• Entire Site
  • Research & Funding
  • Health Information
  • About NIDDK
  • Research Programs & Contacts

Clinical Research in Type 2 Diabetes

Studies in humans aimed at the prevention, treatment, and diagnosis of Type 2 Diabetes and the mechanistic aspects of its etiology.

The Clinical Research in Type 2 Diabetes (T2D) program supports human studies across the lifespan aimed at understanding, preventing and treating T2D. This program includes clinical trials that test pharmacologic, behavioral, surgical or practice-level approaches to the treatment and/or prevention of T2D, including promoting the preservation of beta cell function. Studies may also advance the development of new surrogate markers for use in clinical trials. Studies can be designed to understand the pathophysiology of T2D, including the role of gestational diabetes and metabolic imprinting on the development of T2D, as well as factors influencing the response to treatment. The program also encompasses epidemiologic studies that improve our understanding of the natural history and pathogenesis of T2D, and the development of diagnostic criteria to distinguish type 1 and type 2 diabetes, especially in the pediatric population. The program also supports research to understand and test approaches to accelerate the translation of efficacious interventions into real-world practice and adoption; and to address health equity by reducing health disparities in the incidence and/or clinical outcomes of T2D.

NIDDK Program Staff

  • Shavon Artis Dickerson, Dr.P.H., M.P.H. Health Equity and Implementation Science
  • Henry B. Burch, M.D. Clinical studies utilizing existing digital health technology for the prevention and treatment of type 2 diabetes, clinical and basic science studies involving non-neoplastic disorders of the thyroid, clinical studies involving medical and novel dietary treatment of type 2 diabetes.
  • Maureen Monaghan Center, Ph.D., CDCES Health Psychology, Behavioral Science, Clinical Management of Diabetes
  • Jean M. Lawrence, Sc.D., M.P.H., M.S.S.A. Type 2 diabetes risk and prevention after gestational diabetes; Studies of adults with diabetes/pre-diabetes using secondary data and observational designs, and natural experiments
  • Hanyu Liang, M.D., Ph.D. Hepatic Metabolism; Insulin Resistance; Type 2 Diabetes; Obesity; Bariatric Surgery
  • Barbara Linder, M.D., Ph.D. Type 2 diabetes in children and youth; human studies of metabolic imprinting
  • Saul Malozowski, M.D., Ph.D., M.B.A. Neuroendocrinology of hypothalamic-pituitary axis, neuropeptide signaling and receptors; hormonal regulation of bone and mineral metabolism; HIV/AIDS-associated metabolic and endocrine dysfunction
  • Pamela L. Thornton, Ph.D. Health Equity and Translational Research; Centers for Diabetes Translation Research (P30) Program
  • Theresa Teslovich Woo, Ph.D. Human behavior, developmental cognitive neuroscience, and brain-based mechanisms involved in obesity and diabetes

Recent Funding Opportunities

Continuous ketone monitoring for the safe use of sodium-glucose cotransporter-2 inhibitors in type 1 diabetes (r01 clinical trial required), niddk high risk multi-center clinical study implementation planning cooperative agreements (u34 clinical trial optional), niddk high risk multi-center clinical study cooperative agreement (u01 clinical trial not allowed), niddk high risk multi-center clinical study cooperative agreement (u01 clinical trial required), stephen i. katz early stage investigator research project grant (r01 clinical trial not allowed), related links.

View related clinical trials from ClinicalTrials.gov.

Study sections conduct initial peer review of applications in a designated scientific area. Visit the NIH’s Center for Scientific Review website to search for study sections.

Research Resources

NIDDK makes publicly supported resources, data sets, and studies available to researchers to accelerate the rate and lower the cost of new discoveries.

  • Ancillary Studies to Major Ongoing Clinical Studies to extend our knowledge of the diseases being studied by the parent study investigators under a defined protocol or to study diseases and conditions not within the original scope of the parent study but within the mission of the NIDDK.
  • NIDDK Central Repository for access to clinical resources including data and biospecimens from NIDDK-funded studies.
  • NIDDK Information Network (dkNET) for simultaneous search of digital resources, including multiple datasets and biomedical resources relevant to the mission of the NIDDK.

Additional Research Programs

Research training.

NIDDK supports the training and career development of medical and graduate students, postdoctoral fellows, and physician scientists through institutional and individual grants.

Diversity Programs

The NIDDK offers and participates in a variety of opportunities for trainees and researchers from communities underrepresented in the biomedical research enterprise. These opportunities include travel and scholarship awards, research supplements, small clinical grants, high school and undergraduate programs, and a network of minority health research investigators.

Small Business

Small business programs.

NIDDK participates in the Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs. These programs support innovative research conducted by small businesses that has the potential for commercialization.

Human Subjects Research

NIDDK provides funding for pivotal clinical research, from preliminary clinical feasibility to large multi-center studies.

Translational Research

NIDDK provides funding opportunities and resources to encourage translation of basic discoveries into novel therapeutics.

Meetings & Workshops

dkNET slider card.

Supports researchers with tools to enhance scientific rigor, reproducibility, and transparency, and provides a big data knowledge base for genomic and pathway hypothesis generation.

A man presenting at an NIH seminar

Providing education and training for the next generation of biomedical and behavioral scientist


Stay informed about the latest events, or connect through social media.

View All Meetings

Learn about current projects and view funding opportunities sponsored by the NIH Common Fund .

Registration is required at eRA Commons and grants.gov and can take 4 weeks.

Clinical Trials

Type 2 diabetes.

Displaying 96 studies

The purpose of this study is to identify changes to the metabolome (range of chemicals produced in the body) and microbiome (intestine microbe environment) that are unique to Roux-en-Y gastric bypass surgery and assess the associated effect on the metabolism of patients with type 2 diabetes.

The purpose of this study is to evaluate the impact of a digital storytelling intervention derived through a community-based participatory research (CBPR) approach on type 2 diabetes mellitus (T2D) outcomes among Hispanic adults with poorly controlled type 2 diabetes mellitus (T2D) in primary care settings through a randomized clinical trial.

The purpose of this study is to assess the impact of a whole food plant-based diet on blood sugar control in diabetic patients versus a control group on the American Diabetics Association diet before having a total hip, knee, or shoulder replacement surgery.

The purpose of  this study is to learn more about if the medication, Entresto, could help the function of the heart and kidneys.

The primary aim of this study is to compare the outcome measures of adult ECH type 2 diabetes patients who were referred to onsite pharmacist services for management of their diabetes to similar patients who were not referred for pharmacy service management of their diabetes. A secondary aim of the study is to assess the Kasson providers’ satisfaction level and estimated pharmacy service referral frequency to their patients. A tertiary aim of the study is to compare the hospitalization rates of type 2 diabetes rates who were referred to onsite pharmacist services for management of their diabetes to similar patients ...

To explore the feasibility of conducting a family centered wellness coaching program for patients at high risk for developing diabetes, in a primary care setting.

To determine engagement patterns.

To describe characteristics of families who are likely to participate.

To identify barriers/limitations to family centered wellness coaching.

To assess whether a family centered 8 week wellness coaching intervention for primary care patients at high risk for diabetes will improve self-care behaviors as measured by self-reported changes in physical activity level and food choices.

This study is being done to understand metformin's mechanisms of action regarding glucose production, protein metabolism, and mitochondrial function.

The purpose of this study is to assess the effectiveness of Revita® DMR for improving HbA1c to ≤ 7% without the need of insulin in subjects with T2D compared to sham and to assess the effectiveness of DMR versus Sham on improvement in Glycemic, Hepatic and Cardiovascular endpoints.

The purpose of this study is to evaluate 6 weeks of home use of the Control-IQ automated insulin delivery system in individuals with type 2 diabetes.

This study will evaluate whether bile acids are able to increase insulin sensitivity and enhance glycemic control in T2DM patients, as well as exploring the mechanisms that enhance glycemic control. These observations will provide the preliminary data for proposing future therapeutic as well as further mechanistic studies of the role of bile acids in the control of glycemia in T2DM.

The purpose of this study is to determine if Inpatient Stress Hyperglycemia is an indicator of future risk of developing type 2 Diabetes Mellitus.

The GRADE Study is a pragmatic, unmasked clinical trial that will compare commonly used diabetes medications, when combined with metformin, on glycemia-lowering effectiveness and patient-centered outcomes.

The purpose of this study is to assess the effectiveness of a digital storytelling intervention derived through a community based participatory research (CBPR) approach on self-management of type 2 diabetes (T2D) among Somali adults. 

The overall goal of this proposal is to determine the effects of acute hyperglycemia and its modulation by Glucagon-like Peptide-1 (GLP-1) on myocardial perfusion in type 2 diabetes (DM). This study plan utilizes myocardial contrast echocardiography (MCE) to explore a) the effects of acute hyperglycemia on myocardial perfusion and coronary flow reserve in individuals with and without DM; and b) the effects of GLP-1 on myocardial perfusion and coronary flow reserve during euglycemia and hyperglycemia in DM. The investigators will recruit individuals with and without DM matched for age, gender and degree of obesity. The investigators will measure myocardial perfusion ...

The purpose of this study is to test the hypothesis that patients with T2DM will have greater deterioration in BMSi and in cortical porosity over 3 yrs as compared to sex- and age-matched non-diabetic controls; and identify the circulating hormonal (e.g., estradiol [E2], testosterone [T]) and biochemical (e.g., bone turnover markers, AGEs) determinants of changes in these key parameters of bone quality, and evaluate the possible relationship between existing diabetic complications and skeletal deterioration over time in the T2DM patients.

The purpose of this study is to determine the effect of endogenous GLP-1 secretion on islet function in people with Typr 2 Diabetes Mellitus (T2DM).

GLP-1 is a hormone made by the body that promotes the production of insulin in response to eating. However, there is increasing evidence that this hormone might help support the body’s ability to produce insulin when diabetes develops. 

The purpose of this study is to assess whether psyllium is more effective in lowering fasting blood sugar and HbA1c, and to evaluate the effect of psyllium compared to wheat dextrin on the following laboratory markers:  LDL-C, inflammatory markers such as ceramides and hsCRP, and branch chain amino acids which predict Diabetes Mellitus (DM).

The purpose of this study is to evaluate if breathing pure oxygen overnight affects insulin sensitivity in participants with diabetes.   

The purpose of this study is to determine the impact of patient decision aids compared to usual care on measures of patient involvement in decision-making, diabetes care processes, medication adherence, glycemic and cardiovascular risk factor control, and use of resources in nonurban practices in the Midwestern United States.

This mixed methods study aims to answer the question: "What is the work of being a patient with type 2 diabetes mellitus?" .

The purpose of this study is to assess penile length pre- and post-completion of RestoreX® traction therapy compared to control groups (no treatment) among men with type II diabetes.

This observational study is conducted to determine how the duodenal layer thicknesses (mucosa, submucosa, and muscularis) vary with several factors in patients with and without type 2 diabetes.

This trial is a multi-center, adaptive, randomized, double-blind, placebo- and active- controlled, parallel group, phase 2 study in subjects with Type 2 Diabetes Mellitus to evaluate the effect of TTP399 on HbA1c following administration for 6 months.

The purpose of this study is to find the inheritable changes in genetic makeup that are related to the development of type 2 diabetes in Latino families.

The objective of this early feasibility study is to assess the feasibility and preliminary safety of the Endogenex Divice for endoscopic duodenal mucosal regeneration in patients with type 2 diabetes (T2D) inadequately controlled on 2-3 non-insulin glucose-lowering medications. 

The study is being undertaken to understand how a gastric bypass can affect a subject's diabetes even prior to their losing significant amounts of weight. The hypothesis of this study is that increased glucagon-like peptide-1 (GLP-1) secretion explains the amelioration in insulin secretion after Roux-en-Y Gastric Bypass (RYGB) surgery.

The purpose of this study is to estimate the risk of diabetes related complications after total pancreatectomy.  We will contact long term survivors after total pancreatectomy to obtain data regarding diabetes related end organ complications.

The purpose of this study is to understand nighttime glucose regulation in humans and find if the pattern is different in people with Type 2 diabetes

The study purpose is to understand patients’ with the diagnosis of Diabetes Mellitus type 1 or 2 perception of the care they receive in the Diabetes clinic or Diabetes technology clinic at Mayo Clinic and to explore and to identify the healthcare system components patients consider important to be part of the comprehensive regenerative care in the clinical setting.

However, before we can implement structural changes or design interventions to promote comprehensive regenerative care in clinical practice, we first need to characterize those regenerative practices occurring today, patients expectations, perceptions and experiences about comprehensive regenerative care and determine the ...

The investigators will determine whether people with high muscle mitochondrial capacity produce higher amount of reactive oxygen species (ROS) on consuming high fat /high glycemic diet and thus exhibit elevated cellular oxidative damage. The investigators previously found that Asian Indian immigrants have high mitochondrial capacity in spite of severe insulin resistance. Somalians are another new immigrant population with rapidly increasing prevalence of diabetes. Both of these groups traditionally consume low caloric density diets, and the investigators hypothesize that when these groups are exposed to high-calorie Western diets, they exhibit increased oxidative stress, oxidative damage, and insulin resistance. The investigators will ...

The purpose of this research is to find out how genetic variations in GLP1R, alters insulin secretion, in the fasting state and when blood sugars levels are elevated. Results from this study may help us identify therapies to prevent or reverse type 2 diabetes mellitus.

The purpose of this mixed-methods study is to deploy the tenets of Health and Wellness Coaching (HWC) through a program called BeWell360 model , tailored to the needs of Healthcare Workers (HCWs) as patients living with poorly-controlled Type 2 Diabetes (T2D). The objective of this study is to pilot-test this novel, scalable, and sustainable BeWell360 model that is embedded and integrated as part of primary care for Mayo Clinic Employees within Mayo Clinic Florida who are identified as patients li)ving with poorly-controlled T2D. 

It is unknown how patient preferences and values impact the comparative effectiveness of second-line medications for Type 2 diabetes (T2D). The purpose of this study is to elicit patient preferences toward various treatment outcomes (e.g., hospitalization, kidney disease) using a participatory ranking exercise, use these rankings to generate individually weighted composite outcomes, and estimate patient-centered treatment effects of four different second-line T2D medications that reflect the patient's value for each outcome. 

To determine if the EndoBarrier safely and effectively improves glycemic control in obese subjects with type 2 diabetes.

Can QBSAfe be implemented in a clinical practice setting and improve quality of life, reduce treatment burden and hypoglycemia among older, complex patients with type 2 diabetes?

Questionnaire administered to diabetic patients in primary care practice (La Crosse Mayo Family Medicine Residency /Family Health Clinic) to assess patient’s diabetic knowledge. Retrospective chart review will also be done to assess objective diabetic control based on most recent hemoglobin A1c.    

The purpose of this study is to assess key characteristics of bone quality, specifically material strength and porosity, in patients who have type 2 diabetes. These patients are at an unexplained increased risk for fractures and there is an urgent need to refine clinical assessment for this risk.

This research study is being done to develop educational materials that will help patients and clinicians talk about diabetes treatment and management options.

Muscle insulin resistance is a hallmark of upper body obesity (UBO) and Type 2 diabetes (T2DM). It is unknown whether muscle free fatty acid (FFA) availability or intramyocellular fatty acid trafficking is responsible for muscle insulin resistance, although it has been shown that raising FFA with Intralipid can cause muscle insulin resistance within 4 hours. We do not understand to what extent the incorporation of FFA into ceramides or diacylglycerols (DG) affect insulin signaling and muscle glucose uptake. We propose to alter the profile and concentrations of FFA of healthy, non-obese adults using an overnight, intra-duodenal palm oil infusion vs. ...

The objectives of this study are to identify circulating extracellular vesicle (EV)-derived protein and RNA signatures associated with Type 2 Diabetes (T2D), and to identify changes in circulating EV cargo in patients whose T2D resolves after sleeve gastrectomy (SG) or Roux-en-Y gastric bypass (RYGB).

Assessment of glucose metabolism and liver fat after 12 week dietary intervention in pre diabetes subjects. Subjects will be randomized to either high fat (olive oil supplemented),high carb/high fiber (beans supplemented) and high carb/low fiber diets. Glucose metabolism will be assessed by labeled oral glucose tolerance test and liver fat by magnetic resonance spectroscopy pre randomization and at 8 and 12 week after starting dietary intervention.

To study the effect of an ileocolonic formulation of ox bile extract on insulin sensitivity, postprandial glycemia and incretin levels, gastric emptying, body weight and fasting serum FGF-19 (fibroblast growth factor) levels in overweight or obese type 2 diabetic subjects on therapy with DPP4 (dipeptidyl peptidase-4) inhibitors (e.g. sitagliptin) alone or in combination with metformin.

The purpose of this study is to evaluate whether or not a 6 month supply (1 meal//day) of healthy food choices readily available in the patient's home and self management training including understanding of how foods impact diabetes, improved food choices and how to prepare those foods, improve glucose control.  In addition, it will evaluate whether or not there will be lasting behavior change modification after the program.

The purpose of this study is to learn more about how the body stores dietary fat. Medical research has shown that fat stored in different parts of the body can affect the risk for diabetes, heart disease and other major health conditions.

The purpose of this study is to see why the ability of fat cells to respond to insulin is different depending on body shape and how fat tissue inflammation is involved.

The purpose of this study is to determine the mechanism(s) by which common bariatric surgical procedures alter carbohydrate metabolism. Understanding these mechanisms may ultimately lead to the development of new interventions for the prevention and treatment of type 2 diabetes and obesity.

The purpose of this study is to compare the rate of progression from prediabetes at 4 months to frank diabetes at 12 months (as defined by increase in HbA1C or fasting BS to diabetic range based on the ADA criteria) after transplantation in kidney transplant recipients on Exenatide SR + SOC vs. standard-of-care alone.

The purpose of this study evaluates a subset of people with isolated Impaired Fasting Glucose with Normal Glucose Tolerance (i.e., IFG/NGT) believed to have normal β-cell function in response to a glucose challenge, suggesting that – at least in this subset of prediabetes – fasting glucose is regulated independently of glucose in the postprandial period. To some extent this is borne out by genetic association studies which have identified loci that affect fasting glucose but not glucose tolerance and vice-versa.

A research study to enhance clinical discussion between patients and pharmacists using a shared decision making tool for type 2 diabetes or usual care.

While the potential clinical uses of pulsed electromagnetic field therapy (PEMF) are extensive, we are focusing on the potential benefits of PEMF on vascular health. We are targeting, the pre diabetic - metabolic syndrome population, a group with high prevalence in the American population. This population tends to be overweight, low fitness, high blood pressure, high triglycerides and borderline high blood glucose.

The purpose of this study is to evaluate the effects of improving glycemic control, and/or reducing glycemic variability on gastric emptying, intestinal barrier function, autonomic nerve functions, and epigenetic changes in subjects with type 1 diabetes mellitus (T1DM) and  type 2 diabetes mellitus (T2DM) who are switched to intensive insulin therapy as part of clinical practice.

This study is designed to compare an intensive lifestyle and activity coaching program ("Sessions") to usual care for diabetic patients who are sedentary. The question to be answered is whether the Sessions program improves clinical or patient centric outcomes. Recruitment is through invitiation only.

This is a study to evaluate a new Point of Care test for blood glucose monitoring.

This protocol is being conducted to determine the mechanisms responsible for insulin resistance, obesity and type 2 diabetes.

The purpose of this study is to determine the metabolic effects of Colesevelam, particularly for the ability to lower blood sugar after a meal in type 2 diabetics, in order to develop a better understanding of it's potential role in the treatment of obesity.

The purpose of this study is to test whether markers of cellular aging and the SASP are elevated in subjects with obesity and further increased in patients with obesity and Type 2 Diabetes Mellitus (T2DM) and to relate markers of cellular aging (senescence) and the SASP to skeletal parameters (DXA, HRpQCT, bone turnover markers) in each of these groups.

Integration of Diabetes Prevention Program (DPP) and Diabetes Self Management Program (DSMP) into WellConnect.

The purpose of this study is to assess the effects of a nighttime rise in cortisol on the body's glucose production in type 2 diabetes.

The goal of this study is to evaluate a new format for delivery of a culturally tailored digital storytelling intervention by incorporating a facilitated group discussion following the videos, for management of type II diabetes in Latino communities.

Using stem cell derived intestinal epithelial cultures (enteroids) derived from obese (BMI> 30) patients and non-obese and metabolically normal patients (either post-bariatric surgery (BS) or BS-naïve with BMI < 25), dietary glucose absorption was measured. We identified that enteroids from obese patients were characterized by glucose hyper-absorption (~ 5 fold) compared to non-obese patients. Significant upregulation of major intestinal sugar transporters, including SGLT1, GLU2 and GLUT5 was responsible for hyper-absorptive phenotype and their pharmacologic inhibition significantly decreased glucose absorption. Importantly, we observed that enteroids from post-BS non-obese patients exhibited low dietary glucose absorption, indicating that altered glucose absorption ...

The purpose of this study is to improve our understanding of why gastrointestinal symptoms occur in diabetes mellitus patients and identify new treatment(s) in the future.  

These symptoms are often distressing and may impair glycemic control. We do not understand how diabetes mellitus affects the GI tracy. In 45 patients undergoing sleeve gastrectomy, we plan to compare the cellular composition of circulating peripheral mononuclear cells, stomach immune cells, and interstitial cells of Cajal in the stomach. 

Muscle insulin resistance is a hallmark of upper body obesity (UBO) and Type 2 diabetes (T2DM). It is unknown whether muscle free fatty acid (FFA) availability or intramyocellular fatty acid trafficking is responsible for the abnormal response to insulin. Likewise, we do not understand to what extent the incorporation of FFA into ceramides or diacylglycerols (DG) affect insulin signaling and muscle glucose uptake. We will measure muscle FFA storage into intramyocellular triglyceride, intramyocellular fatty acid trafficking, activation of the insulin signaling pathway and glucose disposal rates under both saline control (high overnight FFA) and after an overnight infusion of intravenous ...

The goal of this study is to evaluate the presence of podocytes (special cells in the kidney that prevent protein loss) in the urine in patients with diabetes or glomerulonephritis (inflammation in the kidneys). Loss of podocyte in the urine may be an earlier sign of kidney injury (before protein loss) and the goal of this study is to evaluate the association between protein in the urine and podocytes in the urine.

Muscle insulin resistance is a hallmark of upper body obesity (UBO) and Type 2 diabetes (T2DM), whereas lower body obesity (LBO) is characterized by near-normal insulin sensitivity. It is unknown whether muscle free fatty acid (FFA) availability or intramyocellular fatty acid trafficking differs between different obesity phenotypes. Likewise, we do not understand to what extent the incorporation of FFA into ceramides or diacylglycerols (DG) affect insulin signaling and muscle glucose uptake. By measuring muscle FFA storage into intramyocellular triglyceride, intramyocellular fatty acid trafficking, activation of the insulin signaling pathway and glucose disposal rates we will provide the first integrated examination ...

The purpose of this study is to evaluate the effects of multiple dose regimens of RM-131 on vomiting episodes, stomach emptying and stomach paralysis symptoms in patients with Type 1 and Type 2 diabetes and gastroparesis.

The purpose of this study is to create a prospective cohort of subjects with increased probability of being diagnosed with pancreatic cancer and then screen this cohort for pancreatic cancer

The purpose of this study is assess the feasibility, effectiveness, and acceptability of Diabetes-REM (Rescue, Engagement, and Management), a comprehensive community paramedic (CP) program to improve diabetes self-management among adults in Southeast Minnesota (SEMN) treated for servere hypoglycemia by the Mayo Clinic Ambulance Services (MCAS).

The purpose of this study is to determine if a blood test called "pancreatic polypeptide" can help distinguish between patients with diabetes mellitus with and without pancreatic cancer.

The purpose of this study is to evaluate the effectiveness and safety of brolucizumab vs. aflibercept in the treatment of patients with visual impairment due to diabetic macular edema (DME).

Women with gestational diabetes mellitus (GDM) are likely to have insulin resistance that persists long after pregnancy, resulting in greater risk of developing type 2 diabetes mellitus (T2DM). The study will compare women with and without a previous diagnosis of GDM to determine if women with a history of GDM have abnormal fatty acid metabolism, specifically impaired adipose tissue lipolysis. The study will aim to determine whether women with a history of GDM have impaired pancreatic β-cell function. The study will determine whether women with a history of GDM have tissue specific defects in insulin action, and also identify the effect of a ...

Although vitreous hemorrhage (VH) from proliferative diabetic retinopathy (PDR) can cause acute and dramatic vision loss for patients with diabetes, there is no current, evidence-based clinical guidance as to what treatment method is most likely to provide the best visual outcomes once intervention is desired. Intravitreous anti-vascular endothelial growth factor (anti-VEGF) therapy alone or vitrectomy combined with intraoperative PRP each provide the opportunity to stabilize or regress retinal neovascularization. However, clinical trials are lacking to elucidate the relative time frame of visual recovery or final visual outcome in prompt vitrectomy compared with initial anti-VEGF treatment. The Diabetic Retinopathy Clinical Research ...

The purpose of this study is to demonstrate feasibility of dynamic 11C-ER176 PET imaging to identify macrophage-driven immune dysregulation in gastric muscle of patients with DG. Non-invasive quantitative assessment with PET can significantly add to our diagnostic armamentarium for patients with diabetic gastroenteropathy.

The purpose of this study is to assess the safety and tolerability of intra-arterially delivered mesenchymal stem/stromal cells (MSC) to a single kidney in one of two fixed doses at two time points in patients with progressive diabetic kidney disease. 

Diabetic kidney disease, also known as diabetic nephropathy, is the most common cause of chronic kidney disease and end-stage kidney failure requiring dialysis or kidney transplantation.  Regenerative, cell-based therapy applying MSCs holds promise to delay the progression of kidney disease in individuals with diabetes mellitus.  Our clinical trial will use MSCs processed from each study participant to test the ...

This study aims to measure the percentage of time spent in hyperglycemia in patients on insulin therapy and evaluate diabetes related patient reported outcomes in kidney transplant recipients with type 2 diabetes. It also aimes to evaluate immunosuppression related patient reported outcomes in kidney transplant recipients with type 2 diabetes.

The purpose of this study is to look at how participants' daily life is affected by their heart failure. The study will also look at the change in participants' body weight. This study will compare the effect of semaglutide (a new medicine) compared to "dummy" medicine on body weight and heart failure symptoms. Participants will either get semaglutide or "dummy" medicine, which treatment participants get is decided by chance. Participants will need to take 1 injection once a week. 

The purpose of this study is to evaluate whether or not semaglutide can slow down the growth and worsening of chronic kidney disease in people with type 2 diabetes. Participants will receive semaglutide (active medicine) or placebo ('dummy medicine'). This is known as participants' study medicine - which treatment participants get is decided by chance. Semaglutide is a medicine, doctors can prescribe in some countries for the treatment of type 2 diabetes. Participants will get the study medicine in a pen. Participants will use the pen to inject the medicine in a skin fold once a week. The study will close when ...

The objectives of this study are to evaluate the safety of IW-9179 in patients with diabetic gastroparesis (DGP) and the effect of treatment on the cardinal symptoms of DGP.

The purpose of this study is to understand why patients with indigestion, with or without diabetes, have gastrointestinal symptoms and, in particular, to understand where the symptoms are related to increased sensitivity to nutrients.Subsequently, look at the effects of Ondansetron on these patients' symptoms.

The purpose of this study is to evaluate the safety, tolerability, pharmacokinetics, and exploratory effectiveness of nimacimab in patients with diabetic gastroparesis.

The purpose of this study is to prospectively assemble a cohort of subjects >50 and ≤85 years of age with New-onset Diabetes (NOD):

  • Estimate the probability of pancreatic ductal adenocarcinoma (PDAC) in the NOD Cohort;
  • Establish a biobank of clinically annotated biospecimens including a reference set of biospecimens from pre-symptomatic PDAC and control new-onset type 2 diabetes mellitus (DM) subjects;
  • Facilitate validation of emerging tests for identifying NOD subjects at high risk for having PDAC using the reference set; and
  • Provide a platform for development of an interventional protocol for early detection of sporadic PDAC ...

The purpose of this study is to demonstrate the performance of the Guardian™ Sensor (3) with an advanced algorithm in subjects age 2 - 80 years, for the span of 170 hours (7 days).

The purpose of this study is to look at the relationship of patient-centered education, the Electronic Medical Record (patient portal) and the use of digital photography to improve the practice of routine foot care and reduce the number of foot ulcers/wounds in patients with diabetes.

Diabetes mellitus is a common condition which is defined by persistently high blood sugar levels. This is a frequent problem that is most commonly due to type 2 diabetes. However, it is now recognized that a small portion of the population with diabetes have an underlying problem with their pancreas, such as chronic pancreatitis or pancreatic cancer, as the cause of their diabetes. Currently, there is no test to identify the small number of patients who have diabetes caused by a primary problem with their pancreas.

The goal of this study is to develop a test to distinguish these ...

The primary purpose of this study is to evaluate the impact of dapagliflozin, as compared with placebo, on heart failure, disease specific biomarkers, symptoms, health status and quality of life in patients with type 2 diabetes or prediabetes and chronic heart failure with preserved systolic function.

The primary purpose of this study is to prospectively assess symptoms of bloating (severity, prevalence) in patients with diabetic gastroparesis.

The purpose of this study is to track the treatment burden experienced by patients living with Type 2 Diabetes Mellitus (T2DM) experience as they work to manage their illness in the context of social distancing measures. 

To promote social distancing during the COVID-19 pandemic, health care institutions around the world have rapidly expanded their use of telemedicine to replace in-office appointments where possible.1 For patients with diabetes, who spend considerable time and energy engaging with various components of the health care system,2,3 this unexpected and abrupt transition to virtual health care may signal significant changes to ...

The purpose of this study is to evaluate the safety and efficacy of oral Pyridorin 300 mg BID in reducing the rate of progression of nephropathy due to type 2 diabetes mellitus.

The purpose of this study is to evaluate the effect of Aramchol as compared to placebo on NASH resolution, fibrosis improvement and clinical outcomes related to progression of liver disease (fibrosis stages 2-3 who are overweight or obese and have prediabetes or type 2 diabetes).

The purpose of this study is to evaluate the ability of appropriately-trained family physicians to screen for and identify Diabetic Retinopathy using retinal camera and, secondarily, to describe patients’ perception of the convenience and cost-effectiveness of retinal imaging.

The primary purpose of this study is to evaluate the impact of dapagliflozin, as compared with placebo, on heart failure disease-specific biomarkers, symptoms, health status, and quality of life in patients who have type 2 diabetes and chronic heart failure with reduced systolic function.

Hypothesis: We hypothesize that patients from the Family Medicine Department at Mayo Clinic Florida who participate in RPM will have significantly reduced emergency room visits, hospitalizations, and hospital contacts.  

Aims, purpose, or objectives: In this study, we will compare the RPM group to a control group that does not receive RPM. The primary objective is to determine if there are significant group differences in emergency room visits, hospitalizations, outpatient primary care visits, outpatient specialty care visits, and hospital contacts (inbound patient portal messages and phone calls). The secondary objective is to determine if there are ...

The purpose of this research is to determine if CGM (continuous glucose monitors) used in the hospital in patients with COVID-19 and diabetes treated with insulin will be as accurate as POC (point of care) glucose monitors. Also if found to be accurate, CGM reading data will be used together with POC glucometers to dose insulin therapy.

The purpose of this study is to evaluate the effect of fenofibrate compared with placebo for prevention of diabetic retinopathy (DR) worsening or center-involved diabetic macular edema (CI-DME) with vision loss through 4 years of follow-up in participants with mild to moderately severe non-proliferative DR (NPDR) and no CI-DME at baseline.

The purpose of this study is to assess painful diabetic peripheral neuropathy after high-frequency spinal cord stimulation.

The purpose of this study is to examine the evolution of diabetic kindey injury over an extended period in a group of subjects who previously completed a clinical trial which assessed the ability of losartan to protect the kidney from injury in early diabetic kidney disease. We will also explore the relationship between diabetic kidney disease and other diabetes complications, including neuropathy and retinopathy.

The purpose of this study is to evaluate the effietiveness of remdesivir (RDV) in reducing the rate of of all-cause medically attended visits (MAVs; medical visits attended in person by the participant and a health care professional) or death in non-hospitalized participants with early stage coronavirus disease 2019 (COVID-19) and to evaluate the safety of RDV administered in an outpatient setting.

Mayo Clinic Footer

  • Request Appointment
  • About Mayo Clinic
  • About This Site

Legal Conditions and Terms

  • Terms and Conditions
  • Privacy Policy
  • Notice of Privacy Practices
  • Notice of Nondiscrimination
  • Manage Cookies


Mayo Clinic is a nonprofit organization and proceeds from Web advertising help support our mission. Mayo Clinic does not endorse any of the third party products and services advertised.

  • Advertising and sponsorship policy
  • Advertising and sponsorship opportunities

Reprint Permissions

A single copy of these materials may be reprinted for noncommercial personal use only. "Mayo," "Mayo Clinic," "MayoClinic.org," "Mayo Clinic Healthy Living," and the triple-shield Mayo Clinic logo are trademarks of Mayo Foundation for Medical Education and Research.

  • Patient Care & Health Information
  • Diseases & Conditions
  • Type 2 diabetes

Type 2 diabetes is usually diagnosed using the glycated hemoglobin (A1C) test. This blood test indicates your average blood sugar level for the past two to three months. Results are interpreted as follows:

  • Below 5.7% is normal.
  • 5.7% to 6.4% is diagnosed as prediabetes.
  • 6.5% or higher on two separate tests indicates diabetes.

If the A1C test isn't available, or if you have certain conditions that interfere with an A1C test, your health care provider may use the following tests to diagnose diabetes:

Random blood sugar test. Blood sugar values are expressed in milligrams of sugar per deciliter ( mg/dL ) or millimoles of sugar per liter ( mmol/L ) of blood. Regardless of when you last ate, a level of 200 mg/dL (11.1 mmol/L ) or higher suggests diabetes, especially if you also have symptoms of diabetes, such as frequent urination and extreme thirst.

Fasting blood sugar test. A blood sample is taken after you haven't eaten overnight. Results are interpreted as follows:

  • Less than 100 mg/dL (5.6 mmol/L ) is considered healthy.
  • 100 to 125 mg/dL (5.6 to 6.9 mmol/L ) is diagnosed as prediabetes.
  • 126 mg/dL (7 mmol/L ) or higher on two separate tests is diagnosed as diabetes.

Oral glucose tolerance test. This test is less commonly used than the others, except during pregnancy. You'll need to not eat for a certain amount of time and then drink a sugary liquid at your health care provider's office. Blood sugar levels then are tested periodically for two hours. Results are interpreted as follows:

  • Less than 140 mg/dL (7.8 mmol/L ) after two hours is considered healthy.
  • 140 to 199 mg/dL (7.8 mmol/L and 11.0 mmol/L ) is diagnosed as prediabetes.
  • 200 mg/dL (11.1 mmol/L ) or higher after two hours suggests diabetes.

Screening. The American Diabetes Association recommends routine screening with diagnostic tests for type 2 diabetes in all adults age 35 or older and in the following groups:

  • People younger than 35 who are overweight or obese and have one or more risk factors associated with diabetes.
  • Women who have had gestational diabetes.
  • People who have been diagnosed with prediabetes.
  • Children who are overweight or obese and who have a family history of type 2 diabetes or other risk factors.

After a diagnosis

If you're diagnosed with diabetes, your health care provider may do other tests to distinguish between type 1 and type 2 diabetes because the two conditions often require different treatments.

Your health care provider will test A1C levels at least two times a year and when there are any changes in treatment. Target A1C goals vary depending on age and other factors. For most people, the American Diabetes Association recommends an A1C level below 7%.

You also receive tests to screen for complications of diabetes and other medical conditions.

More Information

  • Glucose tolerance test

Management of type 2 diabetes includes:

  • Healthy eating.
  • Regular exercise.
  • Weight loss.
  • Possibly, diabetes medication or insulin therapy.
  • Blood sugar monitoring.

These steps make it more likely that blood sugar will stay in a healthy range. And they may help to delay or prevent complications.

Healthy eating

There's no specific diabetes diet. However, it's important to center your diet around:

  • A regular schedule for meals and healthy snacks.
  • Smaller portion sizes.
  • More high-fiber foods, such as fruits, nonstarchy vegetables and whole grains.
  • Fewer refined grains, starchy vegetables and sweets.
  • Modest servings of low-fat dairy, low-fat meats and fish.
  • Healthy cooking oils, such as olive oil or canola oil.
  • Fewer calories.

Your health care provider may recommend seeing a registered dietitian, who can help you:

  • Identify healthy food choices.
  • Plan well-balanced, nutritional meals.
  • Develop new habits and address barriers to changing habits.
  • Monitor carbohydrate intake to keep your blood sugar levels more stable.

Physical activity

Exercise is important for losing weight or maintaining a healthy weight. It also helps with managing blood sugar. Talk to your health care provider before starting or changing your exercise program to ensure that activities are safe for you.

  • Aerobic exercise. Choose an aerobic exercise that you enjoy, such as walking, swimming, biking or running. Adults should aim for 30 minutes or more of moderate aerobic exercise on most days of the week, or at least 150 minutes a week.
  • Resistance exercise. Resistance exercise increases your strength, balance and ability to perform activities of daily living more easily. Resistance training includes weightlifting, yoga and calisthenics. Adults living with type 2 diabetes should aim for 2 to 3 sessions of resistance exercise each week.
  • Limit inactivity. Breaking up long periods of inactivity, such as sitting at the computer, can help control blood sugar levels. Take a few minutes to stand, walk around or do some light activity every 30 minutes.

Weight loss

Weight loss results in better control of blood sugar levels, cholesterol, triglycerides and blood pressure. If you're overweight, you may begin to see improvements in these factors after losing as little as 5% of your body weight. However, the more weight you lose, the greater the benefit to your health. In some cases, losing up to 15% of body weight may be recommended.

Your health care provider or dietitian can help you set appropriate weight-loss goals and encourage lifestyle changes to help you achieve them.

Monitoring your blood sugar

Your health care provider will advise you on how often to check your blood sugar level to make sure you remain within your target range. You may, for example, need to check it once a day and before or after exercise. If you take insulin, you may need to check your blood sugar multiple times a day.

Monitoring is usually done with a small, at-home device called a blood glucose meter, which measures the amount of sugar in a drop of blood. Keep a record of your measurements to share with your health care team.

Continuous glucose monitoring is an electronic system that records glucose levels every few minutes from a sensor placed under the skin. Information can be transmitted to a mobile device such as a phone, and the system can send alerts when levels are too high or too low.

Diabetes medications

If you can't maintain your target blood sugar level with diet and exercise, your health care provider may prescribe diabetes medications that help lower glucose levels, or your provider may suggest insulin therapy. Medicines for type 2 diabetes include the following.

Metformin (Fortamet, Glumetza, others) is generally the first medicine prescribed for type 2 diabetes. It works mainly by lowering glucose production in the liver and improving the body's sensitivity to insulin so it uses insulin more effectively.

Some people experience B-12 deficiency and may need to take supplements. Other possible side effects, which may improve over time, include:

  • Abdominal pain.

Sulfonylureas help the body secrete more insulin. Examples include glyburide (DiaBeta, Glynase), glipizide (Glucotrol XL) and glimepiride (Amaryl). Possible side effects include:

  • Low blood sugar.
  • Weight gain.

Glinides stimulate the pancreas to secrete more insulin. They're faster acting than sulfonylureas. But their effect in the body is shorter. Examples include repaglinide and nateglinide. Possible side effects include:

Thiazolidinediones make the body's tissues more sensitive to insulin. An example of this medicine is pioglitazone (Actos). Possible side effects include:

  • Risk of congestive heart failure.
  • Risk of bladder cancer (pioglitazone).
  • Risk of bone fractures.

DPP-4 inhibitors help reduce blood sugar levels but tend to have a very modest effect. Examples include sitagliptin (Januvia), saxagliptin (Onglyza) and linagliptin (Tradjenta). Possible side effects include:

  • Risk of pancreatitis.
  • Joint pain.

GLP-1 receptor agonists are injectable medications that slow digestion and help lower blood sugar levels. Their use is often associated with weight loss, and some may reduce the risk of heart attack and stroke. Examples include exenatide (Byetta, Bydureon Bcise), liraglutide (Saxenda, Victoza) and semaglutide (Rybelsus, Ozempic, Wegovy). Possible side effects include:

SGLT2 inhibitors affect the blood-filtering functions in the kidneys by blocking the return of glucose to the bloodstream. As a result, glucose is removed in the urine. These medicines may reduce the risk of heart attack and stroke in people with a high risk of those conditions. Examples include canagliflozin (Invokana), dapagliflozin (Farxiga) and empagliflozin (Jardiance). Possible side effects include:

  • Vaginal yeast infections.
  • Urinary tract infections.
  • Low blood pressure.
  • High cholesterol.
  • Risk of gangrene.
  • Risk of bone fractures (canagliflozin).
  • Risk of amputation (canagliflozin).

Other medicines your health care provider might prescribe in addition to diabetes medications include blood pressure and cholesterol-lowering medicines, as well as low-dose aspirin, to help prevent heart and blood vessel disease.

Insulin therapy

Some people who have type 2 diabetes need insulin therapy. In the past, insulin therapy was used as a last resort, but today it may be prescribed sooner if blood sugar targets aren't met with lifestyle changes and other medicines.

Different types of insulin vary on how quickly they begin to work and how long they have an effect. Long-acting insulin, for example, is designed to work overnight or throughout the day to keep blood sugar levels stable. Short-acting insulin generally is used at mealtime.

Your health care provider will determine what type of insulin is right for you and when you should take it. Your insulin type, dosage and schedule may change depending on how stable your blood sugar levels are. Most types of insulin are taken by injection.

Side effects of insulin include the risk of low blood sugar — a condition called hypoglycemia — diabetic ketoacidosis and high triglycerides.

Weight-loss surgery

Weight-loss surgery changes the shape and function of the digestive system. This surgery may help you lose weight and manage type 2 diabetes and other conditions related to obesity. There are several surgical procedures. All of them help people lose weight by limiting how much food they can eat. Some procedures also limit the amount of nutrients the body can absorb.

Weight-loss surgery is only one part of an overall treatment plan. Treatment also includes diet and nutritional supplement guidelines, exercise and mental health care.

Generally, weight-loss surgery may be an option for adults living with type 2 diabetes who have a body mass index (BMI) of 35 or higher. BMI is a formula that uses weight and height to estimate body fat. Depending on the severity of diabetes or the presence of other medical conditions, surgery may be an option for someone with a BMI lower than 35.

Weight-loss surgery requires a lifelong commitment to lifestyle changes. Long-term side effects may include nutritional deficiencies and osteoporosis.

People living with type 2 diabetes often need to change their treatment plan during pregnancy and follow a diet that controls carbohydrates. Many people need insulin therapy during pregnancy. They also may need to stop other treatments, such as blood pressure medicines.

There is an increased risk during pregnancy of developing a condition that affects the eyes called diabetic retinopathy. In some cases, this condition may get worse during pregnancy. If you are pregnant, visit an ophthalmologist during each trimester of your pregnancy and one year after you give birth. Or as often as your health care provider suggests.

Signs of trouble

Regularly monitoring your blood sugar levels is important to avoid severe complications. Also, be aware of symptoms that may suggest irregular blood sugar levels and the need for immediate care:

High blood sugar. This condition also is called hyperglycemia. Eating certain foods or too much food, being sick, or not taking medications at the right time can cause high blood sugar. Symptoms include:

  • Frequent urination.
  • Increased thirst.
  • Blurred vision.

Hyperglycemic hyperosmolar nonketotic syndrome (HHNS). This life-threatening condition includes a blood sugar reading higher than 600 mg/dL (33.3 mmol/L ). HHNS may be more likely if you have an infection, are not taking medicines as prescribed, or take certain steroids or drugs that cause frequent urination. Symptoms include:

  • Extreme thirst.
  • Drowsiness.
  • Dark urine.

Diabetic ketoacidosis. Diabetic ketoacidosis occurs when a lack of insulin results in the body breaking down fat for fuel rather than sugar. This results in a buildup of acids called ketones in the bloodstream. Triggers of diabetic ketoacidosis include certain illnesses, pregnancy, trauma and medicines — including the diabetes medicines called SGLT2 inhibitors.

The toxicity of the acids made by diabetic ketoacidosis can be life-threatening. In addition to the symptoms of hyperglycemia, such as frequent urination and increased thirst, ketoacidosis may cause:

  • Shortness of breath.
  • Fruity-smelling breath.

Low blood sugar. If your blood sugar level drops below your target range, it's known as low blood sugar. This condition also is called hypoglycemia. Your blood sugar level can drop for many reasons, including skipping a meal, unintentionally taking more medication than usual or being more physically active than usual. Symptoms include:

  • Irritability.
  • Heart palpitations.
  • Slurred speech.

If you have symptoms of low blood sugar, drink or eat something that will quickly raise your blood sugar level. Examples include fruit juice, glucose tablets, hard candy or another source of sugar. Retest your blood in 15 minutes. If levels are not at your target, eat or drink another source of sugar. Eat a meal after your blood sugar level returns to normal.

If you lose consciousness, you need to be given an emergency injection of glucagon, a hormone that stimulates the release of sugar into the blood.

  • Medications for type 2 diabetes
  • GLP-1 agonists: Diabetes drugs and weight loss
  • Bariatric surgery
  • Endoscopic sleeve gastroplasty
  • Gastric bypass (Roux-en-Y)

There is a problem with information submitted for this request. Review/update the information highlighted below and resubmit the form.

From Mayo Clinic to your inbox

Sign up for free and stay up to date on research advancements, health tips, current health topics, and expertise on managing health. Click here for an email preview.

Error Email field is required

Error Include a valid email address

To provide you with the most relevant and helpful information, and understand which information is beneficial, we may combine your email and website usage information with other information we have about you. If you are a Mayo Clinic patient, this could include protected health information. If we combine this information with your protected health information, we will treat all of that information as protected health information and will only use or disclose that information as set forth in our notice of privacy practices. You may opt-out of email communications at any time by clicking on the unsubscribe link in the e-mail.

Thank you for subscribing!

You'll soon start receiving the latest Mayo Clinic health information you requested in your inbox.

Sorry something went wrong with your subscription

Please, try again in a couple of minutes

Clinical trials

Explore Mayo Clinic studies testing new treatments, interventions and tests as a means to prevent, detect, treat or manage this condition.

Lifestyle and home remedies

Careful management of type 2 diabetes can reduce the risk of serious — even life-threatening — complications. Consider these tips:

  • Commit to managing your diabetes. Learn all you can about type 2 diabetes. Make healthy eating and physical activity part of your daily routine.
  • Work with your team. Establish a relationship with a certified diabetes education specialist, and ask your diabetes treatment team for help when you need it.
  • Identify yourself. Wear a necklace or bracelet that says you are living with diabetes, especially if you take insulin or other blood sugar-lowering medicine.
  • Schedule a yearly physical exam and regular eye exams. Your diabetes checkups aren't meant to replace regular physicals or routine eye exams.
  • Keep your vaccinations up to date. High blood sugar can weaken your immune system. Get a flu shot every year. Your health care provider also may recommend the pneumonia vaccine. The Centers for Disease Control and Prevention (CDC) also recommends the hepatitis B vaccination if you haven't previously received this vaccine and you're 19 to 59 years old. Talk to your health care provider about other vaccinations you may need.
  • Take care of your teeth. Diabetes may leave you prone to more-serious gum infections. Brush and floss your teeth regularly and schedule recommended dental exams. Contact your dentist right away if your gums bleed or look red or swollen.
  • Pay attention to your feet. Wash your feet daily in lukewarm water, dry them gently, especially between the toes, and moisturize them with lotion. Check your feet every day for blisters, cuts, sores, redness and swelling. Contact your health care provider if you have a sore or other foot problem that isn't healing.
  • Keep your blood pressure and cholesterol under control. Eating healthy foods and exercising regularly can go a long way toward controlling high blood pressure and cholesterol. Take medication as prescribed.
  • If you smoke or use other types of tobacco, ask your health care provider to help you quit. Smoking increases your risk of diabetes complications. Talk to your health care provider about ways to stop using tobacco.
  • Use alcohol sparingly. Depending on the type of drink, alcohol may lower or raise blood sugar levels. If you choose to drink alcohol, only do so with a meal. The recommendation is no more than one drink daily for women and no more than two drinks daily for men. Check your blood sugar frequently after drinking alcohol.
  • Make healthy sleep a priority. Many people with type 2 diabetes have sleep problems. And not getting enough sleep may make it harder to keep blood sugar levels in a healthy range. If you have trouble sleeping, talk to your health care provider about treatment options.
  • Caffeine: Does it affect blood sugar?

Alternative medicine

Many alternative medicine treatments claim to help people living with diabetes. According to the National Center for Complementary and Integrative Health, studies haven't provided enough evidence to recommend any alternative therapies for blood sugar management. Research has shown the following results about popular supplements for type 2 diabetes:

  • Chromium supplements have been shown to have few or no benefits. Large doses can result in kidney damage, muscle problems and skin reactions.
  • Magnesium supplements have shown benefits for blood sugar control in some but not all studies. Side effects include diarrhea and cramping. Very large doses — more than 5,000 mg a day — can be fatal.
  • Cinnamon, in some studies, has lowered fasting glucose levels but not A1C levels. Therefore, there's no evidence of overall improved glucose management.

Talk to your health care provider before starting a dietary supplement or natural remedy. Do not replace your prescribed diabetes medicines with alternative medicines.

Coping and support

Type 2 diabetes is a serious disease, and following your diabetes treatment plan takes commitment. To effectively manage diabetes, you may need a good support network.

Anxiety and depression are common in people living with diabetes. Talking to a counselor or therapist may help you cope with the lifestyle changes and stress that come with a type 2 diabetes diagnosis.

Support groups can be good sources of diabetes education, emotional support and helpful information, such as how to find local resources or where to find carbohydrate counts for a favorite restaurant. If you're interested, your health care provider may be able to recommend a group in your area.

You can visit the American Diabetes Association website to check out local activities and support groups for people living with type 2 diabetes. The American Diabetes Association also offers online information and online forums where you can chat with others who are living with diabetes. You also can call the organization at 800-DIABETES ( 800-342-2383 ).

Preparing for your appointment

At your annual wellness visit, your health care provider can screen for diabetes and monitor and treat conditions that increase your risk of diabetes, such as high blood pressure, high cholesterol or a high BMI .

If you are seeing your health care provider because of symptoms that may be related to diabetes, you can prepare for your appointment by being ready to answer the following questions:

  • When did your symptoms begin?
  • Does anything improve the symptoms or worsen the symptoms?
  • What medicines do you take regularly, including dietary supplements and herbal remedies?
  • What are your typical daily meals? Do you eat between meals or before bedtime?
  • How much alcohol do you drink?
  • How much daily exercise do you get?
  • Is there a history of diabetes in your family?

If you are diagnosed with diabetes, your health care provider may begin a treatment plan. Or you may be referred to a doctor who specializes in hormonal disorders, called an endocrinologist. Your care team also may include the following specialists:

  • Certified diabetes education specialist.
  • Foot doctor, also called a podiatrist.
  • Doctor who specializes in eye care, called an ophthalmologist.

Talk to your health care provider about referrals to other specialists who may be providing care.

Questions for ongoing appointments

Before any appointment with a member of your treatment team, make sure you know whether there are any restrictions, such as not eating or drinking before taking a test. Questions that you should regularly talk about with your health care provider or other members of the team include:

  • How often do I need to monitor my blood sugar, and what is my target range?
  • What changes in my diet would help me better manage my blood sugar?
  • What is the right dosage for prescribed medications?
  • When do I take the medications? Do I take them with food?
  • How does management of diabetes affect treatment for other conditions? How can I better coordinate treatments or care?
  • When do I need to make a follow-up appointment?
  • Under what conditions should I call you or seek emergency care?
  • Are there brochures or online sources you recommend?
  • Are there resources available if I'm having trouble paying for diabetes supplies?

What to expect from your doctor

Your health care provider is likely to ask you questions at your appointments. Those questions may include:

  • Do you understand your treatment plan and feel confident you can follow it?
  • How are you coping with diabetes?
  • Have you had any low blood sugar?
  • Do you know what to do if your blood sugar is too low or too high?
  • What's a typical day's diet like?
  • Are you exercising? If so, what type of exercise? How often?
  • Do you sit for long periods of time?
  • What challenges are you experiencing in managing your diabetes?
  • Professional Practice Committee: Standards of Medical Care in Diabetes — 2020. Diabetes Care. 2020; doi:10.2337/dc20-Sppc.
  • Diabetes mellitus. Merck Manual Professional Version. https://www.merckmanuals.com/professional/endocrine-and-metabolic-disorders/diabetes-mellitus-and-disorders-of-carbohydrate-metabolism/diabetes-mellitus-dm. Accessed Dec. 7, 2020.
  • Melmed S, et al. Williams Textbook of Endocrinology. 14th ed. Elsevier; 2020. https://www.clinicalkey.com. Accessed Dec. 3, 2020.
  • Diabetes overview. National Institute of Diabetes and Digestive and Kidney Diseases. https://www.niddk.nih.gov/health-information/diabetes/overview/all-content. Accessed Dec. 4, 2020.
  • AskMayoExpert. Type 2 diabetes. Mayo Clinic; 2018.
  • Feldman M, et al., eds. Surgical and endoscopic treatment of obesity. In: Sleisenger and Fordtran's Gastrointestinal and Liver Disease: Pathophysiology, Diagnosis, Management. 11th ed. Elsevier; 2021. https://www.clinicalkey.com. Accessed Oct. 20, 2020.
  • Hypersmolar hyperglycemic state (HHS). Merck Manual Professional Version. https://www.merckmanuals.com/professional/endocrine-and-metabolic-disorders/diabetes-mellitus-and-disorders-of-carbohydrate-metabolism/hyperosmolar-hyperglycemic-state-hhs. Accessed Dec. 11, 2020.
  • Diabetic ketoacidosis (DKA). Merck Manual Professional Version. https://www.merckmanuals.com/professional/endocrine-and-metabolic-disorders/diabetes-mellitus-and-disorders-of-carbohydrate-metabolism/diabetic-ketoacidosis-dka. Accessed Dec. 11, 2020.
  • Hypoglycemia. Merck Manual Professional Version. https://www.merckmanuals.com/professional/endocrine-and-metabolic-disorders/diabetes-mellitus-and-disorders-of-carbohydrate-metabolism/hypoglycemia. Accessed Dec. 11, 2020.
  • 6 things to know about diabetes and dietary supplements. National Center for Complementary and Integrative Health. https://www.nccih.nih.gov/health/tips/things-to-know-about-type-diabetes-and-dietary-supplements. Accessed Dec. 11, 2020.
  • Type 2 diabetes and dietary supplements: What the science says. National Center for Complementary and Integrative Health. https://www.nccih.nih.gov/health/providers/digest/type-2-diabetes-and-dietary-supplements-science. Accessed Dec. 11, 2020.
  • Preventing diabetes problems. National Institute of Diabetes and Digestive and Kidney Diseases. https://www.niddk.nih.gov/health-information/diabetes/overview/preventing-problems/all-content. Accessed Dec. 3, 2020.
  • Schillie S, et al. Prevention of hepatitis B virus infection in the United States: Recommendations of the Advisory Committee on Immunization Practices. MMWR Recommendations and Reports. 2018; doi:10.15585/mmwr.rr6701a1.
  • Diabetes prevention: 5 tips for taking control
  • Hyperinsulinemia: Is it diabetes?

Associated Procedures

News from mayo clinic.

  • Mayo study uses electronic health record data to assess metformin failure risk, optimize care Feb. 10, 2023, 02:30 p.m. CDT
  • Mayo Clinic Minute: Strategies to break the heart disease and diabetes link Nov. 28, 2022, 05:15 p.m. CDT
  • Mayo Clinic Q and A: Diabetes risk in Hispanic people Oct. 20, 2022, 12:15 p.m. CDT
  • The importance of diagnosing, treating diabetes in the Hispanic population in the US Sept. 28, 2022, 04:00 p.m. CDT
  • Mayo Clinic Minute: Managing Type 2 diabetes Sept. 28, 2022, 02:30 p.m. CDT
  • Expert Alert: Mayo Clinic Healthcare cardiologist explains link between diabetes, heart disease March 23, 2022, 03:00 p.m. CDT

Products & Services

  • A Book: The Essential Diabetes Book
  • A Book: The Mayo Clinic Diabetes Diet
  • Assortment of Health Products from Mayo Clinic Store
  • Symptoms & causes
  • Diagnosis & treatment
  • Doctors & departments

Mayo Clinic does not endorse companies or products. Advertising revenue supports our not-for-profit mission.

  • Opportunities

Mayo Clinic Press

Check out these best-sellers and special offers on books and newsletters from Mayo Clinic Press .

  • Mayo Clinic on Incontinence - Mayo Clinic Press Mayo Clinic on Incontinence
  • The Essential Diabetes Book - Mayo Clinic Press The Essential Diabetes Book
  • Mayo Clinic on Hearing and Balance - Mayo Clinic Press Mayo Clinic on Hearing and Balance
  • FREE Mayo Clinic Diet Assessment - Mayo Clinic Press FREE Mayo Clinic Diet Assessment
  • Mayo Clinic Health Letter - FREE book - Mayo Clinic Press Mayo Clinic Health Letter - FREE book

Show the heart some love!

Help us advance cardiovascular medicine.

Research Summaries

Keep up with the latest diabetes and diabetes-related studies with these brief overviews. Each summary provides main points, methods, and findings and includes a link to the article.

Diabetes Management and Education

Reaching treatment goals could help people living with type 2 diabetes increase their life expectancy by 3 years or in some cases by as much as 10 years. Read the summary .

Adults who receive diabetes education are more likely to follow recommended preventive care practices that lead to better diabetes management. Read the summary .

In 2017, the total cost of diabetes complications was over $37 billion among Medicare beneficiaries 65 or older with type 2 diabetes. Read the summary .

Kids and teens can get both type 1 and type 2 diabetes. New research shows how diabetes rates in young people may rise by 2060. Read the summary .

New USPSTF and ADA guidelines lower the age for prediabetes and type 2 diabetes screening to 35. This study examined if testing practices aligned with guidelines and which populations were less likely to receive testing. Read the summary .

The SEARCH for Diabetes in Youth study reports trends in young people who are being diagnosed with type 1 and type 2 diabetes. Read the summary .

Recent guidelines recommend newer types of diabetes medications, and most Americans living with type 2 diabetes are eligible. Read the summary .

 Chronic Kidney Disease

End-stage kidney disease—kidney failure that requires dialysis or a kidney transplant—can lead to disability and early death, is expensive to treat, and cases are on the rise. Read the summary .

  • Reports and Publications
  • Research Projects
  • US Diabetes Surveillance System

To receive updates about diabetes topics, enter your email address:

  • Diabetes Home
  • State, Local, and National Partner Diabetes Programs
  • National Diabetes Prevention Program
  • Native Diabetes Wellness Program
  • Chronic Kidney Disease
  • Vision Health Initiative
  • Heart Disease and Stroke
  • Overweight & Obesity

Exit Notification / Disclaimer Policy

  • The Centers for Disease Control and Prevention (CDC) cannot attest to the accuracy of a non-federal website.
  • Linking to a non-federal website does not constitute an endorsement by CDC or any of its employees of the sponsors or the information and products presented on the website.
  • You will be subject to the destination website's privacy policy when you follow the link.
  • CDC is not responsible for Section 508 compliance (accessibility) on other federal or private website.

Cornell Chronicle

  • Architecture & Design
  • Arts & Humanities
  • Business, Economics & Entrepreneurship
  • Computing & Information Sciences
  • Energy, Environment & Sustainability
  • Food & Agriculture
  • Global Reach
  • Health, Nutrition & Medicine
  • Law, Government & Public Policy
  • Life Sciences & Veterinary Medicine
  • Physical Sciences & Engineering
  • Social & Behavioral Sciences
  • Coronavirus
  • News & Events
  • Public Engagement
  • New York City
  • Photos of the Day
  • Big Red Sports
  • Freedom of Expression
  • Student Life
  • Around Cornell
  • All Stories
  • In the News
  • Expert Quotes
  • Cornellians

Large-scale study reveals new genetic details of diabetes

By wynne parry weill cornell medicine.

In experiments of unprecedented scale, investigators at Weill Cornell Medicine and the National Institutes of Health have revealed new aspects of the complex genetics behind Type 2 diabetes. Through these discoveries, and by providing a template for future studies, this research furthers efforts to better understand and ultimately treat this common metabolic disease.

Previous studies have generally examined the influence of individual genes. In research described Oct. 18 in Cell Metabolism, senior co-author Shuibing Chen , the Kilts Family Professor of Surgery at Weill Cornell Medicine, working alongside senior co-author Dr. Francis Collins , a senior investigator at the Center for Precision Health Research within the National Human Genome Research Institute of the U.S. National Institutes of Health, took a more comprehensive approach. Together, they looked at the contribution of 20 genes in a single effort.

“It’s very difficult to believe all these diabetes-related genes act independently of each other,” Chen said. By using a combination of technologies, the team examined the effects of shutting each down. By comparing the consequences for cell behavior and genetics, she said, “we found some common themes.”

As with other types of diabetes, Type 2 diabetes occurs when sugar levels in the blood are too high. In Type 2 diabetes, this happens in part because specialized cells in the pancreas, known as β-cells, don’t produce enough insulin, a hormone that tells cells to take sugar out of the blood for use as an energy source. Over time, high levels of blood sugar damage tissues and cause other problems, such as heart and kidney disease. According to the United States Centers for Disease Control and Prevention, nearly 9% of adults in the United States have been diagnosed with Type 2 diabetes. 

Both genetic and environmental factors, such as obesity and chronic stress, can increase risk for it. Yet evaluating the role of the genetic contributors alone is a massive project. So far, researchers have identified more than 290 locations within the genome where changes to DNA can raise the likelihood of developing the disease. Some of these locations fall within known genes, but most are found in regions that regulate the expression of nearby genes.

For the new research, the team focused on 20 genes clearly identified as contributors. They began their investigation by using the gene editing system CRISPR-Cas9 to shut down these genes, one at a time, within 20 sets of identical stem cells. 

These stem cells had the potential to generate any kind of mature cell, but the researchers coaxed them into becoming insulin-producing β-cells. They then examined the effects of losing each gene on five traits related to insulin production and the health of β-cells. They also documented the accompanying changes in gene expression and the accessibility of DNA for expression.

To make sense of the massive amount of data they collected, the team developed their own computational models to analyze it, leading to several discoveries: By comparing the effects of all 20 mutations on β-cells, they identified four additional genes, each representing a newly discovered pathway that contributes to insulin production. They also found that, of the original 20 genes, only one, called HNF4A, contributed to all five traits, apparently by acting as a master controller that regulates the activity of other genes. In one specific example, they explained how a small variation, located in a space between genes, contributes to the risk of diabetes by interfering with HNF4A’s ability to regulate nearby genes.

Ultimately, this study and others like it hold the promise of benefiting patients, Collins said. “We need to understand all the genetic and environmental factors involved so we can do a better job of preventing diabetes, and to develop new ideas about how to effectively treat it.”

Collins and Chen note that their approach may have relevance beyond diabetes, to other common diseases, such as Alzheimer’s, Parkinson’s and Crohn’s disease, that involve many genetic factors.

The work reported in this newsroom story was supported in part by the United States’ National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases and the American Diabetes Association.

Many Weill Cornell Medicine physicians and scientists maintain relationships and collaborate with external organizations to foster scientific innovation and provide expert guidance. The institution makes these disclosures public to ensure transparency. For this information, see the profile for Shuibing Chen .

Wynne Parry is a freelance writer for Weill Cornell Medicine.

Media Contact

Krystle lopez.

Get Cornell news delivered right to your inbox.

You might also like

research type 2 diabetes

Gallery Heading

  • About Diabetes
  • Warning Signs and Symptoms
  • Prediabetes
  • Diabetes Prevention
  • Gestational Diabetes
  • Understanding A1C
  • Diabetes Complications
  • Devices & Technology
  • Vaccinations
  • Diabetes Statistics
  • More About Diabetes

Are you at risk?

  • Life with Diabetes
  • Newly Diagnosed
  • Living with Type 1
  • Living with Type 2
  • Treatment & Care
  • Health & Wellness
  • Medications
  • Weight Management
  • Substance Use
  • Diabetes & Your Health
  • Mental Health
  • Sexual Health
  • Oral Health
  • Foot Health
  • Heart Health
  • Kidney Health
  • Better Choices for Life

Healthy Living

  • Food & Nutrition
  • Healthy Eating Tips
  • Reading Food Labels
  • Food Insecurity & Diabetes
  • Live Cooking Classes
  • Recipes from Diabetes Food Hub®
  • Diabetes & Food
  • Food & Blood Glucose
  • Food & Carbs
  • Meal Planning
  • Vitamins & Supplements
  • Tools & Resources
  • For Caregivers
  • Find a Diabetes Education Program
  • Driver's License Information
  • Consumer Guide
  • ShopDiabetes
  • For Professionals
  • Managing Health Care Costs
  • Insulin Cost and Affordability
  • Health Insurance Aid
  • Get Connected
  • Find Your Local Chapter

Ask the experts

  • Ways to Contribute
  • Learn About Clinical Trials
  • Become a Supplier
  • Ways To Give
  • Planned Giving
  • Workplace Giving
  • Monthly Giving
  • In-Honor Gift
  • Memorial Gift
  • Vehicle Donation
  • Giving Society
  • Programs & Initiatives
  • Focus on Diabetes®
  • Health Equity
  • Project Power
  • Know Diabetes by Heart
  • Raise Your Voice: Engagement Platform
  • Federal Advocacy
  • State Advocacy
  • Grassroots Advocacy
  • Become an Advocate
  • Know Your Rights
  • Employment Discrimination
  • Safe at School®
  • Resources For Lawyers
  • Advocacy Initiatives
  • Amputation Prevention Alliance
  • Continuous Glucose Monitors (CGMs)—Everything You Need to Know
  • Obesity Advocacy
  • Type 2 Diabetes Research At-a-Glance

research type 2 diabetes

The ADA is committed to continuing progress in the fight against type 2 diabetes by funding research, including support for potential new treatments, a better understating of genetic factors, addressing disparities, and more. For specific examples of projects currently funded by the ADA, see below.

Greg J. Morton, PhD

University of Washington

Project: Neurocircuits regulating glucose homeostasis

“The health consequences of diabetes can be devastating, and new treatments and therapies are needed. My research career has focused on understanding how blood sugar levels are regulated and what contributes to the development of diabetes. This research will provide insights into the role of the brain in the control of blood sugar levels and has potential to facilitate the development of novel approaches to diabetes treatment.”

The problem: Type 2 diabetes (T2D) is among the most pressing and costly medical challenges confronting modern society. Even with currently available therapies, the control and management of blood sugar levels remains a challenge in T2D patients and can thereby increase the risk of diabetes-related complications. Continued progress with newer, better therapies is needed to help people with T2D.

The project: Humans have special cells, called brown fat cells, which generate heat to maintain optimal body temperature. Dr. Morton has found that these cells use large amounts of glucose to drive this heat production, thus serving as a potential way to lower blood sugar, a key goal for any diabetes treatment. Dr. Morton is working to understand what role the brain plays in turning these brown fat cells on and off.

The potential outcome: This work has the potential to fundamentally advance our understanding of how the brain regulates blood sugar levels and to identify novel targets for the treatment of T2D.

Tracey Lynn McLaughlin, MD

Stanford University

Project: Role of altered nutrient transit and incretin hormones in glucose lowering after Roux-en-Y gastric bypass surgery

“This award is very important to me personally not only because the enteroinsular axis (gut-insulin-glucose metabolism) is a new kid on the block that requires rigorous physiologic studies in humans to better understand how it contributes to glucose metabolism, but also because the subjects who develop severe hypoglycemia after gastric bypass are largely ignored in society and there is no treatment for this devastating and very dangerous condition.”

The problem: Roux-en-Y gastric bypass (RYGB) surgery is the single-most effective treatment for type 2 diabetes, with persistent remission in 85% of cases. However, the underlying ways by which the surgery improves glucose control is not yet understood, limiting the ability to potentially mimic the surgery in a non-invasive way. Furthermore, a minority of RYGB patients develop severe, disabling, and life-threatening low-blood sugar, for which there is no current treatment.

The project: Utilizing a unique and rigorous human experimental model, the proposed research will attempt to gain a better understanding on how RYGB surgery improves glucose control. Dr. McLaughlin will also test a hypothesis which she believes could play an important role in the persistent low-blood sugar that is observed in some patients post-surgery.

The potential outcome: This research has the potential to identify novel molecules that could represent targets for new antidiabetic therapies. It is also an important step to identifying people at risk for low-blood sugar following RYGB and to develop postsurgical treatment strategies.

Rebekah J. Walker, PhD

Medical College of Wisconsin

Project: Lowering the impact of food insecurity in African Americans with type 2 diabetes

“I became interested in diabetes research during my doctoral training, and since that time have become passionate about addressing social determinants of health and health disparities, specifically in individuals with diabetes. Living in one of the most racially segregated cities in the nation, the burden to address the needs of individuals at particularly high risk of poor outcomes has become important to me both personally and professionally.”

The problem: Food insecurity is defined as the inability to or limitation in accessing nutritionally adequate food and may be one way to address increased diabetes risk in high-risk populations. Food insecure individuals with diabetes have worse diabetes outcomes and have more difficulty following a healthy diet compared to those who are not food insecure.

The project: Dr. Walker’s study will gather information to improve and then will test an intervention to improve blood sugar control, dietary intake, self-care management, and quality of life in food insecure African Americans with diabetes. The intervention will include weekly culturally appropriate food boxes mailed to the participants and telephone-delivered diabetes education and skills training. It will be one of the first studies focused on the unique needs of food insecure African American populations with diabetes using culturally tailored strategies.

The potential outcome: This study has the potential to guide and improve policies impacting low-income minorities with diabetes. In addition, Dr. Walker’s study will help determine if food supplementation is important in improving diabetes outcomes beyond diabetes education alone.

  • Pathway Publications
  • Collaborate & Communicate
  • Support Pathway
  • Philanthropic Supporters
  • Pathway to Stop Diabetes Awardees
  • Prediabetes/Insulin Resistance Research
  • Cardiovascular Complication Fellowship Awards
  • Funding the Future: Postdoctoral Fellowship Awards
  • Helping the Next Generation: Junior Faculty Awards
  • Innovative Basic Science Awards
  • Innovative Clinical or Translational Science Awards
  • New England Fellowship Awards
  • Type 1 Research Highlights
  • The Impact of Amaranth Supported Research
  • Type 1 and Type 2 Diabetes Research
  • Type 1 Diabetes Research At-a-Glance

Type 2 diabetes


  • 1 Diabetes Research Centre, University of Leicester and the Leicester NIHR Biomedical Research Centre, Leicester General Hospital, Leicester, UK.
  • 2 Department of Internal Medicine, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seongnam, South Korea.
  • 3 Family Medicine Department, Korle Bu Teaching Hospital, Accra Ghana and Community Health Department, University of Ghana Medical School, Accra, Ghana.
  • 4 Diabetes Research Centre, University of Leicester and the Leicester NIHR Biomedical Research Centre, Leicester General Hospital, Leicester, UK. Electronic address: [email protected].
  • PMID: 36332637
  • DOI: 10.1016/S0140-6736(22)01655-5

Type 2 diabetes accounts for nearly 90% of the approximately 537 million cases of diabetes worldwide. The number affected is increasing rapidly with alarming trends in children and young adults (up to age 40 years). Early detection and proactive management are crucial for prevention and mitigation of microvascular and macrovascular complications and mortality burden. Access to novel therapies improves person-centred outcomes beyond glycaemic control. Precision medicine, including multiomics and pharmacogenomics, hold promise to enhance understanding of disease heterogeneity, leading to targeted therapies. Technology might improve outcomes, but its potential is yet to be realised. Despite advances, substantial barriers to changing the course of the epidemic remain. This Seminar offers a clinically focused review of the recent developments in type 2 diabetes care including controversies and future directions.

Copyright © 2022 Elsevier Ltd. All rights reserved.

Publication types

  • Diabetes Mellitus, Type 2* / drug therapy
  • Diabetes Mellitus, Type 2* / epidemiology
  • Pharmacogenetics
  • Precision Medicine
  • Young Adult
  • Alzheimer's disease & dementia
  • Arthritis & Rheumatism
  • Attention deficit disorders
  • Autism spectrum disorders
  • Biomedical technology
  • Diseases, Conditions, Syndromes
  • Endocrinology & Metabolism
  • Gastroenterology
  • Gerontology & Geriatrics
  • Health informatics
  • Inflammatory disorders
  • Medical economics
  • Medical research
  • Medications
  • Neuroscience
  • Obstetrics & gynaecology
  • Oncology & Cancer
  • Ophthalmology
  • Overweight & Obesity
  • Parkinson's & Movement disorders
  • Psychology & Psychiatry
  • Radiology & Imaging
  • Sleep disorders
  • Sports medicine & Kinesiology
  • Vaccination
  • Breast cancer
  • Cardiovascular disease
  • Chronic obstructive pulmonary disease
  • Colon cancer
  • Coronary artery disease
  • Heart attack
  • Heart disease
  • High blood pressure
  • Kidney disease
  • Lung cancer
  • Multiple sclerosis
  • Myocardial infarction
  • Ovarian cancer
  • Post traumatic stress disorder
  • Rheumatoid arthritis
  • Schizophrenia
  • Skin cancer
  • Type 2 diabetes
  • Full List »

share this!

February 22, 2024

This article has been reviewed according to Science X's editorial process and policies . Editors have highlighted the following attributes while ensuring the content's credibility:


trusted source

Researchers uncover potential treatment for cardiovascular complications from type 2 diabetes

by University of Missouri

Researchers uncover potential treatment for cardiovascular complications from type 2 diabetes

New research at the Roy Blunt NextGen Precision Health building has discovered a potential treatment for an underlying cause of cardiovascular disease in people with type 2 diabetes.

More than 30 million Americans live with type 2 diabetes. One common feature of diabetes is the hardening and inflexibility of blood vessels caused by damage to the endothelial cells in the vascular system.

Over time, this can lead to the development and progression of cardiovascular disease, which is the number one cause of death in diabetics. Because endothelial dysfunction is causally linked to cardiovascular disease , there is a considerable need to identify new therapeutic targets to improve endothelial function in type 2 diabetics.

A research team from the University of Missouri has found that neuraminidase activity is elevated in the circulation of type 2 diabetic mice and humans. In a series of mechanistic experiments in cultured endothelial cells and isolated blood vessels, they were able to link increased neuraminidase to endothelial dysfunction.

"Because we know that type 2 diabetics have this increased neuraminidase circulating in their blood and that the presence of it promotes endothelial dysfunction, it is important to target it as a means of addressing the cardiovascular complications faced by those with type 2 diabetes," said Luis Martinez-Lemus, DVM, Ph.D., James O. Davis distinguished professor in cardiovascular research at the University of Missouri School of Medicine.

The team also found that neuraminidase inhibition using zanamivir, an oral inhalation drug used to treat the flu virus, improved endothelial function in diabetic mice.

"This research lays out the molecular mechanisms by which neuraminidase promotes endothelial dysfunction , and these mechanisms can be exploited therapeutically," said Jaume Padilla, Ph.D., an associate professor of nutrition and exercise physiology at MU. "Improving vascular function in people with type 2 diabetes can help them live longer and better lives, which is why this research is so important."

" Neuraminidase inhibition improves endothelial function in diabetic mice " and " Neuraminidase-induced externalization of phosphatidylserine activates ADAM17 and impairs insulin signaling in endothelial cells " were recently published in the American Journal of Physiology-Heart and Circulatory Physiology .

Larissa Ferreira-Santos et al, Neuraminidase-induced externalization of phosphatidylserine activates ADAM17 and impairs insulin signaling in endothelial cells, American Journal of Physiology-Heart and Circulatory Physiology (2023). DOI: 10.1152/ajpheart.00638.2023

Explore further

Feedback to editors

research type 2 diabetes

Researchers create more realistic synthetic human mini hearts

6 hours ago

research type 2 diabetes

New brain stimulation technique shows promise for treating brain disorders

19 hours ago

research type 2 diabetes

New research shows babies use immune system differently, but efficiently

20 hours ago

research type 2 diabetes

Research finds relaxing words heard during sleep can slow the heart down

research type 2 diabetes

New research challenges conventional picture of Parkinson's disease

21 hours ago

research type 2 diabetes

Chemotherapy method uses patient's own cells as trojan horse to direct cancer-killing drugs to tumors

research type 2 diabetes

Study: 'Hexaplex' vaccine aims to boost flu protection

research type 2 diabetes

Model suggests increased use of Paxlovid could cut hospitalizations, deaths and costs

22 hours ago

research type 2 diabetes

Early-life airborne lead exposure associated with lower IQ and self-control: Study

research type 2 diabetes

Genetic signature may predict response to immunotherapy for non-small cell lung cancer

Related stories.

research type 2 diabetes

Using a unique organ-on-a-chip model to unlock answers to vascular aging

Dec 11, 2023

research type 2 diabetes

Study reveals how estrogen exerts its anti-diabetic effects

Oct 23, 2023

research type 2 diabetes

Diabetes researchers discover the potential of CIDEC protein to mitigate obesity-related cardiometabolic disease

Mar 29, 2023

research type 2 diabetes

Endothelial proteins in plasma could reflect vascular health status

Dec 6, 2021

research type 2 diabetes

Study reverses long-held ideas about relationship among diabetes, fat and cardiovascular disease

Aug 5, 2022

research type 2 diabetes

Study identifies a key protein in blood vessel growth

Nov 29, 2023

Recommended for you

research type 2 diabetes

Researchers explore whether gut microbes cause some COVID-19 patients to have higher blood clot risk

23 hours ago

research type 2 diabetes

Substudy of trial reveals how statin therapy prevents cardiovascular disease in people with HIV

Feb 22, 2024

research type 2 diabetes

Air pollution linked to increased hospital admission for major heart and lung diseases

Feb 21, 2024

research type 2 diabetes

Intravascular imaging shown to significantly improve survival, safety and outcomes in stenting procedures

research type 2 diabetes

New research suggests boosting potassium is key for blood pressure control

Let us know if there is a problem with our content.

Use this form if you have come across a typo, inaccuracy or would like to send an edit request for the content on this page. For general inquiries, please use our contact form . For general feedback, use the public comments section below (please adhere to guidelines ).

Please select the most appropriate category to facilitate processing of your request

Thank you for taking time to provide your feedback to the editors.

Your feedback is important to us. However, we do not guarantee individual replies due to the high volume of messages.

E-mail the story

Your email address is used only to let the recipient know who sent the email. Neither your address nor the recipient's address will be used for any other purpose. The information you enter will appear in your e-mail message and is not retained by Medical Xpress in any form.

Newsletter sign up

Get weekly and/or daily updates delivered to your inbox. You can unsubscribe at any time and we'll never share your details to third parties.

More information Privacy policy

Donate and enjoy an ad-free experience

We keep our content available to everyone. Consider supporting Science X's mission by getting a premium account.

E-mail newsletter


  • February 23, 2024 | Redefining Nuclear Magic: Physicists Have Discovered Two New Isotopes
  • February 23, 2024 | The Limits of Math: Study Shows Forests Are More Complex Than Thought
  • February 23, 2024 | Epigenetic Changes Can Cause Type 2 Diabetes, According to New Research
  • February 23, 2024 | Two New Species of Lamprey Fish Discovered in California
  • February 23, 2024 | Tipping Point Alert: The Amazon’s Climate Thresholds Revealed

Epigenetic Changes Can Cause Type 2 Diabetes, According to New Research

By Lund University February 23, 2024

Genetics Breakthrough DNA Change

Research from Lund University indicates that epigenetic changes, not just genetics, can lead to type 2 diabetes, offering new insights for disease prevention and management.

A recent study conducted by Lund University researchers, and published in Nature Communications , suggests that epigenetic changes might be a causal factor in the development of type 2 diabetes, rather than merely occurring as a result of the disease. This new research bolsters the theory that epigenetic modifications can lead to type 2 diabetes, and the team is now focusing on creating strategies for disease prevention.

We inherit our genes from our parents, and they seldom change. However, epigenetic changes that arise due to environmental and lifestyle factors can affect the function of genes.

“Our new extensive study confirms our previous findings from smaller studies, showing that epigenetic changes can contribute to the development of type 2 diabetes. In this study, we have also identified new genes that impact the development of the disease. Our hope is that with the help of these results, we can develop methods that can be used to prevent type 2 diabetes,” says Charlotte Ling, professor of diabetes and epigenetics at Lund University’s Diabetes Centre (LUDC), who led the study.

The same epigenetic changes

The researchers studied epigenetics in insulin -producing cells from donors and found 5584 sites in the genome with changes that differed between 25 individuals with type 2 diabetes and 75 individuals without the disease. The same epigenetic changes found in people with type 2 diabetes were also found in individuals with elevated blood sugar levels, which increase the risk of developing the disease.

“Those of us who study epigenetics, have long tried to understand whether epigenetic changes cause type 2 diabetes or if the changes occur after the disease has already developed. Because we saw the same epigenetic changes in people with type 2 diabetes and individuals at risk for the disease, we conclude that these changes may contribute to the development of type 2 diabetes,” says Tina Rönn, lead author and researcher at LUDC.

The study identified 203 genes with different expression in individuals with type 2 diabetes compared to the control group. The researchers found that the gene RHOT1 showed epigenetic changes in people with type 2 diabetes and that it also played a key role in insulin secretion in insulin-producing cells. When they knocked out the gene expression of RHOT1 in cells from donors without type 2 diabetes, insulin secretion decreased.

“When we examined the same type of cells in rats with diabetes, we found a lack of RHOT1, confirming the gene’s importance for insulin secretion,” says Tina Rönn.

Methods that can prevent the disease

One goal of the research is to develop a blood-based biomarker that can predict who is at risk of developing type 2 diabetes. Therefore, the researchers investigated whether their results from insulin-producing cells in the pancreas were reflected in the blood of living people. They found epigenetic changes in the blood of a group of 540 people without the disease and they linked this to the future development of type 2 diabetes in half of the individuals.

Factors such as unhealthy diet, sedentary lifestyle, and aging increase the risk of type 2 diabetes, and they also affect our epigenetics. With the new study, researchers have identified new mechanisms that may make it possible to develop methods to help prevent type 2 diabetes.

“If we succeed in developing an epigenetic biomarker, we can identify individuals with epigenetic changes before they become ill. These individuals can, for example, receive personalized lifestyle advice that can reduce their risk of disease, or we can develop methods that aim to correct the activity of certain genes using epigenetic editing,” says Charlotte Ling.

Reference: “Genes with epigenetic alterations in human pancreatic islets impact mitochondrial function, insulin secretion, and type 2 diabetes” by Tina Rönn, Jones K. Ofori, Alexander Perfilyev, Alexander Hamilton, Karolina Pircs, Fabian Eichelmann, Sonia Garcia-Calzon, Alexandros Karagiannopoulos, Hans Stenlund, Anna Wendt, Petr Volkov, Matthias B. Schulze, Hindrik Mulder, Lena Eliasson, Sabrina Ruhrmann, Karl Bacos and Charlotte Ling, 12 December 2023, Nature Communications . DOI: 10.1038/s41467-023-43719-9

More on SciTechDaily

New 3D Printing Technique for Research

New 3D Printing Technique Enables Cheaper Patient-Specific Medical Research

Elderly Senior Fall

Risk Factors for Falls in Older Americans Identified – A Growing Public Health Concern

research type 2 diabetes

Orbital Engineering, Yale Engineers Change Electron Trajectories

More research needed to determine “tipping point” of supervolcanoes.

Kids Watching Fireworks

Common Fireworks Emit Toxic Metals Into the Air – Damage Human Cells and Animal Lungs

Quantum Computing Error Correction Art Concept

Finding and Erasing Quantum Computing Errors in Real-Time

COVID Lungs Heart Concept

Novel Molecules Discovered to Combat Asthma and COVID-Related Lung Diseases

Metamorphosed Banded Iron Formation Rock Specimen

The Unseen Spark: How Ancient Microorganisms Helped Cause Massive Volcanic Events

Be the first to comment on "epigenetic changes can cause type 2 diabetes, according to new research", leave a comment cancel reply.

Email address is optional. If provided, your email will not be published or shared.

Save my name, email, and website in this browser for the next time I comment.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • View all journals
  • Explore content
  • About the journal
  • Publish with us
  • Sign up for alerts
  • 19 February 2024

Ambitious survey of human diversity yields millions of undiscovered genetic variants

You can also search for this author in PubMed   Google Scholar

Revellers enjoy themselves at the Wigstock festival which took place in New York City in August 1994.

The All of Us programme aims to recruit one million people from ethnic and socio-economic groups that are typically under-represented in biomedical studies. Credit: Barbara Alper/Getty

A massive US programme that aims to improve health care by focusing on the genomes and health profiles of historically underrepresented groups has begun to yield results.

Analyses of up to 245,000 genomes gathered by the All of Us programme, run by the US National Institutes of Health in Bethesda, Maryland, have uncovered more than 275 million new genetic markers, nearly 150 of which might contribute to type 2 diabetes. The work has also identified gaps in genetics research on non-white populations. The findings were published on 19 February in a package of papers in Nature 1 , 2 , Communications Biology 3 and Nature Medicine 4 .

They are a “nice distillation of the All of Us resource — what it is and what it can do”, says Michael Inouye, a computational genomicist at the University of Cambridge, UK. “This is going to be the go-to data set” for genetics researchers who want to know whether their findings are generalizable to a broad population or apply to only a limited one, he adds.

Bridging the gap

Researchers have long acknowledged the lack of diversity in the genomes available for them to study, says Jibril Hirbo, a geneticist at Vanderbilt University Medical Center in Nashville, Tennessee, who studies the genetics of health disparities. One study 5 that looked at data gathered up until January 2019 found that 78% of people in most large-scale genomic studies of disease were of European descent. This has exacerbated existing health disparities, particularly for non-white individuals, Hirbo says. When researchers choose genetic or molecular targets for new medicines or create models to predict who is at risk of developing a disease, they tend to make decisions on the basis of non-diverse data because that’s all that has been available.

research type 2 diabetes

Facing up to injustice in genome science

The All of Us programme, which has received over US$3.1 billion to date and plans to assemble detailed health profiles for one million people in the United States by the end of 2026, aims to bridge that gap, says Andrea Ramirez, the programme’s chief data officer. It began enrolling people in 2018, and released its first tranche of data — about 100,000 whole genomes — in 2022. By April 2023, it had enrolled 413,000 anonymized participants, 46% of whom belong to a minority racial or ethnic group, and had shared nearly 250,000 genomes. By comparison, the world’s largest whole-genome data set , the UK Biobank, has so far released about half a million genomes, around 88% of which are from white people.

The All of Us data set is “a huge resource, particularly of African American, Hispanic and Latin American genomes, that’s massively missing from the vast majority of large-scale biobank resources and genomics consortia”, says Alicia Martin, a population geneticist at Massachusetts General Hospital in Boston.

In addition to the genomes, the database includes some participants’ survey responses, electronic health records and data from wearable devices, such as Fitbits, that report people’s activity, “making this one of the most powerful resources of genomic data”, Martin says.

An urgent need

A study in Nature on type 2 diabetes 2 is an example of the power of using a database that includes diverse genomes, Ramirez says. The condition, which affects about one in ten people in the United States, can be caused by many distinct biological mechanisms involving various genes. The researchers analysed genetic information from several databases, including All of Us, for a total of more than 2.5 million people; nearly 40% of the data came from individuals not of European ancestry. The team found 611 genetic markers that might drive the development and progression of the disease, 145 of which have never been reported before. These findings could be used to develop “genetically informed diabetes care”, the authors write.

research type 2 diabetes

World’s biggest set of human genome sequences opens to scientists

In another of the studies 3 , researchers used All of Us data to examine pathogenic variants — that is, genetic differences that increase a person’s risk of developing a particular disease. They found that, among the genomes of people with European ancestry, 2.3% had a pathogenic variant. Among genomes from people with African ancestry, however, this fell to 1.6%.

Study co-author Eric Venner, a computational geneticist at Baylor College of Medicine in Houston, Texas, cautions that there should be no biological reason for the differences. He says that the disparity is probably the result of more research having been conducted on people of European ancestry; we simply know more about which mutations in this population lead to disease. In fact, the researchers found more variants of unknown risk in the genomes of people with non-European ancestry than in those with European ancestry, he adds. This underscores the urgent need to study non-European genomes in more detail, Venner says.

Updating models

Gathering and using more genomic and health data from diverse populations will be especially important for generating more accurate ‘polygenic risk scores’. These provide a picture of a person’s risk of developing a disease as a result of their genetics.

research type 2 diabetes

US tailored-medicine project aims for ethnic balance

To calculate a score for a particular disease, researchers develop an algorithm that is trained on thousands of genomes from people who either do or don’t have the disease. A person’s own score can then be calculated by feeding their genetic data into the algorithm.

Previous research 6 has shown that the scores, which might soon be used in the clinic for personalized health care, tend to be less accurate for minority populations than for majority ones. In one of the current papers 4 , researchers used the more-inclusive All of Us data to improve the landscape: they calibrated and validated scores for 23 conditions and recommended 10 to be prioritized for use in the clinic, for conditions including coronary heart disease and diabetes. Martin applauds these efforts, but she hopes that future studies address how physicians and others in the clinic interpret these scores, and whether the scores can improve a person’s health in the long term because of the treatment decisions they elicit.

The All of Us programme plans to release a tranche of data every year, representing new enrolees and genomes, including one later in 2024, Ramirez says. It’s excellent that diverse data are coming in, Hirbo says, adding that he would like to see existing algorithms that were trained mainly on the genomes of people of European ancestry updated soon. “The models are still way behind,” he says.

doi: https://doi.org/10.1038/d41586-024-00502-0

The All of Us Research Program Genomics Investigators Nature https://doi.org/10.1038/s41586-023-06957-x (2024).

Article   Google Scholar  

Suzuki, K. et al. Nature https://doi.org/10.1038/s41586-024-07019-6 (2024).

Venner, E. et al. Commun. Biol . https://doi.org/10.1038/s42003-023-05708-y (2024).

Lennon, N. J. et al. Nature Med . https://doi.org/10.1038/s41591-024-02796-z (2024).

Sirugo, G., Williams, S. M. & Tishkoff, S. A. Cell 177 , 26–31 (2019).

Article   PubMed   Google Scholar  

Martin, A. R. et al. Nature Genet. 51 , 584–591 (2019).

Download references

Reprints and permissions

Related Articles

research type 2 diabetes

  • Research data
  • Personalized medicine
  • Institutions

‘All of Us’ genetics chart stirs unease over controversial depiction of race

‘All of Us’ genetics chart stirs unease over controversial depiction of race

News 23 FEB 24

Open science — embrace it before it’s too late

Open science — embrace it before it’s too late

Editorial 06 FEB 24

In the AI science boom, beware: your results are only as good as your data

In the AI science boom, beware: your results are only as good as your data

Career Column 01 FEB 24

Why citizen scientists are gathering DNA from hundreds of lakes — on the same day

Why citizen scientists are gathering DNA from hundreds of lakes — on the same day

News 21 FEB 24

New Chinese databases are a boost for rare-disease science

Correspondence 20 FEB 24

Super-speedy sequencing puts genomic diagnosis in the fast lane

Super-speedy sequencing puts genomic diagnosis in the fast lane

Technology Feature 19 FEB 24

Medical AI falters when assessing patients it hasn’t seen

Medical AI falters when assessing patients it hasn’t seen

News 11 JAN 24

Australian Indigenous genomes are highly diverse and unlike those anywhere else

Australian Indigenous genomes are highly diverse and unlike those anywhere else

News 13 DEC 23

Tiny robots made from human cells heal damaged tissue

Tiny robots made from human cells heal damaged tissue

News 30 NOV 23

Postdoctoral Fellow

A Postdoctoral Fellow position is immediately available in the laboratory of Dr. Fen-Biao Gao at the University of Massachusetts Chan Medical Schoo...

Worcester, Massachusetts (US)

Umass Chan Medical School - Fen-Biao Gao Lab

research type 2 diabetes

Washing, Sterilisation and Media Preparation Technician

APPLICATION CLOSING DATE: March 7th, 2024 About Human Technopole:  Human Technopole (HT) is an interdisciplinary life science research institute, c...

Human Technopole

research type 2 diabetes

Scientific Officer

ABOUT US The Human Frontier Science Program Organization (HFSPO) is a unique organization, supporting international collaboration to undertake inno...

Strasbourg-Ville, Bas-Rhin (FR)


research type 2 diabetes

Tenure Track Assistant Professor towards Associate Professor in the field of biomedical sciences

UNIL is a leading international teaching and research institution, with over 5,000 employees and 17,000 students split between its Dorigny campus, ...

Lausanne, Canton of Vaud (CH)

University of Lausanne (UNIL)

research type 2 diabetes

Faculty Positions at City University of Hong Kong (Dongguan)

CityU (Dongguan) warmly invites individuals from diverse backgrounds to apply for various faculty positions available at the levels of Professor...

Dongguan, Guangdong, China

City University of Hong Kong (Dongguan)

research type 2 diabetes

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Quick links

  • Explore articles by subject
  • Guide to authors
  • Editorial policies

Phenotype-based targeted treatment of SGLT2 inhibitors and GLP-1 receptor agonists in type 2 diabetes

  • Open access
  • Published: 22 February 2024

Cite this article

You have full access to this open access article

  • Pedro Cardoso   ORCID: orcid.org/0000-0002-1014-9058 1 ,
  • Katie G. Young   ORCID: orcid.org/0000-0003-2570-3864 1 ,
  • Anand T. N. Nair 2 ,
  • Rhian Hopkins   ORCID: orcid.org/0000-0001-6054-3582 1 ,
  • Andrew P. McGovern 1 ,
  • Eram Haider 2 ,
  • Piyumanga Karunaratne 2 ,
  • Louise Donnelly 2 ,
  • Bilal A. Mateen 3 ,
  • Naveed Sattar 4 ,
  • Rury R. Holman 5 , 6 ,
  • Jack Bowden 1 ,
  • Andrew T. Hattersley 1 ,
  • Ewan R. Pearson 2 ,
  • Angus G. Jones 1 ,
  • Beverley M. Shields 1 ,
  • Trevelyan J. McKinley 1 &
  • John M. Dennis   ORCID: orcid.org/0000-0002-7171-732X 1

on behalf of the MASTERMIND consortium

1 Altmetric

Explore all metrics


A precision medicine approach in type 2 diabetes could enhance targeting specific glucose-lowering therapies to individual patients most likely to benefit. We aimed to use the recently developed Bayesian causal forest (BCF) method to develop and validate an individualised treatment selection algorithm for two major type 2 diabetes drug classes, sodium–glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1-RA).

We designed a predictive algorithm using BCF to estimate individual-level conditional average treatment effects for 12-month glycaemic outcome (HbA 1c ) between SGLT2i and GLP1-RA, based on routine clinical features of 46,394 people with type 2 diabetes in primary care in England (Clinical Practice Research Datalink; 27,319 for model development, 19,075 for hold-out validation), with additional external validation in 2252 people with type 2 diabetes from Scotland (SCI-Diabetes [Tayside & Fife]). Differences in glycaemic outcome with GLP1-RA by sex seen in clinical data were replicated in clinical trial data (HARMONY programme: liraglutide [ n =389] and albiglutide [ n =1682]). As secondary outcomes, we evaluated the impacts of targeting therapy based on glycaemic response on weight change, tolerability and longer-term risk of new-onset microvascular complications, macrovascular complications and adverse kidney events.

Model development identified marked heterogeneity in glycaemic response, with 4787 (17.5%) of the development cohort having a predicted HbA 1c benefit >3 mmol/mol (>0.3%) with SGLT2i over GLP1-RA and 5551 (20.3%) having a predicted HbA 1c benefit >3 mmol/mol with GLP1-RA over SGLT2i. Calibration was good in hold-back validation, and external validation in an independent Scottish dataset identified clear differences in glycaemic outcomes between those predicted to benefit from each therapy. Sex, with women markedly more responsive to GLP1-RA, was identified as a major treatment effect modifier in both the UK observational datasets and in clinical trial data: HARMONY-7 liraglutide (GLP1-RA): 4.4 mmol/mol (95% credible interval [95% CrI] 2.2, 6.3) (0.4% [95% CrI 0.2, 0.6]) greater response in women than men. Targeting the two therapies based on predicted glycaemic response was also associated with improvements in short-term tolerability and long-term risk of new-onset microvascular complications.


Precision medicine approaches can facilitate effective individualised treatment choice between SGLT2i and GLP1-RA therapies, and the use of routinely collected clinical features for treatment selection could support low-cost deployment in many countries.

Graphical Abstract

research type 2 diabetes

Similar content being viewed by others

research type 2 diabetes

Management of hyperglycaemia in type 2 diabetes, 2022. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD)

Melanie J. Davies, Vanita R. Aroda, … John B. Buse

research type 2 diabetes

Cardiovascular and mortality outcomes with GLP-1 receptor agonists vs other glucose-lowering drugs in individuals with NAFLD and type 2 diabetes: a large population-based matched cohort study

Arunkumar Krishnan, Carolin V. Schneider, … Saleh A. Alqahtani

research type 2 diabetes

The management of type 1 diabetes in adults. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD)

Richard I. G. Holt, J. Hans DeVries, … Anne L. Peters

Avoid common mistakes on your manuscript.

figure b


A precision medicine approach in type 2 diabetes would aim to target specific glucose-lowering therapies to individual patients most likely to benefit [ 1 ]. Current stratification in type 2 diabetes treatment guidelines involves preferential prescribing of two major drug classes, sodium–glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1-RA), to subgroups of people with or at high risk of cardiorenal disease [ 2 ]. Evidence informing these recommendations comes from average treatment effect (ATE) estimates derived from placebo-controlled cardiovascular and renal outcome trials, which have predominantly recruited participants with advanced atherosclerotic cardiovascular risk or established cardiovascular disease [ 3 , 4 ]. Consequently, there is limited evidence on the benefits of SGLT2i and GLP1-RA for individuals in the broader type 2 diabetes population and, given the lack of head-to-head trials, of the relative efficacy of the two drug classes for individual patients.

Recent studies have demonstrated a clear potential for a precision medicine approach based on glycaemic response, with the TRIMASTER crossover trial establishing a greater efficacy of SGLT2i compared with DPP4 inhibitors (DPP4i) in those with better renal function, and a greater efficacy of thiazolidinedione therapy compared with DPP4i in those with obesity (BMI > 30 kg/m 2 ) compared to those without obesity [ 5 ]. Given these findings, a trial-data-validated prediction model to support individualised treatment selection has recently been developed for SGLT2i vs DPP4i therapy [ 6 ]. For GLP1-RA, although recent studies have identified robust heterogeneity in treatment response based on pharmacogenetic markers and markers of insulin secretion [ 7 , 8 ], the influence of these markers on relative differences in clinical outcomes compared with other drug classes, and therefore their utility for targeting treatment, has not previously been assessed.

Given the lack of evidence to support targeted treatment of SGLT2i compared with GLP1-RA therapies, we aimed to develop and validate a prediction model to provide individual patient-level estimates of differences in 12-month glycaemic (HbA 1c ) outcomes for the two drug classes based on routinely collected clinical features. We also evaluated the downstream impacts of targeting therapy based on glycaemic response on secondary outcomes of weight change, tolerability and longer-term risk of new-onset microvascular complications, macrovascular complications and adverse kidney events.

Study population

Individuals with type 2 diabetes initiating SGLT2i and GLP1-RA therapies between 1 January 2013 and 31 October 2020 were identified in the UK population-representative Clinical Practice Research Datalink (CPRD) Aurum dataset [ 9 ], following our previously published cohort profile [ 10 ] (see https://github.com/Exeter-Diabetes/CPRD-Codelists for all codelists). We excluded individuals prescribed either therapy as first-line treatment (not recommended in UK guidelines) [ 11 ], co-treated with insulin, and with a diagnosis of end-stage renal disease (ESRD) (electronic supplementary material [ESM] Fig. 1 ). Owing to low numbers, we also excluded individuals initiating the GLP1-RA semaglutide ( n =784 study-eligible individuals with outcome HbA 1c recorded) [ 12 ]. The final CPRD cohort was randomly split 60:40 into development and hold-back validation sets, maintaining the proportion of individuals receiving SGLT2i and GLP1-RA in each set. For model development, individuals were excluded from the development and validation sets if they initiated multiple glucose-lowering treatments on the same day; their therapies were initiated less than 61 days since the start of a previous therapy; their baseline HbA 1c was <53 mmol/mol (7%); they had a missing baseline HbA 1c ; or they had a missing outcome HbA 1c (Table 1 , ESM Fig. 1 ).

Additional cohorts

The same eligibility criteria were applied to define an independent cohort in Scotland for model validation (SCI-Diabetes [Tayside & Fife], containing longitudinal observational data including biochemical investigations and prescriptions). To assess reproducibility of differences in HbA 1c response by sex with GLP1-RA therapy, we accessed individual-level data on participants initiating the GLP1-RAs albiglutide and liraglutide in the HARMONY clinical trial programme (sponsored by GlaxoSmithKline [GSK]), an international randomised placebo-controlled trial designed to evaluate the cardiovascular benefit of albiglutide with type 2 diabetes [ 13 ], and the Predicting Response to Incretin Based Agents (PRIBA) prospective cohort study (UK 2011–2013) [ 14 ], designed to test whether individuals with low insulin secretion have lesser glycaemic response to incretin-based treatments.

The primary outcome was achieved HbA 1c at 12 months post drug initiation on unchanged glucose-lowering therapy. Given the variability in the timing of follow-up testing in UK primary care, this outcome was defined as the closest eligible HbA 1c value to 12 months (within 3–15 months) after initiation. To allow for potential differential effects of follow-up duration on HbA 1c , we included an additional covariate to capture the month the outcome HbA 1c was recorded.

Secondary outcomes comprised short-term 12 month weight change after initiation (closest recorded weight to 12 months, within 3–15 months), and, as a proxy for drug tolerability, treatment discontinuation within 6 months of drug initiation (as such short-term discontinuation is unlikely to be related to a lack of glycaemic response), and longer-term outcomes up to 5 years after initiation: new-onset major adverse cardiovascular events (MACE: composite of myocardial infarction, stroke and cardiovascular death); new-onset heart failure; new-onset adverse kidney outcome (a drop of ≥40% in eGFR from baseline or reaching chronic kidney disease [CKD] stage 5 [ 7 ]); and new-onset microvascular complications (ESM Fig. 2 ). We focused on only new-onset cardiorenal events (excluding individuals with pre-existing conditions of interest), as those with pre-existing disease have a clear indication for SGLT2i and GLP1-RA in current guidelines irrespective of differences in glycaemic outcome.

Candidate predictors were selected to represent readily available (available in >75% of individuals) routine clinical features and comprised current age, duration of diabetes, year of therapy start, sex (self-reported), ethnicity (self-reported, categorised into major UK groups: White, South Asian, Black, Mixed, other), social deprivation (index of multiple deprivation quintile), smoking status, the number of current, and ever, prescribed glucose-lowering drug classes, baseline HbA 1c (closest to treatment start date; range in previous 6 months to +7 days), clinical parameters: BMI, eGFR (CKD-EPI formula [ 15 ]), HDL-cholesterol, alanine aminotransferase (ALT), albumin, bilirubin, total cholesterol and mean arterial blood pressure (all defined as closest values to treatment start in the previous two years), microvascular complications: nephropathy, neuropathy, retinopathy, and major comorbidities: angina, atherosclerotic cardiovascular disease, atrial fibrillation, cardiac revascularisation, heart failure, hypertension, ischaemic heart disease, myocardial infarction, peripheral arterial disease, stroke, transient ischaemic attack, CKD and chronic liver disease.

Treatment selection model development

We used the recently proposed Bayesian causal forest (BCF) structure, a framework specifically designed to estimate heterogeneous treatment effects (henceforth: conditional average treatment effects [CATEs]) [ 16 , 17 ] (ESM Methods : Model overview). The CATE for an individual is conditional on their clinical characteristics, and represents the predicted differential effects of the two drug classes on HbA 1c outcome . The BCF framework also minimises confounding from indication bias and allows for flexibility in defining model structure and outputs, and is an extension of Bayesian additive regression tree (BART) counterfactual models [ 18 ]. The model development process consisted of a first step of propensity score estimation to minimise confounding due to prescribing by indication [ 19 ], (ESM Methods : Propensity score estimation), and a second step of model development, using the R packages bcf (version 2.0.1) [ 17 ] and sparseBCF (version 1.0) [ 19 ] packages. Variable selection, based on each variable’s splitting probabilities, was deployed to develop a parsimonious final model whilst maintaining predictive accuracy (ESM Methods : Variable selection). The propensity score was not included in the final predictor set as it did not meet our threshold for variable selection (ESM Methods : Final model fit); however, as a sensitivity analysis, we refitted the final model, including the propensity score in the predictor set and compared predictions across the two models. Currently, the standard BCF software cannot account for missing data [ 20 ], so we used a complete case analysis, informed by our previous study showing a limited impact of missing data on predicting CATE in a similar primary care dataset [ 21 ]. To evaluate the degree of model-predicted treatment effect heterogeneity, differential HbA 1c response—the difference in achieved HbA 1c between drug classes—was extracted from the final model for all individuals.

Variable importance was estimated based on best linear projection (ESM Methods : Variable importance). To assess how CATE estimates varied across major routine clinical features, we also summarised the marginal distributions of key predictor variables (sex, baseline HbA 1c , eGFR, current age and BMI) across subgroups defined by the degree of predicted glycaemic differences (SGLT2i benefit of 0–3, 3–5 or >5 mmol/mol [0–0.3, 0.3–0.5 or >0.5%]; GLP1-RA benefit of 0–3, 3–5 or >5 mmol/mol).

Model validation

Evaluating the accuracy of predicted CATE is a significant challenge since, in practice, true CATE estimates are unobserved as a single individual receives only one therapy, meaning the counterfactual outcome they would have had on the alternative therapy is unobserved [ 22 ]. As such, to validate predicted CATE estimates, we first split validation sets into subgroups based on predicted CATE estimates and then compared the average CATE estimate within each subgroup to estimates derived from a set of alternative models fitted to each of the subgroups in turn. These latter models target the average treatment effect (ATE) within a population of individuals (rather than the conditional average treatment effect [CATE]), with desirable properties justified in the literature [ 23 ]. This validation framework further develops the concordant–discordant approach previously proposed in Dennis et al [ 6 ]. If the average CATE estimates in each subgroup (from the BCF model) align with the ATE estimates from the alternative models, this provides evidence that ATEs are consistent across different inference methods within each subgroup. Restricting the ATE estimates for each subgroup allows for simpler comparison ATE models to be used, since the distribution of covariates in each subgroup is expected to be more consistent within each subgroup than for the complete data. For validation, subgroups were defined by decile of predicted CATE in CPRD and, owing to the smaller cohort size, by quintile in Tayside & Fife.

To estimate the ATEs within subgroups, we used regression adjustment as the primary approach, estimating the ATE as the average difference in HbA 1c outcome between individuals receiving each therapy class within each subgroup Bayesian linear regression, adjusting for the full covariate set used in the HbA 1c treatment selection model (full covariate set; Table 2 ), with all continuous predictors included as 3-knot restricted cubic splines [ 6 ]. As a sensitivity analysis, we estimated CATE using propensity score matching with and without regression adjustment (ESM Methods ).

As our overall dataset predominantly included individuals of white ethnicity, we assessed the accuracy of predicted HbA 1c treatment effects in a subgroup of individuals of South Asian, Black, Other and Mixed ethnicity. We also evaluated accuracy of predicted HbA 1c treatment effects in those with and without cardiovascular disease. We also evaluated the reproducibility of observed differences in HbA 1c response by sex in participants receiving GLP1-RA in the HARMONY clinical trial, the PRIBA prospective study, and Tayside & Fife.

Secondary outcomes

Specific cohorts were defined to evaluate each secondary outcome to mitigate selection bias and maximise the number of individuals available for analysis (ESM Fig. 2 ; ESM Methods : Secondary outcomes). All cohorts required complete predictor data for the HbA 1c -based treatment selection model. To evaluate treatment effect heterogeneities, subgroups were defined by the degree of predicted glycaemic differences (SGLT2i benefit of 0–3, 3–5 or >5 mmol/mol [0–0.3, 0.3–0.5 or >0.5%]; GLP1-RA benefit of 0–3, 3–5 or >5 mmol/mol). As for validation of differences in HbA 1c outcomes, we evaluated subgroup-level ATEs using regression adjustment as the primary approach, with propensity score matching with and without regression adjustment deployed as sensitivity analysis. For evaluation of new-onset cardiovascular and renal outcomes, the propensity score model was refitted incorporating baseline cardiovascular risk as an additional predictor (QRISK2 predicted probability of new-onset myocardial infarction or stroke [ 24 ]). Absolute HbA 1c response was evaluated by drug class as adjusted (full covariate set) HbA 1c change from baseline using Bayesian linear regression. To evaluate differences by drug class in 12 month weight change, we included all individuals with a recorded baseline weight (closest value to 2 years prior to treatment initiation) and a valid outcome weight. Treatment effects were estimated using an adjusted (full covariate set) Bayesian linear regression model with an interaction between the received treatment and the predicted HbA 1c treatment benefit subgroup, with adjustment for baseline weight. Similarly, differences in treatment discontinuation were estimated using adjusted (full covariate set) Bayesian logistic regression with a treatment-by-HbA 1c benefit subgroup interaction.

For longer-term outcomes, we included only individuals without the outcome of interest at therapy initiation, thus evaluating only incident events. Individuals were followed for up to 5 years using an intention-to-treat approach from the date of therapy initiation until the earliest of: the outcome of interest, the date of general practitioner (GP) practice deregistration or death, or the end of the study period. For each outcome, adjusted (full covariate set) Bayesian Cox proportional hazards models with treatment-by-HbA 1c benefit subgroup interactions were fitted with additional adjustment for QRISK2 predicted probability of new-onset myocardial infarction or stroke.

All analyses were conducted using R (version 4.1.2; R Foundation for Statistical Computing, Austria). We followed TRIPOD prediction model reporting guidance (ESM Materials ) [ 25 ].

We included 84,193 people with type 2 diabetes initiating SGLT2i and 28,081 initiating GLP1-RA (ESM Fig. 1 ). The mean age of individuals was 58.2 (SD=10.9) years, 66,248 (59%) were men, and 88,174 (79%) were of white ethnicity. Baseline clinical characteristics by initiated drug class are reported in Table 1 .

Model development

For the development of the 12 month HbA 1c response treatment selection model, individuals with a measured HbA 1c outcome were randomly split 60:40 into development ( n =31,346) and validation ( n =20,865) cohorts (ESM Fig. 1 ; Baseline characteristics by cohort: ESM Table 1 ). Mean unadjusted 12 month HbA 1c response (change from baseline in HbA 1c ) was −12.0 (SD 15.3) mmol/mol (−1.1% [SD 1.4%]) for SGLT2i and −11.7 (SD 17.6) mmol/mol (−1.1% [SD 1.6%]) for GLP1-RA.

After variable selection [ 26 ] (ESM Fig. 3 ), we identified multiple clinical factors predictive of HbA 1c response with SGLT2i (the reference drug class in the model), and multiple factors predictive of differential HbA 1c response with GLP1-RA compared with SGLT2i therapy (Table 2 ). The final BCF model was fitted to 27,319 (87.2% of the starting development cohort) individuals with complete data for all selected clinical factors. In sensitivity analysis, the model predictions for final BCF model were similar to the BCF model with the full covariate set (ESM Fig. 4 ). Overall model fit and performance statistics for predicting achieved HbA 1c outcome in internal validation for both the development and hold-out cohorts are reported in ESM Table 2 . The propensity score did not meet the criteria for variable selection, and model predictions were similar when adding a propensity score as an additional covariate as a sensitivity analysis (ESM Fig. 5 ). The variable selection and performance of the propensity score model are reported in ESM (ESM Fig. 6 – 7 ).

In the development cohort, the mean CATE across all individuals was a 0.1 mmol/mol (95% credible interval [CrI] −0.3, 0.5) (0.01% [95% CrI −0.03, 0.05]) benefit with GLP1-RA over SGLT2i, suggesting similar average efficacy of both therapies. However, between individuals, there was marked heterogeneity in the predicted CATE estimates (Fig. 1 a), with the model predicting a mean HbA 1c benefit on SGLT2i therapy for 13,110 (48%) individuals and on GLP1-RA for 14,209 (52%) individuals. In the development cohort, 4787 (17.5%) had a predicted HbA 1c benefit >3 mmol/mol (0.3%) (3 mmol/mol is used widely as minimally important difference in clinical trials) with SGLT2i over GLP1-RA, and 5551 (20.3%) had a predicted HbA 1c benefit >3 mmol/mol with GLP1-RA over SGLT2i.

figure 1

Predicted CATE effects and model calibration. ( a ) Distribution of CATE estimates for SGLT2i vs GLP1-RA in the CPRD development cohort; negative values reflect a predicted HbA 1c treatment benefit on SGLT2i and positive values reflect a predicted treatment benefit on GLP1-RA. ( b ) Calibration between ATE and predicted CATE estimates, by decile of predicted CATE in the development cohort. ( c ) Calibration of CATE estimates in the validation cohort. ATE estimates are adjusted for all the variables used in the treatment selection model (see Methods )

Model calibration

Calibration by decile of model-predicted CATE estimates was good in the development cohort ( n =27,319; Fig. 1 b), the hold-back CPRD validation cohort ( n =19,075, Fig. 1 c), and in propensity-matched cohorts (ESM Fig. 8 ).

In the external Scottish cohort (Tayside & Fife; n =2252 [1837 initiating SGLT2i, 415 initiating GLP1-RA]; baseline characteristics: ESM Table 1 ), a similar distribution of predicted CATE to CPRD was observed (Fig. 2 a), and there was a clear difference between upper (favouring GLP1-RA) and lower (favouring SGLT2i) quintiles, but modest calibration in middle quintiles (Fig. 2 b). Among 81 (3.6%) individuals with a model-predicted HbA 1c benefit >5 mmol/mol (>0.5%) for SGLT2i over GLP1-RA, there was a 7.4 mmol/mol (95% CrI 0.1, 14.8) (0.7% [95% CrI 0, 1.4]) benefit for SGLT2i (Fig. 2 c). In contrast, among 150 (6.7%) individuals with a model-predicted HbA 1c benefit >5 mmol/mol for GLP1-RA over SGLT2i, there was a 5.6 mmol/mol (95% CrI −0.9, 12.1) (0.5% [95% CrI −0.1, 1.1]) benefit for GLP1-RA.

figure 2

External validation in Tayside & Fife, Scotland ( n =2252). ( a ) Distribution of CATE estimates for SGLT2i vs GLP1-RA; negative values reflect a predicted glucose-lowering treatment benefit on SGLT2i and positive values reflect a predicted treatment benefit on GLP1-RA. ( b ) Calibration between adjusted ATE and predicted CATE estimates, by quintile of predicted CATE. ( c ) ATE estimates within subgroups defined by clinically meaningful CATE thresholds (SGLT2i benefit >5, 3–5 and 0–3 mmol/mol, GLP1-RA benefit >5, 3–5 and 0–3 mmol/mol). Bars represent 95% CrI

Model interpretability

Stratifying the combined development and validation cohorts ( n =46,394 with complete predictor data) into subgroups defined by predicted CATE, there were clear differences in clinical characteristics, with those having a greater predicted HbA 1c benefit with GLP1-RA over SGLT2i being predominantly female and older, with lower baseline HbA 1c , eGFR and BMI (Fig. 3 a–e, ESM Table 1 ). SGLT2i were predicted to have a greater HbA 1c benefit over GLP1-RA for 32% of those with baseline HbA 1c levels <64 mmol/mol (8%), compared to 67% of those with baseline HbA1c ≥86 mmol/mol (≥10%). An evaluation of relative variable importance identified the number of other current glucose-lowering drugs (a higher number of concurrent therapies favouring SGLT2i as the optimal treatment), sex, current age, and to a lesser extent BMI and HbA 1c as the most influential predictors (relative importance ≥3%). In contrast, microvascular complications and cardiovascular comorbidities had very modest effects on differential response (ESM Fig. 9 ).

figure 3

Distributions of major clinical characteristics predicting differential HbA 1c outcome with SGLT2i and GLP1-RA. Distributions of key differential clinical characteristics in the combined development and validation cohorts ( n =46,394 with complete predictor data) for subgroups defined by predicted HbA 1c outcome differences: SGLT2i benefit >5 mmol/mol, 3–5 mmol/mol and 0–3 mmol/mol, GLP1-RA benefit >5 mmol/mol, 3–5 mmol/mol and 0–3 mmol/mol. The box and whisker plots include median, first and third quartile, with outliers laying further than 1.5 times the interquartile range. ( a ) Percentage of male individuals in each of the subgroups. ( b ) Baseline HbA 1c . ( c ) eGFR. ( d ) Current age. ( e ) BMI

Replication of sex differences in glycaemic response in clinical trials

Whilst previous analyses of clinical trials and observational data for SGLT2i have shown a modestly greater HbA 1c response in men compared with women, which we additionally reproduced in Tayside & Fife (Fig. 4 a,b), sex differences in GLP1-RA response have not been clearly established. Here, we focused on individual-level randomised clinical trial data of GLP1-RA from the HARMONY programme (liraglutide [ n =389] and albiglutide [ n =1682]) [ 18 ], the PRIBA prospective cohort study (non-insulin treated participants only: liraglutide [ n =350], exenatide [ n =197], lixisenatide [ n =3]) [ 14 ], and Tayside & Fife ( n =415). Baseline characteristics for the cohorts are reported in ESM Table 1 . Across all studies, there was consistent evidence of a greater baseline HbA 1c adjusted glycaemic response in women vs men; this was most marked for liraglutide in the HARMONY 7 trial [ 7 ] where a 4.4 mmol/mol (95% CrI 2.2, 6.3) (0.4% [95% CrI 0.2, 0.6]) greater response in women vs men was observed.

figure 4

Differences in HbA 1c outcome by sex, in randomised clinical trial and observational datasets. All estimates are adjusted for baseline HbA 1c . Estimates lower than zero represent a greater HbA 1c reduction in male compared with female participants. Bars represent 95% CrI. ( a ) SGLT2i: point estimates for the trials meta-analysis and CPRD are reproduced from Dennis et al (2022) [ 6 ]. ( b ) GLP1-RA

Effect of targeting therapy based on differential HbA 1c outcome on other short- and long-term outcomes

Specific subpopulations were defined for each short-term outcome to maximise the number of eligible individuals for each analysis and based on the availability of observed outcome data (12 month HbA 1c change from baseline [to evaluate absolute response] n =87,835; 12 month weight change n =41,728; treatment discontinuation within 6 months [a proxy for tolerability] n =77,741) (ESM Fig. 2 ). Longer-term outcomes were evaluated up to 5 years from drug initiation, excluding individuals with a history of cardiovascular disease or CKD for MACE, heart failure, and adverse kidney (composite of ≥40% decline in eGFR or kidney failure [ 14 ]) outcomes ( n =52,052) and individuals with a history of retinopathy, neuropathy and nephropathy for microvascular outcome ( n =34,524). (ESM Fig. 2 ).

For HbA 1c change from baseline, of the 6856 individuals (7.8%) with a predicted HbA 1c benefit on SGLT2i of >5 mmol/mol (>0.5%), those who received SGLT2i had a 23.3 mmol/mol (95% CrI 22.6, 24.0) (2.1% [95% CrI 2.1, 2.2]) mean reduction in HbA 1c and those who received GLP1-RA had an 18.4 mmol/mol (95% CrI 17.6, 19.3) (1.7% [95% CrI 1.6, 1.8]) mean reduction in HbA 1c (Fig. 5 a). In contrast, of the 7293 individuals (8.3%) with a predicted HbA 1c benefit on GLP1-RA of >5 mmol/mol, those receiving GLP1-RA had a 15.7 mmol/mol (95% CrI 14.8, 16.6) (1.4% [95% CrI 1.4, 1.5]) mean reduction in HbA 1c , and those receiving SGLT2i had a 9.0 mmol/mol (95% CrI 8.2, 9.7) (0.8% [95% CrI 0.8, 0.9]) mean reduction in HbA 1c . Consistent differences were observed in individuals of South Asian, Black, Other and Mixed ethnicity (ESM Fig. 10 ), and those with and without a history of cardiovascular disease (ESM Fig. 11 ).

figure 5

Differences in short-term and long-term clinical outcomes with SGLT2i and GLP1-RA for subgroups defined by predicted HbA 1c response differences. ( a ) Twelve month HbA 1c change from baseline. ( b ) Twelve month weight change. ( c ) Six month risk of discontinuation. ( d ) HR for 5 year risk of new-onset microvascular complications (retinopathy, nephropathy or neuropathy). ( e ) HR for 5 year relative risk of MACE. ( f ) HR for 5 year risk of heart failure. HRs represent the relative risk for those treated with GLP1-RA in comparison with SGLT2i therapy, with a value under 1 favouring SGLT2i therapy. Data underlying the figure are reported in ESM Table 3 . Bars represent 95% CrI

Observed weight change was consistently greater for individuals treated with SGLT2i compared with GLP1-RA across all subgroups (Fig. 5 b). Short-term discontinuation was lower in those treated with the drugs predicted to have the greatest glycaemic benefit, mainly reflecting differences in SGLT2 discontinuation across predicted levels of differential glycaemic response (Fig. 5 c). Relative risk of new-onset microvascular complications also varied by subgroup, with a lower risk with SGLT2i vs GLP1-RA only in subgroups predicted to have a glycaemic benefit with SGLT2i (Fig. 5 d). HRs for the risk of new-onset MACE were similar overall (HR 1.02 [95% CrI 0.89, 1.18]) and by subgroup (Fig. 5 e). HRs for the risks of both new-onset heart failure and adverse kidney outcomes were lower with SGLT2i (heart failure HR 0.71 [95% CrI 0.59, 0.85]; CKD HR 0.41 [95% CrI 0.30, 0.56]) with no clear evidence of a difference by subgroup (Fig. 5 f, ESM Fig. 12 ). Results for all outcomes were consistent in propensity-matched cohorts (ESM Fig. 13 – 14 ).

Comparison of model predictions with our previously published treatment selection model for SGLT2i and DPP4i therapies

Predictions for HbA 1c response with SGLT2i from the SGLT2i v GLP1-RA treatment selection model were highly concordant ( R 2 >0.92) with those from our recently published SGLT2i vs DPP4i treatment selection model [ 6 ] (ESM Fig. 15 ). Estimating differential HbA 1c responses using both models in our study population with complete data ( n =82,933) suggested SGLT2i is the predicted optimal therapy for HbA 1c in 48.2% ( n =39,975) of individuals, GLP1-RA the predicted optimal therapy in 51.3% ( n =42,519), and DPP4i the optimal therapy for only 0.5% ( n =439).

Prototype treatment selection model

A prototype treatment selection model web calculator providing individualised predictions of differences in HbA 1c outcomes is available at: https://pm-cardoso.shinyapps.io/SGLT2_GLP1_calculator/ .

We have developed and validated a novel treatment selection algorithm using state-of-the-art Bayesian methods to predict differences in one-year glycaemic outcomes for SGLT2i and GLP1-RA therapies. Our evaluation shows that glycaemic response-based targeting of these two major drug classes to individuals with type 2 diabetes based on their characteristics can not only optimise glycaemic control, but may also associate with improved tolerability and reduced risk of new-onset microvascular complications. In contrast, we found limited evidence for heterogeneity in other clinical outcomes, with overall equipoise between the two therapies for new-onset MACE and a clear overall benefit with SGLT2i over GLP1-RA for new-onset heart failure and adverse kidney outcomes independent of differences in glycaemic efficacy (differences which themselves reflect differences in the clinical characteristics of individual patients). Predictions are based on routine clinical characteristics, meaning the model could be deployed in many countries worldwide where these agents are available, without the need for additional testing.

Our approach differs from notable recent studies that have attempted to subclassify people with type 2 diabetes or used dimensionality reduction to represent type 2 diabetes heterogeneity [ 6 , 27 , 28 ]. Whilst these approaches can provide important insight into underlying heterogeneity of type 2 diabetes, they, by definition, lose information about the specific characteristics of individual patients, meaning they could be suboptimal for accurately predicting the treatment or disease progression outcomes for individuals [ 29 ]. If subclassification approaches based on clinical features are to have potential clinical utility, they will need to be updated over time as an individual’s phenotype evolves [ 30 ]. In contrast, our ‘outcomes-based’ approach enables the prediction of optimal therapy when a treatment decision is made, uses the specific information available for a patient at that point in time and avoids subclassification.

Although BCF models are only causal under specific assumptions [ 31 ], our study might provide insights into differences in the possible underlying mechanisms of action of GLP1-RA and SGLT2i, and the clinical utility of these differences. The strongest predictor of a differential glycaemic response was the number of currently prescribed glucose-lowering therapies, which is a likely proxy of the degree of diabetes progression (and, therefore, underlying beta cell failure) of an individual. A plausible biological explanation for this proxy is an attenuated GLP1-RA response in individuals with markers of beta cell failure including longer diabetes duration and lower fasting C-peptide, as previously demonstrated in a prospective population-based analysis [ 7 ], with no evidence of differences for SGLT2i [ 31 ]. Whilst in contrast, post hoc analyses of clinical trials have found type 2 diabetes duration and beta cell function do not modify glycaemic outcomes with GLP1-RA [ 19 , 32 , 33 ], this may reflect trial inclusion criteria as participants had relatively higher beta cell function compared with population-based cohorts [ 34 ]. The favouring of GLP1-RA over SGLT2i in women is novel but is supported by our trial validation and recent pharmacokinetic data demonstrating higher circulating GLP1-RA drug concentrations and, consequently, greater HbA 1c reduction in female compared with male participants [ 33 ]. For SGLT2i, increased urinary glucose excretion likely explains the greater relative glycaemic efficacy with higher baseline HbA 1c and eGFR, which, in concordance with our analysis, has been previously demonstrated in trial data [ 35 ]. Given the lack of previous studies evaluating whether the relative glucose-lowering efficacy of the two drug classes is altered by baseline HbA 1c [ 6 ], an interesting finding is that our model suggests a greater relative glycaemic benefit with SGLT2i over GLP1-RA at higher baseline HbA 1c levels, which warrants further study. Of note, the comorbidities included in the final model had modest effects on HbA 1c and are likely to be proxy measures of factors underlying differential response to these therapies.

A further interesting finding is that mean HbA 1c response on both drug classes was similar, and weight loss slightly greater with SGLT2i, in contrast to RCTs where network meta-analysis suggests a greater glycaemic and weight efficacy of most individual GLP1-RA over SGLT2i [ 12 , 36 , 37 ]. The relative average equipoise between the two drug classes in our study is likely indicative of a diminished real-world response to GLP1-RA, a phenomenon also documented in other real-world studies [ 37 , 38 ], which may relate to reduced real-world adherence to GLP1-RA [ 38 ].

Our study represents the second application of our novel validation framework for precision medicine models, which, in the absence of true observed outcomes (for an individual patient on one therapy, the counterfactual outcome they would have had on an alternative therapy cannot be observed [ 39 ]), evaluates accuracy in subgroups defined by predicted CATE. The previous study developed a treatment selection model for SGLTi2 vs DPP4i therapy in an independent dataset. Although this previous model demonstrated marked heterogeneity in the relative glycaemic outcome, most (84%) individuals had a greater glycaemic reduction with SGLT2i. In contrast, this GLP1-RA/SGLT2i model shows greater heterogeneity in treatment effects but with equipoise on ATE between the two therapies (52% favouring GLP1-RA). Furthermore, we demonstrate that optimising therapy based on predicted glycaemic response may lower microvascular complication risk, a finding concordant with evidence from the UKPDS study on the importance of good glycaemic control to lower the risk of microvascular disease [ 23 , 40 ].

Further developments to this model could include the incorporation of non-routine and pharmacogenetic markers (recently identified for GLP1-RA) [ 41 ], and additional glucose-lowering drug classes, in particular, off-patent sulfonylureas and pioglitazone, to support the deployment of the algorithm in lower-income countries where the availability of newer medications may be limited. Assessment of semaglutide, a GLP1-RA with potent glycaemic effect excluded here due to low numbers prescribed during the period of data availability, and tirzepatide, a dual glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 receptor agonist not currently available in the UK, is an important area for future research as our model may benefit from recalibration for these newer therapies. Although our ethnicity-specific validation suggests good performance in individuals of South Asian, Black, Other and Mixed ethnicity, setting and ethnicity-specific validation and optimisation would also improve future clinical utility. Given the possibility of selection bias due to non-random treatment assignment, validation in a dataset where individuals were randomised to therapy would further strengthen the evidence for model deployment. However, few active comparator trials of these two drug classes have been conducted [ 8 ] and, to our knowledge, none are available for data sharing. Ultimately, research, likely in even larger datasets, is needed on whether individualised models for other short- and long-term outcomes beyond glycaemia, particularly cardiorenal disease, can further improve current prescribing approaches [ 42 ]. Finally, a limitation of our study is that despite being state-of-the-art and with a key advantage of allowing estimation of predictions with uncertainty, and so facilitating more transparent evaluation, the BCF methods we applied are subject to ongoing development in several key areas such as variable selection [ 18 , 19 ], scalability and handling of missing data [ 20 ].

In conclusion, our study demonstrates a clear potential for targeted prescribing of GLP1-RA and SGLT2i to individual people with type 2 diabetes based on their clinical characteristics to improve glycaemic outcomes, tolerability and risk of microvascular complications. This provides an important advance on current type 2 diabetes guidelines, which only recommend preferentially prescribing these therapies to individuals with, or at high risk of, cardiorenal disease, with no clear evidence to choose between the two drug classes. Precision type 2 diabetes prescribing based on routinely available characteristics has the potential to lead to more informed and evidence-based decisions on treatment for people with type 2 diabetes worldwide in the near future.


95% Credible interval

Alanine aminotransferase

Average treatment effect

Bayesian causal forest

Conditional average treatment effect

Chronic kidney disease

UK Clinical Practice Research Datalink

DPP4 inhibitors

Glucagon-like peptide-1

Glucagon-like peptide-1 receptor agonists

Major adverse cardiovascular events

Sodium–glucose cotransporter 2 inhibitors

Dennis JM (2020) Precision medicine in type 2 diabetes: using individualized prediction models to optimize selection of treatment. Diabetes 69(10):2075–2085. https://doi.org/10.2337/dbi20-0002

Article   CAS   PubMed   PubMed Central   Google Scholar  

Davies MJ, Aroda VR, Collins BS et al (2022) Management of hyperglycemia in type 2 diabetes, 2022. A consensus report by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetologia 65(12):1925–1966

Cefalu WT, Kaul S, Gerstein HC et al (2018) Cardiovascular outcomes trials in type 2 diabetes: where do we go from here? Reflections from a diabetes care editors’ expert forum. Diabetes Care 41(1):14–31. https://doi.org/10.2337/dci17-0057

Article   CAS   PubMed   Google Scholar  

McGovern A, Feher M, Munro N, de Lusignan S (2017) Sodium-glucose co-transporter 2 (SGLT2) inhibitor: comparing trial data and real-world use. Diabetes Ther 8:365–376. https://doi.org/10.1007/s13300-017-0254-7

Shields BM, Dennis JM, Angwin CD et al (2023) Patient stratification for determining optimal second-line and third-line therapy for type 2 diabetes: the TriMaster study. Nat Med 29:376–383. https://doi.org/10.1038/s41591-022-02120-7

Dennis JM, Young KG, McGovern AP et al (2022) Development of a treatment selection algorithm for SGLT2 and DPP-4 inhibitor therapies in people with type 2 diabetes: a retrospective cohort study. Lancet Digital Health 4(12):873–883. https://doi.org/10.1016/S2589-7500(22)00174-1

Article   Google Scholar  

Jones AG, McDonald TJ, Shields BM et al (2016) Markers of β-cell failure predict poor glycemic response to GLP-1 receptor agonist therapy in type 2 diabetes. Diabetes Care 39(20):250–257. https://doi.org/10.2337/dc15-0258

Dawed AY, Mari A, Brown A et al (2023) Pharmacogenomics of GLP-1 receptor agonists: a genome-wide analysis of observational data and large randomised controlled trials. Lancet Diabetes Endocrinol 11(1):33–41. https://doi.org/10.1016/S2213-8587(22)00340-0

Wolf A, Dedman D, Campbell J et al (2019) Data resource profile: Clinical Practice Research Datalink (CPRD) Aurum. Int J Epidemiol 48(6):1740–1740g. https://doi.org/10.1093/ije/dyz034

Article   PubMed   PubMed Central   Google Scholar  

Rodgers LR, Weedon MN, Henley WE, Hattersley AT, Shields BM (2017) Cohort profile for the MASTERMIND study: using the clinical practice research datalink (CPRD) to investigate stratification of response to treatment in patients with type 2 diabetes. BMJ Open 7:e017989. https://doi.org/10.1136/bmjopen-2017-017989

National Institute for Health and Care Excellence (2015) Type 2 diabetes in adults: management. Available from: https://www.nice.org.uk/guidance/ng28 . Accessed 5 Apr 2023

Tsapas A, Avgerinos I, Karagiannis T et al (2020) Comparative effectiveness of glucose-lowering drugs for type 2 diabetes. Ann Int Med 173(4):278–286. https://doi.org/10.7326/M20-0864

Article   PubMed   Google Scholar  

Pratley RE, Nauck MA, Barnett AH et al (2014) Once-weekly albiglutide versus once-daily liraglutide in patients with type 2 diabetes inadequately controlled on oral drugs (HARMONY 7): a randomised, open-label, multicentre, non-inferiority phase 3 study. Lancet Diabetes Endocrinol 2(4):289–297. https://doi.org/10.1016/S2213-8587(13)70214-6

Grams ME, Brunskill NJ, Ballew SH et al (2022) Development and validation of prediction models of adverse kidney outcomes in the population with and without diabetes. Diabetes Care 45(9):2055–2063. https://doi.org/10.2337/dc22-0698

Inker LA, Eneanya ND, Coresh J et al (2021) New creatinine- and cystatin c-based equations to estimate GFR without race. N Engl J Med 385(19):1737–1749. https://doi.org/10.1056/NEJMoa2102953

Hahn PR, Murray JS, Carvalho CM (2020) Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (with discussion). Bayesian Anal 15(3):965–1056. https://doi.org/10.1214/19-BA1195

Article   MathSciNet   Google Scholar  

Caron A, Baio G, Manopoulou I (2021) Shrinkage Bayesian causal forests for heterogeneous treatment effects estimation. J Comput Graph Stat 31(4):1202–1214. https://doi.org/10.1080/10618600.2022.2067549

Hill JL (2011) Bayesian nonparametric modelling for causal inference. Bayesian Nonparametric 20(1):217–240. https://doi.org/10.1198/jcgs.2010.08162

Caron A (2020) SparseBCF: sparse Bayesian causal forest for heterogeneous treatment. R package version 1.0

Kapelner A, Bleich J (2016) bartMachine: machine learning with Bayesian additive regression trees. J Stat Softw 70(4):1–40

Cardoso P, Dennis JM, Bowden J, Shields BM, McKinley TJ (2024) Dirichlet process mixture models to impute missing predictor data in counterfactual prediction models: an application to predict optimal type 2 diabetes therapy. BMC Med Inform Decis Mak 24(1):12. https://doi.org/10.1186/s12911-023-02400-3

Holland PW (1985) Statistics and causal inference. J Am Stat Assoc 81(396):945–960. https://doi.org/10.1080/01621459.1986.10478354

Harrell FE (2016) Regression modeling strategies. Springer International Publishing

Hippisley-Cox J, Coupland C, Vinogradova Y et al (2008) Predicting cardiovascular risk in England and Wales: prospective derivation and validation of QRISK2. BMJ 336(7659):1475–1482. https://doi.org/10.1136/bmj.39609.449676.25

Collins GS, Reitsma JB, Altman DG, Moons KG (2015) Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. Ann Int Med 162(1):55–63. https://doi.org/10.7326/M14-0697

Ahlqvist E, Storm P, Käräjämäki A et al (2018) Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol 6(5):361–369. https://doi.org/10.1016/S2213-8587(18)30051-2

Nair ATN, Wesolowska-Andersen A, Brorsson C et al (2022) Heterogeneity in phenotype, disease progression and drug response in type 2 diabetes. Nat Med 28:982–988. https://doi.org/10.1038/s41591-022-01790-7

Udler MS, Kim J, von Grotthuss M et al (2018) Type 2 diabetes genetic loci informed by multi-trait associations point to disease mechanisms and subtypes: a soft clustering analysis. PLOS Med 15(9):e1002654. https://doi.org/10.1371/journal.pmed.1002654

Dennis JM, Shields BM, Henley WE, Jones AG, Hattersley AT (2019) Clusters provide a better holistic view of type 2 diabetes than simple clinical features - author’s reply. Lancet Diabetes Endocrinol 7(9):669. https://doi.org/10.1016/S2213-8587(19)30250-5

Zaharia OP, Strassburger K, Strom A et al (2019) Risk of diabetes-associated diseases in subgroups of patients with recent-onset diabetes: a 5-year follow-up study. Lancet Diabetes Endocrinol 7(9):684–694. https://doi.org/10.1016/S2213-8587(19)30187-1

Gallwitz B, Dagogo-Jack S, Thieu V et al (2018) Effect of once-weekly dulaglutide on glycated haemoglobin (HbA1c) and fasting blood glucose in patient subpopulations by gender, duration of diabetes and baseline HbA1c. Diabetes Obes Metab 20(2):409–418. https://doi.org/10.1111/dom.13086

Mathieu C, Prato S, Botros F et al (2018) Effect of once weekly dulaglutide by baseline beta-cell function in people with type 2 diabetes in the AWARD programme. Diabetes Obes Metab 20(8):2023–2028. https://doi.org/10.1111/dom.13313

Bonadonna R, Blonde L, Antsiferov M et al (2017) Lixisenatide as add-on treatment among patients with different β-cell function levels as assessed by HOMA-β index. Diabetes Metab Res Rev 33(6):e2897. https://doi.org/10.1002/dmrr.2897

Overgaard RV, Hertz CL, Ingwersen SH, Navarria A, Drucker DJ (2021) Levels of circulating semaglutide determine reductions in HbA1c and body weight in people with type 2 diabetes. Cell Rep Med 2(9):100387. https://doi.org/10.1016/j.xcrm.2021.100387

Young KG, McInnes EH, Massey RJ et al (2023) Treatment effect heterogeneity following type 2 diabetes treatment with GLP1-receptor agonists and SGLT2-inhibitors: a systematic review. Commun Med 3(1):131. https://doi.org/10.1038/s43856-023-00359-w

Edelman S, Polonsky W (2017) Type 2 diabetes in the real world: the elusive nature of glycemic control. Diabetes Care 40(11):1425–1432. https://doi.org/10.2337/dc16-1974

Weiss T, Yang L, Carr R et al (2022) Real-world weight change, adherence, and discontinuation among patients with type 2 diabetes initiating glucagon-like peptide-1 receptor agonists in the UK. BMJ Open Diabetes Res Care 10(1):e002517. https://doi.org/10.1136/bmjdrc-2021-002517

Carls G, Tuttle E, Tan R-D et al (2017) Understanding the gap between efficacy in randomized controlled trials and effectiveness in real-world use of GLP-1 RA and DPP-4 therapies in patients with type 2 diabetes. Diabetes Care 40(11):1469–1478. https://doi.org/10.2337/dc16-2725

Stratton IM, Adler AI, Neil HAW et al (2000) Association of glycaemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. Br Med J 321(7258):405–412. https://doi.org/10.1136/bmj.321.7258.405

Article   CAS   Google Scholar  

Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HAW (2008) 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med 359(15):1577–1589. https://doi.org/10.1056/NEJMoa0806470

Patoulias DI, Katsimardou A, Kalogirou MS et al (2020) Glucagon-like peptide-1 receptor agonists or sodium–glucose cotransporter-2 inhibitors as add-on therapy for patients with type 2 diabetes? A systematic review and meta-analysis of surrogate metabolic endpoints. Diabetes Metab 46(4):272–279. https://doi.org/10.1016/j.diabet.2020.04.001

McMurray JJ, Sattar N (2022) Heart failure: now centre-stage in diabetes. Lancet Diabetes Endocrinol 10(10):689–691. https://doi.org/10.1016/S2213-8587(22)00249-2

Download references

Author information

Authors and affiliations.

Institute of Biomedical & Clinical Science, University of Exeter Medical School, Exeter, UK

Pedro Cardoso, Katie G. Young, Rhian Hopkins, Andrew P. McGovern, Jack Bowden, Andrew T. Hattersley, Angus G. Jones, Beverley M. Shields, Trevelyan J. McKinley & John M. Dennis

Division of Molecular & Clinical Medicine, Ninewells Hospital and Medical School, University of Dundee, Dundee, UK

Anand T. N. Nair, Eram Haider, Piyumanga Karunaratne, Louise Donnelly & Ewan R. Pearson

Institute of Health Informatics, University College London, London, UK

Bilal A. Mateen

Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK

Naveed Sattar

Diabetes Trials Unit, Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, Oxford, UK

Rury R. Holman

Oxford NIHR Biomedical Research Centre, Churchill Hospital, Oxford, UK

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to John M. Dennis .

Ethics declarations


This article is based in part on data from the CPRD obtained under license from the UK Medicines and Healthcare products Regulatory Agency. CPRD data are provided by patients and collected by the UK National Health Service (NHS) as part of their care and support. Approval for CPRD data access and the study protocol was granted by the CPRD Independent Scientific Advisory Committee (eRAP protocol number: 22_002000). This publication is based in part on research using data from GSK that has been made available through secured access. GSK has not contributed to or approved, and is not in any way responsible for, the contents of this publication. The PRIBA study was funded by a National Institute for Health Research (UK) Doctoral Research Fellowship (DRF-2010-03-72, AGJ) and supported by the National Institute for Health Research (NIHR) Clinical Research Network. The authors thank the members of the Predicting Response to Incretin Based Agents (PRIBA) study group and all cohort participants (see ESM for a list of PRIBA study group members). ATH and BMS are supported by the NIHR Exeter Clinical Research Facility; the views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. PC, KGY, JB, TJM and JMD are supported by Research England’s Expanding Excellence in England (E3) fund. The authors acknowledge contributions from the wider MASTERMIND consortium who supported this work (see ESM for a list of MASTERMIND consortium members). The authors acknowledge support from the National Institute for Health and Care Research Exeter Biomedical Research Centre. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

Data availability

The UK routine clinical data analysed during the current study are available in the CPRD repository (CPRD; https://cprd.com/research-applications ), but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. For re-using these data, an application must be made directly to CPRD. Data from Scotland are anonymised real-world medical records available by request through the Scottish Care Information-Diabetes Collaboration, Tayside & Fife, Scotland unit ( https://www.sci-diabetes.scot.nhs.uk/ ). Clinical trial data are not publicly available for access an application must be made directly to GSK and www.ClinicalStudyDataRequest.com .

Code availability

All R code used for the analysis is provided at https://github.com/Exeter-Diabetes/CPRD-Pedro-SGLT2vsGLP1 .

This research was funded by the Medical Research Council (UK) (MR/N00633X/1) and a BHF-Turing Cardiovascular Data Science Award (SP/19/6/34809).

Authors’ relationships and activities

APM declares previous research funding from Eli Lilly and Company, Pfizer and AstraZeneca. BAM holds an honorary post at University College London for the purposes of carrying out independent research, and declares payments to their institution from the Medical Research Council (MRC), Health Data Research UK (HDRUK) and British Heart Foundation (BHF). NS declares personal fees from Abbott Diagnostics, Afimmune, Amgen, Astra Zeneca, Boehringer Ingelheim, Eli Lilly, Hanmi Pharmaceuticals, Merck Sharp & Dohme, Novartis, Novo Nordisk, Pfizer and Sanofi and grants to his University from AstraZeneca, Boehringer Ingelheim, Novartis and Roche Diagnostics. RRH reports research support from AstraZeneca, Bayer and Merck Sharp & Dohme, and personal fees from Anji Pharmaceuticals, Bayer, Novartis and Novo Nordisk. JB is an employee of Novo Nordisk, outside of the submitted work. ERP has received honoraria for speaking from Lilly, Novo Nordisk and Illumina. AGJ has received research funding from the Novo Nordisk foundation. Representatives from GSK, Takeda, Janssen, Quintiles, AstraZeneca and Sanofi attend meetings as part of the industry group involved with the MASTERMIND consortium. No industry representatives were involved in the writing of the manuscript or analysis of data. For all authors these are outside the submitted work. All other authors declare that there are no relationships or activities that might bias, or be perceived to bias, their work

Contribution statement

PC, JMD, BMS, TJM, ATH, AGJ and ERP conceived and designed the study. PC, with support from JMD, BMS and TJM analysed the data and developed the code. KGY, RH and APM helped with curating the CPRD dataset. ATNN, PK, EH and LD helped analyse the Scottish independent dataset. All authors contributed to the writing of the article, provided support for the analysis and interpretation of results, critically revised the article and approved the final article. TJM and JMD are the guarantors of this work.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Trevelyan J. McKinley and John M. Dennis are joint senior authors.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (PDF 3535 KB)

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cardoso, P., Young, K.G., Nair, A.T.N. et al. Phenotype-based targeted treatment of SGLT2 inhibitors and GLP-1 receptor agonists in type 2 diabetes. Diabetologia (2024). https://doi.org/10.1007/s00125-024-06099-3

Download citation

Received : 17 October 2023

Accepted : 04 January 2024

Published : 22 February 2024

DOI : https://doi.org/10.1007/s00125-024-06099-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • Bayesian non-parametric modelling
  • GLP1-receptor agonists
  • Heterogeneous treatment effects
  • Precision medicine
  • SGLT2-inhibitors
  • Type 2 diabetes
  • Find a journal
  • Publish with us
  • Track your research


Gargling away the 'bad' bacteria in type 2 diabetes

More than bad breath, there is growing evidence that ongoing inflammation in the mouth, like with gum disease, is associated with serious diseases such as Alzheimer's disease or type 2 diabetes. Now, researchers from Osaka University have identified an easy way to fight bacteria that might cause such problems.

In a study published this month in Scientific Reports , the researchers reported that when people with type 2 diabetes gargled with an antiseptic mouthwash, the numbers of periodontitis-related bacteria decreased. Excitingly, some patients with reduced bacteria also achieved much better control of their blood sugar, hinting at promising future clinical applications.

"There are three highly virulent bacterial species that are linked to periodontitis, or diseases of the tissues surrounding the teeth," explains lead author of the study Saaya Matayoshi. "We decided to see if we could reduce these three species -- Porphyromonas gingivalis , Treponema denticola , and Tannerella forsythia -- in patients with type 2 diabetes using a mouthwash containing the antiseptic chlorhexidine gluconate."

To do this, the researchers took monthly or bimonthly saliva and blood samples from 173 patients over an entire year. With the saliva, the researchers noted the presence or absence of the three bacterial species, and with the blood samples, they measured HbA1c levels as a marker of blood-sugar control. Importantly, for the first 6 months of the study, the patients gargled with water, whereas for the second 6 months they gargled with the antiseptic mouthwash. In this way, the research team could see whether gargling itself was effective for reducing bacteria, or whether mouthwash was more effective.

"We were unsurprised to see that gargling with water had no effects on bacterial species or HbA1c levels," explains Kazuhiko Nakano, senior author of the study. "However, there was an overall reduction in bacterial species when the patients switched to mouthwash, as long as they were gargling at least twice a day."

The researchers also found that, although there were no overall changes in HbA1c levels when patients gargled with the antiseptic mouthwash, there appeared to be large variations in individual responses. For example, when they split the group into younger and older patients, younger patients had greater reductions in bacterial species and significantly better blood-sugar control with the mouthwash compared with water.

Given that poor oral health is linked to serious disease, simple methods to improve oral hygiene have important ramifications. If researchers can identify patients who are likely to respond well to antiseptic mouthwash, this easy-to-use treatment may improve the lives of people with periodontitis-linked diseases such as diabetes, dementia, cardiovascular disease, and respiratory tract infections.

  • Wounds and Healing
  • Diseases and Conditions
  • Alzheimer's
  • Multiple Sclerosis
  • South Beach diet
  • Alzheimer's disease
  • Tropical disease
  • Microorganism
  • Diabetic diet

Story Source:

Materials provided by Osaka University . Note: Content may be edited for style and length.

Journal Reference :

  • Saaya Matayoshi, Fumikazu Tojo, Yuto Suehiro, Makoto Okuda, Misato Takagi, Marin Ochiai, Maika Kadono, Yusuke Mikasa, Rena Okawa, Ryota Nomura, Yoshito Itoh, Naoto Itoh, Kazuhiko Nakano. Effects of mouthwash on periodontal pathogens and glycemic control in patients with type 2 diabetes mellitus . Scientific Reports , 2024; 14 (1) DOI: 10.1038/s41598-024-53213-x

Cite This Page :

  • Anchors Holding Antarctic Land-Ice Shrinking
  • Compound Vital for All Life and Life's Origin
  • How Competition Can Help Promote Cooperation
  • Giant New Snake Species Identified in the Amazon
  • Chemists Synthesize Unique Anticancer Molecules
  • Neutron Star at Heart of Supernova Remnant
  • The Search for More Temperate Tatooines
  • Steering Light With Supercritical Coupling
  • Dramatic Improvements in Crohn's Patients
  • Record-Breaking Quasar Discovered

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List

Logo of nutrients

Reversing Type 2 Diabetes: A Narrative Review of the Evidence

Sarah j hallberg.

1 Virta Health, 535 Mission Street, San Francisco, CA 94105, USA; moc.htlaehatriv@einimahs

2 Indiana University Health Arnett, Lafayette, IN 47904, USA; gro.htlaehui@nubzaht

3 Indiana University School of Medicine, Indianapolis, IN 46202, USA

Victoria M Gershuni

4 Department of Surgery, Perelman School of Medicine University of Pennsylvania, Philadelphia, PA 19104, USA; moc.liamg@dminuhsregairotciv

Tamara L Hazbun

Shaminie j athinarayanan.

Background: Type 2 diabetes (T2D) has long been identified as an incurable chronic disease based on traditional means of treatment. Research now exists that suggests reversal is possible through other means that have only recently been embraced in the guidelines. This narrative review examines the evidence for T2D reversal using each of the three methods, including advantages and limitations for each. Methods: A literature search was performed, and a total of 99 original articles containing information pertaining to diabetes reversal or remission were included. Results: Evidence exists that T2D reversal is achievable using bariatric surgery, low-calorie diets (LCD), or carbohydrate restriction (LC). Bariatric surgery has been recommended for the treatment of T2D since 2016 by an international diabetes consensus group. Both the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD) now recommend a LC eating pattern and support the short-term use of LCD for weight loss. However, only T2D treatment, not reversal, is discussed in their guidelines. Conclusion: Given the state of evidence for T2D reversal, healthcare providers need to be educated on reversal options so they can actively engage in counseling patients who may desire this approach to their disease.

1. Introduction

According to 2017 International Diabetes Federation (IDF) statistics, there are approximately 425 million people with diabetes worldwide [ 1 ]. In the United States, there are an estimated 30.3 million adults living with diabetes, and its prevalence has been rising rapidly, with at least 1.5 million new diabetes cases diagnosed each year [ 2 ]. Diabetes is a major public health epidemic despite recent advances in both pharmaceutical and technologic treatment options.

Type 2 diabetes (T2D) has long been identified as an incurable chronic disease. The best outcome that has been expected is amelioration of diabetes symptoms or slowing its inevitable progression. Approximately 50% of T2D patients will need insulin therapy within ten years of diagnosis [ 3 ] Although in the past diabetes has been called chronic and irreversible, the paradigm is changing [ 4 , 5 ].

The recent 2016 World Health Organization (WHO) global report on diabetes added a section on diabetes reversal and acknowledged that it can be achieved through weight loss and calorie restriction [ 4 ]. “Diabetes reversal” is a term that has found its way into scientific articles and the lay press alike; “remission” has also been used. While the exact criteria are still debated, most agree that a hemoglobin A1c (HbA1c) under the diabetes threshold of 6.5% for an extended period of time without the use of glycemic control medications would qualify [ 6 ]. Excluding metformin from the glycemic control medications list, as it has indications beyond diabetes, may also be a consideration [ 7 , 8 ]. Likewise, terms such as “partial” (HbA1c <6.5 without glycemic control medications for 1 year) or “complete” (HbA1c <5.7 without glycemic control medications for 1 year) remission have been defined by an expert panel as more evidence accumulates that points to the possibility of avoiding the presumably progressive nature of T2D [ 9 ]. It is important to note that the term “cure” has not been applied to T2D, as there does exist the potential for re-occurrence, which has been well documented in the literature.

Despite the growing evidence that reversal is possible, achieving reversal is not commonly encouraged by our healthcare system. In fact, reversal is not a goal in diabetes guidelines. Specific interventions aimed at reversal all have one thing in common: they are not first-line standard of care. This is important, because there is evidence suggesting that standard of care does not lead to diabetes reversal. This raises the question of whether standard of care is really the best practice. A large study by Kaiser Permanente found a diabetes remission rate of 0.23% with standard of care [ 10 ]. The status quo approach will not reverse the health crisis of diabetes.

A significant number of studies indicate that diabetes reversal is achievable using bariatric surgery, while other approaches, such as low-calorie diets (LCD) or carbohydrate restriction (LC), have also shown effectiveness in an increasing number of studies. This review will examine each of these approaches, identifying their beneficial effects, supporting evidence, drawbacks, and degree of sustainability.

2. Materials and Methods

A literature search was performed as appropriate for narrative reviews, including electronic databases of PubMed, EMBASE, and Google Scholar from 1970 through December 2018. We reviewed English-language original and review articles found under the subject headings diabetes, bariatric surgery, metabolic surgery, very low-calorie diet, calorie restriction, low carbohydrate diet, ketogenic diet, diabetes remission, and diabetes reversal. References of the identified publications were searched for more research articles to include in this review. Selected studies were reviewed and evaluated for eligibility for inclusion in this review based on their relevance for diabetes reversal and remission. Either remission or reversal needed to be discussed in the paper or the results were consistent with these terms for inclusion. Randomized clinical trials and intervention-based studies were given emphasis for inclusion.

A total of 99 original articles containing information pertaining to diabetes reversal or remission were included in this narrative review.

3. Results and Discussion

3.1. bariatric surgery.

Bariatric surgery has long been recognized as a potential treatment for both morbid obesity and the metabolic processes that accompany it, specifically T2D. While the efficacy of T2D reversal depends on the choice of procedure, there is unilateral improvement in glycemia following operation [ 11 ], and bariatric surgery has been found to be superior to intensive T2D medical management. Accordingly, in 2016, the second Diabetes Surgery Summit (DSS-II) released recommendations, endorsed by 45 medical and scientific societies worldwide, to use bariatric surgery as a treatment for T2D (bariatric surgery is currently approved by the 2016 recommendations for adults with a body mass index (BMI) >40, or >35 kg/m 2 with obesity-related comorbidities) [ 12 ]. Of interest is the consistent finding that glycemic improvements occur rapidly, often within hours to days, and precede weight loss, which likely represents the enteroendocrine responses to altered flow of intestinal contents (i.e., bile acid signaling and changes in microbiota and their metabolome) [ 13 , 14 , 15 , 16 , 17 , 18 , 19 ].

The most commonly performed bariatric surgeries in the United States include laparoscopic and robotic Roux-en-Y Gastric Bypass (RYGB) or Sleeve Gastrectomy (SG). While surgical treatment is based on the principles of restriction and intestinal malabsorption, evidence suggests that there are more complex mechanisms at play. Bariatric surgery has consistently been shown to dramatically and rapidly improve blood glucose [ 20 ] while allowing decreased oral hypoglycemic medications and insulin use, effectively reversing diabetes in up to 80% of patients [ 21 ] in the short term. In addition to early post-operative improvement in blood glucose and insulin sensitivity, bariatric surgery has also been shown to cause alterations in GI hormone release, including ghrelin, leptin, cholecystokinin (CCK), peptide-tyrosine-tyrosine (PYY), and glucagon-like peptide 1 (GLP-1), that may impact feeding behavior via the gut–brain axis in addition to modulating euglycemia [ 22 ]. Furthermore, microbial changes in the human gut have been linked to obesity, and surgical alterations to gastrointestinal anatomy have been associated with dramatic changes in gut microbiota populations with reversion from an “obesogenic” to a lean bacterial population [ 13 , 14 , 16 , 19 , 23 , 24 ].

Long-term outcomes from bariatric surgery depend on multiple factors, including type of surgery performed, patient comorbidities, patient readiness for lifelong dietary change, and ongoing surveillance. While bariatric surgery has been demonstrated to be safe and effective overall, it is important to recognize that it is not without risks. Each patient must weigh the risks and benefits associated with untreated morbid obesity versus those associated with surgery or effective dietary management and choose accordingly. Surgery of any type can be associated with complications leading to morbidity or mortality; the complication rates have been stated to be as high as 13% and 21% for SG and RYGB, respectively. The postoperative mortality rate is 0.28–0.34% for SG and 0.35–0.79% for RYGB; in comparison, an elective laparoscopic cholecystectomy is associated with overall complication rates of 9.29% and with a 30-day mortality rate of 0.15–0.6%, depending on the series [ 25 , 26 ]. Significant complications include anastomotic leak or hemorrhage, post-operative readmission, need for reoperation, post-operative hypoglycemia, dumping syndrome, worsening acid reflux, marginal ulceration, and micronutrient deficiencies [ 25 , 26 , 27 , 28 , 29 ].

It is important to consider that while short-duration studies have shown early resolution of comorbidities following bariatric procedures, when followed for multiple decades, there may be decreased efficacy of disease resolution and increased incidence of hospital admission long-term. Long-term reversal of T2D and true glucose homeostasis remain uncertain. Weight loss after surgery is a significant predictor of a return to euglycemia post-operatively. Multiple studies have reported initial T2D remission rates as high as 80% [ 30 , 31 ], however, long-term remission is less durable. The five-year follow-up outcomes of the SLEEVEPASS RCT found complete or partial remission of T2D in 37% of SG and 45% of RYGB patients, which is similar to other studies showing long-term T2D remission in up to a third of patients [ 32 ]. In the large prospective cohort study Longitudinal Assessment of Bariatric Surgery 2 (LABS-2), the investigators found that long-term diabetes remission after RYGB was higher than predicted by weight loss alone, which suggests that the surgery itself impacts metabolic factors that contribute to disease management [ 31 ]. Similarly, the STAMPEDE trial—an RCT that followed 150 patients with T2D who were randomized to intensive medical intervention (IMT) versus IMT plus RYGB versus IMT plus SG for diabetes resolution (defined as HbA1c <6.0%) and followed for five years—revealed increased rates of T2D resolution with RYGB (29%) and SG (23%) compared to IMT alone (5%) ( Figure 1 ). The surgery cohort also demonstrated greater weight loss and improvements in triglycerides, HDL, need for insulin, and overall quality of life [ 33 , 34 , 35 ].

An external file that holds a picture, illustration, etc.
Object name is nutrients-11-00766-g001.jpg

( A ) Mean changes of hemoglobin A1c (HbA1c) from baseline to last published date for each study retrieved to represent the three methods of reversal; ( B ) mean changes of weight from baseline to last published date for each studies retrieved to represent the three methods of reversal. Note: We chose these three studies to represent the three methods of reversal based on publication date and relevance to diabetes reversal. Note that baseline characteristics differ. Surgery trial examined by sleeve gastrectomy and Roux-en-Y gastric bypass separately and were represented as sleeve and bypass in the graph. Surgery: STAMPEDE [ 34 , 35 ]. Low-calorie diets (LCD): DIRECT [ 65 , 66 ]; carbohydrate restriction (LC): IUH [ 99 , 107 ].

Despite the likelihood of improved glycemic control, there are significant financial costs for the patient, health system, and insurance companies associated with bariatric surgery (U.S. average of $14,389) [ 36 ]. Despite the high initial cost of surgery, Pories and colleagues found that prior to surgery, patients spend over $10,000 per year on diabetes medications; after RYGB, the annual cost falls to less than $2000, which represents an $8000 cost savings at the individual level [ 30 ]. Furthermore, economic analyses show that surgery is likely to be cost-effective, especially in patients who are obese [ 37 , 38 ]. In a clinical effectiveness review of the literature that included 26 trials extracted from over 5000 references, Picot et al. found that bariatric surgery was a more effective intervention for weight loss than non-surgical options; however, there was extreme heterogeneity and questionable long-term adherence to the non-surgical interventions [ 39 ]. After surgery, metabolic syndrome improved, and there were higher rates of T2D remission compared to the non-surgical groups [ 39 ]. Further, while there were improvements in comorbidities after surgery independent of bariatric procedure, there was also an increased likelihood of adverse events. While the overall event rate remained low, major adverse events included medication intolerance, need for reoperation, infection, anastomotic leakage, and venous and thromboembolic events [ 39 ].

It is imperative to consider that one of the requirements of qualifying for bariatric surgery is demonstration of at least six months of unsuccessful attempts at weight loss using traditional dietary and exercise advice according to the 2016 Recommendations [ 12 ]. There are, however, no requirements as to what weight loss strategy is employed, which may represent a time point where dietary intervention, including low-calorie, ketogenic, or carbohydrate-restricted diets, should be utilized. At least two recent clinical trials have demonstrated safety and efficacy in pre-operative very low-carbohydrate ketogenic diets before bariatric surgery for increasing weight loss and decreasing liver volume [ 40 , 41 ].

Furthermore, despite technically adequate surgery, an alarming number of patients may still experience weight regain and/or recurrence of comorbid obesity-associated conditions. In these patients, effective strategies for dietary intervention are even more important. Approximately 10–15% of patients fail to lose adequate weight (failure defined as <50% of excess weight) or demonstrate significant weight regain after bariatric surgery without evidence of an anatomic or technical reason [ 42 ]. Additionally, in 25–35% of patients who undergo surgery, significant weight regain (defined as >15% of initial weight loss) occurs within two to five years post-operatively [ 43 ]. These patients often require further medical management with weight loss medications, further dietary and behavioral intervention, and, for some, reoperation. Reoperation can be for either revision for further weight loss (narrowing of the gastric sleeve, conversion of VSG to RYGB , and increasing the length of the roux limb) or reversal of RYGB due to health concerns, most commonly associated with malnutrition. A small cohort of patients (4%) may experience severe weight loss with significant malnutrition leading to hospitalization in over 50%, mortality rates of 18%, and need for reversal of RYGB anatomy. While the incidence of RYGB reversal is unknown, based upon a systematic review that included 100 patients spanning 1985–2015, the rate of reversal parallels the increasing rate of bariatric surgery [ 44 ].

In the short term, T2D reversal rates with surgery have been reported to be as high as 80%, with an additional 15% demonstrating partial improvement in T2D despite still requiring medication [ 17 ]. Within one week after RYGB, patients experience improved fasting hepatic insulin clearance, reduced basal de novo glucose production, and increased hepatic insulin sensitivity; by three months and one year after surgery, patients have improved beta-cell sensitivity to glucose, increased GLP-1 secretion from the gut, and improved insulin sensitivity in muscle and fat cells [ 45 ]. Over time, T2D remission rates remain high but do decline; Purnell and colleagues reported three-year remission rates of 68.7% after RYGB [ 29 ]. However, Pories published results from a 14-year prospective study with mean follow-up of 7.6 years, and found 10-year remission rates remained around 83% [ 46 ]. In a 10-year follow-up study of participants from the Swedish Obese Subjects (SOS) study that prospectively followed patients who underwent bariatric surgery, the authors reported a 72% ( n = 342) and 36% ( n = 118) recovery rate from T2D for RYGB at two years and 10 years, respectively [ 47 ].

The long-term metabolic impact and risk reduction from surgery remain high in a substantial number of patients and this route to reversal clearly has the most robust data to support its use. As evidenced by the dramatic improvements in metabolic state that precede weight loss, bariatric surgery is far more than merely a restrictive and/or malabsorptive procedure. Large shifts in bile acid signaling in the lumen of the small intestine, gut nutrient sensing, and changes in the microbiota community appear to greatly impact overall host health. Further research is ongoing using both basic and translational science models to identify the role of these various hormones and metabolites; perhaps there will be a way to one day harness the beneficial effects of bariatric surgery without the need for anatomic rearrangement.

3.2. Low-Calorie Diets (LCD)

As diabetes rates have risen to unprecedented levels [ 1 , 2 ], the number of studies examining diabetes reversal using non-surgical techniques has increased. A handful of studies have reported successful weight loss with decreased insulin resistance, plasma glucose, and medication use following a LCD. As early as 1976, Bistrian et al. [ 48 ] reported that a very low-calorie protein-sparing modified fast allowed for insulin elimination in all seven obese patients with T2D. The average time to insulin discontinuation was only 6.5 days, and the longest was 19 days. In a study by Bauman et al., a low-calorie diet of 900 kcal, including 115 g of protein, led to significant improvement in glycemic control that was mainly attributed to improvements in insulin sensitivity [ 49 ]. Furthermore, a study conducted in obese T2D patients found that a LCD and gastric bypass surgery were equally effective in achieving weight loss and improving glucose and HbA1c levels in the short term [ 50 ]. Weight loss, however, persisted in the diet-treated patients only for the first three months, indicating difficulty with long-term maintenance [ 47 ]. Similarly, other studies also reported similar pattern of early blood glucose normalization without medication use, but the improvements were not sustained long-term [ 51 , 52 , 53 ]. Likewise, the study by Wing et al., even though reported significant and greater improvements of HbA1c at 1 year in the intermittently delivered very low-calorie diet, the HbA1c improvement was not significantly different than what was reported in the patients receiving low-calorie diet (LCD) throughout the one year period [ 54 ]. Furthermore, the glycemic improvements observed at 1 year were not maintained through 2-years, even though the group with intermittent very low-calorie diet had less medication requirement than the group in the LCD arm at 2 years [ 54 ]. Lastly, micronutrient deficiencies with the use of calorie restricted diets has been shown and supplementation and monitoring for deficiencies is a consideration with their use [ 55 , 56 ].

While these previous studies were not assessing diabetes remission or reversal rate per se, they demonstrated the effectiveness of calorie restriction in achieving weight loss and improved glycemic control, which are the core goals of reversal. In 2003, the Look AHEAD trial randomized 5145 overweight or obese patients with T2D to an intervention group that received either an intensive lifestyle intervention (ILI) including calorie restriction and increased physical activity or to a control group that included diabetes support and education (DSE) [ 57 ]. Post hoc analysis of this study revealed that at one year, 11.5% of the participants in the ILI group achieved remission (partial or complete); however, remission rates subsequently decreased over time (9.2% at year two and 7.3% at year four). Nevertheless, the remission rates achieved through ILI were three to six times higher than those achieved in the DSE group. Lower baseline HbA1c, greater level of weight loss, shorter duration of T2D diagnosis, and lack of insulin use at baseline predicted higher remission rate in ILI participants [ 58 ].

Following the Look AHEAD study, other studies have evaluated a LCD for diabetes remission [ 59 , 60 , 61 ]. Most of these studies assessed remission over a short period of time in a small study sample. Bhatt et al. reported that six of the 12 individuals achieved partial remission at the end of the three-month intervention [ 61 ]. Ades et al. studied an intensive lifestyle program including calorie restriction and exercise, and reported that eight of the 10 individuals with recently diagnosed T2D achieved partial remission at six months, including one with complete remission [ 60 ]. The study ended at six months, therefore long term sustainability was not assessed. Another study assessing a one-year diabetes remission retrospectively among those undergoing 12 weeks of the intensive weight loss program “Why Wait” had a much lower remission rate of 4.5%, with 2.3% of them achieving partial remission, while another 2.3% had complete remission [ 59 ]. This study suggests that long-term maintenance of remission is a challenge. Moreover, diabetes remission was more likely reported in those who had a shorter diabetes duration, lower baseline HbA1c, and were taking fewer hypoglycemic medications [ 59 , 61 ].

An initial 2011 diabetes reversal study by Taylor and colleagues showed that a very low-calorie diet of 600 Kcal/day not only normalized glucose, HbA1c, and hepatic insulin sensitivity levels within a week, but also led to decreased hepatic and pancreatic triacylglycerol content and normalization of the insulin response within eight weeks [ 62 ]. At 12 weeks post-intervention, many of the improvements were maintained, but over a quarter of the patients had an early recurrence of diabetes. Further, average weight regain during the 12 weeks post-intervention was 20% [ 62 ]. As a follow-up to the 2011 study, the same group performed a larger and longer study with eight weeks of a very low-calorie meal replacement (624–700 kcal/day) followed by two weeks of solid food replacement and a weight maintenance program of up to six months [ 63 ]. In this study, those who achieved a fasting blood glucose of <7 mmol/L (<126 mg/dL) were categorized as responders, while others were categorized as non-responders. At six months, 40% of participants who initially responded to the intervention were still in T2D remission which was defined by achieving a fasting plasma glucose of <7mmol/L; the majority of those who remitted (60%) had a shorter diabetes duration (<4 years) [ 63 ].

These short-term studies were the foundation for a community-based cluster-randomized clinical trial called DiRECT (Diabetes Remission Clinical Trial). DiRECT enrolled a sample of 306 relatively healthy participants with T2D (people on insulin or with a diabetes duration longer than six years were excluded) [ 64 ] ( Figure 1 ). They were cluster randomized to either standard diabetes care or an intervention using low-calorie meal replacement diet (825–853 kcal/day) for three to five months, followed by stepwise food re-introduction and a long-term weight maintenance program. At one-year follow-up, 46% of patients met the study criteria of diabetes remission (HbA1c <6.5% without antiglycemic medications) [ 64 ] and at two years the remission rate was 36% [ 65 ]. The DiRECT study has extended their follow-up an additional three years to assess the long-term impact on remission.

Taken together, evidence suggests that a LCD is effective in reversing diabetes in the short term up to two years, and its effectiveness was predominantly demonstrated in those with shorter duration since diabetes diagnosis. It is important to note that a substantial level of calorie restriction is needed to generate a sufficient level of weight loss for reversing diabetes. Short-term intervention with moderate energy restriction and metformin for modest weight loss was not as effective in reversing diabetes as compared to standard diabetes care [ 66 ]. Lifestyle intervention with severe energy restriction may have some deleterious effect on the body composition and physiology, which poses a concern for long-term health [ 67 ]. Furthermore, long-term achievement of diabetes remission, adherence to the diet, and weight loss maintenance after the diet remain a challenge. Studies have also suggested that physiological and metabolic adaptation of the body in response to caloric restriction may shift energy balance and hormonal regulation of weight toward weight regain after weight loss [ 67 , 68 ]. Thus, it is crucial that future studies are directed towards assessing the long-term sustainability of diabetes remission led by LCD and feasibility of this diet on the physiological adaptation and body composition changes.

3.3. Carbohydrate-Restricted Diets (LC)

Before the discovery of insulin in 1921, low carbohydrate (LC) diets were the most frequently prescribed treatment for diabetes [ 69 , 70 ]. The paradigm shifted both with the development of exogenous insulin and later with the emergence of the low-fat diet paradigm. A diet low in fat, which by default is high in carbohydrate, became the standard recommendation in guidelines around the globe [ 71 ]. Rather than preventing elevations in glucose, the goal became maintenance of blood sugar control via the increased use of glycemic control medications, including insulin [ 72 ]. Over the last decade, clinical studies have begun to resurrect the pre-insulin LC dietary approach. In response to the new evidence on the efficacy of carbohydrate restriction, low-carbohydrate has recently been endorsed as an eating pattern by the ADA and the European Association for the Study of Diabetes (EASD) [ 5 , 73 ]. In addition, the Veterans Affairs/Department of Defense (VA/DOD) guidelines now recommend carbohydrate restriction as low as 14% of energy intake in its most recent guidelines for treatment of diabetes (VA) [ 74 ].

LC diets are based on macronutrient changes rather than a focus on calorie restriction [ 75 ]. Although the exact definition varies, a low-carbohydrate diet usually restricts total carbohydrates to less than 130 grams per day, while a very low-carbohydrate or ketogenic diet usually restricts total carbohydrates to as low as 20–30 grams per day. Protein consumption is generally unchanged from a standard ADA diet (around 20% of intake), with the remaining energy needs met by fat from either the diet or mobilized body fat stores. Carbohydrate sources are primarily non-starchy vegetables with some nuts, dairy, and limited fruit [ 75 ].

A total of 32 separate trials examining carbohydrate restriction as a treatment for T2D were found when our search was performed [ 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 , 98 , 99 , 100 , 101 , 102 , 103 , 104 , 105 , 106 , 107 , 108 ]. However, for reasons that may include varied levels of carbohydrate restriction and differing levels of support given, not all studies had results that would be consistent with diabetes reversal. A number of shorter-term trials have found a significant between-group advantage of a low-carbohydrate intervention for T2D [ 80 , 84 , 92 , 97 ]. Data from longer-term trials are limited, and in some follow-up studies, the between-group advantage seen initially was lost or reduced, although it often remains significantly improved from baseline. This raises the question of long-term sustainability using this approach. Due to heterogeneity in methodology and definition of carbohydrate restriction, the ability to fully examine T2D reversal based on the existing studies is limited. Based upon a recent systematic review of LC, it appears that the greatest improvements in glycemic control and greatest medication reductions have been associated with the lowest carbohydrate intake [ 109 ]. In consideration of these limitations, it appears important to assess the level of carbohydrate restriction, support or other methods given to encourage sustainability, and length of follow-up.

A study comparing an ad libitum very low-carbohydrate (<20 g total) diet to an energy-restricted low-glycemic diet in T2D found greater reduction in HbA1c, weight, and insulin levels in the low-carbohydrate arm [ 89 ]. Additionally, 95% of participants in the low-carbohydrate arm reduced or eliminated glycemic control medications, compared to 62% in the low glycemic index arm at 24 weeks. Instruction was given in a one-time session with a dietician and included take-home materials for reference. A slightly longer study (34 weeks) trial [ 85 ] found that a very low-carbohydrate ketogenic diet intervention (20–50 g net carbs per day) resulted in HbA1c below the threshold for diabetes in 55% of the patients, compared to 0% of patients in the low-fat arm. The education sessions were all online and included behavior modification strategies and mindful eating which was aimed to address binge eating. New lessons were emailed to the patients weekly for the first 16 weeks and then every two weeks for the remainder of the study.

A small (34 participants) one-year study of an ad libitum, very low-carbohydrate diet compared to a calorie-restricted moderate carbohydrate diet found a significant reduction in HbA1c between groups favoring the low-carbohydrate arm [ 86 ]. At one year, 78% of participants who began the trial with a HbA1c above 6.5% no longer met the cutoff for the diagnosis of diabetes, no longer required any non-metformin medication, and significantly reduced or eliminated metformin. Total kilocalorie intake was not significantly different between the two groups, even with moderate carbohydrate restriction. Despite equal energy intake, the low carbohydrate group lost significantly more weight and had improved glycemic control, which indicates a potential mechanistic role for carbohydrate restriction itself. The support given was 19 classes over the 12-month period, tapering in frequency over time.

Another one-year trial [ 76 ] found significant HbA1c reduction in the subset of patients with diabetes ( n = 54) assigned to an ad libitum low-carbohydrate diet (<30 total grams per day), compared to an energy-restricted low-fat diet. These results remained significant after adjusting the model for weight loss, indicating an effect of the carbohydrate reduction itself. The support given was four weekly sessions during the first month, followed by monthly sessions for the remaining 11 months.

A metabolic ward study on 10 patients with T2D [ 96 ] found that 24-h glucose curves normalized within two weeks on a very low-carbohydrate diet (<21 g total per day). This was in addition to medication reduction and elimination including insulin and sulfonylureas After accounting for body water changes, the average weight loss during the two-week period was 1.65 kg (average of <2% total body weight which is similar to the results of bariatric surgery, where normoglycemia is seen prior to significant weight loss. Interestingly, despite the diet being ad libitum other than the carbohydrate limit, the average energy intake decreased by 1000 kcal per day. Assuming no further change in glycemic control, HbA1c would be 5.6% after eight weeks, which would represent a reduction of 23% from baseline. The fact that HbA1c reductions were greater than in other, longer-term outpatient studies may indicate that support of dietary changes is the key to longer-term success.

In our published trial providing significant support through the use of a continuous care intervention (CCI), we examined using a low-carbohydrate diet aimed at inducing nutritional ketosis in patients with T2D ( n = 262), compared with usual care T2D patients ( n = 87) [ 98 ] ( Figure 1 ). At one year, the HbA1c decreased by 1.3% in the CCI, with 60% of completers achieving a HbA1c below 6.5% without hypoglycemic medication (not including metformin). Overall, medications were significantly reduced, including complete elimination of sulfonylureas and reduction or elimination of insulin therapy in 94% of users. Most cardiovascular risk factors showed significant improvement [ 110 ]. The one-year retention rate was 83%, which indicates that a non-calorie-restricted, low-carbohydrate intervention can be sustained. Improvements were not observed in the usual care patients. The newly released two-year results of this trial [ 106 ] show sustained improvements in normoglycemia, with 54% of completers maintaining HbA1c below 6.5% without medication or only on metformin. The retention rate at two years was 74%, further supporting the sustainability of this dietary intervention for diabetes reversal. Weight loss of 10% was seen at 2-years despite no intentional caloric restriction instruction. Additionally, this trial involved participants with a much longer duration of diabetes (8.4 years on average) than other nutrition trial interventions [ 58 , 64 , 65 ] and did not exclude anyone taking exogenous insulin. As duration of T2D and insulin use have both been identified to be negative factors in predicting remission after bariatric surgery [ 111 , 112 ], the 2-year results of this trial may be even more significant.

It is interesting to note that most studies utilize ad libitum intake in the carbohydrate-restricted arm. Despite this, in studies that have tracked energy intake, spontaneous calorie restriction has occurred [ 113 , 114 ]. In many trials where energy intake has been prescribed or weight loss has been equal, an advantage has been seen in glycemic control, weight, or both in the low-carbohydrate arm [ 86 , 91 , 107 ]. A better understanding of the role that caloric intake, whether prescribed or spontaneous, plays in the overall success is important. In cases of spontaneous energy intake reduction, elucidating the specific mechanism behind this reduction would help in the overall personalization of this approach.

Multiple studies have evaluated side effects or potential complications of carbohydrate restriction. The diet has been found to be safe and well tolerated although long term hard outcome data is lacking and should be a focus of future research. A transient rise in uric acid early in very low-carbohydrate restriction without an associated increase in gout or kidney stones has been documented [ 84 , 98 , 100 ]. Blood urea nitrogen (BUN) has been found to increase and decrease in different studies without an associated change in kidney function [ 87 , 98 , 100 , 115 , 116 ]. Recently, bone mineral density has been found to be unchanged despite significant weight loss after two years of a ketogenic diet intervention in patients with T2D [ 108 ]. While most studies show an improvement or no change in LDL-C levels in patients with T2D on a low-carbohydrate diet, there have been two studies that have found an increase in LDL-C in participants with T2D [ 99 , 111 ]. In one of the studies that found an increase in calculated LDL-C, a non-significant reduction in measured ApoB lipoproteins and unchanged non-HDL cholesterol were seen. Monitoring LDL-C or a measured value of potentially atherogenic lipoproteins such as ApoB should be considered. Lastly, micronutrient deficiency has been seen with a carbohydrate restricted diet, supplementation and monitoring should be given consideration with this intervention [ 56 ].

Although the use of very low-carbohydrate diets for diabetes reversal shows promising results, the lack of longer-term follow-up studies remains a limitation. Follow up is limited to two years, and therefore longer-term studies are needed to determine the sustainability of the metabolic improvements. Determining the appropriate method of support may be a key to the overall success with disease reversal.

Additional evidence has become available in recent years suggesting that diabetes reversal is a possible alternative to consider in place of traditional diabetes treatment and management. In this paper, we provide a review of three methods that have been shown to successfully reverse type 2 diabetes. The current body of evidence suggests that bariatric surgery is the most effective method for overall efficacy and prolonged remission, even though concerns associated with surgical complications, treatment cost and complete lifestyle modification after surgery remain challenges for wide adoption of this approach. While both the LCD and LC dietary approaches are convincing for reversing diabetes in the short term (up to two years), long term maintenance of diabetes remission is still unproven. There are limited available data supporting long term maintenance of weight loss and its associated glycemic improvements in response to LCD; similarly, long-term adherence to a low carbohydrate diet will likely remain an obstacle without the development of proper patient education and optimal support for long-term behavioral change. Moreover, research in understanding the mechanism of diabetes reversibility in all three approaches and its overlapping mechanistic pathways are lacking; this is an area for future research emphasis.

There are similar identified negative predictors of remission for all three approaches. These factors include longer diabetes duration and increased severity, lower BMI, advanced age, poor glycemic control, and low C-peptide levels (indicating decreased endogenous insulin production) [ 117 ]. Further exploration into the heterogeneity of these factors will help personalize the approach, determine realistic goals for each patient, and should be considered during treatment discussions. Ongoing research into algorithm development will be helpful in this regard.

5. Conclusions

Overall, as a society we can no longer afford or tolerate the continued rising rates of diabetes. Despite many barriers within the healthcare system as a whole, providers are responsible on a daily basis for the lives of patients caught up in this unprecedented epidemic. The current standard of care may be suitable for some, but others would surely choose reversal if they understood there was a choice. The choice can only be offered if providers are not only aware that reversal is possible but have the education needed to review these options in a patient-centric discussion.


We thank James McCarter and Stephen Phinney for their edits, which greatly improved the manuscript.


Author contributions.

Conceptualization, S.J.H. and S.J.A. Investigation, S.J.H., V.M.G., T.L.H., S.J.A. Writing—original draft, S.J.H., V.M.G., S.J.A. Writing—review and editing, S.J.H., V.M.G., T.L.H., S.J.A. All authors approved of the final manuscript.

Conflicts of Interest

S.J.H. is an employee and shareholder of Virta Health, a for-profit company that provides remote diabetes care using a low-carbohydrate nutrition intervention, and serves as an advisor for Atkins Corp. V.M.G. has no conflicts of interest to declare. T.L.H. is an employee of Virta Health. S.J.A. is an employee and shareholder of Virta Health.


  1. (PDF) Research of Type 2 Diabetes Patients’ Problem Areas and Affecting

    research type 2 diabetes

  2. Researchers Find a Molecular Mechanism Involved in Type 2 Diabetes

    research type 2 diabetes

  3. The Early Treatment of Type 2 Diabetes

    research type 2 diabetes

  4. Diabetes Type 2

    research type 2 diabetes

  5. Surveys Find Adults with Type 2 Diabetes Are More Willing to Take

    research type 2 diabetes

  6. New Research Reveals Five Types of Diabetes, Not Two

    research type 2 diabetes


  1. Understanding Type 2 Diabetes

  2. What Is Type 2 Diabetes?

  3. Diagnosis of Type 2 Diabetes

  4. Treatment and Management of Type 2 Diabetes

  5. Type 2 Diabetes

  6. Treating type II diabetes


  1. Type 2 diabetes

    Type 2 diabetes - Latest research and news | Nature Type 2 diabetes articles from across Nature Portfolio Atom RSS Feed Type 2 diabetes mellitus, the most frequent subtype of diabetes, is...

  2. Type 2 Diabetes

    About 38 million Americans have diabetes (about 1 in 10), and approximately 90-95% of them have type 2 diabetes. Type 2 diabetes most often develops in people over age 45, but more and more children, teens, and young adults are also developing it. What Causes Type 2 Diabetes?

  3. Type 2 Diabetes Research At-a-Glance

    Type 2 Diabetes Research At-a-Glance The ADA is committed to continuing progress in the fight against type 2 diabetes by funding research, including support for potential new treatments, a better understating of genetic factors, addressing disparities, and more. For specific examples of projects currently funded by the ADA, see below.

  4. Clinical Research in Type 2 Diabetes

    Henry B. Burch, M.D. Clinical studies utilizing existing digital health technology for the prevention and treatment of type 2 diabetes, clinical and basic science studies involving non-neoplastic disorders of the thyroid, clinical studies involving medical and novel dietary treatment of type 2 diabetes.

  5. Pathophysiology of Type 2 Diabetes Mellitus

    Type 2 Diabetes Mellitus (T2DM), one of the most common metabolic disorders, is caused by a combination of two primary factors: defective insulin secretion by pancreatic β-cells and the inability of insulin-sensitive tissues to respond appropriately to insulin.

  6. Type 2 Diabetes Clinical Trials

    Rochester, MN This study is being done to understand metformin's mechanisms of action regarding glucose production, protein metabolism, and mitochondrial function. Evaluation of the Efficacy and Safety of Duodenal Mucosal Resurfacing Using the Revita® System in Subjects With Type 2 Diabetes on Insulin Therapy Scottsdale/Phoenix, AZ

  7. Clinical Research on Type 2 Diabetes: A Promising and Multifaceted

    Type 2 diabetes constitutes an imposing epidemiological, economic, and scientific global challenge. The chronic complications of type 2 diabetes are a major cause of mortality and disability worldwide [ 1, 2 ]. Clinical research is the main way to gain knowledge about long-term diabetic complications and reduce the burden of diabetes.

  8. Type 2 diabetes

    Type 2 diabetes is usually diagnosed using the glycated hemoglobin (A1C) test. This blood test indicates your average blood sugar level for the past two to three months. Results are interpreted as follows: Below 5.7% is normal. 5.7% to 6.4% is diagnosed as prediabetes. 6.5% or higher on two separate tests indicates diabetes.

  9. Research Summaries

    An Additional 12 Million US Adults Become Eligible for Diabetes Screening. New USPSTF and ADA guidelines lower the age for prediabetes and type 2 diabetes screening to 35. This study examined if testing practices aligned with guidelines and which populations were less likely to receive testing. Read the summary.

  10. Epidemiology of Type 2 Diabetes

    This research aimed to analyze the global epidemiology of type 2 diabetes. We analyzed the incidence, prevalence, and burden of suffering of diabetes mellitus based on epidemiological data from the Global Burden of Disease (GBD) current dataset from the Institute of Health Metrics, Seattle.

  11. A promising new pathway to treating type 2 diabetes

    June 29, 2021 Source: University of Arizona Summary: Researchers believe the liver may hold the key to new, preventative Type 2 diabetes treatments. FULL STORY

  12. Large-scale study reveals new genetic details of diabetes

    By Wynne ParryWeill Cornell Medicine. October 19, 2023. In experiments of unprecedented scale, investigators at Weill Cornell Medicine and the National Institutes of Health have revealed new aspects of the complex genetics behind Type 2 diabetes. Through these discoveries, and by providing a template for future studies, this research furthers ...

  13. Diet and exercise in the prevention and treatment of type 2 diabetes

    The worldwide prevalence of type 2 diabetes mellitus (T2DM) in adults has increased from ~150 million affected people in 2000 to >450 million in 2019 and is projected to rise further to ~700 ...

  14. Recent Advances

    Amelioration of Both Central and Peripheral Neuropathy in Mouse Models of Type 1 and Type 2 Diabetes by the Neurogenic Molecule NSI-189. Diabetes, 68(11), 2143-2154. Read more. ADA-funded researcher studying link between ageing and type 2 diabetes. One of the most important risk factors for developing type 2 diabetes is age.

  15. Diabetes

    A meta-analysis of genome-wide association studies of type 2 diabetes (T2D) identifies more than 600 T2D-associated loci; integrating physiological trait and single-cell chromatin accessibility ...

  16. Type 2 Diabetes Research At-a-Glance

    Type 2 Diabetes Research At-a-Glance The ADA is committed to continuing progress in the fight against type 2 diabetes by funding research, including support for potential new treatments, a better understating of genetic factors, addressing disparities, and more. For specific examples of projects currently funded by the ADA, see below.

  17. Study reveals what causes type 2 diabetes and how to reverse it

    In the U.K., the National Health Service (NHS) will roll out a program that will test the weight loss therapy in thousands of people living with type 2 diabetes. Breakthrough research shows that ...

  18. Changing our Future Through Research

    ADA research provides critical funding for diabetes research. With 100% of donations directed to research, our goal is to ensure adequate financial resources to support innovative scientific discovery that will translate to life-changing treatments and eventual cures. ... Type 2 Diabetes Research Project topics include support for potential new ...

  19. New Aspects of Diabetes Research and Therapeutic Development

    Type 1 and type 2 diabetes (T1D and T2D, respectively) make up the majority of diabetes cases with T1D characterized by autoimmune destruction of the insulin-producing pancreatic beta cells.

  20. Type 2 diabetes

    PMID: 36332637 DOI: 10.1016/S0140-6736 (22)01655-5 Abstract Type 2 diabetes accounts for nearly 90% of the approximately 537 million cases of diabetes worldwide. The number affected is increasing rapidly with alarming trends in children and young adults (up to age 40 years).

  21. Type 2 Diabetes in Youngsters May Be a Genetically Distinct Form of the

    However, new genetics research focused on a form of type 2 diabetes that is becoming more common in adolescents suggests a more complicated picture. Get more HMS news here. The study, led by researchers at Harvard Medical School, the Broad Institute of MIT and Harvard, and Boston Children's Hospital, found that youth-onset type 2 diabetes is ...

  22. Researchers uncover potential treatment for cardiovascular

    New research at the Roy Blunt NextGen Precision Health building has discovered a potential treatment for an underlying cause of cardiovascular disease in people with type 2 diabetes.

  23. Type 2 Diabetes

    Type 2 diabetes is the most common form. Approximately 90 percent of those with diabetes have type 2. Unlike type 1 diabetes, in which all the insulin-producing cells are destroyed, people with type 2 diabetes are able to produce some of their own insulin, but their bodies are unable to use this insulin to completely control blood sugar levels.

  24. Management of Type 2 Diabetes: Current Strategies, Unfocussed Aspects

    Type 2 diabetes is a multifactorial disorder that leads to a disturbed glucose homeostasis. Lifestyle management along with pharmacological approaches is crucial to achieve a successful management of diabetes. Complex interplays between genetics and environmental factors play important roles in the development of diabetes.

  25. Epigenetic Changes Can Cause Type 2 Diabetes, According to New Research

    A recent study conducted by Lund University researchers, and published in Nature Communications, suggests that epigenetic changes might be a causal factor in the development of type 2 diabetes, rather than merely occurring as a result of the disease.This new research bolsters the theory that epigenetic modifications can lead to type 2 diabetes, and the team is now focusing on creating ...

  26. Ambitious survey of human diversity yields millions of undiscovered

    A study in Nature on type 2 diabetes 2 is an example of the power of using a database that includes diverse genomes, Ramirez says. The condition, which affects about one in ten people in the ...

  27. Phenotype-based targeted treatment of SGLT2 inhibitors and GLP-1

    A precision medicine approach in type 2 diabetes would aim to target specific glucose-lowering therapies to individual patients most likely to benefit [].Current stratification in type 2 diabetes treatment guidelines involves preferential prescribing of two major drug classes, sodium-glucose cotransporter 2 inhibitors (SGLT2i) and glucagon-like peptide-1 receptor agonists (GLP1-RA), to ...

  28. Gargling away the 'bad' bacteria in type 2 diabetes

    Effects of mouthwash on periodontal pathogens and glycemic control in patients with type 2 diabetes mellitus. Scientific Reports , 2024; 14 (1) DOI: 10.1038/s41598-024-53213-x

  29. Reversing Type 2 Diabetes: A Narrative Review of the Evidence

    Abstract Background: Type 2 diabetes (T2D) has long been identified as an incurable chronic disease based on traditional means of treatment. Research now exists that suggests reversal is possible through other means that have only recently been embraced in the guidelines.

  30. Type 2 diabetes: Red light therapy could help lower blood sugar levels

    The majority of diabetes diagnoses — between 90 % to 95% — are type 2 diabetes. Unlike type 1 diabetes, which is an autoimmune disease, type 2 diabetes is manageable and potentially reversible ...