Stakeholder Analysis: Medicare Diabetes Prevention Program Awareness and Implementation

Abstract

Objectives:

The Medicare Diabetes Prevention Program (MDPP) provides unprecedented coverage of a behavior change program for older-adult Medicare beneficiaries, but uptake has been extremely limited; only 1.5 sites deliver the program per 100,000 beneficiaries nationwide. Inadequate reach and utilization of the MDPP threaten its long-term success; thus, the purpose of this project was to determine facilitators and barriers to MDPP implementation and use in western Pennsylvania.

Design:

We conducted a qualitative stakeholder-analysis project with suppliers of the MDPP and healthcare providers.

Methods:

Using an implementation science framework, we conducted individual interviews with five program suppliers and three healthcare providers (N=8) to determine their perspectives on positive aspects of the program and reasons for MDPP unavailability and lack of use. Data were analyzed using Thorne and colleagues’ approach of interpretive description.

Results:

Three main themes emerged: 1) facilitators and attributes of the MDPP, 2) barriers to MDPP implementation, and 3) suggestions for improvement. Facilitators of the program included technical support and webinars from Medicare to assist with the application process. Barriers such as financial reimbursement constraints and a lack of a systematic referral process were noted. Stakeholders suggested refinements to participant eligibility and performance-based payments, a seamless method of flagging and referring patients through the electronic health record, and ongoing virtual program delivery options.

Conclusion:

Findings from this project can be used to improve implementation of the MDPP in Western Pennsylvania, support Medicare policy refinement, and inform implementation research to promote broader adoption of the MDPP across the U.S.

Précis:

This project describes facilitators and barriers to the underutilized Medicare Diabetes Prevention Program from the perspectives of healthcare providers and program suppliers in Western Pennsylvania.

Background

Population-level strategies to prevent type 2 diabetes are urgently needed for the >24 million older adults with prediabetes in the US. ¹ Evidence-based lifestyle interventions can prevent diabetes onset, per evidence from the landmark Diabetes Prevention Program trial. ² Thus, the Centers for Disease Control and Prevention (CDC) launched the National Diabetes Prevention Program (NDPP) in 2010. ³ Significant reductions in weight and medical spending were observed among Medicare beneficiaries who participated in the NDPP, ⁴ prompting the Centers for Medicare and Medicaid Services (CMS) to fully cover the Medicare Diabetes Prevention Program (MDPP) starting in 2018. ⁵

Despite unprecedented Medicare coverage for a disease prevention program, MDPP uptake is limited. Only 1 MDPP site existed per 100,000 Medicare beneficiaries in 2019, ⁶ increasing to 1.5 sites per 100,000 beneficiaries in 2021. ⁷ Distance is another challenge; 43% of Medicare beneficiaries live >25 miles from an MDPP, ⁸ an insurmountable barrier for many older adults. Regarding awareness, national guidelines recommend referral to lifestyle intervention for adults aged 40–70 years with prediabetes. ⁹ Yet, <5% of adults eligible for a NDPP reported receiving a referral, ¹⁰ which may stem from limited awareness among healthcare providers. ¹¹ Thus, we conducted a qualitative stakeholder analysis to learn about regional awareness of; referral to; and facilitators and barriers to the MDPP.

Method

We recruited program suppliers and healthcare providers in Western Pennsylvania through local outreach among CDC-registered programs and referrals from participants. Program suppliers included organizations that are MDPP suppliers and organizations that are NDPP suppliers but not Medicare-certified. The initial contact was with the chronic disease prevention program director of a YMCA. Subsequent stakeholder interviews stemmed from colleague introductions and word-of-mouth. Eight interviewees included five program directors (three from YMCAs, one from a private organization, and one from a hospital system) and three healthcare providers (two family physicians and one dietician). Interviews were conducted via Zoom^® from March-September 2021. Our project was designated as quality improvement in consultation with our Institutional Review Board. Participants provided verbal consent for interviews, knowing that aggregate, de-identified information would be used for program improvement.

We used the Consolidated Framework for Implementation Research (CFIR) ¹² to inform our interview guide. CFIR has 5 domains—intervention characteristics (e.g., cost), outer setting (e.g., patient needs), inner setting (e.g., resources), individual characteristics (e.g., self-efficacy), and implementation processes (e.g., executing). CFIR suggests interacting individual, organizational, and societal-level constructs may influence the MDPP’s reach and use. ¹² Semi-structured interview guides were developed for each stakeholder group. For example, to assess program design quality, program suppliers were asked, “What supports, such as online resources or marketing materials, are available to help you implement and use the program?” “Is cost a consideration for implementation?” was asked of program suppliers to determine the financial aspects of MDPP implementation. Interviews of healthcare providers focused on knowledge of the MDPP and referrals. For example, providers were asked, “Do the healthcare providers within the organization refer older adults with prediabetes to the MDPP? What is the process for that to happen?” and “Do you believe there are things that should be changed to increase referrals to the MDPP?”

Individual interviews lasted 35–60 minutes, were recorded with permission, and transcribed verbatim. Data were analyzed using interpretive description¹³ to understand patterns and individual variations in complex clinical phenomena, with CFIR as the organizing framework. Data analysis was concurrent with data collection. Two authors coded the interviews using broad-based coding to organize data inductively (derived from the data) and deductively (derived from CFIR constructs). By comparing data elements within these coding structures, patterns and themes emerged. Notes kept during analysis aided in interpretation and facilitated consensus between analyzers.

Results

Three main themes emerged: 1) facilitators and attributes of the MDPP, 2) barriers to MDPP implementation, and 3) suggestions for improvement (Table 1).

Table 1.

Themes Associated with Implementation of the Medicare Diabetes Implementation Program

Themes	Main Ideas	Supporting quotes
Facilitators and Attributes of MDPP	Well organized, relatable curriculum	“…if people didn’t feel that the curriculum was achievable and relatable, I don’t think they would be coming back.”
	In-person format provided personal relationships; online format worked well during COVID restrictions	“We had older adults that wanted to repeat DPP. They wanted to keep doing it because they liked the interaction… being part of a group.”
		“We had to switch everything to virtual, but it’s been amazing for our participants, not only Medicare, but our Medicaid Participants have so many other barriers, just getting to a class.”
	Successful recruitment strategies	“So, we’ve done churches and schools… in senior centers, libraries. Pretty much any place that was willing to give us a room at no cost.”
		“I just spent $4,000 on print ads. We have a digital ad campaign. We have a text ad campaign. I have collaborated with the Area Office on Aging, the Department of Health, a residential home.”
	Use of electronic systems for referral to the program	From hospital-based supplier: “And we have an EMR, we have EPIC. So, we have all kinds of decision trees made, both on the acute care and the ambulatory side, anybody who triggers positive for DPP eligibility, our referral is sent to the primary care physician. They don’t have to refer, but we’re trying to build that collaboration and ongoing education.”
		From health care provider: “We have a registry of patients with prediabetes that, every Sunday night, spits out a report of everyone with prediabetes who’s coming into the office over the next week for a visit; and then our nurse navigators, the day before the visit, when they were doing chart prep, run the calculator on that patient that day before.”
Barriers to MDPP Implementation	Challenging CMS application process	“So, the biggest challenge for us was getting our board members social security numbers… As well as, you know, their birthdates, their addresses, that kind of information.”
		“Why are they making it so cumbersome? It’s irritating to me, you know. I don’t see why Medicare wouldn’t recognize our diabetes education approved program.
	Upfront funding and reimbursement criteria from CMS	“…the majority of the costs are upfront- session one, four, and nine trigger the biggest payments.”
		“I gotta tell you, I’ve been at this for five years. It is not worth it. It is not going to make one bit of difference on your P & L [profit and loss] for your organization.”
	Inadequate number of MDPPs for face-to face delivery; travel	“I have a strong opinion on the biggest barrier, why a lot of people haven’t started off in the program is the CMS rules— ‘we [CMS] will pay you after the fact, if you prove to us that you make the patient lose weight.’ People are saying, ‘we’re not investing the money in this… there’s no guarantee that the patients are going to perform.’ I think that’s the biggest barrier.”
		“Some barriers that we’ve seen are that they [participants] don’t like to drive at night. So having an evening class can be tough sometimes, but we try and offer daytime… we even have offered morning classes in the past.”
	No systematic referral process across health care providers	“we just do a referral system with the physicians’ offices currently. I have had a few people call in to me that they’ve seen that we are a Medicare supplier in their booklet.”
		“… [I’m] going out to some of these primary cares, I mean, some of them are still on paper medical records, and they’re so resource limited that, it’s really hard.”
Suggestions for Improvement	Refine eligibility and payment criteria from CMS	“I think they need to decrease all those crazy regulations. it’s not like we are giving out opioids here. We are providing health, disease prevention, information. Why make it so difficult?”
		“It’s hard for me to say this in the middle of a pandemic knowing where all our federal resources are going, but you know, the coverage for the program would have to drastically change.”
	Allow for MDPP delivery using both face-to-face and virtual options	“…a lot of seniors don’t want to drive at night, you know. So, meeting them where they are. I know that sounds redundant, but we have to meet the seniors where they are and make it entertaining…”
		“…so if you’re focusing on the senior population, we need to make access so simple that we’re pushing it out where they are—churches, senior centers, their cell phones, if they are more savvy, the tablets, or whatever. It’s probably going to take government funding to make it smoother.
	Develop systematic referral system across all providers and hospital systems	…“listen, we did this at our little pocket of [agency name] in the Eastern suburbs. This is something that needs to be implemented in the [health care system computer system]. This is an effort that we need to do all over Western Pennsylvania, and the [health care system] footprint.”
	Educate patients and providers about MDPP	“My plan before COVID was to pair with the Area Agency on Aging. And to really have them, you know, help us get the word out, which is in the works.”
		“We would go out to physicians’ offices, specialty offices, and provide lunch for them, and then tell them what we have. We offer seven disease prevention or management programs”

Facilitators and Attributes of MDPP

All program directors reported positive aspects of the program based on personal experiences and participant and facilitator feedback. They perceived that the curriculum allowed for topic reinforcement throughout the program. In-person sessions facilitated positive relationships between coaches and participants. MDPP participants enjoyed the support, connection to others, and motivation from group members. Program directors reported many older adults were not comfortable enough with computers to engage in an online program and preferred face-to-face sessions. Yet, transitioning to virtual delivery during the COVID-19 pandemic was reported as successful. Suppliers further noted that MDPP participants appreciated connecting to others virtually during a time of greater social isolation.

MDPP suppliers and healthcare providers perceived increasing referrals in recent years (except during the COVID-19 pandemic). This uptick was partly attributed to new protocols to flag eligible patients within electronic health records (EHRs). One provider incorporated an automated approach of flagging and referring patients to a prediabetes program. Other providers stated they were informing patients of their prediabetes more frequently.

MDPP suppliers dedicated financial and personnel resources to marketing their program. They conducted information sessions at senior centers, health fairs, and community centers to recruit participants. In some instances, an incentive was offered. One supplier offered a 1-year free YMCA membership after participants were enrolled. Former MDPP participants also promoted recruitment through positive feedback to friends and family.

Barriers to MDPP Implementation

Program suppliers cited many bureaucratic difficulties. Some suppliers perceived the Medicare-designation process as complicated, including a lengthy process to become a MDPP provider (even if already a general Medicare provider), certify each program coach and have them obtain a National Provider Identifier. and provide social security numbers for all board members. ¹⁴ However, CMS did provide technical support and helpful webinars to assist with the application process if they actively sought assistance. Suppliers reported needing substantial resources to provide the program, e.g., an employee to manage enrollment, physical space, personnel, and financial support. Different standards for participant eligibility were also challenging. For example, the NDPP does not require a blood test documenting prediabetes, whereas the MDPP does.

Barriers also included issues with payments. Program suppliers reported that CMS payments were not always timely or sufficient to cover costs, making it challenging to recoup initially uncompensated upfront costs and manage their budgets. All MDPP suppliers we interviewed also provided the NDPP, which was reimbursed by private insurance companies; that reimbursement was seen as essential for balancing overall budgets because of inadequate Medicare payment. CMS maintains strict performance-based payments that prioritize ≥5% weight loss. However, participants often achieve less weight loss, ¹⁵ leaving suppliers to absorb an increasing share of program costs. Additionally, CMS will only cover the MDPP once in a beneficiary’s lifetime (e.g., one supplier stated, “one and done”), and no payments are received if suppliers serve a beneficiary who needs to re-enroll.

MDPP suppliers described difficulties faced by older-adult participants. Participants were expected to attend in-person sessions (per CMS requirements, except during the COVID-19 pandemic). Due to the limited number of MDPPs, this requirement likely necessitated driving a fair distance. Many older adults were unable to drive at night, requiring daytime sessions, even if inconvenient for others. Furthermore, MDPP suppliers reported participants were often unaware that prediabetes is a diagnosed condition or that a prevention program is available at no cost.

Healthcare providers reported a lack of MDPP awareness among their colleagues. Even if providers were aware of the MDPP, the limited number of local programs impeded patient referrals. They also stated that some providers may be biased against lifestyle interventions, leading to fewer referrals. Additionally, there was often no systematic process for generating referrals based on diagnosed prediabetes. One provider indicated that recording a prediabetes diagnosis may even be avoided because patients are uncomfortable being labeled as such; thus, patient referral may be difficult if alternate diagnosis codes are used, e.g., impaired fasting glucose.

Suggestions for Improvement

Stakeholders recommended multiple systematic changes for the program. They believed healthcare providers needed to commit to recommending diabetes prevention programs. Consistently diagnosing prediabetes, such as using a risk calculator within the EHR to predict low, medium, or high risk of developing diabetes within 3 years, and referring eligible patients to specific program suppliers were considered essential. Stakeholders commented that a systematic referral process utilizing EHRs to seamlessly connect providers directly to suppliers would be beneficial.

MDPP suppliers suggested CMS improve eligibility and payment criteria. Removing the once-in-a-lifetime limit would help those who dropout due to unforeseen circumstances but later wish to re-enroll. Additionally, program suppliers recommended more flexible payment criteria. For example, if a participant achieved 4% weight loss instead of 5%, a payment schedule could be prorated accordingly (vs. all-or-nothing payments for weight loss currently). Increasing payments to better align with private payers and providing payments to support upfront costs were suggested.

MDPP stakeholders believed that Medicare beneficiaries needed to be informed about prediabetes and the no-cost MDPP. Specific activities endorsed by stakeholders included community outreach to present at senior centers or local health fairs, and meeting with local primary care and specialist physician offices to promote their awareness of and referral to the MDPP. The COVID-19 pandemic required suppliers to deliver the MDPP virtually. Suppliers noted that virtual delivery was successful overall, and especially helped participants who needed to maintain social distancing or experienced other barriers to in-person attendance. Because there are few or no suppliers who provide the MDPP near where many Medicare beneficiaries live,^6,7 continuing to allow virtual MDPP delivery appears both necessary and appropriate to increase program access and uptake.

Discussion

This project aimed to learn about awareness of, referrals to, and facilitators/barriers for the MDPP from program suppliers and healthcare providers in Western Pennsylvania. Results are largely consistent with previous findings, including major challenges of insufficient reimbursement, low awareness of and referrals to the program, and the need to accommodate priority populations. ^16–18 Newer insights include implementation facilitators/barriers regarding lagged payments, recommendations about formally diagnosing prediabetes among MDPP participants, and noted benefits specific to virtual delivery. For one, our findings support continued virtual delivery of the MDPP, which facilitated access and social support for older participants. Despite evidence supporting the effectiveness of virtual delivery, ¹⁹ including among Medicare beneficiaries, ²⁰ CMS recently refrained from allowing virtual MDPP delivery after the COVID-19 public health emergency. ²¹ In turn, our findings may aid future rulemaking, and policy recommendations from this stakeholder analysis are summarized in Table 2.

Table 2.

Stakeholder Recommendations to Improve MDPP Policy

Program Reimbursement  • Provide payment upfront for the start-up of an MDPP.  • Provide payments in a timely manner.  • Increase payment amounts to better align with private insurance providers.  • Adjust performance-based payments based on weight loss, e.g., provide a prorated payment for 4% weight loss. Program Requirements  • Align CMS’s MDPP participant eligibility with the CDC’s NDPP eligibility requirements, e.g., eliminate blood test documenting prediabetes.  • Allow beneficiaries to participate more than once if they need to re-enroll in the future.  • Streamline the process to become a MDPP supplier, e.g., remove requirements to obtain the social security number of each member from the board of directors of the organization.  • Allow virtual program delivery beyond the timeframe of the current public health emergency.

The need for a systematic process to refer patients to the MDPP was highlighted. EHR systems that identify patients with prediabetes, flag them for the healthcare provider, and directly connect patients to an MDPP supplier could improve referral and use. Using EHRs to create a diabetes-risk registry²² of patients who meet MDPP eligibility criteria, along with decision trees to aid referrals is recommended. All healthcare providers must demonstrate meaningful use of EHRs to receive Medicare reimbursement, and these EHR functions support meaningful use through care coordination. ¹⁴ Automated messages can also encourage providers to document a prediabetes diagnosis, which is associated with improved outcomes in the general population. ²³

CMS solicits public comments and uses reputable evidence for program refinement, so our data from stakeholders could be used accordingly. For example, the gap between MDPP costs and reimbursement became increasingly well-documented following the initial launch of the MDPP. ^17,18 Subsequent recent policy change lowered costs and increased payments by eliminating requirements for a second year of MDPP sessions and shifting those payments to the first year of program delivery. ²¹ Based on findings here, further recommendations include continuing to increase payments to fully cover costs and reducing payment wait times. In turn, the overall impact of any changes in CMS rulemaking may be assessed by examining whether or not (and to what extent) there is a corresponding increase in the number of participating suppliers and beneficiaries.

Project limitations include focusing on perspectives from regional program suppliers and healthcare providers. Nonetheless, issues and suggestions raised have potential to be broadly applicable and may still contribute to the body of literature that aids future CMS rulemaking. Future directions also include assessing the perspectives of older adults. Lastly, our findings on barriers and facilitators can help inform research on the MDPP within the field of implementation science, which focuses on adoption of evidence-based interventions into real-world practice.

Conclusion

Medicare beneficiary enrollment will peak over the next 10 years, ²⁴ making MDPP reach and use even more important. The MDPP’s success in preventing diabetes among eligible beneficiaries is dependent on robust enrollment in this evidence-based program. Our stakeholder analysis can help increase implementation of the MDPP in Western Pennsylvania and inform policy refinement, so that at-risk older adults may take advantage of this no-cost program to prevent type 2 diabetes.

Takeaway Points.

Four years after the Centers for Medicare and Medicaid Services enacted coverage of the CDC-approved Medicare Diabetes Prevention Program (MDPP) in 2018, little is known about why MDPP uptake is so limited.

• Our stakeholder analysis with program suppliers and healthcare providers reinforced existing evidence on insufficient reimbursement and low awareness of the program.

• Newer insights include recommendations about lagged payments, ongoing virtual delivery, and formally diagnosing prediabetes among MDPP participants.

• Our findings on barriers and facilitators can inform policy to refine the MDPP and research on the MDPP, particularly within the field of implementation science.