Several causes for the extreme increase in incidence of T2DM in the late 20th century have been noted. First, the U.S. population is expanding annually, resulting in an increase in the number of people with T2DM [10]. Second, rising rates of overweight and obesity correlate with the rising incidence of T2DM, as these factors have a positive association [5,11]. Rates of obesity were 15% in 1980, 23% in 1994, 31% in 2000, 35% in 2011–2012, and more than 42% in 2017–2018 [5,12,13].Among those with diabetes, 89.8% of patients are overweight or obese, with 45.8% being obese (body mass index [BMI] >30 kg/m² to ≤39.9 kg/m²) and 16.2% being classified as severely obese (BMI >40 kg/m²) [1,6]. Third, because T2DM disproportionately affects elderly individuals, rising rates of T2DM are anticipated to reflect the steadily growing population of adults 65 years of age and older [10,14]. It has also been observed that T2DM appears to be occurring at a greater frequency in younger adults. Between 1988–1994 and 1999–2000, the mean age at diagnosis of T2DM decreased from 52 years to 46 years, and in 2015, more than one-half of new cases were among adults 45 to 64 years of age [5,15]. Rates of incidence, both increasing and decreasing, can be attributed to changing diagnostic criteria, improved physician recognition of T2DM, and increased public awareness, making additional research necessary to determine if changes in incidence rates are due to prevention and treatment strategies or if the rates reflect a shifting trend toward earlier onset of T2DM [15].

T2DM is marked by a number of disparities among affected groups, most notably within specific racial and ethnic populations. For example, while the number of non-Hispanic whites in the United States with T2DM far exceeds the number of non-Hispanic blacks with this condition, the percentage of non-Hispanic blacks with T2DM (11.5%) is significantly greater than that of non-Hispanic whites (7.2%) (Figure 1) [1,16]. Hispanic Americans also appear to be disproportionately affected by T2DM, with studies indicating rates of T2DM to be significantly higher in this population compared to non-Hispanic whites (11.8% and 7.2%, respectively) [1]. Asian Americans have a slightly higher rate (8.9%) compared to non-Hispanic whites (7.2%) [1]. In addition, the group reporting the largest percentage of T2DM, at 14.5% of the total adult population, is American Indians and Alaska Natives, with rates ranging from 6.0% among Alaska Natives to as high as 22.2% among American Indians located in certain areas of the Southwest [1]. Increasing numbers of members of high-risk minority groups in the United States contribute to the overall projected rise of T2DM [8]. Therefore, targeting these populations for prevention, intervention, and education is a strategy that can be employed to mitigate the impact of T2DM across the nation.

A number of predisposing risk factors have been attributed to the development of T2DM, most notably obesity, which affects cellular metabolism, serum free fatty acids, and adipocyte hormone production (Figure 2) [17,18,19]. Other environmental factors, including poor diet (evidenced by increased caloric intake as well as decreased food quality) and decreased activity, amplify these effects, as can certain medications and ongoing stress [17,20]. However, these factors alone are not adequate to initiate T2DM; certain genetic characteristics must also be present. To date, a large number of loci possessing common variants have been implicated in diabetes susceptibility, aided in great part by genome-wide association—a hypothesis-generating approach directed at linking new loci with a disease or trait of interest (in this instance, T2DM) [21]. The exact combination of genetic and environmental factors that generates T2DM is as yet unknown, but research continues to elucidate potential contributors.

Impaired insulin secretion is a key mechanism in the development of T2DM. The framework for this process is built on the understanding that insulin is secreted from beta cells and hyperglycemia occurs when beta-cell secretion of insulin is inadequate respective to the glucose load [22]. In healthy individuals, glucose ingestion (and resultant increase in plasma glucose concentration) triggers the production and release of insulin by pancreatic beta cells [22]. In those with T2DM, insulin response to glucose declines as a result of a functional beta-cell deficiency [23]. This defect has been demonstrated by findings that islet function appears to be approximately 50% of normal at the time of diagnosis and is supported by data from the United Kingdom Prospective Diabetes Study (UKPDS), in which patients who achieved beta-cell function greater than 55% with sulfonylurea monotherapy were less likely to require additional therapy to maintain glycemic targets [24]. Thus, it is now understood that insulin sensitivity is inversely and proportionally related to beta-cell function [24]. These processes are further exacerbated when the inability of the pancreas to adapt beta-cell mass to insulin demand (referred to as pancreas plasticity) results in a decrease in functional beta-cell mass [17]. Studies have revealed an approximate 40% reduction in beta-cell mass in patients with impaired glucose tolerance and a 60% reduction in beta-cell mass in patients with T2DM compared to healthy individuals [24].

Incretin hormones, hormones released from gut endocrine cells during meals, also play a significant role in insulin secretion [25]. Glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) may be responsible for as much as 70% of postprandial insulin secretion in healthy individuals, but demonstrate severely reduced or even absent effects in those with T2DM, often referred to as the incretin effect. This incretin effect is believed to contribute to the impaired insulin regulation and glucagon secretion that are the hallmarks of T2DM, and studies have demonstrated that improved glycemic control in patients with T2DM partially restores the impaired action of GLP-1 and GIP [17].

Another central mechanism of T2DM is insulin resistance, which is characterized by the failure of insulin to decrease plasma glucose levels through hepatic glucose suppression and stimulation of glucose uptake in skeletal muscle and adipose tissue [22]. Consequently, inefficient glucose utilization is replaced by cellular utilization of fats and proteins. Factors that contribute to insulin resistance are complex and may include defective insulin-mediated cell signaling pathways, decreased muscle glycogen syntheses, and reduced numbers of skeletal muscle, liver, and adipose tissue insulin receptors (particularly in obese individuals). In many cases, insulin resistance may be the earliest detectable marker for T2DM [22]. Patients who transition from normal glucose tolerance to T2DM typically experience a 40% decrease in insulin sensitivity, which is further accentuated by chronic hyperglycemia and elevated free fatty acids.

Another central mechanism of T2DM is insulin resistance, which is characterized by the failure of insulin to decrease plasma glucose levels through hepatic glucose suppression and stimulation of glucose uptake in skeletal muscle and adipose tissue [22]. Consequently, inefficient glucose utilization is replaced by cellular utilization of fats and proteins. Factors that contribute to insulin resistance are complex and may include defective insulin-mediated cell signaling pathways, decreased muscle glycogen syntheses, and reduced numbers of skeletal muscle, liver, and adipose tissue insulin receptors (particularly in obese individuals). In many cases, insulin resistance may be the earliest detectable marker for T2DM [22]. Patients who transition from normal glucose tolerance to T2DM typically experience a 40% decrease in insulin sensitivity, which is further accentuated by chronic hyperglycemia and elevated free fatty acids.

According to the ADA, a diagnosis of T2DM can be confirmed in patients whose A1c level is 6.5% or higher. T2DM can also be diagnosed with FPG ≥126 mg/dL, two-hour PG ≥200 mg/dL during an OGTT, or random plasma glucose ≥200 mg/dL in patients with classic symptoms of hyperglycemia or hyperglycemia crisis [28]. Prediabetes—a condition in which glucose levels are below diagnosis criteria but higher than a normal range—can also be diagnosed in those with FPG 100–125 mg/dL, two-hour PG 140–199 mg/dL in OGTT, or A1c level of 5.7% to 6.4%, although the World Health Organization and several other diabetes groups define prediabetes at a FPG of 110 mg/dL [28].

As a simple rule, to induce weight loss, patients' caloric intake should be reduced by 500–1,000 calories per day from the current level. This reduction will produce the recommended weight loss of one to two pounds per week in most patients [35].

Biguanides, specifically metformin, are the most widely used first-line T2DM medication. Both the AACE/ACE and the ADA/EASD recommend metformin as first-line therapy in most patients with T2DM, except where contraindicated [29,30]. Biguanides inhibit expression of hepatic gluconeogenic genes, thereby reducing fasting plasma glucose levels [2]. Metformin offers several benefits, including an oral route of administration (to be dosed once or twice daily) and a weight-neutral, non-hypoglycemia-inducing side-effect profile [30]. Metformin may also provide some cardiovascular benefit, although a meta-analysis does not support this claim [30,40]. In addition, generic formulations of metformin are available, which may be preferred in patients who are unable to afford the cost of this agent due to high co-pays or lack of prescription insurance [2].

The various secretagogues have different profiles. Sulfonylureas are relatively long-acting secretagogues that can decrease both fasting plasma and postprandial glucose levels, whereas glinides are short-acting secretagogues that decrease postprandial glucose levels only [2]. These oral agents are intended to be taken with meals; the morning meal if dosed daily, or the morning and evening meals (or bedtime with food) for twice-daily dosing [2]. Both types of secretagogues are associated with modest weight gain and risk of hypoglycemia, although glinides may carry a lower risk of the latter [29]. In addition, some studies have suggested these agents may be associated with a secondary failure rate in excess of other T2DM drugs resulting from exacerbation of islet dysfunction [30]. The central benefit of secretagogues is cost. These agents have generic equivalents, and many are available in combination with other drugs, such as metformin or the TZD pioglitazone [2].

DPP-4 inhibitors (alogliptin, saxagliptin, linagliptin, sitagliptin) also target GLP-1 and the incretin effect. But unlike GLP-1 agonists, DPP-4 inhibitors restrict DPP-4 to prevent enzymatic inactivation of endogenous GLP-1, thereby prolonging the availability of endogenous GLP-1 and increasing GLP-1 concentrations in the gastrointestinal tract [2]. By inhibiting more than 80% of DPP-4 activity over a 24-hour period, these agents produce an estimated 1.5-to 4-fold increase in active postprandial GLP-1 levels, resulting in increased insulin and amylin secretion from beta cells and decreased glucagon secretion and liver glucose production [2]. DPP-4 inhibitors offer many benefits, including oral route of administration (with or without food), low hypoglycemia risk, a weight-neutral profile, and a potential cardiovascular benefit [2]. They are also associated with a lower incidence of nausea and vomiting compared to GLP-1 agonists. However, like TZDs, DPP-4 inhibitors require the presence of insulin (either endogenous or exogenous) for efficacy [2]. These second-or third-line drugs are often used in combination with other agents; they have been studied and are often used with metformin, pioglitazone, sulfonylureas, and basal insulin [2].

Exogenous insulin works much like endogenous insulin in that it causes cells in the liver, muscle, and fat tissue to take up glucose from the blood and store it as glycogen. Due to the progressive nature of T2DM, injectable or inhaled insulin therapy is typically required at some point in the disease course, with the goal of creating as normal a glycemic profile as possible without causing unacceptable weight gain or hypoglycemia [30]. Basal insulin is usually offered first when insulin therapy becomes necessary, as this approach offers relatively uniform insulin coverage over a 24-hour period, with peakless time-action curves producing a more consistent effect [29,30]. Intermediate-acting, long-acting, and insulin detemir formulations are available, with the latter two offering a slightly lower risk for overnight hypoglycemia and weight gain.

The ACE, the AACE, and the ADA have each provided guidelines for the step-wise selection of therapies for T2DM (Figure 3 and Figure 4) [28,29,30,31]. Lifestyle modification (e.g., healthy diet, weight control, increased physical activity) is recommended as a first step in glycemic control, although a consensus statement from the AACE/ACE states that lifestyle optimization efforts should not delay needed pharmacotherapy in higher risk individuals [29]. The AACE/ACE emphasize that minimizing risk of weight gain and promoting weight loss in patients with adiposity-based chronic disease should be a high priority [29].

Should pharmacotherapy become necessary, metformin should be the initial drug of choice, although an AACE/ACE algorithm denotes that TZDs, sulfonylureas, DPP-4 inhibitors, GLP-1 agonists, or alpha-glucosidase inhibitors (AGIs) can be used alone or together for therapeutic initiation depending on the presenting A1c level and with consideration of other factors [29]. Two-or three-drug combinations with metformin plus other oral antidiabetes drugs are recommended when first-line therapy is deemed inadequate, with all organizations agreeing that insulin should be reserved for failure of triple combination therapy. To measure efficacy, all groups recommend regular assessment of A1c. Although target A1c levels vary slightly according to organization, all agree that elevated A1c indicates treatment inadequacy and warrants treatment escalation.

The AACE recommends an A1c target level of ≤6.5% unless contraindicated by specific factors, such as age and hypoglycemia risk [29]. This goal is different than that set forth by the ADA and EASD, which recommend a target A1c of <7.0% [30]. In addition, in 2018, the ACP controversially recommended a target A1c of 7% to 8% for most patients with diabetes [60,61]. The recommendation for tight glycemic control by the AACE is based on data from numerous older trials. The Diabetes Control and Complications Trial, the Epidemiology of Diabetes Interventions and Complications Study, and the UKPDS trial data between 1993 and 2008 are supportive of an A1c target of ≤6.5%, as are findings from the 2000 Kumamoto Study and the Steno-2 trial by Gaede et al. [62,63,64,65,66,67,68,69,70]. Although some of these trials are relatively dated, further evidence was provided by three landmark trials: the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study, the Action in Diabetes and Vascular disease (ADVANCE) trial, and the VA Diabetes Trial (VADT) [71,72,73]. In these studies, lower A1c levels correlated with reduced incidence of microvascular and sometimes macrovascular complications.

Of these studies, the ADVANCE trial provides a particularly strong rationale for tight glycemic control, as it was designed specifically to evaluate the effect of intensive glycemic control on death from cardiovascular disease, nonfatal myocardial infarction or nonfatal stroke, and new or progressing neuropathy or retinopathy [59,72]. More than 11,000 patients with T2DM were enrolled in this trial, with a median age of 67 years, a median duration of T2DM of 8 years, and a mean A1c of 7.5% upon study entry. Patients were required to have a history of major macrovascular or microvascular disease or at least one risk factor for macrovascular disease. Outcomes of this study revealed no adverse effects of intensive glycemic therapy (with an A1c target of ≤6.5%). In fact, tight glycemic control was associated with a non-significant trend toward lower cumulative mortality and a beneficial effect on macrovascular and microvascular complications (when assessed collectively) [59]. Both intensive and standard glycemic control groups also experienced improvements in blood pressure and low-density lipoprotein cholesterol levels.

Alternatively, some data challenge the importance of tight glycemic control in T2DM. In the VADT study, no beneficial effect of intensive glycemic control was observed when the entire study population was taken into consideration [59]. In fact, a deleterious effect was observed in association with tight glycemic control in patients who had T2DM for 15 years or more. This is suggestive of a potential "window of opportunity" early in the disease course for intensive control that closes over time [59]. Moreover, in the ACCORD study, the intensively treated group experienced a number of negative effects, including a three-fold increased risk of hypoglycemia and severe hypoglycemia and excessive weight gain—a well-known risk factor for increased mortality [59]. However, it is important to note that in the ACCORD trial, a 1.4% reduction in A1c was achieved within four months in the intensive control group [59]. The more gradual reduction in A1c observed in the ADVANCE trial may have influenced the apparent beneficial effect of intensive glycemic control on outcomes [59]. Still, in the ADVANCE trial, much like in the ACCORD study, excessive weight gain (and associated morbidity and mortality) was positively correlated with intensive glycemic control. Likewise, in a 2011 meta-analysis of 14 clinical trials involving approximately 28,600 patients with T2DM, intensive glycemic control did not appear to significantly impact the relative risk of all-cause or cardiovascular mortality, and study authors were unable to find sufficient evidence for the beneficial effect of intensive control on microvascular outcomes [77]. Given the severe risk of hypoglycemia that was found to accompany intensive glycemic control (with a 30% increased relative risk), authors concluded that intensive glycemic control does not appear to reduce all-cause mortality and that data fail to clearly demonstrate a benefit on cardiovascular mortality, non-fatal myocardial infarction, or microvascular complications. Other questions remain regarding the benefit of intensive glycemic control in T2DM, including duration of follow-up required to demonstrate beneficial effects, extent of influence glycemic control has on outcomes (given the numerous risk factors in T2DM), and extent of influence glycemic control has on pre-existing macrovascular outcomes [59].

As with many chronic diseases, medication adherence is a significant barrier to optimal outcomes in T2DM. Adequate adherence to T2DM therapies, which is typically defined as collecting and/or taking >80% of prescribed medication, appears to vary significantly, although the vast majority of studies on this topic report some level of nonadherence among most patients with T2DM. An estimated 36% to 93% of individuals with T2DM are believed to practice inadequate adherence to oral antidiabetic agents, a considerably broad percentage [80]. Insulin nonadherence is similarly indefinite, reported to affect between 19% and 46% of patients with T2DM [81,82]. Paradoxically, anticipated patient nonadherence actually dissuades a significant number of providers from initiating insulin [83]. Research suggests that nonadherence is a relatively immediate response in many patients with T2DM. Rather than tapering off medication over a substantial period of time, evidence suggests that patients often fail to fill second or additional prescriptions, and many discontinue medication within a year of prescription, actions suggestive of causative factors beyond simple medication fatigue [84]. The consequences of medication nonadherence are significant, ultimately resulting in hundreds of billions of dollars in additional economic cost and increased risk of mortality among nonadherent individuals compared to adherent patients with T2DM [82,85,86]. To reduce the personal and societal burden of nonadherence to antidiabetes medication, it is important that healthcare professionals understand the factors that contribute to suboptimal compliance, as well as strategies to overcome these barriers (Table 3).