients with established kidney disease typically need.

What changed

Previous randomised trials that established the kidney-protective reputation of GLP-1 and SGLT-2 drugs enrolled patients who were disproportionately high-risk: many had existing cardiovascular disease, and most had albuminuria. The CREDENCE and DAPA-CKD trials, for example, enrolled only patients with albuminuria. The FLOW trial, which tested semaglutide specifically for kidney outcomes, had 97% of participants with albuminuria. When those results were published, the natural assumption was that the benefit extended broadly to people with type 2 diabetes.

The BESTMED study, published September 16, 2026 in the BMJ, is the first large observational analysis to directly compare outcomes in patients with and without albuminuria using a target trial emulation framework — a method designed to approximate what a randomised trial would show, using real-world electronic health records and insurance claims from 10 US health systems and two national insurers. The researchers used DPP-4 inhibitors as the comparison group because randomised trials have consistently shown those drugs have no effect on kidney disease progression, making them a neutral benchmark.

Among the 13,872 patients with albuminuria, the five-year risk of serious kidney deterioration (defined as a doubling of serum creatinine or an estimated glomerular filtration rate dropping below 15 mL/min/1.73 m2) was 3.2% for those on GLP-1 or SGLT-2 drugs versus 5.4% for those on DPP-4 inhibitors — a risk ratio of 0.60, meaning roughly 40% lower risk. Among the 61,583 patients without albuminuria, the five-year risk was 2.4% versus 2.1%, a difference that was not statistically significant (risk ratio 1.10, 95% confidence interval 0.90 to 1.38).

A secondary finding drew attention to sulfonylureas (glimepiride, glyburide, glipizide), an older and cheaper drug class still widely used in Canada. In patients without albuminuria, sulfonylurea users had a 31% higher five-year risk of kidney deterioration compared with DPP-4 inhibitor users (risk ratio 1.31, 95% CI 1.11 to 1.58). This signal has appeared in prior observational work, including a 2023 CMAJ Open study, but the BESTMED analysis is the first to isolate it specifically in patients without albuminuria.

What Canadian patients should know

If you have type 2 diabetes and are taking or considering a GLP-1 receptor agonist or SGLT-2 inhibitor, ask your family doctor or endocrinologist whether you have had a urine ACR test. This is a routine urine test, not a specialist procedure, and it is covered under provincial health insurance across Canada. The result tells you whether albumin is spilling into your urine at levels that suggest early kidney stress.

If your ACR is 30 mg/g or higher, this study adds to a substantial body of evidence that GLP-1 and SGLT-2 drugs are likely protecting your kidneys, not just managing your blood sugar or weight. That is a meaningful reason to prioritize access to these medications, including pursuing provincial formulary coverage if your drug plan has not yet approved them.

If your ACR is below 30 mg/g, the kidney-protection argument for these drugs is weaker based on current evidence. That does not mean the drugs are wrong for you — they have cardiovascular benefits and, for GLP-1 receptor agonists, weight management benefits that may be independently relevant — but kidney protection alone is not a strong justification at this stage.

If you are currently on a sulfonylurea and have not had a recent urine ACR test, this study is a reason to ask for one. The signal linking sulfonylureas to faster kidney decline in patients without albuminuria is observational and not definitive, but it is consistent enough to warrant a conversation with your prescriber.

Provincial coverage differences are real. Patients in Ontario with an ODB card who have documented chronic kidney disease may qualify for SGLT-2 inhibitor coverage; those without that documentation may face out-of-pocket costs that can reach CAD $150 to $250 per month for branded SGLT-2 inhibitors, and higher for GLP-1 receptor agonists. Generic empagliflozin is not yet available in Canada as of September 2026.

Limitations and open questions

The BESTMED study is observational, not a randomised trial. Despite the researchers' use of target trial emulation and careful confounding adjustment, unmeasured factors could still explain some of the results. The study population was drawn entirely from US health systems and insurers, and the authors acknowledge that findings may not generalize to populations with different healthcare access or different racial and ethnic distributions. Canada's population, including its Indigenous communities who face disproportionately high rates of type 2 diabetes and kidney disease, was not represented.

The median follow-up was only 32 months, and most GLP-1 and SGLT-2 users had even shorter follow-up (median 23 months). Whether the absence of kidney benefit in patients without albuminuria holds over five or ten years is unknown. The study also could not separate GLP-1 receptor agonists from SGLT-2 inhibitors in its primary analysis; confidence intervals for the two drug classes overlapped substantially, so it is not possible to say from this data whether one class is better than the other for kidney protection.

Health Canada has not issued updated prescribing guidance based on this study. Diabetes Canada's clinical practice guidelines were last updated in 2023 and do not yet reflect this specific albuminuria-stratified finding. The SOGC has not issued a position statement on this topic. Whether Canadian provincial formularies will update their coverage criteria to incorporate albuminuria status more explicitly is not yet known.

For patients without albuminuria who are already on GLP-1 or SGLT-2 drugs for cardiovascular or weight-related reasons, this study is not a reason to stop. It is a reason to have a more precise conversation with your prescriber about what the drug is and is not doing for your kidneys specifically.

This article is for informational purposes only and is not medical advice. Consult your Canadian healthcare provider about your situation.

Editorial note

Hormone Journal articles are written by our editorial team and reviewed against published clinical guidelines, with a focus on Canadian patient access. We do not promote specific clinics or providers.