Sage Journals: Discover world-class research

Abstract

Objective:

We assessed the economic impact of using the newest flash continuous glucose monitoring (CGM) among Medicaid beneficiaries with diabetes treated with intensive insulin therapy (IIT).

Research Design and Methods:

A budget impact analysis was created to assess the impact of increasing the proportion of Medicaid beneficiaries with diabetes on IIT, who use flash CGM by 10%. The analysis included glucose monitoring device costs, cost savings due to reductions in glycated hemoglobin, severe hypoglycemia events, and hyperglycemic emergencies such as diabetic ketoacidosis. The net change in costs per person to adopt flash CGM for three populations treated with IIT (adults with type 1 diabetes [T1D] or type 2 diabetes [T2D], and children and adolescents with T1D or T2D) was calculated; these costs were used to estimate the impact of increasing flash CGM use by 10% to the U.S. Medicaid budget over 1–3 years.

Results:

The analysis found that flash CGM demonstrated cost savings in all populations on a per patient basis. Increasing use of flash CGM by 10% was associated with a $19.4 million overall decrease in costs over the year and continued to reduce costs by $25.3 million in years 2 and 3.

Conclusions:

Our results suggest that the new flash CGM system can offer cost savings compared to blood glucose monitoring in Medicaid beneficiaries treated with IIT, especially T1D adults, and children and adolescents. These findings support expanding access to CGM by Medicaid plans.

Introduction

Medicaid is a public insurance program that provides health care coverage to low-income families and individuals. Administered by the Centers for Medicare and Medicaid Services (CMS), the program is jointly funded by the federal government and the states, with each state operating its own Medicaid program within federal guidelines.

According to 2020 data from CMS, 73,469,597 adults and 6,681,177 children and adolescents were enrolled in the Medicare and Children's Health Insurance Program (CHIP).¹ Within the Medicaid/CHIP population, 7,373,871 people carry a diagnosis of diabetes: 400,754 adults with type 1 diabetes (T1D),² 6,812,815 adults with type 2 diabetes (T2D),² and 176,331 children and adolescents with T1D (n = 160,302)³ or T2D (n = 16,029).³ An estimated 93% of the T1D⁴ and 11% of the T2D⁵ beneficiaries are treated with intensive insulin therapy (IIT), either multiple daily insulin injections or insulin pump.

Although current clinical guidelines recommend glycated hemoglobin (HbA1c) levels of <7.0% as the measure of glycemic control for children and adolescents and most adults with diabetes,⁶ the median percentage of Medicaid/CHIP beneficiaries with HbA1c levels >9.0% within the 29 states that report this measure is estimated to be 38.8%.⁷ This is considerably higher than the national average of 15.5%,⁸ who have HbA1c levels >9.0%, and most concerning given the severe consequences of poorly controlled diabetes.

As demonstrated in large clinical studies, persistent elevation of HbA1c levels is associated with the acute complications of diabetes, including severe hypoglycemia events (SHEs) and diabetic ketoacidosis (DKA),^9
–11 as well as chronic microvascular and macrovascular disease.^12

–15 While the chronic complications of diabetes develop over time, SHE and DKA occur rapidly and can lead to severe morbidities and mortality if not treated immediately.^16,17

Because each state differs in their reporting, it is difficult to accurately assess the total diabetes-related costs among Medicaid beneficiaries. Data from a 2018 report showed higher DKA hospitalization rates over time for Medicaid beneficiaries compared with privately insured individuals.¹⁸ Analyses of the 2012 Medicaid claims data¹⁹ and MarketScan multistate Medicaid database²⁰ also report significantly higher costs for adults and children and adolescents with diabetes (with and without a disability) compared with those without diabetes.

In a recent report by Ng et al., the diabetes-related cost was significantly higher among adults with diabetes and no disability compared with no diabetes or disability ($9,530 vs. $4,545, P < 0.001).¹⁹ Similar findings were reported by Shrestha et al., who found even greater cost disparities between children and adolescents with diabetes and those with no diabetes ($24,093 vs. $14,149, P < 0.001).²⁰

These disparities are likely related to differences in access to care between individuals living in low-income versus high-income communities.²¹ A recent study by Nguyen et al. reported that individuals living in low-income urban and rural areas were more likely to have fewer primary care physicians in their communities (0.5% and 7.4%, respectively) than those living in high-income areas.²²

Moreover, Medicaid beneficiaries have greater difficulty accessing specialists, such as endocrinologists, than those with private insurance.²¹ This is significant because many primary care physicians are challenged to provide adequate diabetes care to individuals treated with IIT and often delay intensifying insulin therapy due to the complexity of these regimens.²³

Safe use of IIT regimens requires patients to perform frequent blood glucose testing throughout the day to calculate their insulin dosages and monitor their glucose levels to detect hypoglycemia and hyperglycemia. Currently, patients have two options for testing: blood glucose monitoring (BGM) with a meter and test strips and continuous glucose monitoring (CGM).

Although use of BGM remains the most common method for glucose testing, this approach is not effective in detecting asymptomatic hypoglycemia unless the individual is willing to test many times daily. Because many individuals find this both inconvenient and painful, adherence to prescribed testing regimens is poor. One study found that 61.9% of participants reported that they did not use BGM at home,²⁴ whereas CGM measures glucose continuously and provides information to the patient about their current glucose level, glucose trend, and information about the direction and velocity of changing glucose by rate-of-change arrows.^25
–27 Importantly, CGM devices alert patients when glucose levels are crossing above or below safe glucose thresholds. This allows patients to take immediate action to mitigate current or impending hypoglycemia and hyperglycemia.

Numerous clinical trials and real-world studies have shown that use of flash CGM systems (FreeStyle Libre, FreeStyle Libre 14-Day, Abbott Diabetes Care, Alameda, CA) in T1D and T2D treated with IIT improves glycemic outcomes^28

–33 with reductions in hypoglycemia,^32

–37 improved treatment satisfaction,^33,38 less hypoglycemia fear,^32,38,39 improvements in other health-related measures,^{28,30,31,33,38} and decreased rates of diabetes-related events requiring hospitalization and/or emergency department (ED) services.^31,33

Given that the hospitalization costs to Medicaid for SHE and DKA average $4068⁴⁰ and $9733,^41,42 respectively, it was expected that reductions in these severe events would result in cost savings. Commercial plans pay even higher costs for SHE and DKA. Still, many Medicaid plans have remained hesitant to provide coverage for CGM technologies, despite rapid adoption by patients and commercial payers. As cost may be a concern, we assessed the economic impact of increasing adoption of the FreeStyle Libre^® 2 flash glucose monitoring system, the newest flash CGM, by Medicaid members with diabetes treated with IIT over 1–3 years. This new flash CGM has real-time, optional alarms that can warn the user about low or high glucose levels when turned on. As the evidence supporting flash CGM was with prior versions of the FreeStyle Libre systems, this analysis assumes the alarms have been turned off.

Methods

This budget impact analysis was created using a hypothetical population covered under a Medicaid insurance plan. The results are presented as a net cost difference per person relative to the flash CGM compared with BGM. The analysis reports total costs for a current and alternative scenario, as well as the annual net budget reduction and cost per member per year (PMPY) over 3 years. A 3-year time horizon was applied to report difference in costs in the first year of flash CGM system use, which includes setup costs. Costs associated with flash CGM system are slightly lower in two subsequent years where no further setup costs are required. Costs were reported from a Medicaid payer perspective, considering the total U.S. Medicaid population of adult and pediatric persons with diabetes treated with IIT.¹

A Microsoft Excel^®-based cost calculator was constructed to estimate the budget impact of increasing utilization by 10% of flash CGM to Medicaid adult beneficiaries with T1D or T2D and pediatric beneficiaries (4 to 18 years of age), treated with IIT. The cost calculator compared the flash CGM with BGM for managing T1D and T2D adults and children and adolescents.

The prevalence of diabetes in the general population and prevalence by insulin delivery modality were used to estimate the size of the adult population.^2

–5 The pediatric population was calculated using prevalence, assuming that all children and adolescents between 4 and 18 years of age treated with IIT may be eligible for the flash CGM.

Clinical inputs.

Diabetes prevalence and all clinical inputs are presented in Table 1. Hospital costs due to SHEs were calculated using a baseline rate for BGM and adjusting this for a treatment effect with CGM use. Baseline rates for SHE in the T1D and T2D adult populations and pediatric population using IIT were sourced from a U.S. real-world study in a commercial T1D population⁴³ and a real-world study in U.S. pediatric patients,⁴⁴ respectively. These rates were adjusted to apply a wider definition of SHE⁴⁵ and reflect a Medicaid population.¹⁸ Details of the adjustments applied to calculate baseline risk are described in Supplementary Table S1. Hospital costs were calculated applying the proportions for treating SHEs in T1D and T2D⁴⁵ based on a large phase 3 clinical program.⁴⁵

Table 1.

Diabetes Prevalence and Clinical Inputs

Input measures	Population	Values	References
Plan size: national	All	Adults: 73,469,597 Pediatric: 6,681,177	Medicaid,¹ for pediatric this is size of the CHIP population
Prevalence by patient subgroup
Population prevalence:	Adults T1D	0.5%	Xu et al. (2018)² Xu et al. (2018)²
	Adults T2D	8.5%	Xu et al. (2018)² Xu et al. (2018)²
	Pediatrics, T1D	0.2%	Pettitt et al. (2014)³ Pettitt et al. (2014)³
	Pediatrics, T2D	0.02%	Pettitt et al. (2014)³ Pettitt et al. (2014)³
% of IIT users	T1D	100%	Assumption
% of IIT users	T2D	11.0%	Li et al. (2012)⁵ and internal report^a
	Pediatrics	100%	All pediatric insulin users are assumed to be IIT
SHEs, events PPPY
Baseline incidence	T1D	0.872	Pettus et al. (2020)^43,b Pettus et al. (2020)^43,b
	T2D	0.445	Pettus et al. (2020)^43,b Pettus et al. (2020)^43,b
	Pediatrics	0.913	Katz et al. (2012)^44,b
Flash CGM treatment effect (rate ratio)	T1D	0.560	Bolinder et al. (2020)^51,c
	T2D	0.390	Bergenstal et al. (2020)^47,c
	Pediatrics	0.560	Assumed to be the same as T1D
% Treated by ambulance call	T1D	31.0%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	T2D	25.3%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	Pediatrics	31.0%	Assumed to be the same as T1D
% Treated by ER visit	T1D	9.5%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	T2D	23.2%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	Pediatrics	9.5%	Assumed to be the same as T1D
% Treated by hospital admission	T1D	5.0%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	T2D	5.3%	Heller et al. (2015)⁴⁵ Heller et al. (2015)⁴⁵
	Pediatrics	5.0%	Assumed to be the same as T1D
Hospitalization for DKA, events PPPY
Baseline incidence	T1D	0.147	Pettus et al. (2020)^43,b
	T2D	0.091	Pettus et al. (2020)^43,d
	Pediatrics	0.147	Assumed to be the same as T1D
Flash CGM treatment effect (rate ratio)	T1D	0.46	Roussel et al. (2021)⁴⁸ Roussel et al. (2021)⁴⁸
	T2D	0.57	Roussel et al. (2021)⁴⁸ Roussel et al. (2021)⁴⁸
	Pediatrics	0.46	Assumed to be the same as T1D
Change in HbA1c
Flash CGM treatment effect (change in %)	T1D	−0.55%	Evans et al. (2019)⁴⁹
	T2D	−0.90%	Kröger et al. (2019)²⁹
	Pediatrics	−0.37%	Laffel et al. (2020)⁵⁰

Li et al. (2012)⁵ reported that 22% of the U.S. T2DM population are insulin users. An unpublished Gfk-Roper report, reports 11% of the T2 DMI population are IIT.

Adjusted assuming a 2.4285 times increased risk of event in Medicaid population compared with commercially insured population and applying distribution reported in Heller et al.⁴⁵ Refer to supplement for further details of adjustment calculation.

Treatment effect on ADE used as proxy for treatment effect on SHE. Refer to supplement for further details of adjustment calculation.

Adjusted assuming a 2.4285 times increased risk of event in Medicaid population compared with commercially insured population and adjusted assuming a ratio of 0.619 between DKA hospital events for T1 and T2, sourced from MarketScan database. Refer to supplement for further details of adjustment calculation.

DKA or HHS, ER; IIT multiple daily injections or continuous subcutaneous insulin infusion.

ADE, acute diabetes-related events; CGM, continuous glucose monitoring; CHIP, Children's Health Insurance Program; DKA, diabetic ketoacidosis; ER, emergency room; IIT, intensive insulin therapy; PPPY, per patient per year; SHE, severe hypoglycemic event; T1D, type 1 diabetes; T2D, type 2 diabetes.

The rate ratios for reduction in SHE with the flash CGM system use were obtained from U.S. real-world studies, before and after using flash CGM for T1D⁴⁶ and T2D,⁴⁷ respectively, using reduction in acute diabetes-related events (ADE) as a proxy for reduction in SHE events. The treatment effect in the pediatric population was assumed to be the same as for the T1D adult population. The clinical inputs are reported in Table 1.

Hospital admissions for DKA and hyperosmolar hyperglycemia syndrome events were calculated using a similar approach to SHE hospital costs. For simplicity, these hyperglycemic emergencies were labeled as DKA. Similar to the baseline rate for SHE, the baseline rate of DKA admission in the T1D population was obtained from Pettus et al.⁴³ and adjusted for a Medicaid population.¹⁸ To reflect risk in a T2D population, a query of a large payer administrative claims database, IBM Watson MarketScan, was performed using the dates Q4 2017 – Q2 2019. Rates for hospitalizations for hyperglycemia in T1D (ICD-9 250.x1, 250.x3; ICD-10 E10) and T2D (ICD-9 250.x0, 250.x2; ICD-10 E11.xx) on rapid- or short-acting insulin were calculated. The ICD-10 codes used to identify the hyperglycemic complications are listed in Supplementary Table S2.

Details of the adjustments applied are reported in Supplementary Table S1. The rate ratios for reduction in DKA events with the flash CGM system use were obtained from the French real-world studies,⁴⁸ before and after starting use of flash CGM. The baseline rate of DKA and treatment effect in the T1D population were assumed to be the same in the pediatric population (Table 1).

Baseline rates of HbA1c and treatment effect with the flash CGM system in adult T1D populations were sourced from a meta-analysis.⁴⁹ Baseline rates of HbA1c and treatment effect with the flash CGM system in adult T2D populations were sourced from Kröger et al.,²⁹ a retrospective review of medical records in 18 European centers in T2D patients before and after starting flash CGM. The baseline rate and treatment effect in the pediatric population were sourced from a randomized trial of 153 individuals 14 to 24 years of age with T1D⁵⁰ (Table 1).

Cost inputs

The costs considered in the calculator include glucose monitoring costs, physician setup and follow-up costs, hospital costs for SHE, hospital costs for admission with DKA, and health care costs avoided for HbA1c reduction. Costs associated with BGM include test strips ($0.14/strip) and lancets ($0.07/lancet) at a usage rate of 5.6/day for T1D beneficiaries,⁵¹ 3.8/day for T2D beneficiaries,³⁶ and 7.3/day for pediatric beneficiaries.⁵²

Costs associated with the flash CGM system cover the reader unit ($70.00), which has a 3-year lifetime, 14-day sensor ($54.43 Wholesale Acquisition Cost), and cost of BGM as needed (test strips and lancets). The anticipated costs of BGM use in conjunction with the flash CGM associated include test strips and lancets at a usage rate of 0.5/day for T1D beneficiaries,⁵¹ 0.3/day for T2D beneficiaries,³⁶ and 0.5/day for pediatric beneficiaries.⁵²

Because these usage rates are based on clinical trials that assessed the first-generation flash CGM, we anticipate a lower usage rate due to the functionality of the optional alarms that integrated into the flash CGM system. Unit costs were sourced from 2020 price lists or inflated to 2020 prices using the consumer price index for medical care published by the Health Resources and Services Administration.⁵³ Unit costs and resource use assumptions are provided in Supplementary Table S3.

The 3-year costs for physician visits include the setup cost ($55.58⁵⁴), a one-time expense, and clinic visits ($518.63⁵⁵ per visit) at an assumed rate of four visits per year. Setup costs only apply to the flash CGM system (not BGM) and are only applied in the first year of treatment. All other costs are assumed to be the same in year 1 and subsequent years. All costs are adjusted to 2020 prices.

The health care utilization costs for SHE include ambulance service ($571.00),⁵⁶ ED service ($2,000.05),⁵⁷and hospital admission ($4,068.00).⁴⁰ The cost per DKA event is $9,733.00.⁴¹ The annual cost savings per 1.0% HbA1c reduction is $804.67.⁵⁸ All costs are inflated to 2020 prices.

Penetration projections

The penetration projections were hypothetical. The current scenario analyzed the costs assuming 23% flash CGM penetration, although this estimate was not based on market share data. The alternative scenario assumed a 10% absolute increase in flash CGM penetration. Costs for one cohort were calculated over 2 and 3 years, keeping market distribution constant to show that costs with flash CGM are higher in year 1 due to setup costs and decrease in subsequent years (year 2 onward) because no further setup costs are incurred.

One-way sensitivity analysis

One-way, deterministic sensitivity analysis (DSA) was performed on the primary inputs, varying each individually within an upper and lower bound of 20% in either direction.

Results

Base case analysis

Table 2 presents a summary of the difference in cost per patient per year (PPPY) treated with the flash CGM system relative to BGM in the T1D, T2D, and pediatric populations in year 1. In the T1D population, the annual acquisition costs are estimated to be $1052 PPPY higher with the flash CGM system compared to BGM. When health care resource costs (SHE and DKA) and cost off-sets from improved HbA1c are considered, the flash CGM system results in lower costs and the net difference is estimated to be $327 PPPY.

Table 2.

Difference in Cost Per Patient Per Year Treated with the Flash Continuous Glucose Monitoring System Relative to Blood Glucose Monitoring in the Type 1 Diabetes, Type 2 Diabetes, and Pediatric Populations in Year 1

	Flash CGM	BGM	Cost difference
	Annual	Annual	Cost difference
T1D
Glucose monitoring acquisition costs	$1482	$430	−$1052
Total health care resource costs	$3063	$4000	$936
SHE: ambulance calls	$86	$154	$68
SHE: emergency room visits	$93	$166	$73
SHE: hospital admissions	$99	$177	$78
DKA: hospital admissions	$655	$1428	$773
Physician appointments	$2130	$2075	−$56
HbA1c cost avoided	−$443	$0	$443
Total (PPPY)	$4103	$4429	$327
T2D
Glucose monitoring acquisition costs	$1466	$291	−$1175
Total health care resource costs	$2772	$3324	$551
SHE: ambulance calls	$25	$64	$39
SHE: emergency room visits	$81	$206	$126
SHE: hospital admissions	$37	$95	$58
DKA: hospital admissions	$500	$883	$384
Physician appointments	$2130	$2075	−$56
HbA1c cost avoided	−$724	$0	$724
Total (PPPY)	$3515	$3615	$101
Pediatric
Glucose monitoring acquisition costs	$1482	$560	−$922
Total health care resource costs	$3077	$4023	$947
SHE: ambulance calls	$90	$161	$71
SHE: emergency room visits	$97	$174	$77
SHE: hospital admissions	$104	$186	$82
DKA: hospital admissions	$655	$1428	$773
Physician appointments	$2130	$2075	−$56
HbA1c cost avoided	−$298	$0	$298
Total (PPPY)	$4261	$4583	$323

Note due to rounding (e.g., $518.63 × 4 was rounded up to $2075 vs. $2074.52) some numbers may not add up as presented.

BGM, blood glucose monitoring; SHE, severe hypoglycemia event.

Similarly, in the T2D population, the annual acquisition costs are estimated to be $1175 PPPY higher with the flash CGM system compared to BGM. When health care resource costs, physician costs, and cost savings from improved HbA1c are considered, flash CGM is cost saving, resulting in cost decrease of $101 PPPY. In the pediatric population, the annual acquisition costs are estimated to be $922 PPPY higher with the flash CGM system compared to BGM. When cost offsets are considered, the cost savings with the flash CGM system are $323 PPPY. Table 2 presents the annual per patient cost, comparing the flash CGM system with BGM in the T1D, T2D, and pediatric populations.

Impact to budget

The base case simulated the budget impact for the total U.S. Medicaid population (plan size 73.5 million adults and 6.7 million children and adolescents) using the default values described in Table 1, which projected the plan would cover 367,348 T1D IIT users, 686,941 T2D IIT users, and 14,231 children and adolescents. These populations were assumed to be static over time. The market share assumptions assume that flash CGM system uptake would increase from 23% to 33% in all three populations with the rest assumed to use routine BGM.

The total costs for glucose monitoring, DKA, hypoglycemia, elevated HbA1c, and the provider visits described in the methods for the whole population over 1 year were $4131 million in the current scenario and $4112 million in the alternative scenario of increased utilization of the flash CGM system. Increasing the proportion treated with the flash CGM system was estimated to result in annual savings of $19.4 million in year 1. The cost savings increase to $25.3 million in years 2 and 3 on treatment because no further setup costs with the flash CGM system are required. The results of the budget impact analysis are illustrated in Figure 1. The net budget reduction was estimated to be $0.24 PMPY in year 1 and the reduction increases to $0.32 PMPY in subsequent years.

FIG. 1.

Net annual budget reduction when flash CGM market uptake is increased. Costs include glucose monitoring, hypoglycemia, DKA, elevated HbA1c, and the provider visits described Methods. The cost reductions amount to $0.27 in year 1 and $0.35 in years 2 and 3 on a PMPY basis. CGM, continuous glucose monitoring; DKA, diabetic ketoacidosis; PMPY, per member per year; HbA1c, glycated hemoglobin.

Impact on utilization

The current and alternative scenarios are also compared in terms of difference in the number of acute diabetes events (Table 3). The results suggest that increasing the proportion of the Medicaid population using the flash CGM system by 10% reduces the number of hospital events for SHE or DKA by 5.9% from 382,500 to 360,100.

Table 3.

Breakdown of Acute Diabetes Events When Flash Continuous Glucose Monitoring System Market Uptake Is Increased (1000 Events in U.S. Medicaid Population)

Events (in 1000s)	Current	Alternative	Difference	Difference (%)
SHE ambulance call outs	159.1	149.9	−9.3	5.8%
SHE ER visits	89.5	83.8	−5.7	6.4%
SHE hospitalizations	28.8	27.1	−1.7	6.0%
DKA hospitalizations	105.1	99.4	−5.7	5.5%
Total	382.5	360.1	−22.4	5.9%

One-way sensitivity analysis

The results of the one-way DSA in the base-case per person analysis for T1D, T2D, and pediatrics comparing the flash CGM system with BGM are illustrated in Supplementary Figures S1–S3. Varying the following inputs had the greatest impact on the analysis in the T1 population: number of physician visits per year with the flash CGM and BGM, respectively, treatment effect with flash CGM on DKA, and the baseline incidence of DKA events. Flash CGM is more expensive when there are more physician visits than BGM per year, which is not expected. Varying any other input 20% in favor of BGM decreases the savings with flash CGM system, yet had no impact on the conclusion of cost savings associated with increased use of flash CGM.

This analysis was repeated for the T2D and pediatric populations and similar trends were observed. A small number of inputs were sensitive to a 20% variation in the T2D population because the difference in cost was closer to zero.

Discussion

Numerous clinical trials and real-world studies have shown that use of earlier generation flash CGM systems in T1D and T2D treated with IIT improves both outcomes glycemic outcomes^{28

–37} improved treatment satisfaction,^33,38 enhanced quality of life,^{28,30

–33,38} and lower rates of severe diabetes-related events.^31,33 Real-world studies have also shown that flash CGM increases the users' time in glucose range (TIR),^59,60 with reductions in time below range (TBR).^36,51 Both TIR (70–180 mg/dL) and TBR (<70 mg/dL and <54 mg/dL) are new CGM metrics of glycemic control endorsed by several international diabetes societies, including the American Diabetes Association, European Association for the Study of Diabetes, American Association of Clinical Endocrinologists, and International Society for Pediatric and Adolescent Diabetes.⁶¹

Our analysis demonstrates both the clinical and economic impact of improvements in HbA1c, with reductions in SHE and DKA hospitalization rates to be expected with flash CGM use compared with BGM.

Findings from our analysis suggest that increasing use of flash CGM by Medicaid beneficiaries with diabetes treated with IIT will lead to significant cost savings. Although the glucose monitoring costs during the first year are notably higher for flash CGM compared with BGM (for example, for T1D, it is $1482 vs. $430, respectively), the overall cost difference clearly favors flash CGM use in the T1D, T2D, and pediatric populations due to the potential reductions in both ADE and HbA1c levels. These cost savings would be expected in the first year and then persist through the second and third years.

In addition to the benefits of achieving and maintaining near-normal HbA1c in preventing or delaying the progression of long-term complications,^12

–15 even short-term reductions in HbA1c levels and HbA1c variability have been shown to reduce the risk for cardiovascular events and all-cause mortality.⁶² The anticipated HbA1c reduction may also lead to increased treatment satisfaction among beneficiaries,⁶³ which strongly correlates with enhanced medication adherence and decreased health care resource utilization.^64

–67 A recent study by Ueno et al. reported significant reductions in HbA1c and improvements in treatment satisfaction in adult T1D patients using flash CGM.⁶⁸ Moreover, use of flash CGM reduces the burden of diabetes on patients and their caregivers,^30,69 one of the intangible benefits that is not often considered.

Strengths of this analysis were the incorporation of data from actual use of flash CGM in the real world along with randomized controlled trial (RCT) data to backup other conclusions. It is known that participants in RCTs represent only a small fraction of patients who use the intervention. Another strength was the use of realistic costs for complications based on publicly available data and BGM devices. Finally, this study looked at various types of cost offsets related to hypoglycemia, hyperglycemia, and change in HbA1c.

Our analysis has notable limitations. Baseline rates for SHE and DKA were estimated using the Pettus et al. study.⁴³ Because these data looked at the population overall, an inflator had to be used to adjust for anticipated higher rates of complications in the Medicaid population versus the general population. This may have overestimated or underestimated the true burden for Medicaid patients. The baseline risk of DKA hospitalization, sourced from Pettus et al.,⁴³ included both ED visits and admissions. This approach may therefore overestimate the cost of DKA admissions by applying an admission cost to ED visits and underestimate the wider costs of DKA events by excluding ambulance call outs.⁴⁸

An additional limitation was the use of a cost offset for HbA1c reduction. There is potential for some overlap between the cost savings captured from reduced risk of SHE and DKA events and cost savings from improved HbA1c because the source applied to calculate the reduction in health care costs per 1% change in HbA1c captured all health care costs, including DKA and SHE events. However, the cost attributed to a 1% change in HbA1c of $670 is conservative compared to similar economic analyses⁴² that attribute cost savings ranging between $1111⁷⁰ and $2073⁷¹ per 1% HbA1c reduction and applied in addition to cost savings due to reduced DKA and SHE events associated with CGM use.

In addition, treatment effects of flash CGM on SHE hospitalizations were estimated using studies that looked at reductions in overall acute diabetes events.^46,47 However, we believe this assumption is reasonable because the treatment effect applied is similar to the treatment effect on SHE reported in a UK real-world study in T1D populations before and after starting the flash CGM system.⁷² The analysis did not include impact of flash CGM on hospitalizations for non-diabetes-related events. However, we believe the cost offset for HbA1c reduction may have captured some of this impact.

In addition, the studies incorporated into the cost calculator used prior versions of flash CGM, which did not have the optional, real-time alarm feature. As such, we potentially may see even greater improvements in clinical outcomes using the newest generation of flash CGM. This analysis also did not incorporate improvements to productivity, quality measures due to improvements associated with use of flash CGM as reported in several recent real-world studies.^{28,30,31,33,38}

Conclusions

Despite these limitations, this study suggests that the new flash CGM system can offer cost savings compared to BGM in Medicaid beneficiaries with T1D or T2D adults and children and adolescents treated with IIT. Although system costs are higher for flash CGM, cost offsets related to reducing health care utilization (e.g., reducing hospitalizations for hypoglycemia) can help Medicaid plans recover the higher system cost within the first year. These findings support expanding access to flash CGM by Medicaid plans.

Footnotes

Acknowledgments

The authors wish to thank Christopher G. Parkin, MS, CGParkin Communications, Inc., for medical writing support, and Gregory Roberts and Yelena Nabutovsky (both Abbott employees) for support with the MarketScan data.

Author Disclosure Statement

R.H. is a contractor and N.V. is an employee of Abbott Diabetes Care. D.B. and J.F. are consultants to Abbott.

Funding Information

Funding for the development of this article was provided by Abbott Diabetes Care.

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

References

Medicaid.gov. November 2020 Medicaid & CHIP Enrollment Data Highlights. https://www.medicaid.gov/medicaid/program-information/medicaid-and-chip-enrollment-data/report-highlights/index.html (accessed March 30, 2021 ).

, Liu

, Sun

, et al.: Prevalence of diagnosed type 1 and type 2 diabetes among US adults in 2016 and 2017: population based study. BMJ, 2018; 362:k1497.

Pettitt

, Talton

, Dabelea

, et al.: Prevalence of diabetes in U.S. youth in 2009: the SEARCH for Diabetes in Youth study. Diabetes Care, 2014; 37:402–408.

Foster

, Beck

, Miller

, et al.: State of type 1 diabetes management and outcomes from the T1D exchange in 2016–2018. Diabetes Technol Ther, 2019; 21:66–72.

, Oldenburg

, Chamberlain

, et al.: Diabetes prevalence and determinants in adults in China mainland from 2000 to 2010: a systematic review. Diabetes Res Clin Pract, 2012; 98:226–235.

American Diabetes Association. 6. Glycemic Targets: standards of medical care in diabetes—2021. Diabetes Care, 2021; 44 (Supplement 1):S73–S84.

Medicaid.gov. Comprehensive Diabetes Care: Hemoglobin A1c (HbA1c) Poor Control (>9.0%): Ages 18 to 75. https://www.medicaid.gov/state-overviews/scorecard/comprehensive-diabetes-care/index.html (accessed April 20, 2021 ).

Carls

, Huynh

, Tuttle

, et al.: Achievement of glycated hemoglobin goals in the US remains unchanged through 2014. Diabetes Ther, 2017; 8:863–873.

Cengiz

, Xing

, Wong

, et al.: Severe hypoglycemia and diabetic ketoacidosis among youth with type 1 diabetes in the T1D Exchange clinic registry. Pediatr Diabetes, 2013; 14:447–454.

10.

Weinstock

, DuBose

, Bergenstal

, et al.: Risk factors associated with severe hypoglycemia in older adults with type 1 diabetes. Diabetes Care, 2016; 39:603–610.

11.

Karges

, Rosenbauer

, Holterhus

P-M

, et al.: Hospital admission for diabetic ketoacidosis or severe hypoglycemia in 31,330 young patients with type 1 diabetes. Eur J Endocrinol, 2015; 173:341–350.

12.

UK Prospective Diabetes Study (UKPDS) Group: Intensive blood-glucose control with sulphonlylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet, 1998; 352:837–853.

13.

Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin-dependent diabetes mellitus. N Engl J Med, 1993; 329:977–986.

14.

Holman

, Paul

, Bethel

, et al.: 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med, 2008; 359:1577–1589.

15.

Nathan

, DCCT/ EDIC Research

Group

: The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications study at 30 years: overview. Diabetes Care, 2014; 37:9–16.

16.

Bonds

, Miller

, Bergenstal

, et al.: The association between symptomatic, severe hypoglycaemia and mortality in type 2 diabetes: retrospective epidemiological analysis of the ACCORD study. BMJ, 2010; 340:b4909.

17.

Realsen

, Goettle

, Chase

: Morbidity and mortality of diabetic ketoacidosis with and without insulin pump care. Diabetes Technol Ther, 2012; 14:1149–1154.

18.

Benoit

, Zhang

, Geiss

, et al.: Trends in diabetic ketoacidosis hospitalizations and in-hospital mortality—United States, 2000–2014. Weekly 2018;67;362–365.

19.

, Laxy

, Shrestha

, et al.: Prevalence and medical expenditures of diabetes-related complications among adult Medicaid enrollees with diabetes in eight U.S. states. J Diabetes Complications, 2021; 35:107814.

20.

Shrestha

, Zhang

, Thompson

, et al.: Medical expenditures associated with diabetes among youth with medicaid coverage. Med Care, 2017; 55:646–653.

21.

Nguyen

, Sommers

: Access and quality of care by insurance type for low-income adults before the affordable care act. Am J Public Health, 2016; 106:1409–1415.

22.

Nguyen

, Chernew

, Ostrer

, Beaulieu

: Comparison of Healthcare Delivery Systems in Low- and High-Income Communities. AMJC, 2019; 7:11–18.

23.

Bin rsheed

, Chenoweth

: Barriers that practitioners face when initiating insulin therapy in general practice settings and how they can be overcome. World J Diabetes, 2017; 8:28–39.

24.

Siddiqui

, Khan

, Moyeen

, Chaudhary

: Identifying barriers to therapeutic adherence in type 2-diabetes: a complex and multidimensional clinical issue. Asp Biomed Clin Case Rep, 2019; 2; 22–28.

25.

Edelman

, Argento

, Pettus

, Hirsch

: Clinical implications of real-time and intermittently scanned continuous glucose monitoring. Diabetes Care, 2018; 41:2265–2274.

26.

Adolfsson

, Parkin

, Thomas

, Krinelke

: Selecting the appropriate continuous glucose monitoring system—a practical approach. Eur Endocrinol, 2018; 14:24–29.

27.

Kruger

, Edelman

, Hinnen

, Parkin

: Reference guide for integrating continuous glucose monitoring into clinical practice. Diabetes Educ, 2019; 45(suppl 1):3S–20S.

28.

Pintus

: Freestyle libre flash glucose monitoring improves patient quality of life measures in children with Type 1 diabetes mellitus (T1DM) with appropriate provision of education and support by healthcare professionals. Diabetes Metab Syndr, 2019; 13:2923–2926.

29.

Kröger

, Fasching

, Hanaire

: Three European retrospective real-world chart review studies to determine the effectiveness of flash glucose monitoring on HbA1c in adults with type 2 diabetes. Diabetes Ther, 2020; 11:279–291.

30.

Fokkert

, van Dijk

, Edens

, et al.: Improved well-being and decreased disease burden after 1-year use of flash glucose monitoring (FLARE-NL4). BMJ Open Diabetes Res Care, 2019; 7:e000809.

31.

Tyndall

, Stimson

, Zammitt

, et al.: Marked improvement in HbA1c following commencement of flash glucose monitoring in people with type 1 diabetes. Diabetologia, 2019; 62:1349–1356.

32.

Paris

, Henry

, Pirard

, et al.: The new FreeStyle Libre Flash Glucose Monitoring System improves the glycaemic control in a cohort of people with type 1 diabetes followed in real-life conditions over a period of one year. Endocrinol Diab Metab, 2018; 1:e00023.

33.

Charleer

, De Block

, Van Huffel

, et al.: Quality of life and glucose control after 1 year of nationwide reimbursement of intermittently scanned continuous glucose monitoring in adults living with type 1 diabetes (FUTURE): a prospective observational real-world cohort study. Diabetes Care, 2020; 43:389–397.

34.

Aasa

, Henriksson

, Dellborg

, et al.: Cost and health outcome of primary percutaneous coronary intervention versus thrombolysis in acute ST-segment elevation myocardial infarction—results of the Swedish Early Decision reperfusion Study (SWEDES) trial. Am Heart J, 2010; 160:322–328.

35.

Oskarsson

, Antuna

, Geelhoed-Duijvestijn

, et al.: Impact of flash glucose monitoring on hypoglycaemia in adults with type 1 diabetes managed with multiple daily injection therapy: a pre-specified subgroup analysis of the IMPACT randomised controlled trial. Diabetologia, 2018; 61:539–550.

36.

Haak

, Hanaire

, Ajjan

, et al.: Use of flash glucose sensing technology for 12 months as a replacement for blood glucose monitoring in insulin-treated type 2 diabetes. Diabetes Ther, 2017; 8:573–586.

37.

NHS Improvement. 2017 Reference Costs and Guidance. https://webarchive.nationalarchives.gov.uk/ukgwa/20200501111106mp_/https:/improvement.nhs.uk/documents/6468/201718_reference_costs_data_and_guidance.zip (accessed August 16, 2021 ).

38.

Al Hayek

, Al Dawish

: The potential impact of the FreeStyle Libre flash glucose monitoring system on mental well-being and treatment satisfaction in patients with type 1 diabetes: a prospective study. Diabetes Ther, 2019; 10:1239–1248.

39.

Messaaoui

, Tenoutasse

, Crenier

: Flash glucose monitoring accepted in daily life of children and adolescents with type 1 diabetes and reduction of severe hypoglycemia in real-life use. Diabetes Technol Ther, 2019; 21:329–335.

40.

Liu

, Wang

, Ganz

, et al.: The burden of severe hypoglycemia in type 1 diabetes. Curr Med Res Opin, 2018; 34:171–177.

41.

Tieder

, McLeod

, Keren

, et al.: Variation in resource use and readmission for diabetic ketoacidosis in children's hospitals. Pediatrics, 2013; 132:229–236.

42.

Minshall

, Isitt

, Graham

, Lynch

: Budget impact analysis (BIA) comparing real-time continuous glucose monitoring (RT-CGM) with self-monitoring of blood glucose (SMBG) for all U.S. medicaid enrollees with type 1 diabetes. Diabetes, 2020; 69(Suppl 1):174-OR.

43.

Pettus

, Zhou

, Shepherd

, et al.: Incidences of severe hypoglycemia and diabetic ketoacidosis and prevalence of microvascular complications stratified by age and glycemic control in US adult patients with type 1 diabetes: a real-world study. Diabetes Care, 2019; 42:2220–2227.

44.

Katz

, Volkening

, Anderson

, Laffel

: Contemporary rates of severe hypoglycaemia in youth with type 1 diabetes: variability by insulin regimen. J Diabetic Med, 2012; 29:926–932.

45.

Heller

, Frier

, Hersløv

, et al.: Severe hypoglycaemia in adults with insulin-treated diabetes: impact on healthcare resources. Diabet Med, 2016; 33:471–477.

46.

Kerr

, Roberts

, Souto

, et al.: CGM use associated with reduction in acute diabetes complications, even in patients using less than four test strips per day. Diabetes Technol Ther, 2020; 22(Suppl 1):A193.

47.

Bergenstal

, Kerr

MSD

, Roberts

, et al.: Flash CGM is associated with reduced diabetes events and hospitalizations in insulin-treated type 2 diabetes. J Endocr Soc, 2021; 5:bvab013.

48.

Roussel

, Bruno Guerci

, Vicaut

, et al.: Dramatic drop in ketoacidosis rate after FreeStyle Libre™ system initiation in type 1 and type 2 diabetes in France, especially in people with low self-monitoring of blood glucose (SMBG): a nationwide study. Diabetes, 2020; 69(Supplement 1):68-OR.

49.

Evans

, Welsh

, Ells

, Seibold

: The impact of flash glucose monitoring on glycaemic control as measured by HbA1c: a meta-analysis of clinical trials and real-world observational studies. Diabetes Ther, 2020; 11:83–95.

50.

Laffel

, Kanapka

, Beck

, et al.: Effect of continuous glucose monitoring on glycemic control in adolescents and young adults with type 1 diabetes: a randomized clinical trial. JAMA, 2020; 323:2388–2396.

51.

Bolinder

, Antuna

, Geelhoed-Duijvestijn

, et al.: Novel glucose-sensing technology and hypoglycemia in type 1 diabetes: a multicentre, non-masked, randomised controlled trial. Lancet, 2016; 388:2254–2263.

52.

Campbell

, Murphy

, Stewart

, et al.: Outcomes of using flash glucose monitoring technology by children and young people with type 1 diabetes in a single arm study. J Pediatric Diabetes, 2018; 19:1294–1301.

53.

Health Resources & Services Administration (HRSA). Consumer Price Index (CPI) for Medical Care. 2020. https://www.hrsa.gov/get-health-care/affordable/hill-burton/cpi.html (accessed March 30, 2021 ).

54.

Miller

: Using continuous glucose monitoring in clinical practice. Clin Diabetes, 2020; 38:429–438.

55.

Moses

, Pedroza

, Bara

, et al.: Funding and services needed to achieve universal health coverage: applications of global, regional, and national estimates of utilisation of outpatient visits and inpatient admissions from 1990 to 2016, and unit costs from 1995 to 2016. Lancet, 2019; 4:e49–e73.

56.

U.S. Government Accountability Office. GAO@100. Ambulance Providers: Costs and Medicare Margins Varied Widely; Transports of Beneficiaries Have Increased. 2012. https://www.gao.gov/products/gao-13-6 (accessed March 30, 2021 ).

57.

Quilliam

, Simeone

, Ozbay

, et al.: The incidence and costs of hypoglycemia in type 2 diabetes. Am J Manag Care, 2011; 17:673–680.

58.

Bansal

, Shah

, Reilly

, et al.: Impact of reducing glycated hemoglobin on healthcare costs among a population with uncontrolled diabetes. Appl Health Econ Health Policy, 2018; 16:675–684.

59.

Gomez-Peralta

, Dunn

, Katherine Landuy

, et al.: Flash glucose monitoring reduces glycemic variability and hypoglycemia: real-world data from Spain. BMJ Open Diabetes Res Care, 2020; 8:e001052.

60.

Calliari

LEP

, Krakauer

, Vianna

AGD

, et al.: Real-world flash glucose monitoring in Brazil: can sensors make a difference in diabetes management in developing countries?. Diabetol Metab Syndr, 2020; 12:3.

61.

Battelino

, Danne

, Bergenstal

, et al.: Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care, 2019; 42:1593–1603.

62.

Cavero-Redondo

, Peleteiro

, Álvarez-Bueno

, et al.: Glycated haemoglobin A1c as a risk factor of cardiovascular outcomes and all-cause mortality in diabetic and non-diabetic populations: a systematic review and meta-analysis. BMJ Open, 2017; 7:e015949.

63.

Haas

, Persson

, Hagström Toft

, et al.: Treatment satisfaction correlated with glycaemic control and burden of diabetes in Swedish adolescents with type 1 diabetes. Acta Paediatr, 2020; 109:573–580.

64.

Asche

, LaFleur

, Conner

: A review of diabetes treatment adherence and the association with clinical and economic outcomes. Clin Ther, 2011; 33:74–109.

65.

Boye

, Curtis

, Lage

, Garcia-Perez

: Associations between adherence and outcomes among older, type 2 diabetes patients: evidence from a Medicare Supplemental database. Patient Prefer Adherence, 2016; 10:1573–1581.

66.

Hsu

, Lemon

, Wong

, et al.: Factors affecting medication adherence: patient perspective from five veterans affairs facilities. BMC Health Serv Res, 2014; 14:533–541.

67.

Emechebe

, Agu

, Malm

, Zgibor

: Examining the association between treatment satisfaction and medication adherence among patients with poorly controlled type 2 diabetes. Diabetes, 2018; 67 (Supplement 1):724-P.

68.

Ueno

, Chujo

, Takahashi

, et al.: Patient satisfaction and clinical efficacy of flash glucose monitoring in patients with type 1 diabetes: a prospective, single-center, single-arm study. Diabetes Ther, 2020; 11:1883–1890.

69.

Ontario Health (Quality): Flash glucose monitoring system for people with type 1 or type 2 diabetes: a health technology assessment. Ont Health Technol Assess Ser [Internet], 2019; 19:1–108.

70.

Gilmer

, O'Connor

, Rush

, et al.: Predictors of health care costs in adults with diabetes. Diabetes Care, 2005; 28:59–64.

71.

Wagner

, Sandhu

, Newton

, et al.: Effect of improved glycemic control on health care costs and utilization. JAMA, 2001; 285:182–189.

72.

Deshmukh

, Wilmot

, Gregory

, et al.: Effect of flash glucose monitoring on glycemic control, hypoglycemia, diabetes-related distress, and resource utilization in the Association of British Clinical Diabetologists (ABCD) nationwide audit. J Diabetes Care, 2020; 43:2153–2160.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.09 MB

Budget Impact of the Flash Continuous Glucose Monitoring System in Medicaid Diabetes Beneficiaries Treated with Intensive Insulin Therapy

Abstract

Objective:

Research Design and Methods:

Results:

Conclusions:

Introduction

Methods

Clinical inputs.

Cost inputs

Penetration projections

One-way sensitivity analysis

Results

Base case analysis

Impact to budget

Impact on utilization

One-way sensitivity analysis

Discussion

Conclusions

Footnotes

Acknowledgments

Author Disclosure Statement

Funding Information

Supplementary Material

References

Supplementary Material