According to the IDF Diabetes Atlas, “Diabetes is one of the fastest growing global health emergencies of the 21st century.”1 Patients with diabetes often have micro- and macrovascular complications, necessitating a multidisciplinary approach to care.2-4 In a recent AJMC Stakeholder Summit series, experts who specialize in treating patients with type 2 diabetes (T2D) were joined by experts from the insurance industry to discuss the burden of disease, how to prevent clinical inertia in early treatment management, and future considerations for clinical practice. The session was moderated by Yehuda Handelsman, MD, FACP, FNLA, FASPC, MACE, medical director and principal investigator at The Metabolic Institute of America and chair of the Scientific Advisory Board for the Diabetes, Cardiorenal & Metabolic Institute (DCMi).
The information discussed in this series is based on output from the DCMi Think Tank, a workshop in which 17 medical experts created treatment consensuses based on data and expert opinion for patients with complex multimorbid diseases plateauing or worsening in outcomes despite the emergence and availability of numerous therapeutic options. The DCMi will soon have a published consensus paper on early intervention and intensive treatment of the patient with diabetes and cardio-renal-metabolic diseases, which complements the diabetes, cardiorenal, and/or metabolic multispecialty practice recommendations published in 2022.
Diabetes mellitus, commonly referred to as diabetes, is a serious, long-term condition characterized by hyperglycemia due to inadequate production of insulin (type 1 diabetes [T1D]) or to insulin resistance (T2D).1 Most patients with diabetes have T2D (> 90%).1 Efforts to stratify individuals who either have T2D or are at a high risk of developing T2D have shown some success in predicting the risk of developing complications.5,6 Analysis using these newer stratification efforts provides evidence for the heterogeneity and complexity of T2D and indicates that there is pathophysiological variation among individuals even prior to the development of T2D.6,7
The global prevalence of diabetes (including all diagnosed and undiagnosed T1D and T2D) in adults aged 20 to 79 years was estimated to be 537 million in 2021, representing 10.5% of all adults in this age group. Furthermore, the number of patients with diabetes is rapidly increasing. While the world’s population is estimated to grow by 20% by 2045, the number of individuals with diabetes is estimated to increase by 46% (783 million by 2045).1 Diabetes was the seventh leading cause of death in the United States in 2019 (crude rate, 26.7 per 100,000 persons), with an estimated crude prevalence of 14.7% (37.1 million) among individuals 18 years or older and an undiagnosed rate of 3.4% (8.5 million).8,9
Complications traditionally associated with diabetes include macrovascular conditions such as coronary heart disease, stroke, and peripheral arterial disease, and microvascular conditions, including diabetic kidney disease, retinopathy, and peripheral neuropathy.2,3 Another common initial manifestation of cardiovascular disease (CVD) in patients with T2D is heart failure (HF), which has a high risk of mortality.10,11 Complications whose associations with diabetes are emerging include cancer, infections, liver disease, functional disability, cognitive disability, and affective disorders.3,12-15 Such complications are associated with a substantial burden, and there is a need to increase awareness of these T2D-related complications among primary care physicians (PCPs). Future clinical and public health strategies should also ensure screening for these emerging complications, adding strength training to physicians’ recommendations to reduce functional disability in aging populations, and ongoing high-quality surveillance of diabetes outcomes to identify areas of need.16
Handelsman; Jaime Murillo, MD, senior vice president and chief medical officer at UnitedHealth Group; and Jennifer B. Green, MD, diabetes and metabolism specialist and endocrinologist at the Duke Endocrinology Clinic and Duke University Hospital and a professor of medicine at Duke University School of Medicine, discussed the importance of considering more than just hyperglycemia when treating patients with T2D. “We need to think very broadly when we consider and determine a treatment plan to either prevent or manage diabetes-related complications,” Green said. “Traditionally, we focused more on what were called microvascular complications: retinopathy, neuropathy, and nephropathy. But I think we all have come to appreciate the fact that [patients] with diabetes also are at very significant risk for cardiovascular complications.” Green further explained that patients with recently diagnosed T2D “often have established microvascular, if not macrovascular, complications, so we need to be vigilant and ensure that we are screening appropriately from the outset of the disease.” Eugene E. Wright Jr, MD, medical director of performance improvement at the Charlotte Area Health Education Center and consulting associate in the Department of Medicine at Duke University School of Medicine, agreed with the need to evaluate and manage more than just hyperglycemia. He emphasized that “the reason we [manage] diabetes is to prevent or significantly delay all these bad things, these complications we know are ahead [for patients]. It’s a different approach. It’s trying to prevent a disease and an illness in its earliest stages.”
Handelsman highlighted the evolution and debate about how to classify or name T2D based on the understanding of the disease itself. When its link to CVD first became clear, T2D was called a CVD. Later, T2D’s association with renal complications, then hepatic complications and obesity, became widely recognized, leading to complex names such as that of the DCMi. Handelsman emphasized the need to look at the whole patient instead of just dealing with 1 organ or 1 issue at a time. Wright agreed, saying that the treatment of patients with T2D requires “a coordinated, collaborative approach in identifying and [managing the] many aspects of this disease.” Wright also summarized common comorbidities associated with T2D—including obesity, nonalcoholic fatty liver disease, kidney disease, and CVD—as well as mental health aspects such as high stress levels and depression.
Murillo discussed the impact that T2D has on CVD. CVD is a major cause of mortality among patients with T2D. However, not all patients with T2D have cardiovascular complications. In fact, 30% to 40% of patients with T2D have no signs of arteriosclerosis, 30% to 40% have just mild complications, and only 6% to 13% have severe complications—all suggesting the wide variations among patients with T2D.
Green agreed that there are differences among patients, but she cautioned that “it’s important we remember that no [patient] with diabetes is without risk for complications. [However, some patients] appear to be at greater risk and progress faster in their development of complications to a certain extent, which may be due to a greater burden of hyperglycemia.” Other risk factors for more severe disease include obesity, poorly controlled hypertension, marked dyslipidemia, and genetic risk factors that predispose the patient to an accelerated development of diabetes-related complications, as well as a greater burden of other cardiovascular or kidney risk factors. Wright added that the patient’s life experience and other socioeconomic determinants of health also play a role in the progression of the disease.
In terms of the differences among patients and the risk factors related to T2D, Handelsman highlighted 3 aspects. First, he stressed that all patients with T2D have an increased risk for CVD, albeit not to the same extent. Therefore, he advises all his patients to get a coronary artery calcium test to assess for the presence of arteriosclerosis. Second, analyses of populations such as the Pima in Arizona—who have high rates of obesity, dyslipidemia, and uncontrolled hypertension, but a lower incidence of diabetes than might be expected—suggest that high insulin levels may at least delay the onset of diabetes. Finally, although no randomized controlled trial has shown a relationship between hyperglycemia and CVD in patients with T2D, that link has been demonstrated in patients with T1D.
Finally, Green highlighted that when discussing cardiovascular complications, focus should expand beyond coronary heart disease, because “many [patients] with diabetes, particularly women, may in fact have microvascular changes or damage in the heart and not much involvement of the large coronary vessels.” Handelsman agreed, saying that the causes of HF in patients with diabetes include ischemic heart disease, hypertension, cardiomyopathy, and possibly hyperglycemic cardiomyopathy. Patients with hyperglycemia who have hemoglobin A1c (HbA1c) levels between 5.5 and 6.5 have increased protein glycation, which in turn leads to changes in the heart muscle that can make it more susceptible to developing HF, Handelsman explained.
A limitation of traditional treatment guidelines is that they are focused on responding to poor metabolic control after it has developed, but they do not predict which patients will need more intensive treatment.5 This is particularly worrisome because there is evidence that target tissues “remember” poor metabolic control even decades later (so-called metabolic memory), which results in life-shortening complications that could have been avoided by early treatment.17
Preventing complications and optimizing quality of life are the goals of care established by the 2022 American Diabetes Association (ADA) and European Association for the Study of Diabetes (EASD) consensus report on the management of hyperglycemia in T2D, which recommends the use of a holistic, person-centered approach.4 This report recognizes that the management of this chronic, complex disease requires multifactorial behavioral and pharmacological treatments, including the management of blood glucose levels, weight, cardiovascular risk factors, comorbidities, and complications. As such, care should be delivered by a multidisciplinary team in an organized and structured manner, taking that person-centered approach to enhance the patient’s engagement in self-care activities. Furthermore, individualized treatment goals and strategies should be informed by a patient’s preferences, as well as by a careful consideration of social determinants of health (SDOH), barriers to care, comorbid conditions, degree of hyperglycemia, risks of complications, and susceptibility to medication adverse effects.4
The holistic person-centered approach described by the ADA/EASD consensus is organized around 4 components of care: medication for glycemic management, weight management, cardiovascular risk factor management, and cardiovascular protection. More specifically, it includes 13 principles of diabetes care4:
Nonpharmacological interventions to manage diabetes include having trained diabetes care and education specialists offer ongoing diabetes self-management education and support.4 Additionally, practitioners should regularly promote healthy behaviors, including balanced nutrition, regular physical activity, adequate sleep, and smoking cessation.4
The classes of medications for glycemic control include metformin, sodium-glucose cotransporter 2 inhibitors (SGLT2is), glucagonlike peptide 1 receptor agonists (GLP1-RAs), gastric inhibitory polypeptide, dipeptidyl peptidase 4 inhibitors (DPP4is), thiazolidinediones (TZDs), second-generation sulfonylureas, and human and analogue insulin.4 Selection of the appropriate medication will depend on individual factors. Clinicians should choose an approach that provides efficacy to achieve individual glycemic goals while avoiding hypoglycemia. Medications with high to very high efficacy include high-dose dulaglutide, semaglutide, tirzepatide (each of which is a GLP1-RA), insulin (can be given alone or in combination), other GLP1-RAs, metformin, SGLT2is, sulfonylurea, and TZDs.4
For patients with diabetes who are overweight or have obesity, weight reduction is a strategy that can improve HbA1c levels, risk factors for cardiometabolic disease, and quality of life, as well as reduce the risk of weight-related complications. A weight loss of 5% to 15% should be a primary target for most patients with T2D, with a higher magnitude of weight loss conferring higher benefits; a 5% to 10% reduction confers metabolic improvement, and a 10% to 15% reduction can have a disease-modifying effect and lead to remission of diabetes.4
The consensus report recommendations start with the setting of individualized weight management goals and include lifestyle interventions (eg, medical nutrition therapy, eating patterns, and physical activity, as well as the use of intensive evidence-based structured weight management programs), medications, and metabolic surgery. If weight loss is necessary and the clinician is prescribing medication in that pursuit, they should opt for glucose-lowering regimens with high to very high dual glucose and weight efficacy, such as semaglutide, tirzepatide, dulaglutide, and liraglutide. Other GLP1-RAs, SGLT2is, DPP4is, and metformin are not as effective for weight reduction.4
In addition to comprehensive cardiovascular risk management, the goal for patients with HF, established CVD, or multiple risk factors for CVD is to reduce the risk of cardiorenal disease. Initiation of a GLP1-RA or an SGLT2i should be done independently of background use of metformin.4
Atherosclerotic Cardiovascular Disease
Individuals with established CVD (eg, myocardial infarction [MI], stroke, any revascularization procedure, transient ischemic attack, unstable angina, coronary artery disease) and those at high risk of developing CVD (aged ≥ 55 years and having ≥ 2 additional risk factors: obesity, hypertension, smoking, dyslipidemia, or albuminuria) should be treated with either a GLP1-RA or an SGLT2i with proven CVD benefit. If the HbA1c is above target, the combination of GLP1-RA and SGLT2i therapy, or the addition of a TZD, is recommended.4
SGLT2is are recommended for patients with current or prior risk of HF or with documented HF with preserved or reduced ejection fraction.4
Chronic Kidney Disease
The preferred medication for patients with chronic kidney disease (CKD) (estimated glomerular filtration rate < 60 mL/min per 1.73 m2, or albuminuria [albumin-creatinine ratio ≥ 3.0 mg/mmol]), who are on the maximally tolerated dose of either an angiotensin-converting enzyme inhibitor (ACEI) or an angiotensin receptor blocker (ARB), is an SGLT2i or, in case of intolerance, a GLP1-RA. If the HbA1c level is above target, clinicians should consider an SGLT2i/GLP1-RA combination.4
Green and Wright expressed a preference for interventions that improve multiple conditions simultaneously. Green opined, “When it comes down to how early we should be intervening in T2D, we really should be intervening before [patients receive a diagnosis of] T2D, identifying [patients] with increased risk for progression to T2D, and introducing interventions at that time to delay or potentially prevent progression to T2D.” Moreover, medical treatment must be started when the diagnosis is made because lifestyle and other nonmedical interventions are not effective treatment strategies, Green said. Blood pressure and lipid levels should be managed aggressively and screening for complications should begin immediately.
“When I was president [of the American Association of Clinical Endocrinology], we changed our guidelines from the management of diabetes to a comprehensive approach of [treating] the patient, addressing how we see risk factors,” Handelsman said. He agreed that physicians have a responsibility to treat the whole person. On the other hand, he pointed out that this could mean that a patient leaves an appointment with a prescription for multiple medications, which can cause problems with insurers.
Murillo pointed out that, from the perspective of payers, there should not be a problem with physicians taking a more holistic approach. What often happens is that specialists do not have enough time to provide this type of care. He also pointed out that one stakeholder alone cannot change the whole system, and that opportunities therefore exist for insurance companies to partner with health systems, doctors, hospitals, and other stakeholders to change how the disease is treated and optimize patient care. For example, he explained, we could move to a value-based care approach where the measure of success is not about how many patients you can see in an hour, but rather about the best way you can treat an individual patient in order to prevent complications down the road and lower the total cost of care.
From a primary care perspective, making CKD or T2D a health care priority would engage all stakeholders in the process, Wright said. It is important to start with a readiness assessment, prioritize the patient experience, and use validated and endorsed guidelines so all specialists are in agreement with the treatment plan. Ideally, the PCP should lead with the support of specialist colleagues. Information technology (IT) systems should be leveraged to improve the identification of at-risk patients and monitor adherence to guidelines. Finally, community outreach can be used to raise awareness, education, and engagement; this can be followed by appropriate screening. However, Handelsman cautioned that insurance companies often do not cover screening and early identification procedures.
In the traditional model for managing hyperglycemia, also known as the treat-to-failure model, patients are typically prescribed metformin as the first therapy, Wright said. If the target HbA1c goal is reached after a predetermined amount of time, the therapy is maintained. When the HbA1c level rises above the goal, the patient advances to either a GLP1-RA, if they have a cardiovascular risk factor, or an SGLT2i, if they have CKD or are at risk for HF.
Whereas Wright believes sulfonylurea should be a drug of last resort, Green pointed out that metformin and sulfonylurea are the most commonly prescribed drugs around the world for T2D. However, there is growing incentive to use the newer medications. “It’s not just to reduce cardiovascular or kidney risk,” Green said. “It’s also because these medicines have little to no inherent risk of causing hypoglycemia, and they often provide favorable effects on weight.”
Green and Handelsman agreed that most patients with T2D would need more than 1 medication to control hypertension, with ACEIs and ARBs being the preferred first-line choices. Other medications should be customized to the individual; options include diuretics, calcium channel blockers, and β-blockers.
Green and Wright both use statins for lipid management. “I try to explain to patients that we’re not just treating cholesterol,” Wright said. Green explained that low-dose statins are recommended for most patients who are 40 years or older and have T2D and dyslipidemia. For patients with atherosclerotic CVD (ASCVD), a high-intensity statin is recommended. Handelsman argued that the cutoff age of 40 years is outdated because many patients already have CVD in their 30s and would benefit from a statin. In his opinion, patients should start taking a high-dose statin, with dose lowering if necessary.
Murillo argued that artificial intelligence (AI) algorithms offer an opportunity to clinicians to help them determine the best drug combinations for individual patients. He shared results from a proof-of-concept study for hypertension in which patients were treated either following traditional guidelines or based on an AI algorithm that moved some medications up front. The latter strategy resulted in much faster control of blood pressure. Wright said that the same is true for therapies used for glycemic control.
Murillo emphasized another important point: Some patients start developing the precursors to these conditions at a much earlier age than previously recognized. For example, some patients with T1D already have arterial calcium deposits at age 36 years, meaning that they started developing plaque at least 6 to 8 years earlier, highlighting the need for earlier screening and better risk-assessment algorithms. Handelsman added that there is a need to move from treat-to-failure to treat-to-success models in diabetes, which may even change how we perceive diabetes: as a curable condition rather than a progressive disease.
Initiating combination therapy for blood pressure, lipids, and hyperglycemia, based on the patient’s condition and how far they are from their goals, can already be done, Handelsman said. The objective is to use 2 or 3 drugs to effectively treat the conditions. Moreover, the use of combination therapy has already been incorporated into guidelines for close to a decade. Green added that a benefit of early combination therapy is that it allows for the use of lower doses of each drug, minimizing the potential for adverse effects or intolerance.
Wright said that this approach is feasible in the primary care setting: “Identifying [these systems] early [on] and targeting them early [on] is not dissimilar to what we do with diseases like cancer. We think about all the different targets we have to treat. We treat them early [on], and we treat them aggressively, intending to induce remission. If we took that approach with diabetes and in the cardiorenal system, we’d see different outcomes.” Murillo added that remission in diabetes is an attractive idea for insurers and mentioned a value-based group called Level2 that is working with a cohort of approximately 50,000 patients with diabetes across 25 states, treating them with the aim of achieving remission. Handelsman added that newer drugs make the prospect of remission more likely, specifically mentioning proprotein convertase subtilisin/kexin type 9 inhibitors, which can lower low-density lipoprotein cholesterol levels much more effectively than high-dose statins.
Regarding the management of HbA1c levels, Handelsman said that advancements have accompanied “the invention [and use] of more potent drugs that don’t cause hypoglycemia, namely the GLP1-RAs. SGLT2is may not be as potent, but they’re potent enough early on in the disease without causing hypoglycemia.” These agents can decrease HbA1c levels to 5.7% or lower, which is significantly lower than the standard goal of 6.5%.
The medical community continues to debate exactly what is considered remission of diabetes, Green explained. One faction believes that remission exists only when a patient remains in a normal glycemic range following the discontinuation of the intervention (medication or surgery) that allowed them to achieve a normal glycemic range. Another group believes that the normalization of glucose levels in and of itself can be considered a remission, regardless of whether the intervention is continuous.
Green noted that few interventions—bariatric surgery and intensive insulin therapy are examples—can result in normal glycemic levels without continuous implementation. However, the duration of remission among patients who do not adhere to medication regimens and lifestyle changes is often only a few months, Handelsman added, and “even among bariatric [surgery] patients, up to 50% return to diabetes again in 3 to 5 years.” Furthermore, in his opinion, only a sustained HbA1c level lower than 5.4% for 3 years will allow β islet cells to regenerate enough for patients to be drug free for a couple of years.
Although these sustained HbA1c levels are generally very difficult to reach with lifestyle interventions, Green said, she added that “some of these newer agents are so effective in glucose lowering and promoting weight loss [that they] make it easier for the person to make effective lifestyle changes. Some of it is a drug effect, but some of it is the fact that individuals are less hungry and eat considerably less.” Handelsman agreed, citing studies of patients who had had diabetes for 14 years that support this conclusion; the patients were able to reduce their insulin doses after treatment with GLP1-RAs and tirzepatide. Wright and Handelsman agreed that these medications should be moved to the front line to help prevent complications.
In fact, “the contemporary approach to managing diabetes is not only to control hyperglycemia but also to truly prevent the next event and protect the patient from complications,” Handelsman said. Green explained that there is a clear dichotomy in the ADA/EASD algorithm for managing hyperglycemia in T2D, with one side focused on individuals with ASCVD, HF, or CKD. These patients are given specific medications to reduce the risk of cardiovascular and kidney outcomes. Especially for patients with a prior MI, it is important to include these medications in treatment, regardless of glucose control needs. On the other hand, for lower-risk patients, glucose control and weight management are the initial focus.
“[Eugene] Braunwald, [MD], the most revered cardiologist alive, said SGLT2is today are [like what] statins [were in] the 1990s,” Handelsman said. Wright agreed, saying that SGLT2is have the potential to have a positive impact on all areas that are at risk in patients with diabetes, conferring cardiorenal protection in addition to glycemic control. He added that from a primary care perspective, it is much more impactful to prevent, rather than treat, these complications.
Green noted that “the absolute benefit or reduction in risk with these newer agents is more clearly evident in individuals who are at higher risk with more advanced complications. However, it’s clear that treating individuals with diabetes earlier with these medications can also reduce risk in a person who, for example, has multiple risk factors but not established disease.” However, the cost of treating patients with these newer agents has to be considered.
From the payer’s perspective, Murillo explained, value analysis is necessary to determine “the value in terms of benefit for the patient outcomes in a way that’s not going to continue to burden the health care system.” The thinking has evolved from the treat-to-failure framework. “Now we’re thinking that we should introduce several [medication] classes at a lower dose, so we can get multiple effects and the patient can benefit earlier,” he said. However, clearly stated treatment guidelines carry significant weight in guiding insurance policies regarding what to reimburse and cover.
Findings from some studies have shown rapid improvements in patients with SGLT2is, but the data are not as clear as with GLP1-RAs, Handelsman explained. For example, “in EMPEROR-Preserved [NCT03057951], I believe there are data to show an improvement in 18 days. In EMPEROR-Reduced [NCT03057977], evidence shows an improvement in 12 days. These very early improvements are important. When we’re looking at the GLP1-RAs and SGLT2is…the data show that, independent of glucose[-lowering] and other medications, they will get patients to their goal,” he said. Therefore, he would recommend one of these agents, as opposed to metformin, for a hypothetical patient with diabetes who is adherent to recommended diet, has an HbA1c level of 6.4%, and has a heart problem.
Green pointed out that “the vast majority of prescriptions for SGLT2is and GLP1-RAs [in the United States] are for these lower-risk individuals. They’re prescribed primarily for the control of hyperglycemia. In fact, implementation of these beneficial treatments in the higher-risk group is very low.” Moreover, “it would be absolutely appropriate, and probably would positively affect population health, if we started using those drugs very early on in the course of diabetes.” Wright agreed, saying that the message from recent guidelines has not “made its way onto the clinical practice protocols.” He suggested that all stakeholders, including insurers and IT program developers, should reinforce the idea that high-risk patients should be treated with the newer agents. Handelsman and Murillo added that the cost also plays a role in these decisions because not all patients are able to afford co-pays.
Handelsman summarized the consensus, saying that treating high-risk patients “very early on with everything we can is very important. In this case, if somebody has an established CVD, they’re at risk for heart attack, stroke, HF, and maybe also CKD. In that respect, SGLT2is and GLP1-RAs would both work in a patient [like this]. The earlier [on] we give it, the better [off] the patient will be.”
Addressing SDOH and implementing best evidence-based strategies of care more effectively provide a major opportunity to improve diabetes outcomes in the near future.4 Additionally, broad support for basic and implementation science is necessary, including efforts to advance learning health care systems, which integrate published evidence with internal data to drive quality improvement. Another promising area for research is in precision medicine, both omics-based and focused on SDOH, to support the better understanding of these approaches to the full spectrum of diabetes interventions.4
Both Green and Handelsman said that their patients often have multiple risk factors and would not be considered low risk. Green added that patients in the endocrinology clinic often have some kind of abnormality that can be detected with further testing, but that this testing is not always done. Therefore, there is a need to more effectively identify patients who are at a higher risk for complications. Green suggested that a team-based approach could help identify and reduce risk factors, and Wright said that questionnaires used in primary care can also help. IT systems can be used to risk stratify and flag at-risk patients who are not being treated with the appropriate medications. Handelsman added that classifying diseases on a spectrum, as was done for CKD, may help clinicians realize the severity of some advanced stages.
Murillo suggested that a multistakeholder approach, including community health workers and clinics, to incentivize holistic treatment may be successful in incentivizing providers to improve patient outcomes. Moreover, there is an increasing shortage of physicians and specialists. Therefore, the answer lies in empowering primary care providers and utilizing a team approach, which includes pharmacists, social workers, and community health workers. Wright concurred that incentivizing providers through guidelines and payment systems can also drive better practice behavior.
Handelsman pointed out that some insurance companies continue to prefer older guidelines and treatments, but Murillo believes that the companies’ role has been evolving to be more collaborative and involved in finding solutions. Handelsman said that the issue of when to intervene and when to manage is often lost in guidelines. Therefore, the DCMi Think Tank, which includes specialists and PCPs, is drawing up a consensus recommendation, addressing early intervention and intensity of management among other topics, that should be published soon. Murillo added that screening recommendations should also be included in any future guidelines.
Green finished by saying that good diabetes care and cardiovascular risk reduction should be available to everyone, not just those who can afford it.