Therapeutic Inertia and Delays in Insulin Intensification in Type 2 Diabetes: A Literature Review

For this review, we conducted a comprehensive search of recent human-related literature on type 2 diabetes and delay in initiating or intensifying insulin published from January 2012 to March 2022 using the Allied & Complementary Medicine, BIOSIS Previews, Embase, EMCare, International Pharmaceutical Abstracts, MEDLINE, and ToxFile databases. Keywords in titles and abstracts were combined using “OR”/“AND” operators and included search terms for type 2 diabetes and delay in initiating or intensifying insulin. Additionally, references cited in key articles were also searched manually.

After the initial search was performed, studies were screened for eligibility. Studies were eligible for inclusion if they were clinical studies, meta-analyses, or observational research pertaining to humans. Exclusion criteria included articles that were 1) not relevant to the subject; 2) published >10 years before the search date range; 3) pertaining to pregnancy; 4) not clinical studies, meta-analyses, or observational research; and 5) nonhuman studies.

A total of 116 publications were initially identified. Ninety-nine publications were excluded based on the exclusion criteria. The final inclusion set consisted of 22 publications (13–34), of which 17 were based on the search criteria and an additional five were detected through a review of references included in key studies. Results are presented by categorical organization of findings.

Therapeutic inertia in diabetes care is multifactorial, with clinician, provider, and health system contributions (13–16). Major drivers of therapeutic inertia include fear of hypoglycemia, concerns about weight gain, and the perceived complexity of insulin regimens (13–16). For example, the results of a physician survey–based study conducted by Peyrot et al. (13) found that 87.6% of physicians agreed that insulin-treated patients do not have adequate glucose control and that 75.5% would treat more aggressively if not for concerns over hypoglycemia. Results from a more recent physician survey–based study conducted by Leto et al. (17) revealed that the five most inappropriate actions contributing to therapeutic inertia in clinical practice were 1) nonuse of sulfonylureas/glinides, 2) failure to appreciate the risk of lack of hypoglycemia awareness, 3) reluctance to initiate intensification of diabetes therapy, 4) failure to initiate treatment changes for patients who are not at glycemic target, and 5) failure to train patients for hypoglycemia treatment.

In a review of published literature on the subject, Giugliano et al. (18) concluded that the greatest contributing factors to therapeutic inertia were failure to initiate or intensify insulin treatment. In support of this assertion, several studies have documented long delays in the time until insulin intensification. Escalada et al. (19) found that, in patients with poorly controlled diabetes, 31–46% of general practitioners, internists, and endocrinologists waited 3–6 months before initiating insulin, and 58–71% confirmed elevated A1C levels twice before initiating insulin. Other studies have found that, for patients with type 2 diabetes on two or three antihyperglycemic agents, the median time to insulin initiation was from 5 years to >7.1 years (20–22). Furthermore, the mean A1C for patients with type 2 diabetes on two to three antihyperglycemic agents who were initiating insulin was observed to be 8.7–9.8% (20,21,23–25), indicating a significant delay in insulin initiation. Khunti et al. (26) documented a median time until 3.7 years for insulin intensification from basal insulin, despite having an A1C >7.5%. Other studies have detected low rates of insulin intensification (5.1–30.9%) among clinically eligible (A1C ≥7.5%) patients with type 2 diabetes (16,20,26,27). A study following type 2 diabetes patients who were newly initiated on basal insulin therapy found that, after 6 months, 81% remained in poor glycemic control, and after 12 months, only 34% of the patients with uncontrolled glycemia underwent the addition of another antihyperglycemic agent (28).

A study by Paul et al. (27) identified significantly increased risks of myocardial infarction (MI) (67%), stroke (51%), heart failure (64%), and composite macrovascular events (62%) after a 1-year delay in treatment intensification. Hosomura et al. (29) found that clinically eligible patients who initiated insulin for the first time as recommended by physicians had a median time to an A1C of 7.0% of 18 months compared with 30 months for those who did not initiate insulin the first time. A meta-analysis of the ACCORD (Action to Control Cardiovascular Risk in Diabetes), UKPDS (UK Prospective Diabetes Study), ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation), and VADT (Veterans Affairs Diabetes Trial) cardiovascular trials showed reductions of 15% in MI and 9% in cardiovascular events in the tight glycemic control group (30). It has also been shown that therapeutic inertia in patients with type 2 diabetes was associated with a shorter median time of 46 months to progression of diabetic retinopathy (DR) compared with the median not being reached during a 4-year study in the noninertia group, in addition to the finding of a higher incidence of DR (10 vs. 2.2 cases per 1,000 person-months, in the inertia vs. noninertia groups, respectively) (31).

A retrospective study conducted by Hersi et al. (32) compared groups with early insulin intensification (<1 year), delayed insulin intensification (>1 year), and no insulin intensification. Findings indicated that early treatment intensification was linked to 18, 23, and 28% reductions in the risk of heart failure, MI, and stroke, respectively. The results from one study in patients with type 2 diabetes initiating basal insulin found a decrease in A1C of 1.4% and a decrease of severe hypoglycemic events from 4 to <1 per 100 person-years (33).

Using Medicare Benefits Schedule codes (general practitioner [GP], specialist, allied health, nurse, procedures, imaging, and pathology), Johnson et al. (34) detected an increase in health service usage (HSU) events from a median of 18–23 6 months after insulin initiation. The investigators reported that this increase was largely attributable to an increase in GP consultations from six to eight. HSU and GP consultations subsequently returned to baseline at 12 months (34).

This literature review revealed a time to insulin intensification of at least 5 years in patients with type 2 diabetes with an average A1C before intensification >8%. These findings are strongly suggestive of significant delays in insulin initiation and intensification for clinically eligible patients with type 2 diabetes.

The main reason for this delay appears to be fear of hypoglycemia; however, there was even a decrease in hypoglycemia as insulin was intensified. This counterintuitive finding is corroborated by the ACCORD study; the highest rates of severe hypoglycemia were observed in those with the highest baseline A1C and the smallest A1C reductions (35,36).

Additional findings suggest that such delays pose a variety of serious health risks, including the risk of DR progression and potentially fatal cardio- and cerebrovascular events. Several of the reviewed studies have shown that these risks can be dramatically reduced when insulin intensification is initiated promptly, indirectly highlighting the dangers of therapeutic inertia. In addition to decreasing health risks, insulin intensification was not found to be associated with long-term increases in HSU, which should stimulate a greater interest in reducing therapeutic inertia at the payer level.

Recognizing the lack of improvement of outcomes in type 2 diabetes, the American Diabetes Association in 2020 launched its Overcoming Therapeutic Inertia initiative (37). Part of addressing therapeutic inertia is understanding and addressing the patient and physician attitudes contributing to less timely treatment modification. Survey-based publications on patient and provider attitudes pertaining to insulin initiation and intensification identified in this literature search indicated that fear of hypoglycemia and a history of severe hypoglycemic events contribute importantly to therapeutic inertia (15,19).

However, patient-related factors must also be considered. For example, many patients are resistant to initiating and/or intensifying insulin because of their beliefs and perceptions about this mode of therapy, a condition first referred to by Polonsky et al. (38) as “psychological insulin resistance.” In addition to fear of hypoglycemia, some patients are resistant because they believe insulin therapy will be harmful (e.g., will cause blindness), or they may believe that starting insulin means their diabetes is worsening (38). Other reasons can include the desire to avoid the pain of injections, low confidence in their ability to safely use insulin, and the worry that insulin therapy will be too restrictive (38). Lack of social support (39) and psychological conditions such as depression, distress, and anxiety can also affect patient acceptance and adherence to insulin therapy (40).

Treatment with newer medications such as GLP-1 receptor agonists and SGLT2 inhibitors provide an alternative to insulin in managing hyperglycemia with a very low risk of hypoglycemia. However, these medications may not be acceptable for some patients because of intolerance of side effects and/or cost.

It is noteworthy that the advent of new diabetes-related technology specifically focused on assessing glucose levels in a continuous manner to avoid hypoglycemic events (41) and improve quality of life and psychological conditions (42,43) may alleviate these concerns and contribute to decreased therapeutic inertia. Numerous studies conducted over the past decade have shown that use of CGM is associated with significant reductions in A1C and severe hypoglycemic events in patients with diabetes receiving insulin via a variety of delivery methods (41–44). Gavin and Bailey (43) recently noted that, in addition to improvements in clinical outcomes for patients with type 1 or type 2 diabetes using CGM, the results of several studies have also indicated significant improvements in patient psychosocial measures, including reduced fear of hypoglycemia.

Additionally, associations between CGM use and significant reductions in hospitalizations for severe hypoglycemia and diabetic ketoacidosis (DKA) have been detected in large observational registry and database studies (45), strongly suggesting a level of cost-effectiveness that may be valuable for consideration by payers interested in reducing the huge financial burden associated with diabetes.

Given that new diabetes technology has been found to improve diabetes management, better understanding of how racial/ethnic group identity and socioeconomic status affect access to these new tools may be useful to help mitigate therapeutic inertia. The 2021 U.S. Census found that Hispanics and Blacks have lower median incomes than non-Hispanic Whites ($51,560, $41,361, and $70,642, respectively) (46). Lower socioeconomic status and minority race/ethnicity have been associated with higher A1C levels (12,47,48) and greater risks for negative health outcomes (e.g., DKA, DR, lower limb amputation, major cardiovascular disease, stroke, and end-stage renal disease) in patients with diabetes (6,49–52), and these populations are also the most likely to experience therapeutic inertia (53,54). Non-Hispanic Blacks and people with the lowest annual incomes also have the highest rates of severe hypoglycemia (55–58). A study conducted by Miller et al. (47) showed that, regardless of income level and type of insurance (i.e., none, state/federal, or private), individuals using CGM had lower A1C levels than nonusers of CGM. In addition, initiating CGM has been found to reduce severe hypoglycemia (41,59). Recent evidence suggests that patients with diabetes from minority racial/ethnic groups are less likely than their non-Hispanic White counterparts to use CGM (8–12), and likewise that those with lower socioeconomic status are less likely to use diabetes technology (47,48).

Collectively, the results of this comprehensive literature review on the causes and consequences of delay in initiating and intensifying insulin for patients with type 2 diabetes support an urgent need for overcoming the therapeutic inertia that places this patient population at high risk for negative health outcomes. New diabetes technology, including CGM, may contribute to a dramatic decline in the therapeutic inertia driven by concerns about increased hypoglycemia with insulin initiation or intensification, which can now be readily assessed and more comprehensively addressed.

Therapeutic inertia remains a barrier to the optimal care of patients with uncontrolled type 2 diabetes. Patients, providers, and payers are uncomfortable with insulin initiation and intensification largely because of concerns about provoking hypoglycemia, which can lead to hospitalizations and an ever-increasing cost burden. Based on recent developments in diabetes technology, including the widespread use of CGM, hypoglycemia may become less of a persistent consequence of intensified insulin treatment, and this in turn may reduce therapeutic inertia via improved confidence among all stakeholders. Barriers to CGM access will need to be addressed to decrease therapeutic inertia and provide effective treatment for all patients regardless of their racial/ethnic identity or socioeconomic status.

The authors thank William Perlman, PhD, CMPP, for his editorial assistance in developing this article.

This study was funded by Abbott Diabetes Care, Alameda, CA.

J.R.G. has served on advisory boards and/or speaker bureaus for Abbott Diabetes Care, Boehringer Ingelheim, Medtronic, and Novo Nordisk. R.M.A. and N.S.V. are employees of Abbott Diabetes Care. No other potential conflicts of interest relevant to this article were reported.

J.R.G. contributed to the concept and design of the study, provided data regarding health disparities, participated in writing the initial draft, and provided critical revision of the manuscript. R.M.A. contributed to the acquisition of data, analysis and interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. N.S.V. participated in the acquisition of the data, analysis and interpretation of data, and drafting of the manuscript and provided administrative, technical, or logistic support and supervision. All authors reviewed the manuscript and approved its submission. J.R.G. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.