AMCP Partnership Forum: Navigating Innovations in Diabetes Care

New developments that provide opportunities to enhance cost-effective diabetes care include advances in the pharmacologic treatment of diabetes, new drug delivery devices, innovations in patient management strategies, contracting strategies that incentivize effective interventions, and mobile health technologies. Payers must carefully consider the utility of these advances when making coverage decisions and designing benefits. To engage national stakeholders in a discussion about how to utilize innovations in diabetes care to optimize patient outcomes, the Academy of Managed Care Pharmacy organized the Partnership Forum on Navigating Innovations in Diabetes Care in Arlington, Virginia, on July 19-20, 2016. The forum explored current trends and advances in diabetes treatments and engaged in discussions about how organizations can leverage these emerging options to develop strategies that improve coordination of care and patient outcomes, while managing limited health resources. Additionally, stakeholders were tasked with identifying gaps in evidence that hinder decision making around novel therapies and other advances that are of direct relevance to managed care organizations.

A s a chronic condition that requires complex and multifaceted interventions to achieve optimal outcomes and avoid expensive complications, diabetes is an important disease state for managed care. As of 2014, there were approximately 29 million individuals in the United States with diabetes. 1 The effect of diabetes on overall population health is dramatic. It is the seventh-leading cause of death and is associated with annual costs of $245 billion, representing more than $1,000 per person in the country. 2,3 The prevalence of diabetes is growing and is expected to continue rising as populations at risk for type 2 diabetes (T2D) increase and as people with diabetes live longer. The influence of the obesity epidemic is a contributing factor challenging diabetes care and comorbid cardiovascular conditions. 4 This proceedings summarizes recommendations for how to optimize the implementation of advances in diabetes care that were discussed by stakeholders during a partnership forum conducted by the Academy of Managed Care Pharmacy (AMCP) in Arlington, Virginia, on July 19-20, 2016.

SUMMARY
New developments that provide opportunities to enhance cost-effective diabetes care include advances in the pharmacologic treatment of diabetes, new drug delivery devices, innovations in patient management strategies, contracting strategies that incentivize effective interventions, and mobile health technologies. Payers must carefully consider the utility of these advances when making coverage decisions and designing benefits.
To engage national stakeholders in a discussion about how to utilize innovations in diabetes care to optimize patient outcomes, the Academy of Managed Care Pharmacy organized the Partnership Forum on Navigating Innovations in Diabetes Care in Arlington, Virginia, on July 19-20, 2016. The forum explored current trends and advances in diabetes treatments and engaged in discussions about how organizations can leverage these emerging options to develop strategies that improve coordination of care and patient outcomes, while managing limited health resources. Additionally, stakeholders were tasked with identifying gaps in evidence that hinder decision making around novel therapies and other advances that are of direct relevance to managed care organizations.

P R O C E E D I N G S
be approved in the coming years. Of note, the first follow-on biologic insulin was approved in 2015, and several biosimilar insulin products are anticipated to reach the market in 2017. Although biosimilars and follow-on biologics may not provide the same cost savings as small molecule generics, they have the potential to provide substantial savings compared with currently available branded products; however, issues such as the ease of substitution of biosimilars for branded products will likely affect the uptake and pricing. Other medications in the pipeline include novel glucagon-like peptide 1 (GLP-1) analogs, including those with novel administration forms.
Another groundbreaking therapeutic advancement is the development of artificial pancreas systems, the first of which was submitted to the U.S. Food and Drug Administration (FDA) for approval in 2016. 5 It is anticipated that these systems, which combine blood glucose monitoring with automated insulin release, will reduce the diabetes management burden, improve patient quality of life, and decrease the risk of significant comorbidities.
New treatment options, combined with an increased focus on improving glycemic control have led to a reduction in the rate of diabetes complications over the last 20 years. Importantly, clinical trials of some newer treatments, including the sodium-glucose cotransporter 2 (SGLT2) inhibitor empagliflozin and the GLP-1 analog liraglutide, have shown modest improvement in cardiovascular outcomes and reduce the risk of death in certain high-risk populations. 6,7 However, additional data will be needed to determine whether these are class effects of SGLT2 inhibitors and GLP-1 analogs and if the same benefits will be seen in larger populations using these medications under real-world conditions.

Challenges Facing the Incorporation of Therapeutic Advances in Benefit Design
A key challenge regarding the treatment of T2D is the lack of comparative efficacy data for available pharmacotherapeutic options. Forum stakeholders noted that the data generated via the current drug approval process does not provide the level of comparative evidence that payers find helpful when making coverage decisions. Requiring head-to-head comparisons of active interventions before drug approval would be an improvement in the evidence base for novel diabetes treatments. Stakeholders identified the most important clinical trial outcomes beyond glycosylated hemoglobin (A1c) that would be of value to managed care decision makers when evaluating the evidence for novel diabetes treatments and tools (Table 1).
Stakeholders agreed that the development and maintenance of a data repository to generate and evaluate real-world evidence around outcomes associated with various pharmacotherapeutic options would be valuable. It was suggested that the FDA could convene a multistakeholder task force, which would include managed care decision makers, to develop a sustainable national repository of longitudinal clinical patient data to facilitate rapid learning approaches. The national repository would be built and made accessible to stakeholders with appropriate de-identification, data security, and patient consent measures. Such a database would be invaluable for determining the real-world effect of novel medications and would support the development of sensible third-party payer benefit structures, According to recommendations from the American Diabetes Association (ADA) and American Association of Clinical Endocrinologists/American College of Endocrinology, metformin is the recommended first-line treatment for patients with T2D. 8,9 For patients who require dual or triple therapy to achieve glycemic control, clinicians may select from one of several other classes of medications. No specific preference is given for any class of medications, and there is a lack of scientific data to assess the efficacy and safety of various dual and triple therapy approaches. As a result, payers do not have the data necessary to drive utilization management strategies in a manner that will support improved patient outcomes.

Top Outcomes of Interest Other than A1c Participant Discussion/Rationale Comparative effectiveness of treatments
There is a lack of evidence about the comparative effectiveness of antidiabetes medications to help managed care decision makers understand a treatment's relative place in therapy, including the impact on clinically meaningful outcomes such as reduction in cardiovascular risk and mortality.
Comparative effectiveness of treatments is especially relevant to diabetes given the multitude of available interventions for its management, the sensitivity of the condition to multiple intrinsic and extrinsic factors, variability in second-and third-line treatment approaches, and a marketplace that sees new diabetes treatments emerging at a rapid pace.
Requiring head-to-head comparisons of active interventions before drug approval would be a clear improvement in the evidence base for novel diabetes treatments. Key to this recommendation would be for the FDA to establish a standard set of active comparators based on the intended use of the novel treatment, which will make it more feasible to conduct indirect treatment comparisons and make it easier to combine study results in a systematic review.
For example, a drug seeking approval for monotherapy could have 1 active comparator (e.g., metformin), while a drug seeking approval as add-on therapy would be compared with a 2-drug therapy (e.g., metformin plus second agent). The comparator should be rationally selected (dose, delivery, and duration of treatment) and based on the current standard of care or guidelines so as to avoid any bias in the comparison and provide useful information to decision makers.

Hypoglycemia
There was a notable emphasis on the importance of measuring hypoglycemia, the current gap in evidence regarding patient concerns about hypoglycemia impacting adherence to treatment regimens, and in seeking medical treatment for hypoglycemia. It was strongly recommended that studies for all novel treatments continue to measure hypoglycemia and that data be gathered to understand the relative risks of hypoglycemia among various treatment options.

Patient-centered care/patient-reported outcomes • Time patient is within glycemic range • Depression scores/mental health assessment • Sleep time and quality • Patient self-management assessment • Quality of life • Patient satisfaction with treatment
Because patient-centered care is increasingly recognized as crucial to improving diabetes care, diabetes studies should incorporate the measurement of valid patient-reported outcomes (PROs). Some of these measures focus on the patient's perspective on treatment and care and, therefore, provide valuable information regarding the real-world utility of a medication under study.
Key to this recommendation would be for the FDA to establish a standard set of PROs, similar to the NIH Patient Reported Outcomes Measurement Information System, which will make it more feasible to conduct indirect treatment comparisons.

Hospitalizations and readmissions • Hospitalization/ER visits • 30-day readmissions
Hospitalizations and readmissions are high-priority health care quality measures and targets for cost reduction. The burden of hospitalization among patients with diabetes is substantial, growing, and costly, and readmissions contribute a significant portion of this burden. Therefore, measuring a novel treatment's effect on hospitalization and readmission rates in patients with diabetes provides the opportunity to better inform health care decision makers regarding the potential effect of a treatment option on the total cost of care. Adherence and national quality measures • Impact (improvement) to HEDIS and star measures • Adherence Patients must be adherent to medications to experience benefit. HEDIS has published quality measures for health plans on adherence to diabetes medications, but adherence is currently not reported routinely and uniformly in diabetes studies. Standardized reporting on adherence that aligns with national quality measures will enhance usability of study results for payers and providers. Adherence and barriers to adherence should be reported for all study arms.  coverage decisions, risk-sharing contracts, medical policies, guidelines, and pathways.
Stakeholders recognized the need for medical and pharmacy benefit integration for better focus on the patient health care experience, care coordination, and the total cost of diabetes care. However, they also identified several other challenges currently affecting care coordination, including variations in benefit designs, silos within the coverage determination process under medical and pharmacy benefits, uncertainty regarding the value of newer technology-supported care delivery models (e.g., telemedicine), and the need for additional data to guide coverage determinations.

Harmonizing Metrics Used in Managed Care Contracting
Quality and value measures that incentivize specific outcomes are used by managed care organizations to support the triple aim of improving the patient's experience of care, improving population health, and slowing cost growth. To this end, the use of value-based contracting has increased in recent years as a way to engage different stakeholders (e.g., hospitals, physicians, and pharmaceutical manufacturers) to align payment with quality, accountability, and care coordination. Currently, however, there is substantial heterogeneity among how quality is measured in diabetes, which creates data challenges with implementing these contracts.
Forum stakeholders recommended standardizing quality improvement measures in diabetes associated with value-based contracts so that the same measures can consistently be used throughout the health care system. This would reduce the burden of different contracts' measures on physicians, delivery systems, and payers. Standardized metrics would also generate usable real-world evidence and allow for meaningful comparisons across various treatment approaches.
Stakeholders suggested that regional coalitions (made up of payers and care delivery systems) be created to develop standardized metrics. These coalitions would facilitate consensus on the use of measures that could track the impact of therapies and create more opportunities for data integration and analysis. They suggested that these coalitions could use the measures from the National Committee for Quality Assurance that are included in America's Health Insurance Plans and the 7 core sets of quality measures from the Centers for Medicare & Medicaid Services (CMS) to support greater quality improvement and reporting across health systems. 10 Two factors were identified as key components in the strategy to encourage continuous improvement around standardized metrics: (1) transparency (data on performance needs to be transparent and unblinded) and (2) the use of benchmarking. Stakeholders acknowledged that multiple stakeholders are engaging in quality improvement activities, mainly with the same goals, such as improving HEDIS (Healthcare Effectiveness Data and Information Set) performance. However, there are often slight variations among populations, plans, or provider groups.
Standardizing clinical and nonclinical measures are equally important. Nonclinical measures include those that capture the effect on patient wellness and function, such as activities of daily living, fall prevention, and work attendance. The inclusion of measures such as annual visits with nonphysician team members (which would reflect the quality of care provided by pharmacists, nurses, dieticians, and ancillary health care providers) was another recommendation of the group.
Stakeholders also recommended that AMCP partner with physician associations to develop continuing medical education programs that focus on updating physician groups about the payer environment and managed care processes and priorities to help them be successful in a value-based contracting arrangement.

■■ Driving Patient Engagement to Improve the Quality of Diabetes Care
Patient self-management is the cornerstone of diabetes care. Effective self-management empowers patients to sustain the behavior changes needed for a lifetime of diabetes care. To manage diabetes successfully, patients must be able to set goals and make frequent daily decisions around numerous activities related to diet, physical activity, self-monitoring of blood glucose levels, and medication adherence. Participants emphasized that weight management is also crucial for effective diabetes management. Yet, even when patients are educated about the value of self-management, behavioral change remains challenging.

The Importance of Adherence to Diabetes Management
In 2010, 47% of adults with diabetes in the United States had an A1c above 7.0%. 11 Despite advances in care, there has been no change in the percentage of patients who have achieved the treatment goal established by the ADA of an A1c less than 7%. Since then, rates of very poor control (an A1c of > 9%) have hovered at nearly one third of patients. 12 A key barrier to optimizing diabetes outcomes is poor adherence to treatment regimens. Nearly one third of new prescriptions for diabetes medications are never filled; of those filled, even fewer are taken or refilled as prescribed. Barriers to adherence may include affordability of medications, treatment complexity, distrust of providers or treatments, literacy (including health literacy, numeracy, and understanding benefit design), chronic disease, and personal challenges such as depression or stress.
Payers are increasing the emphasis on patient-centered care, which is tailored to individual needs, to address issues such as lack of patient engagement that may lead to poor adherence. For instance, direct patient interaction with health care providers for education and support is increasingly being seen as a crucial component of diabetes care and is a focus in innovative care models. Other strategies that stimulate patient engagement include new reimbursement models that use payment or incentives for services that give patients more autonomy and control over their diabetes, such as waived copays for evidencebased treatments and mobile heath (mHealth) technologies such as remote monitoring and telehealth. Such technologies can be used to connect real-time, analyzed, actionable data from glucose meters and smartphone applications with appropriately trained team members (e.g., nurses and pharmacists) who can monitor data and intervene when necessary.

Leveraging Care Model Innovations
Forum stakeholders discussed several strategies to maximize the potential of innovative care models such as those previously mentioned. They noted the importance of using contracting effectively, including the development and use of quality metrics and financial incentives that support team-based care as previously discussed. Such metrics could include process

Factor
Questions and Considerations Target population • Is the app designed to be used by patients, members of the health care team, or a combination?
• Is the app tailored to patients with specific clinical characteristics?
• How will issues such as affordability and literacy be addressed? Product usability • What is the ease of use for downloading, installing, and configuring the app?
• Is there any data that support patient adherence to the use of the product? Product lifecycle • Will the timeline required to develop and implement patient support programs align with the lifecycle of technology products?
• How can patient support programs adapt to product innovations?
• Will manufacturers provide health care providers with training to adopt new products and upgrades? Product features • Does it include coaching or messaging based on behavioral theories and/or evidence-based clinical algorithms?
• Does it provide clinical decision support?
• Does it collect and analyze data to share with the health care team?
• Does it include options to create patient communities through social media? Outcomes • What outcomes does the app address and have any outcomes been evaluated in clinical studies?
• How should evidence be generated when outcomes of interest may take longer to track than the lifecycle of the mHealth product?
• What are appropriate study designs for evaluating mHealth products? Benefit design • How should payers determine which patients are the most appropriate candidates for the use of mHealth products?
• Are mHealth products covered under the pharmacy benefit, the medical benefit, or as durable medical equipment?
• Are providers compensated for their time evaluating data from mHealth products and responding to alerts that are generated? Provider communication • How is data analyzed and presented so that it is meaningful and actionable for patients and providers?
• What information will be communicated with the health care team and at what frequency (e.g., A1c levels, adherence to medication and lifestyle interventions, blood glucose levels, and hypoglycemic events)?
• Do certain types of information trigger alerts?
• Will products and programs need to be tailored to address interstate differences in practice acts, particularly for pharmacists and nurses? Data management • How is data encrypted for transmission?
• Can, or should, the data be integrated with EHR systems, and if so, how will this be done?
• Can data be used to perform predictive analytics and segment the population for targeted interventions?
• How should data be analyzed across different products and different demographics?
• What are the data-sharing policies?
• How are privacy and security for the user addressed? Legal issues • Is the mHealth product HIPAA compliant, particularly if the product interacts with social media?
• Are there liability issues related to cybersecurity and/or modification of mHealth products by users who are able to access the source code?
• Provider liability regarding data that is transmitted to the providers; what are the expectations for providers regarding the application of the data to patient care? EHR = electronic health record; HIPAA = Health Insurance Portability and Accountability Act of 1996; mHealth = mobile health. and outcomes measures, as well as incentives for diabetes selfmanagement education and prevention programs.
Strategies that increase pharmacist engagement in patient care teams were also seen as beneficial. Stakeholders recommended supporting collaborative practice agreements that allow pharmacists to perform certain diabetes management activities under a protocol and developing payment models that support pharmacists' services, as well as telemedicine, mHealth products, and virtual visits, to cost-effectively manage patients with diabetes between visits. Currently, barriers to pharmacists receiving compensation for patient care services (e.g., lack of provider status in Medicare Part B) hinder the integration of pharmacists in many settings, but some models and payment strategies are addressing these barriers to support integration of pharmacists in the care of patients with diabetes. For example, pharmacists' services may be cost justified if they result in improved performance of measures included in valuebased payment models, such as quality measures that are used by accountable care organizations.

Use of mHealth Tools in Diabetes Care
Mobile apps that are operated by handheld devices, such as smartphones, can be used for a variety of health-related functions. App functions may include allowing patients to track behaviors (e.g., maintain food diaries), linking to wearable devices (e.g., pedometers) to track daily activity, providing disease management decision support (e.g., calculating insulin dosages), and triggering interventions (e.g., regulating the release of insulin from a continuous insulin infusion device in response to real-time data transmitted from a continuous blood glucose monitor). Examples of benefits associated with such apps include the generation of real-time blood glucose data, enhancing patient empowerment and confidence in selfmanagement skills, and allowing for remote interaction and support with health care providers. 13 Some, but not all, mHealth products are regulated by the FDA. The FDA regulates mHealth products that meet the definition of a medical device and that could pose a risk to patient safety if they do not function properly. 14 For example, an app that controls an insulin pump would likely be subject to regulation. The FDA applies the same risk-based approach to ensure the safety and effectiveness of such products as it does for other medical devices. In addition to requiring clinical studies to demonstrate safety and effectiveness, the FDA may include postmarketing surveillance plans that monitor parameters that are specific to mHealth products, including cybersecurity and issues related to the use of a smartphone as a medical device (e.g., compatibility across device platforms and assurance that the app will continue to operate properly through software upgrades).
Technological innovations, including mHealth products, are revolutionizing patient care, but there is uncertainty about how to evaluate the value and leverage of such products. Stakeholders explored options for evaluating use of mHealth products, including the need for clinical and scientific differentiation. Forum stakeholders discussed a conceptual framework for evaluating mHealth products in managed care to improve outcomes and affordability ( Table 2). Use of an evaluation format, similar to the evidence-based review process for pharmaceuticals, was also suggested.
An important challenge associated with applying managed care strategies to mHealth products is the sheer number of options. As of 2015, there were over 165,000 digital health products, including more than 1,000 products intended for use in managing diabetes. 15 Forum stakeholders felt strongly that there was a need for a central repository of information about available mHealth products that would include product details, safety, accuracy, usability, and other parameters, • Engage in strategies to explore the effect of various treatment approaches on patient outcomes, such as cardiovascular disease.
• Develop a data repository to generate and evaluate real-world evidence around outcomes associated with various pharmacotherapeutic options.
• Design clinical trials for novel diabetes treatments to include outcomes beyond A1c and use of active standard-of-care comparators.
• Standardize quality and value measures used in value-based contracting for use consistently throughout the health care system.
• Consider the use of nonclinical measures that capture the impact of diabetes on patient wellness and function, including patient-reported outcomes.
• Use regional coalitions (made up of payers and care delivery systems) to develop consensus on which quality measures will be used to ease the burden on data integration and analysis. Population management strategies should be used to segment patients into subgroups that are likely to obtain the most value from various interventions.
• Managed care organizations should implement strategies and payment models that increase pharmacist engagement in patient care teams.
• Financial incentives for patients, such as waived copays, should be considered to stimulate engagement with various interventions.
• A central repository of information about various mHealth tools, including data regarding the effect on outcomes, should be developed. A1c = glycosylated hemoglobin; mHealth = mobile health.

Key Stakeholder Recommendations for Optimizing the Value of Advances in Diabetes Care
including data regarding the effect of the mHealth product on outcomes. Existing tools for assessing apps include the World Health Organization's mHealth Evaluation Reporting and Assessment Checklist, the Mobile App Rating Scale, and a guide for the evaluation of diabetes mobile apps published by Diabetes Spectrum. [16][17][18] Managed care pharmacy principles are based on the use of scientific evidence to make informed decisions in order to achieve healthy outcomes. Stakeholders expressed a desire for the creation of a framework for manufacturers to submit the evidence or information needed by health care decision makers when reviewing the use or coverage of mHealth products in health care plans. Some stakeholders suggested that the addition of reviews and recommendations based on end-user experiences could also enhance the evaluation framework. It was suggested that the AMCP Format could be used to convey evidentiary requirements for medical devices cleared by the FDA and that a database, such as the AMCP eDossier System, could be used as the database for mHealth tools. 19 Stakeholders felt that mHealth manufacturers should be encouraged to develop and make available mHealth dossiers for health care decision makers and health systems. Stakeholders identified a general need for managed care decision makers to better understand factors that affect how mHealth solutions are covered, including CMS coverage determination, the effect of regulatory approval status (e.g., legend, over the counter, and durable medical equipment), and the available resources for health technology evaluations to guide better evidence-based decisions and recommendations on mHealth solutions.

■■ Conclusions
Managed care stakeholders have offered several key recommendations for optimizing the value of advances in diabetes care, including new pharmacologic options, new management strategies, evolving care models, and novel technologies such as mHealth products (Table 3), since many of these products and innovations have demonstrated a meaningful effect on patient outcomes. Stakeholders have also identified that managed care organizations have significant opportunities to design programs that leverage contracting and incentives to improve the care of patients with diabetes, particularly those who struggle with adherence and self-management. Increased harmonization of metrics and aggregation of data will allow for identification of interventions that are effective for various patient subgroups and facilitate the design and implementation of effective programs.