INTRODUCTION
Cardiovascular disease (CVD) has long been the leading cause of death among individuals with type 2 diabetes mellitus (T2DM), who are twice as likely to develop CVD than those without diabetes. Despite the increased CV risk associated with T2DM, much of CVD can be prevented with aggressive risk factor management. Indeed, persons with T2DM who have all five risk factors (glycated hemoglobin [HbA 1c ], low-density lipoprotein cholesterol [LDL-C], albuminuria, tobacco use, and blood pressure) within the target ranges have minimal excess risk of death, myocardial infarction, or stroke compared with those without T2DM. Beyond traditional risk factor management, we now have increasing therapeutic options with which to combat this increased CV risk. In addition to the significant benefit that high-intensity statins, BP control, ACE inhibitors (ACEi), and angiotensin receptor blockers (ARB) provide in CV risk reduction, glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium glucose cotransporter-2 inhibitors (SGLT2i) have more recently been shown to impart significant CV benefit in T2DM, independent of their glycemic effect. Thus with appropriate risk factor control and use of effective therapies, we have within our grasp the ability to significantly reduce the burden of CVD among persons with T2DM.
However, the difference between what we know and what we actually do in clinical practice is unfortunately stark. This is the so-called quality chasm, a term coined by the Institute of Medicine in the landmark 2001 report “Crossing the Quality Chasm: A New Health System for the 21st Century.” In CV risk reduction in T2DM, this quality chasm manifests as too few persons actually achieving complete risk factor control or being prescribed and taking all evidence-based therapies for which they are eligible. The reasons behind these gaps in care for persons with T2DM are multiple, varied, and as complex as the persons and the healthcare ecosystem itself. To begin to understand and address why persons with T2DM are not receiving optimal care, we must apply the same scientific rigor that we use to develop effective therapies in the first place.
Implementation science provides this backbone for scientific inquiry into the reasons behind and remedies for the quality chasm. The National Institutes of Health defines implementation science as “the study of methods to promote the adoption and integration of evidence-based practices, interventions, and policies into routine health care and public health settings to improve our impact on population health.” Indeed, implementation science can been defined in many ways, but at its core, it is the science of systematically getting “what works” to the people who need it. It requires a commitment to producing generalizable knowledge about applying the evidence that can be scaled and used to improve health on a population level.
Here, current challenges facing providers of and persons with T2DM in achieving optimal care will be reviewed, and ways in which implementation science can provide opportunities to improve the management of this high-risk population will be explored.
The Quality Chasm in CV Risk Reduction in Type 2 Diabetes
The quality chasm in CV risk reduction in T2DM exists on multiple levels, including achievement of guideline-recommended risk factor targets and prescription of and adherence to evidence-based therapies. With regard to achieving optimal risk factor levels, data from three diverse US longitudinal cohorts between 1990 and 2004 revealed that only 7.2% of individuals with T2DM were at target for a composite of three risk factors (BP < 130/80 mm Hg, LDL-C < 100 mg/dL, HbA 1c < 7%). However, those who did achieve this composite were at significantly decreased risk (HR 0.38 [95% CI 0.25–0.58]) for CVD events compared with those without any risk factors at target. Subsequently, an examination of the US National Health and Nutrition Examination Survey between 1999 and 2010 revealed that even though there were modest improvements over time, only 25% of individuals with T2DM in 2009–2010 had all three of the above risk factors at goal. From a global perspective, within the context of an international trial in 2008–2012, only 29.9% of persons with T2DM and ASCVD achieved five secondary prevention goals (aspirin therapy, LDL-C < 70 mg/dL or statin therapy, BP < 140/90 mm Hg, ACEi or ARB use, nonsmoking status). More recently, a study from the US Diabetes Collaborative Registry found that only 13% of individuals with T2DM achieved a composite of four risk factors at target (HbA 1c < 7% or < 8% with ASCVD, LDL-C < 100 mg/dL or < 70 mg/dL with ASCVD, BP < 130/80 mm Hg, and nonsmoking status). Perhaps even more striking were the discrepancies by socioeconomic status and race: individuals with high median household income were 29% more likely to achieve the composite than those with low income, and White individuals were 43% more likely to achieve the composite than non-White individuals.
Contributing to this low level of risk factor control are significant deficiencies in the prescription of evidence-based therapies to reduce CV risk among pesons with T2DM. These gaps persist even with older therapies that are supported by several decades’ worth of data. For example, among 155,958 persons with T2DM and ASCVD in a commercially insured population, only 24.7% were on a high-intensity statin, and only 53.1% were on an ACEi or ARB. When SGLT2i and GLP-1RAs were also taken into account, only 2.7% of the eligible population were on all three classes of therapies ( Fig. 27.1 ). An examination of health systems across the United States yields similar findings: among over 320,000 persons with T2DM and ASCVD in 12 large health systems, only 4.6% were on all three classes of therapies; 42.6% were on none.
Proportions of persons on all three, two, one, or none of the evidence-based therapies: high-intensity statin, ACEi or ARB, SGLT-2i or GLP-1RA. ACEI , Angiotensin-converting enzyme inhibitor; ARB , angiotensin receptor blocker; SGLT2i , sodium-glucose cotransporter 2 inhibitors; GLP1-RA , glucagonlike peptide-1 receptor a gonists.
Nelson AJ, Ardissino M, Haynes K, et al. Gaps in evidence-based therapy use in insured persons in the United States with type 2 diabetes mellitus and atherosclerotic cardiovascular disease. J Am Heart Assoc . 2021;10:e016835.
SGLT2i and GLP-1RA appear to have particularly poor usage in the United States. Despite increasing uptake overall, 80% of clinicians who prescribed metformin to Medicare beneficiaries in 2018 did not prescribe SGLT2i, and sulfonylurea prescriptions were three times more frequent than SGLT2i, despite no evidence of CV benefit. Prescription rates are also low in the Veterans Affairs (VA) health system, despite coverage of these agents. In 2020, 11.2% of persons with T2DM and ASCVD in the VA system received an SGLT2i, while 8.0% received a GLP-1RA. These individuals were more likely to be young and White, again highlighting disparities in use. Despite the magnitude of ASCVD and HF benefit from these agents, cardiologists appear to be particularly hesitant to prescribe these medications. This hesitancy is reflected in national prescribing data: in 2020, cardiologists represented only 1.5% of SGLT2i prescriptions and 0.4% of GLP-1RA prescriptions.
Adherence to evidence-based therapies in T2DM is also quite low, with significant disparities by race. For example, persistent statin use at 1 year after initiation is significantly lower among younger, Black, and Hispanic persons versus older White persons. With regard to SGLT2i, approximately 55% of persons are not highly adherent (defined as proportion of days covered ≥80%), and those with lower adherence are more likely to be female and/or Black. Rates of adherence to GLP-1RA appear to be similar.
Thus, while new, effective therapies continue to be developed, unacceptably low prescription of and adherence to these therapies persist, and disparities widen. Reasons for this quality chasm are multifactorial and include society-level social determinants of health, as well as health system level, provider level, patient level, and cost barriers. The optimal way to overcome all of these barriers is not clear, but one thing is absolutely certain—what we as a medical community have been doing thus far is not working. It takes approximately 17 years for new evidence from clinical trials to be incorporated into routine clinical practice, and even then application is inequitable, as is supported by the data above. If we are to improve the quality of care for persons with T2DM, we need to apply a rigorous, iterative, scientific process to achieve this goal—such as the one provided by implementation science.
IMPLEMENTATION SCIENCE PRINCIPLES
Implementation science can be thought of as the science of turning what we know into what we do . By applying research findings to real-world settings, implementation science seeks to integrate knowledge and best practices into practice and policy in a sustainable way. This discipline is distinct from quality improvement, which focuses on a specific problem in a specific healthcare setting; implementation science, in contrast, starts with an underutilized evidence-based practice and seeks to create knowledge that can be generalized to other settings.
While clinical studies assess the impact of a therapy or evidence-based practice on health outcomes, implementation studies focus on the rate or quality of use of the therapy or practice. For example, a clinical study may evaluate cardiovascular and renal outcomes of a given SGLT2i, but an implementation study may evaluate the rates of prescription and adherence to SGLT2i across broad populations. As such, outcomes of implementation studies generally focus on process metrics and on the use of the target therapy or practice. As opposed to clinical studies, in which the study teams are often blinded to the outcome during the trial, data in implementation studies can be fed back to the study team to improve the implementation process. This so-called formative evaluation allows for an iterative process that is central to implementation science but must be specified a priori as part of the research plan or study hypothesis. Of note, elements of clinical and implementation studies can be combined in hybrid effectiveness-implementation designs, which both evaluate the effectiveness of a given therapy or practice and evaluate the implementation strategies to increase or improve its use.
An essential component of the implementation science process is early, frequent, and meaningful engagement of all relevant stakeholders. This begins as early as research question formulation. Gaining a deep understanding of the barriers and facilitators that stakeholders face when utilizing a given therapy or practice is a critical first step to the implementation process. In this context, the term stakeholder is used broadly to indicate any individual or organization who is relevant to the use of the therapy of interest, including persons, families, caregivers, clinicians of varying types, health systems, payers, policymakers, and others. Importantly, particular thought should be given to those who should be involved in the process but are not because they have been historically marginalized.
Once a gap in care has been identified and stakeholders have been engaged to better understand the barriers and facilitators of the use of the evidence-based therapy or practice, implementation strategies that specifically address those barriers and facilitators must be chosen. Implementation strategies are, quite simply, the “stuff we do” to increase the use of a given therapy or practice. Significant efforts have been made to codify discrete implementation strategies, leading to a compilation of 73 discrete implementation strategies. Examples of implementation strategies as they pertain to patient-, provider-, and system-level barriers to optimal CV care for persons with T2DM are outlined in Fig. 27.2 .
Framework of barriers to optimal management of T2D and CVD and corresponding implementation strategies. CVD , Cardiovascular disease; T2DM , type 2 diabetes mellitus.
Gao Y, Peterson E, Pagidipati N. Opportunities for improving use of evidence-based therapy in persons with type 2 diabetes and cardiovascular disease. Clin Cardiol. 2019;42:1063–1070.
Finally, a critically important component of the implementation process is rigorous evaluation of outcomes. Many potential implementation outcomes exist, each with specific measurement tools and metrics available to assess them. Such outcomes include acceptability, adoption, appropriateness, costs, feasibility, fidelity, penetration, and sustainability. Several frameworks have been developed to guide the evaluation process for implementation studies, perhaps the most well-known among them being the RE-AIM framework. The RE-AIM framework focuses on concepts that are important for the external validity of a given program: Reach (the proportion or representativeness of individuals who are willing to participate in the program), Effectiveness (the impact of the program on important outcomes), Adoption (the proportion or representativeness of settings and people delivering the intervention who are willing to participate in the program), Implementation (fidelity to the intervention protocol and consistency of delivery), and Maintenance (long-term sustainability). Utilizing a systematic approach to evaluating implementation efforts is critical for the generalizability of this knowledge.
IMPLEMENTATION SCIENCE IN CV RISK REDUCTION IN TYPE 2 DIABETES
Several examples of implementation studies to improve the care of high-risk persons with T2DM can illustrate the principles outlined above.
Provider Level
As outlined above, there are significant deficiencies in the degree to which evidence-based agents are prescribed to persons with T2DM and CVD. The COORDINATE-Diabetes Trial (COOrdinating CaRDIology CliNics RAndomized Trial of Interventions to Improve OutcomEs) was a cluster-randomized implementation trial to increase the prescription of evidence-based therapies in persons with T2DM and CVD (NCT03936660). The trial tested whether a multifaceted intervention that incorporates collaboration between cardiologists and diabetes specialists, provider education, site-tailored care pathways, and audit and feedback to providers could increase the prescription of ACEi/ARB, high-intensity statin, and SGLT2i or GLP-1RA. Randomization to this intervention versus standard education occurred at the level of the cardiology clinic. After 6 to 12 months, individuals in the intervention arm were more likely to be prescribed all three therapies (173/457 [37.9%]) versus those in the usual arm (85/588 [14.5%]; absolute difference of 23.4%; adjusted odds ratio [OR], 4.38 [95% CI, 2.49–7.71]; P < 0.001). This large, randomized trial proved that a coordinated, multifaceted intervention could improve the prescription of evidence-based therapies across cardiology clinics in the United States.
System Level
Several implementation studies targeting system-level aspects of care delivery to high-risk persons with T2DM have been evaluated. A study within the VA health system evaluated the implementation of an intervention that had already been well-proven to prevent T2DM, the Diabetes Prevention Program (DPP). Using the RE-AIM framework to assess outcomes, they found that their method of implementing the DPP, called the VA-DPP, had better fidelity and participant satisfaction than standard of care but experienced significant barriers related to reach. Another study in the Mass General Brigham Network randomly assigned individuals with T2DM and increased CV or kidney risk to simultaneous virtual patient education and prescription of SGLT2i or GLP-1RA versus 2 months of virtual patient education followed by medication prescription. The intervention was innovative because it was completely remotely delivered and was led by nonlicensed navigators and clinical pharmacists. After 6 months, those in the Simultaneous arm were more likely to have a prescription for SGLT2i or GLP-1RA (70%) than those in the Education-First arm (56%). Of note, prescription of either of these therapies was 64% across both arms, indicating a high rate of success of the overall intervention, regardless of whether the prescriptions were given early or later.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
