Impact of Medicare Continuous Subcutaneous Insulin Infusion Policies in Patients With Type 1 Diabetes

Abstract

Background:

The Centers for Medicare and Medicaid Services (CMS) has numerous requirements for coverage of continuous subcutaneous insulin infusion (CSII; insulin pump). Due to recent improvements in diabetes treatment, people with type 1 diabetes are living longer, resulting in an increase in the number of individuals who are eligible for Medicare and are impacted by CMS policies regarding CSII.

Methods:

Two hundred forty-one adults with type 1 diabetes who had been on CSII with CMS coverage for at least 6 months were surveyed. Median age was 67 years, mean A1c was 7.0%, 64% were women, 93% were white, and the median type 1 diabetes duration was 42 years. Participants reported median CSII use of 15 years and 82% were on CSII before starting CMS.

Results:

Of those starting CSII while on CMS, challenges included cost of supplies (29%) or the insulin pump (24%). The majority (57.5%) reported issues with obtaining supplies, the most common problems being delays in release of supplies (29%), difficulty getting paperwork completed (23.5%), and seeing a health care provider every 90 days (18%). Participants reported changing their CSII behaviors because of supply delays (39%) including leaving site in place >3 days (64%), and reusing pump supplies (34%). Consequently, participants reported adverse outcomes including more erratic (48%) or higher (42%) blood glucose and pain or irritation at sites (34%).

Conclusion:

This study concluded that current CMS CSII policies promote adverse CSII behaviors and outcomes in type 1 diabetes and thus call for changes in the CMS CSII policies.

The Centers for Medicare and Medicaid Services (CMS) provide health care to 59.1 million beneficiaries (16.7% of the United States population). ¹ Of these beneficiaries, one in five has either type 1 or type 2 diabetes. ² Improvement of blood glucose management has been proven to reduce the risk of long term complications in both type 1 diabetes and type 2 diabetes.^3,4 The goal of diabetes management is to keep blood glucose levels in the target range (70-180 mg/dL) and minimize hypoglycemia in order to avoid and/or delay complications. ⁵

Individuals with type 1 diabetes who are on the Medicare program must follow CMS policy requirements in order to receive the disposable supplies needed for continued CSII use. These supplies are expensive and many patients cannot afford to pay for them out of pocket. ⁶ Although cost is a possible barrier to CSII therapy, there are many benefits of CSII therapy including fewer insulin injections, flexibility in both meal timing and daily schedule, and most importantly improved glycemic management. ⁷ Effective CSII therapy requires that CSII sites be changed every two to three days. ⁸ Not changing CSII sites can cause infection, ⁹ lipohypertrophy, and scar tissue formation, which in turn can cause reduced or delayed insulin absorption and adversely impact blood sugar management in both the short and long term. ¹⁰

Due to recent improvements in diabetes treatment and technology, people with type 1 diabetes are living longer, resulting in an increase in the number of individuals with type 1 diabetes who are eligible for Medicare and therefore are impacted by CMS policies regarding CSII. ¹¹ CSII is only covered if the patient has either a low fasting C-peptide level when the blood glucose was at or below 225 mg/dL, or evidence of islet cell antibodies using an outmoded assay. ¹² In addition, CMS requires patients to attend quarterly face-to-face visits (quarterly visit requirement or QVR) with their health care provider in order to qualify to receive the supplies that are necessary for uninterrupted CSII use. ¹²

The QVR applies regardless of individual patient needs or circumstances, which violates the emphasis on the individualization of diabetes care and may not be possible or feasible for every patient, especially those who vary their residence with the seasons or live in remote areas. Furthermore, it creates a substantial administrative burden on patients, suppliers, and providers. These policies may delay the shipment of durable medical equipment and result in patients trying to extend or reuse their supplies due to lack of the adequate number of reservoirs and infusion sets for uninterrupted CSII therapy. Individuals with type 1 diabetes who use CMS have reported frequently practicing adverse CSII behaviors due to challenges resulting from the CMS QVR requirement. ¹³

Minimal research exists regarding the impact of CMS policies on individuals with type 1 diabetes, their diabetes related treatment outcomes, and CSII usage.^14,15 Existing research has found evidence of CMS beneficiaries with diabetes experiencing worse general health, more inpatient admissions, and higher out of pocket health care costs than CMS beneficiaries without diabetes. ² In addition, research examining CMS and diabetes (type 1 and type 2) often examines the cost of diabetes supplies to the government in terms of cost effectiveness, but does not examine how these policies impact the patient and their health care provider. This study evaluated how CMS CSII policies affect patient diabetes self-management behaviors and health outcomes related to type 1 diabetes. This study also aimed to understand perceptions of CMS in adults with type 1 diabetes who are on CSII and currently in the CMS program.

Methods

A survey was conducted on T1D Exchange Glu (www.myglu.org), an online community of people with type 1 diabetes for research and engagement, between October 9 and November 3, 2017. The survey contained 59 multiple choice or open-ended questions regarding respondents’ experience with CSII use while under Medicare coverage, as well as demographic information (see the online Appendix).

Potentially interested participants on Glu were invited to respond to the survey via email. An article about the survey was also posted on Glu during October 2017 to recruit eligible participants. Eligibility criteria included adults (>18 years old) who had type 1 diabetes for at least 1 year, were currently on, or previously on, CSII as their primary therapy for at least 6 months, and had Medicare coverage for CSII therapy for at least 6 months. Adult caregivers (>18 years old) who cared for pediatric patients and who met the above criteria were also eligible. Institutional review board approval was obtained and informed consent was acquired before beginning study participation.

Statistical analyses were performed using R version 3.4.1. ¹⁶ Frequencies and percentages were reported for categorical responses; for responses with missing data, percentages were calculated based on the number of participants who did respond. Means and standard deviations (SD) were reported for continuous responses; median and range were reported if distribution was skewed. Open-ended survey responses were analyzed using QDA Miner 5; ¹⁷ responses were categorized into groups based on similar themes, frequencies and percentages were reported for each theme.

Results

Two hundred seventy-eight people responded to the survey. Twenty-eight did not meet eligibility criteria; 9 caregivers of pediatric patients were excluded from analysis due to the small numbers. The following results were based on responses from 241 adults with type 1 diabetes.

Demographic and diabetes management information for the respondents are shown in Table 1. The adults in this survey had a median age of 67 years (range 19-86), had lived with type 1 diabetes for an average of 38.2 years (SD 18.4), had an average A1c of 7.0% (SD 1.1), and had been using CSII for an average of 14.8 years (SD 8.6).

Table 1.

Demographic and Diabetes Management Information for Adult Respondents.

Characteristic	Mean (±SD), median (range), or %	N
Age, years	67 (19-86)	235
Sex	63.4% female, 36.1% male, 0.4% other	238
Race and ethnicity	93.2% white 96.2% non-Hispanic	235
Education	32.8% below bachelor’s degree 30.2% bachelor’s degree 37.1% advanced degree	232
Work status	60.8% retired 17.3% working 16.0% on disability	237
Household income, US dollar	44.7% <50,000 39.6% 50-100,000 15.7% >100,000	197
Years with type 1 diabetes	38.2 (±18.4)	240
Current CSII user	94.8%	233
Duration of CSII, years	14.8 (±8.6)	216
Last A1c (%)	7.0 (±1.1)	234
On CSII before Medicare	82.4% yes 17.6% no	193
Stopped CSII while on Medicare	92.7% no 7.3% yes	232
Medicare as primary payer for CSII supplies	70.5% yes 29.5% no	237

One hundred fifty-nine (82%) respondents were already using CSII prior to going on Medicare as their primary insurance. Of those 34 adults who started CSII while on Medicare, 17 (50%) reported encountering challenges in obtaining coverage. The most common challenges endorsed were cost of supplies (29%), cost of insulin pump (24%), and issues with receiving the required blood testing (18%). Twelve percent reported challenges in finding a health care provider who participated in Medicare and prescribes pumps and supplies, or noted that the preferred pump (OmniPod) was not covered by Medicare.

Of those 159 adults who were already using CSII before going on Medicare, 115 (72%) reported experiencing problems when switching to Medicare as the primary payer for their insulin pump and insulin pump supplies. Twenty-five percent had to switch brand of insulin pump, type of insulin, or supplier due to coverage restriction; 24% reported excessive blood lab work and paperwork as challenges; 12% reported increased cost. Other problems included unclear instructions for patients and pharmacists on filing process (11%), denied coverage (11%), restrictions on the amount of supplies approved at a time (9%), and increased time spent on insurance related issues (4%).

Two hundred fifteen current CSII users answered about possible problems getting supplies. One hundred twenty-one (56%) reported that they experienced difficulties obtaining supplies, the most common of which were delays releasing supplies (30%). Participants also reported difficulty getting paperwork completed (24%), problems getting a face-to-face visit with a provider every 90 days (18%), difficulty getting needed appointments (12%), difficulties informing suppliers that they had been seen for the QVR (11%), and problems keeping scheduled pump health care provider appointments due to weather, illness, lack of transportation, or health care providers cancelling or rescheduling appointments (8%).

When asked if they had changed CSII related behaviors as a result of actual or possible delays, of 238 who answered, 92 (39%) indicated that they had changed CSII related behaviors. The most common actions were to leave sites in place beyond 3 days to avoid running out of supplies (66%) or reduce cost (33%), reuse CSII supplies (34%), use injections to make supplies last longer (23%), reduce insulin pump basal rates or boluses (9%), or temporarily stopping use of CSII therapy (18%). Of those that made changes based on these requirements, 24% of the 88 respondents reported no adverse reactions or issues, while many reported one or more of the following adverse outcomes: more erratic (48%) or higher (42%) blood glucose, more hypoglycemia (17%), increased anxiety or frustration (44%), pain or irritation at sites (34%), more scarring of sites (28%), required emergency supply shipments (19%), or received emergency supplies from a local source (15%).

When asked to rate their overall experience of using CSII while on Medicare, in comparison to using CSII while on their prior insurance, of 200 patients who answered, 10% rated it much better (6%) or somewhat better (4%), 31% similar, 28% somewhat worse and 31% much worse than prior insurance. Free text comments were wide ranging, but many expressed frustration or concern with supply delays, limited coverage for certain pumps such as those that are tubeless or integrated with CGM, only covering a pump every 5 years when the manufacturer warranty only covers 4 years, payment denials, and the QVR.

Discussion

This study examined the effect of CMS policies with regards to CSII. We found that (1) more than half of patients using CSII reported difficulties getting insulin pump supplies in a timely manner; (2) patients endorsed compensatory behaviors that are necessary when experiencing delays in receiving supplies, such as leaving sites in longer than clinically appropriate, reusing CSII supplies (which are FDA labeled for single use), or stopping CSII therapy; (3) patients reported that these changed CSII behaviors (including compensatory behaviors) lead to important and significant adverse outcomes, such as more erratic or higher blood glucose, more hypoglycemia, increased anxiety or frustration, pain or irritation at sites, and more scarring of sites. It is also notable that over half of the participants rated their experience in dealing with CMS with regard to CSII as being worse or much worse than their previous experience with commercial insurance.

This study highlighted a few issues of the QVR policy for Medicare patients to receive the supplies necessary for continued CSII therapy. (1) This policy is not evidence-based. Data from the T1D Exchange Clinic Registry showed no correlation between visit frequency and A1c in patients over 65. ¹² Importantly, respondents in this study were consistently well managed for years before they went on Medicare, mandating a visit frequency violates the principle of individualization of diabetes care espoused by ADA ¹⁸ and AACE. ¹⁰ (2) The QVR policy adds an undue administrative and practical burden on patients, providers and suppliers. It also likely increases direct costs by requiring visits that may not be medically necessary because CSII therapy is a well-established evidence-based treatment that is often started prior to Medicare coverage. (3) More importantly, our data showed that the QVR policy may lead to delays in receiving CSII supplies, which places individuals with type 1 diabetes at risk for harm. Finally, by mandating a face-to-face visit at a fixed frequency, the CMS CSII policy indirectly prohibits developing new options for medical care including remote assessment, which might be the only feasible assessment available to patients in remote areas, or patients who might temporarily change their place of residence based on family or seasonal-weather considerations.

In addition to QVR, many participants expressed that C-peptide tests and other blood work mandated by CMS were burdensome. CMS policies should allow for flexibility to continue their individual treatment regimen without imposing unnecessary paperwork and clinic visit requirements.

To avoid these and other unfortunate consequences of the current CMS CSII policy, we would recommend that obtaining CSII supplies should not require visits at an arbitrary frequency. To avoid critical interruptions of insulin pump supplies, we suggest continued coverage of CSII be mandated when patients are seen and evaluated by the treating physician at a frequency that is appropriate for the patient based on the managing physician’s plan of care. This would allow safe and patient specific recommendations, without the adverse consequences of the current policy.

Some limitations of this study should be taken into consideration when interpreting the survey results. First, the survey respondents were recruited from the online community T1D Exchange Glu, which consists of a cohort of well-educated, predominantly Caucasian, and well-managed individuals with type 1 diabetes. This cohort is not representative of people with type 1 diabetes at large. Second, there is a potential that individuals who took this online survey may have participated in the study in order to explicitly express their negative experiences with CMS. Future research should seek to further examine the role that health policy has on health behaviors in patients with type 1 diabetes with more diverse socio-economic backgrounds.

Conclusion

The current study provided evidence that the current CMS policies on CSII use present challenges to a high percentage of people with type 1 diabetes and often lead to delays or interruptions for individuals to receive essential CSII supplies on time; these unintended delays promote adverse CSII behaviors and practices among many patients that consequently result in adverse outcomes. As such the majority of patients negatively perceived CMS CSII coverage policies in comparison to their experiences with commercial insurers. Based on findings from this study, changes are much needed in the CMS CSII policies in order to allow a more flexible and personalized treatment plan for individuals, to reduce medically unnecessary clinic visits, lab works, and paperwork, and to ensure continuous coverage of CSII treatment and supplies.

Supplemental Material

Online_Appendix – Supplemental material for Impact of Medicare Continuous Subcutaneous Insulin Infusion Policies in Patients With Type 1 Diabetes

Supplemental material, Online_Appendix for Impact of Medicare Continuous Subcutaneous Insulin Infusion Policies in Patients With Type 1 Diabetes by Nicholas B. Argento, Jingwen Liu, Allyson S. Hughes and Alicia H. McAuliffe-Fogarty in Journal of Diabetes Science and Technology