Abstract
Purpose
To compare the impact of sodium-glucose cotransporter 2 inhibitor (SGLT2 inhibitor), glucagon-like peptide-1 receptor agonist (GLP-1 RA), with or without metformin, on out of pocket and total prescription expenditure and health-related quality of life (HRQoL) for patients with type 2 diabetes mellitus (T2DM).
Patients and Methods
This observational study utilized 2017–2021 Medical Expenditure Panel Survey (MEPS) data from patients with T2DM (≥18 years) on SGLT2 inhibitor, GLP-1 RA, with or without metformin, from payer and self-perspective. HRQoL was assessed using physical (PCS) and mental component summary (MCS) scores based on Veterans Rand 12. This study estimated survey-weighed out-of-pocket (OOP) costs for prescription refills and total prescription expenditures. Propensity score matching was used to mitigate selection bias and health expenditures, and HRQoL were compared using the Mann–Whitney U-test. P-value thresholds were recalculated using Bonferroni adjustment (Total prescription expenditure or OOP, PCS, and MCS: p=0.017).
Results
Patients on GLP-1 RA alone had significantly higher OOP costs than those on SGLT2 inhibitor alone (median: $166.50 vs $81.00, p<0.01). No significant difference existed between the two treatments for total prescription expenditures (median: $9831.53vs. $9458.80, p=0.059), MCS (median:52.41 vs 53.48, p=0.40), or PCS (median: 45.22 vs 44.54, p=0.19). Patients on metformin with GLP-1 RA had higher OOP costs compared to those on SGLT2 inhibitor with metformin (median: $140.40 vs $107.33, p <0.01). There is a significant difference between the combination treatments for total prescription expenditure (median: $9453.96 vs $6711.47, p<0.01), MCS (median: 54.19 vs 54.30, p=0.70), or PCS (median: 45.69 vs 46.08, p=0.55).
Conclusion
Even though patients on GLP-1 RA have higher OOP costs, the difference in PCS or MCS scores between GLP-1 RA and SGLT2 inhibitor was not significant. Further investigation is needed to study the long-term impact on HRQoL and clinical outcomes.
Keywords: SGLT2 inhibitor, GLP-1 receptor agonist, medical expenditure panel survey, health care expenditure, health-related quality of life
Introduction
Type 2 diabetes mellitus (T2DM) is a chronic metabolic disease that occurs when the pancreas fails to produce enough insulin to regulate glucose, or the body develops insulin resistance. In patients with T2DM excess glucose circulation in their blood for longterm, lead to complications affecting the circulatory, nervous, and immune systems. In 2022, 25.5 million Americans were diagnosed with diabetes, contributing to 107,000 deaths. The direct medical cost for diabetes in 2022 was $306.6 billion. The cost for glucose-lowering medications was $84.5 billion in 2022, an increase of 26% from 2017. The excess medical cost per person in 2022 was $12,022. On average, patients with diabetes have a medical expenditure 2.6 times higher than those without the condition.1 In addition to metformin, the 2024 American Diabetes Association (ADA) guideline recommends the use of sodium-glucose cotransporter 2 inhibitors (SGLT2 inhibitor) and/or glucagon-like peptide-1 receptor agonist (GLP-1 RA) for treating T2DM.2 SGLT2 inhibitor and GLP-1 RA are relatively new classes of antidiabetic medications, expecting significant improvements in major adverse cardiovascular events (MACE) in patients with T2DM.3–6
SGLT2 inhibitor decreases blood glucose by reducing glucose reabsorption at the proximal tubules in the kidney, which can lead to urine tract and genital infection.7 GLP-1 RA decreases blood glucose by enhancing insulin secretion from β-cells and decreasing glucagon release from α-cells in the pancreas. GLP-1 RA can cause gastrointestinal disorders, including nausea, vomiting, and diarrhea.7,8 However, these two medication classes are more expensive than traditional medications. SGLT2 inhibitor costs around $600, and GLP-1 RA costs around $1000 for one month’s supply. In contrast, metformin and sulfonylureas cost less than $100 for one month’s supply. While showing significant benefits in clinical outcomes, the high cost of SGLT2 inhibitor and GLP-1 RA has become a significant burden for patients who require these medications but cannot afford them.
Studies have analyzed the cost-effectiveness of GLP-1 RA and SGLT2 inhibitor.9–11 Most of these studies used the Center for Outcomes Research (CORE) Diabetes Model and involved hypothetical patients. Several of the studies were funded by companies that manufactured one of these medications.12–16 The results of these cost-effective analyses were controversial. Furthermore, most of the studies compared a single medication to another instead of comparing between these two medication classes.
Luo et al conducted a study comparing the out-of-pocket (OOP) costs of SGLT2 inhibitor and GLP-1 RA using Medicare Advantage claims data. Their analysis found that GLP-1 RA were slightly more expensive than SGLT2 inhibitor, with an average OOP costs of $118 vs $91, respectively. However, the study only included patients with Medicare Advantage insurance, making the findings less generalizable to populations with other insurance types. The generalization of these findings to the wider type 2 diabetes population are limited as this study excluded patients not receiving metformin.17
To address these gaps in the literature, our study aimed to assess the expenditures and health-related quality of life (HRQoL) associated with SGLT2 inhibitor vs GLP-1 RA, either alone or in combination with metformin, for patients with T2DM from payer’s and self-perspectives.18 By using the nationally representative Medical Expenditure Panel Survey - a real-world dataset, the aim of this study is to provide an assessment of the expenditure and -HRQoL for these two medication classes. This study hypothesized that SGLT2 inhibitor would have lower OOP costs and higher MCS or PCS scores than GLP-1 RA.
Methods
Data Source
MEPS is one of the most complete and publicly available data sources on health expenditure, supported by the Agency for Healthcare Research and Quality (AHRQ).18 MEPS used large-scale surveys to collect health expenditure-related information from families, individuals, medical providers, and employers across the US. MEPS includes a household component and insurance component. The household component collects demographic characteristics, medical conditions, utilized medical services, and associated costs for families and individuals. The insurance component focused on private insurance plans, including premiums and benefits.18 With the information it collects, MEPS is a valuable resource for researchers to better understand healthcare usage and costs in the US. Current study extracted patient characteristics, physical (PCS) and mental component summary (MCS) scores, and total prescription expenditure from full-year consolidated files. Additionally, this study analyzed out-of-pocket (OOP) prescription payments from prescribed medicines files, specifically using the “amount paid, self or family (imputed)” variableThis study also used the CLINK file to link the full-year consolidated file, medical conditions file, and prescribed medicines files, consolidating all patient information into a single file. We utilized data from 2017 to 2021 to evaluate total prescription expenditure, OOP. Additionally, PCS, and MCS based on the Veterans Rand 12 (VR 12) were also assessed.
Population and Study Sample
Patients that were ≥ 18 years old with T2DM on SGLT2 inhibitor or GLP-1 RA, either alone or in combination with metformin were included in this study. Publicly available MEPS full-year consolidated, medical conditions, and prescribed medicines data files from 2017 to 2021 (N=146,994) were analyzed.18 Subsequently, patients’ diagnoses with diabetes by professional healthcare providers using variable DSIDA53 from the full-year consolidated file (N=9083) were identified. Patients diagnosed with T2DM using variable ICD10CDX = E11 from medical conditions files (N=8222) (Figure 1) were separated. For OOP analysis, this study only includes patients with positive OOP costs.
Figure 1.
Flow diagram of adults with type 2 diabetes, medical expenditure panel survey, 2017–2021.
Abbreviation: DSDIA53, Diabetes diagnosis by health professional.
Measures
Outcome Measures
Both types of expenditures utilized in this study were direct medical expenditures. This study identified OOP and total prescription expenditures from the full-year consolidated file and adjusted both expenditures to 2021-dollar value according to the Bureau of Labor Statistics.19–21 This study analyzed PCS using VPCS42, and MCS using VMCS42, as variables in MEPS from the full-year consolidated file.
Independent Variable and Covariates
Current study classified patients with T2DM into two groups: patients on SGLT2 inhibitor either alone or with metformin and patients on GLP-1 RA either alone or with metformin (Table 1). Medications were identified using the variable TC1S1_1. All groups’ covariates according to patients’ baseline characteristics from the full-year consolidated file were balanced. These characteristics included age, sex, race, region, education, marriage state, income level, employment, insurance type, perceived health status, whether had diabetes caused eye or kidney problems, and whether had MI or stroke. Additionally, total number of medical conditions a patient had from the medical conditions file using the variable ICD10CDX were also balanced. Then total conditions were divided into 1–2, 3–5, 6–10, 11–20, 21–30, and 31–40 categories. Meanwhile, this study used the Elixhauser comorbidity score, a commonly used measurement for comorbidity, healthcare utilization, and expenditures.22 Comorbidity score is calculated using Comorbidity package in R studio (version 2023.09.1. Build 494).23 Based on the number of comorbidity scores identified, this study further categorized the comorbidity score into 0, 1, 2, and ≥ 3.
Table 1.
OOP, Total Prescription Expenditure, PCS, MCS of SGLT2 Inhibitor and GLP-1 RA Either Alone or with Metformin: Medical Expenditure Panel Survey 2017–2021
| SGLT2 Inhibitor (N = 117) Median (Range) |
GLP-1 RA (N = 117) Median (Range) |
P value | SGLT2 Inhibitor (N = 160) Median (Range) |
GLP-1 RA (N = 160) Median (Range) |
P value | |
|---|---|---|---|---|---|---|
| OOP ($) | 81.00 (1.07–7218.80) | 166.50 (4.17–10,119.57) | <0.001* | |||
| MCS | 53.48 (22.65–65.43) | 52.32 (28.14–69.32) | 0.15 | |||
| PCS | 47.80 (16.59–58.10) | 45.88 (17.33–65.84) | 0.10 | |||
| Total prescription expenditure ($) | 9458.80 (415.76–90,040.81) | 9831.53 (870.67–91,219.55) | 0.059 | |||
| MCS | 53.48 (22.65–65.43) | 52.41 (27.15–69.32) | 0.40 | |||
| PCS | 44.54 (16.59–58.10) | 45.22 (16.15–65.84) | 0.19 | |||
|
SGLT2 Inhibitor + Metformin (N = 271) Median (Range) |
GLP-1 RA + Metformin (N= 271) Median (Range) |
P value |
SGLT2 Inhibitor + Metformin (N = 313) Median (Range) |
GLP-1 RA + Metformin (N= 313) Median (Range) |
P value | |
| OOP ($) | 107.33 (3.85–4643.17) | 140.40 (1.36–9126.79) | <0.001* | |||
| MCS | 54.37 (20.46–65.97) | 54.24 (20.89–66.75) | 0.59 | |||
| PCS | 46.53 (17.20–60.88) | 44.12 (14.75–64.92) | 0.088 | |||
| Total prescription expenditure ($) | 6711.47 (393.87–80,660.56) | 9453.96 (914.72–79,833.30) | <0.001* | |||
| MCS | 54.30 (20.46–65.97) | 54.19 (21.40–66.75) | 0.69 | |||
| PCS | 46.08 (17.20–60.88) | 45.69 (15.09–66.46) | 0.55 |
Notes: *Statistical significance. P = 0.017 (after Bonferroni adjustment).
Abbreviations: OOP, out-of-pocket; PCS, physical component summary; MCS, mental component summary; SGLT2 inhibitor, SGLT2 inhibitor; GLP-1 RA, GLP-1 receptor agonist.
Statistical Analysis
Since this was an observational study, it lacked the randomized allocations of patients between SGLT2 inhibitor and GLP-1 RA groups. To minimize selection bias and improve the likelihood that outcomes were only due to different treatments, this study used a 1:1 propensity score matching. This study matched and compared patients’ baseline characteristics between groups using a chi-squared test. Before analyzing health expenditures, this study used the Kolmogorov–Smirnov test to confirm that the expenditures were not normally distributed. Consequently, this study used the Mann–Whitney U-test to compare health expenditures and HRQoL between treatments. This study also used Bonferroni adjustment for multiple comparisons to recalculate the p-value thresholds (Baseline characteristics: p=0.0031. Total prescription expenditure or OOP, PCS, and MCS: p=0.017). This study extracted and analyzed data using R studio (version 2023.09.1. Build 494).
Results
Patient Characteristics
For OOP, PCS, and MCS comparison, after propensity matching, both the SGLT2 inhibitor and GLP-1 RA groups have 117 patients each (Table 1). Additionally, 271 patients were included in both the SGLT2 inhibitor with metformin and GLP-1 RA with metformin group (Table 1). When comparing OOP, PCS, and MCS, before propensity score matching, the only statistical difference between the SGLT2 inhibitor and GLP-1 RA groups was whether patients had diabetes caused kidney problems (Table 2). However, after propensity matching, all baseline characteristics were well balanced (Table 2). Similarly, all baseline characteristics were well balanced for patients receiving SGLT2 inhibitor and GLP-1 RA in combination with metformin after propensity matching (Table 2). When comparing OOP, GLP-1 RA (GLP-1 RA alone, with metformin: $166.50, $140.40) was significantly higher than SGLT2 inhibitor (SGLT2 inhibitor alone, with metformin: $81.00, $107.33) either alone or with metformin (P <0.001) (Table 1). The median prescription number for GLP-1 RA and SGLT2 inhibitor was 4 (p=0.21). The median prescription number for GLP-1 RA with metformin is 9 and for SGLT2 inhibitor with metformin is 8 (p=0.18). While there was no significant difference between PCS and MCS (Table 1).
Table 2.
Baseline Characteristics of Adults with T2DM on SGLT2 Inhibitor and GLP-1 RA Without and with Metformin for OOP, PCS, and MCS Before and After Matching Using MEPS Data 2017–2021. (N=234 vs N = 542*)
| Baseline Characteristics | Before Matching | After Matching | Before Matching | After Matching | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SGLT2 Inhibitor N = 117 n (%) |
GLP-1 RA N = 224 n (%) |
P value | SGLT2 Inhibitor N = 117 n (%) |
GLP-1 RA N = 117 n (%) |
P value | SGLT2 Inhibitor + Metformin N = 271 n (%) |
GLP-1 RA + Metformin N = 358 n (%) |
P value | SGLT2 Inhibitor + Metformin N = 271 n (%) |
GLP-1 RA + Metformin N = 271 n (%) |
P value | |||
| Age | 18–44 | 14 (11.97) | 17 (7.59) | 0.57 | 14 (11.97) | 11 (9.40) | 0.88 | 20 (7.38) | 26 (7.26) | 0.75 | 20 (7.38) | 18 (6.64) | 0.97 | |
| 45–64 | 52 (44.44) | 104 (46.43) | 52 (44.44) | 51 (43.59) | 131 (48.34) | 169 (47.21) | 131 (48.34) | 132 (48.71) | ||||||
| ≥65 | 51 (43.59) | 103 (45.98) | 51 (43.59) | 55 (47.01) | 120 (44.28) | 163 (45.53) | 120 (44.28) | 121 (44.65) | ||||||
| Sex | Male | 54 (46.15) | 87 (38.84) | 0.38 | 54 (46.15) | 56 (47.86) | 0.86 | 145 (53.51) | 161 (44.97) | 0.053 | 145 (53.51) | 137 (50.55) | 0.65 | |
| Female | 63 (53.85) | 137 (61.16) | 63 (53.85) | 61 (52.14) | 126 (46.49) | 197 (55.03) | 126 (46.49) | 134 (49.45) | ||||||
| Race | White | 83 (70.94) | 172 (76.79) | 0.031 | 83 (70.94) | 88 (75.21) | 0.36 | 217 (80.07) | 285 (79.61) | 0.0094** | 217 (80.07) | 217 (80.07) | 0.86 | |
| Black | 23 (19.66) | 41 (18.30) | 23 (19.66) | 25 (21.37) | 27 (9.96) | 44 (12.29) | 27 (9.96) | 33 (12.18) | ||||||
| Multiple | 1 (0.86) | 7 (3.13) | 1 (0.86) | 0 (0) | 9 (3.32) | 15 (4.19) | 9 (3.32) | 7 (2.58) | ||||||
| Other | 10 (8.55) | 4 (1.79) | 10 (8.55) | 4 (3.42) | 18 (6.64) | 14 (3.91) | 18 (6.64) | 14 (5.17) | ||||||
| Region | Northeast | 13 (11.11) | 29 (12.95) | 0.81 | 13 (11.11) | 9 (7.69) | 0.88 | 41 (15.13) | 56 (15.64) | 0.95 | 41 (15.13) | 38 (14.02) | 0.99 | |
| Midwest | 20 (17.09) | 40 (17.86) | 20 (17.09) | 24 (20.51) | 72 (26.57) | 90 (25.14) | 72 (26.57) | 72 (26.57) | ||||||
| South | 70 (59.83) | 123 (54.91) | 70 (59.83) | 70 (59.83) | 105 (38.75) | 150 (41.90) | 105 (38.75) | 109 (40.22) | ||||||
| West | 14 (11.97) | 32 (14.29) | 14 (11.97) | 14 (11.97) | 53 (19.56) | 62 (17.32) | 53 (19.56) | 52 (19.19) | ||||||
| Education | No school | 0 (0) | 0 (0) | 0.67 | 0 (0) | 0 (0) | 0.93 | 0 (0) | 0 (0) | 0.32 | 0 (0) | 0 (0) | 0.59 | |
| Grade 1–8 | 3 (2.56) | 3 (1.34) | 3 (2.56) | 3 (2.56) | 26 (9.59) | 18 (5.03) | 26 (9.59) | 17 (6.27) | ||||||
| Grade 9–12 | 62 (52.99) | 101 (45.09) | 62 (52.99) | 59 (50.43) | 98 (36.16) | 132 (36.87) | 98 (36.16) | 101 (37.27) | ||||||
| Grade >12 | 52 (44.44) | 120 (53.57) | 52 (44.44) | 55 (47.01) | 147 (54.24) | 208 (58.10) | 147 (54.24) | 153 (56.46) | ||||||
| Marriage | Never married | 14 (11.97) | 23 (10.27) | 0.62 | 14 (11.97) | 11 (9.40) | 0.86 | 29 (10.70) | 49 (13.69) | 0.31 | 29 (10.70) | 30 (11.07) | 0.81 | |
| Widowed | 39 (33.33) | 77 (34.38) | 39 (33.33) | 38 (32.48) | 99 (36.53) | 104 (29.05) | 99 (36.53) | 89 (32.84) | ||||||
| Married | 64 (54.70) | 124 (55.36) | 64 (54.70) | 68 (58.12) | 143 (52.77) | 205 (57.26) | 143 (52.77) | 152 (56.09) | ||||||
| Income | Negative/poor/near poor | 25 (21.37) | 56 (25.00) | 0.93 | 25 (21.37) | 26 (22.22) | 0.91 | 57 (21.03) | 54 (15.08) | 0.084 | 57 (21.03) | 49 (18.08) | 0.84 | |
| Low | 16 (13.68) | 26 (11.61) | 16 (13.68) | 12 (10.26) | 29 (10.70) | 57 (15.92) | 29 (10.70) | 36 (13.28) | ||||||
| Middle | 33 (28.21) | 65 (29.02) | 33 (28.21) | 37 (31.62) | 72 (26.57) | 105 (29.33) | 72 (26.57) | 71 (26.20) | ||||||
| High | 43 (36.75) | 77 (34.38) | 43 (36.75) | 42 (35.90) | 113 (41.70) | 142 (39.66) | 113 (41.70) | 115 (42.44) | ||||||
| Employment | Not employed | 63 (53.85) | 141 (62.95) | 0.48 | 63 (53.85) | 67 (57.26) | 0.71 | 147 (54.24) | 210 (58.66) | 0.34 | 147 (54.24) | 150 (55.35) | 0.87 | |
| Employed | 54 (46.15) | 83 (37.05) | 54 (46.15) | 50 (42.74) | 124 (45.76%) | 148 (41.34%) | 124 (45.76%) | 121 (44.65%) | ||||||
| Insurance | Uninsured | 3 (2.56) | 5 (2.23) | 0.45 | 3 (2.56) | 3 (2.56) | 0.94 | 8 (2.95) | 3 (0.84) | 0.30 | 8 (2.95) | 3 (1.11) | 0.38 | |
| Private | 69 (58.97) | 132 (58.93) | 69 (58.97) | 66 (56.41) | 165 (60.89) | 232 (64.80) | 165 (60.89) | 174 (64.21) | ||||||
| Public | 45 (38.46) | 87 (38.84) | 45 (38.46) | 48 (41.03) | 98 (36.16%) | 123 (34.36%) | 98 (36.16%) | 94 (34.69%) | ||||||
| Perceived health status | Poor | 2 (1.71) | 26 (11.61) | 0.17 | 2 (1.71) | 1 (0.85) | 0.92 | 15 (5.54) | 26 (7.26) | 0.11 | 15 (5.54) | 15 (5.54) | 0.94 | |
| Fair | 36 (30.77) | 69 (30.80) | 36 (30.77) | 35 (29.91) | 56 (20.66) | 78 (21.79) | 56 (20.66) | 55 (20.30) | ||||||
| Good | 49 (41.88) | 86 (38.39) | 49 (41.88) | 54 (46.15) | 112 (41.33) | 157 (43.85) | 112 (41.33) | 121 (44.65) | ||||||
| Very good | 22 (18.80) | 37 (16.52) | 22 (18.80) | 21 (17.95) | 80 (29.52) | 84 (23.46) | 80 (29.52) | 71 (26.20) | ||||||
| Excellent | 8 (6.84) | 6 (2.68) | 8 (6.84) | 6 (5.13) | 8 (2.95) | 13 (3.63) | 8 (2.95) | 9 (3.32) | ||||||
| Comorbidity score | 0 | 24 (20.51) | 40 (17.86) | 0.93 | 24 (20.51) | 26 (22.22) | 0.99 | 40 (14.76) | 45 (12.57) | 0.96 | 40 (14.76) | 38 (14.02) | 0.99 | |
| 1 | 55 (47.01) | 78 (34.82) | 55 (47.01) | 53 (45.30) | 135 (49.82) | 165 (46.09) | 135 (49.82) | 135 (49.82) | ||||||
| 2 | 25 (21.37) | 57 (25.45) | 25 (21.37) | 25 (21.37) | 70 (25.83) | 108 (30.17) | 70 (25.83) | 70 (25.83) | ||||||
| ≥3 | 13 (11.11) | 49 (21.88) | 13 (11.11) | 13 (11.11) | 26 (9.59) | 40 (11.17) | 26 (9.59) | 28 (10.33) | ||||||
| Total condition | 1–2 | 12 (10.26) | 18 (8.04) | 0.14 | 12 (10.26) | 11 (9.40) | 0.99 | 16 (5.90) | 20 (5.59) | 0.73 | 16 (5.90) | 15 (5.54) | 0.98 | |
| 3–5 | 29 (24.79) | 48 (21.43) | 29 (24.79) | 32 (27.35) | 84 (31.00) | 96 (26.82) | 84 (31.00) | 85 (31.37) | ||||||
| 6–10 | 56 (47.86) | 75 (33.48) | 56 (47.86) | 55 (47.01) | 127 (46.86) | 149 (41.62) | 127 (46.86) | 120 (44.28) | ||||||
| 11–20 | 19 (16.24) | 68 (30.36) | 19 (16.24) | 18 (15.38) | 41 (15.13) | 83 (23.18) | 41 (15.13) | 48 (17.71) | ||||||
| 21–30 | 1 (0.86) | 14 (6.25) | 1 (0.86) | 1 (0.85) | 2 (0.74) | 9 (2.51) | 2 (0.74) | 2 (0.74) | ||||||
| 31–40 | 0 (0) | 1 (0.45) | 0 (0) | 0 (0) | 1 (0.37) | 1 (0.28) | 1 (0.37) | 1 (0.37) | ||||||
| Eye | No | 91 (77.78) | 162 (72.32) | 0.12 | 91 (77.78) | 93 (79.49) | 0.81 | 221 (81.55) | 285 (79.61) | 0.75 | 221 (81.55) | 218 (80.44) | 0.81 | |
| Yes | 26 (22.22) | 62 (27.68) | 26 (22.22) | 24 (20.51) | 50 (18.45) | 73 (20.39) | 50 (18.45) | 53 (19.56) | ||||||
| Kidney | No | 105 (89.74) | 163 (72.77) | 0.0028** | 105 (89.74) | 98 (83.76) | 0.28 | 249 (91.88) | 319 (89.11) | 0.40 | 249 (91.88) | 245 (90.41) | 0.66 | |
| Yes | 12 (10.26) | 61 (27.23) | 12 (10.26) | 19 (16.24) | 22 (8.12%) | 39 (10.89) | 22 (8.12) | 26 (9.59) | ||||||
| MI | No | 104 (88.89) | 196 (87.50) | 0.40 | 104 (88.89) | 102 (87.18) | 0.79 | 235 (86.72) | 315 (87.99) | 0.42 | 235 (86.72) | 236 (87.08) | 0.92 | |
| Yes | 13 (11.11) | 28 (12.50) | 13 (11.11) | 15 (12.82) | 36 (13.28) | 43 (12.01) | 36 (13.28) | 35 (12.92) | ||||||
| Stroke | No | 106 (90.60) | 196 (87.50) | 0.98 | 106 (90.60) | 107 (91.45) | 0.85 | 250 (92.25) | 317 (88.55) | 0.26 | 250 (92.25) | 250 (92.25) | 1 | |
| Yes | 11 (9.40) | 28 (12.50) | 11 (9.40) | 10 (8.55) | 21 (7.75) | 41 (11.45) | 21 (7.75) | 21 (7.75) | ||||||
Notes: *N=sample size after propensity score matching. **Statistical significance. P = 0.0031 (after Bonferroni adjustment).
Abbreviations: T2DM, type 2 diabetes mellitus; SGLT2 inhibitor, SGLT2 inhibitor; GLP-1 RA, GLP-1 receptor agonist; OOP, out-of-pocket; PCS, physical component summary; MCS, mental component summary; MEPS, Medical Expenditure Panel Survey.
After propensity matching, for total prescription expenditure, PCS, and MCS comparison, both SGLT2 inhibitor and GLP-1 RA groups have 160 patients each, and 313 patients were included for both SGLT2 inhibitor with metformin and GLP-1 RA with metformin groups (Table 1). Before propensity matching, the only statistical difference between SGLT2 inhibitor and GLP-1 RA, either alone or in combination with metformin, was race (P<0.001, P= 0.0016, respectfully) (Table 3). All baseline characteristics were well balanced after propensity matching (Table 3). Although the median total prescription expenditure for GLP-1 RA ($9453.96) was higher than for SGLT2 inhibitor ($9458.80), there was no significant difference (P=0.0059) (Table 1). The PCS (P=0.19) and MCS (P=0.40) scores were similar for both groups (Table 1). The median total prescription expenditure for GLP-1 RA was statistically significantly higher than the one for SGLT2 inhibitor ($ 9453.96 vs $ 6711.47. P< 0.001). There was also no significant difference in PCS, and MCS for SGLT2 inhibitor and GLP-1 RA with metformin (Table 1).
Table 3.
Baseline Characteristics of Adults with T2DM on SGLT2 Inhibitor and GLP-1 RA Without and with Metformin for Total Prescription Expenditure, PCS, and MCS Before and After Matching Using MEPS Data 2017–2021 (N=320 vs N = 626*)
| Baseline Characteristics | Before Matching | After Matching | Before Matching | After Matching | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SGLT2 Inhibitor N=160 n (%) |
GLP-1 RA N = 285 n (%) |
P value | SGLT2 Inhibitor N = 160 |
GLP-1 RA N = 160 n (%) |
P value | SGLT2 Inhibitor + Metformin N=313 n (%) |
GLP-1 RA + Metformin N = 404 n (%) |
P value | SGLT2 Inhibitor + Metformin N = 313 n (%) |
GLP-1 RA + Metformin N = 313 n (%) |
P value | ||
| Age | 18–44 | 17 (10.63) | 25 (8.77) | 0.43 | 17 (10.63) | 15 (9.38) | 0.93 | 24 (7.67) | 31 (7.67) | 0.91 | 24 (7.67) | 21 (6.71) | 0.89 |
| 45–64 | 82 (51.25) | 128 (44.91) | 82 (51.25) | 80 (50.00) | 154 (49.20) | 198 (49.01) | 154 (49.20) | 162 (51.76) | |||||
| ≥65 | 61 (38.13) | 132 (46.32) | 61 (38.13) | 65 (40.63) | 135 (43.13) | 175 (43.32) | 135 (43.13) | 130 (41.53) | |||||
| Sex | Male | 76 (47.50) | 106 (37.19) | 0.08 | 76 (47.50) | 68 (42.50) | 0.56 | 166 (53.04) | 178 (44.06) | 0.054 | 166 (53.04) | 159(50.80) | 0.72 |
| Female | 84 (52.50) | 179 (62.81) | 84 (52.50) | 92 (57.50) | 147 (46.96) | 226 (55.94) | 147 (46.96) | 154 (49.20) | |||||
| Race | White | 110 (68.75) | 212 (74.39) | <0.001** | 110 (68.75) | 117 (73.13) | 0.20 | 250 (79.87) | 312 (77.23) | 0.0016** | 250 (79.87) | 245 (78.27) | 0.58 |
| Black | 28 (17.50) | 56 (19.65) | 28 (17.50) | 33 (20.63) | 32 (10.22) | 60 (14.85) | 32 (10.22) | 45 (14.38) | |||||
| Multiple | 4 (2.50) | 11 (3.86) | 4 (2.50) | 4 (2.50) | 10 (3.19) | 18 (4.46) | 10 (3.19) | 9 (2.88) | |||||
| Other | 18 (11.25) | 6 (2.11) | 18 (11.25) | 6 (3.75) | 21 (6.71) | 14 (3.47) | 21 (6.71) | 14 (4.47) | |||||
| Region | Northeast | 19 (11.88) | 39 (13.68) | 0.80 | 19 (11.88) | 20 (12.50) | 0.94 | 50 (15.97) | 61 (15.10) | 0.87 | 50 (15.97) | 47 (15.02) | 0.96 |
| Midwest | 33 (20.63) | 51 (17.89) | 33 (20.63) | 32 (20.00) | 82 (26.20) | 101 (25.00) | 82 (26.20) | 81 (25.88) | |||||
| South | 86 (53.75) | 149 (52.28) | 86 (53.75) | 91 (56.88) | 116 (37.06) | 168 (41.58) | 116 (37.06) | 125 (39.94) | |||||
| West | 22 (13.75) | 46 (16.14) | 22 (13.75) | 17 (10.63) | 65 (20.77) | 74 (18.32) | 65 (20.77) | 60 (19.17) | |||||
| Education | No school | 0 (0) | 0 (0) | 0.83 | 0 (0) | 0 (0) | 0.76 | 0 (0) | 0 (0) | 0.15 | 0 (0) | 0 (0) | 0.41 |
| Grade 1–8 | 6 (3.75) | 11 (3.83) | 6 (3.75) | 9 (5.63) | 34 (10.86) | 22 (5.45) | 34 (10.86) | 21 (6.71) | |||||
| Grade 9–12 | 84 (52.50) | 131 (45.96) | 84 (52.50) | 82 (51.25) | 111 (35.46) | 157 (38.86) | 111 (35.46) | 120 (38.34) | |||||
| Grade >12 | 70 (43.75) | 143 (50.18) | 70 (43.75) | 69 (43.13) | 168 (53.67) | 225 (55.69) | 168 (53.67) | 172 (54.95) | |||||
| Marriage | Never married | 21 (13.13) | 38 (13.33) | 0.81 | 21 (13.13) | 19 (11.88) | 0.87 | 35 (11.18) | 58 (14.36) | 0.31 | 35 (11.18) | 38 (12.14) | 0.74 |
| Widowed | 53 (33.13) | 107 (37.54) | 53 (33.13) | 58 (36.25) | 118 (37.70) | 123 (30.45) | 118 (37.70) | 104 (33.23) | |||||
| Married | 86 (53.75) | 140 (49.12) | 86 (53.75) | 83 (51.88) | 160 (51.12) | 223 (55.20) | 160 (51.12) | 171 (54.63) | |||||
| Income | Negative/poor/near poor | 39 (24.38) | 87 (30.53) | 0.63 | 39 (24.38) | 42 (26.25) | 0.94 | 76 (24.28) | 77 (19.06) | 0.15 | 76 (24.28) | 65 (20.77) | 0.81 |
| Low | 20 (12.50) | 36 (12.63) | 20 (12.50) | 17 (10.63) | 36 (11.50) | 63 (15.59) | 36 (11.50) | 42 (13.42) | |||||
| Middle | 45 (28.13) | 74 (25.96) | 45 (28.13) | 49 (30.63) | 78 (24.92) | 115 (28.47) | 78 (24.92) | 83 (26.52) | |||||
| High | 56 (35.00) | 88 (30.88) | 56 (35.00) | 52 (32.50) | 122 (38.98) | 149 (36.88) | 122 (38.98) | 123 (39.30) | |||||
| Employment | Not employed | 85 (53.13) | 189 (65.61) | 0.20 | 85 (53.13) | 88 (55.00) | 0.81 | 179 (57.19) | 241 (59.65) | 0.70 | 179 (57.19) | 179 (57.19) | 1 |
| Employed | 75 (46.88) | 98 (34.39) | 75 (46.88) | 72 (45.00) | 134 (42.81) | 163 (40.35) | 134 (42.81) | 134 (42.81) | |||||
| Insurance | Uninsured | 3 (1.88) | 8 (2.81) | 0.82 | 3 (1.88) | 3 (1.88) | 0.95 | 8 (2.56) | 3 (0.74) | 0.32 | 8 (2.56) | 3 (0.96) | 0.41 |
| Private | 90 (56.25) | 147 (51.58) | 90 (56.25) | 87 (54.38) | 176 (56.23) | 246 (60.89) | 176 (56.23) | 187 (59.74) | |||||
| Public | 67 (41.88) | 130 (45.61) | 67 (41.88) | 70 (43.75) | 129 (41.21) | 155 (38.37) | 129 (41.21) | 123 (39.30) | |||||
| Perceived health status | Poor | 6 (3.75) | 36 (12.63) | 0.17 | 6 (3.75) | 4 (2.50) | 0.74 | 23 (7.35) | 31 (7.67) | 0.17 | 23 (7.35) | 20 (6.39) | 0.95 |
| Fair | 52 (32.50) | 89 (31.23) | 52 (32.50) | 51 (31.88) | 66 (21.09) | 89 (22.03) | 66 (21.09) | 64 (20.45) | |||||
| Good | 62 (38.75) | 109 (38.25) | 62 (38.75) | 72 (45.00) | 126 (40.26) | 176 (43.56) | 126 (40.26) | 137 (43.77) | |||||
| Very good | 31 (19.38) | 45 (15.79) | 31 (19.38) | 28 (17.50) | 89 (28.43) | 94 (23.27) | 89 (28.43) | 84 (26.84) | |||||
| Excellent | 9 (5.63) | 6 (2.11) | 9 (5.63) | 5 (3.13) | 9 (2.88) | 14 (3.47) | 9 (2.88) | 8 (2.56) | |||||
| Comorbidity score | 0 | 33 (20.63) | 50 (17.54) | 0.93 | 33 (20.63) | 32 (20.00) | 0.99 | 48 (15.34) | 51 (12.62) | 0.90 | 48 (15.34) | 46 (14.70) | 0.98 |
| 1 | 70 (43.75) | 97 (34.04) | 70 (43.75) | 69 (43.13) | 157 (50.16) | 180 (44.55) | 157 (50.16) | 154 (49.20) | |||||
| 2 | 35 (21.88) | 74 (25.96) | 35 (21.88) | 34 (21.25) | 75 (23.96) | 119 (29.46) | 75 (23.96) | 81 (25.88) | |||||
| ≥3 | 22 (13.75) | 64 (22.46) | 22 (13.75) | 25 (15.63) | 33 (10.54) | 54 (13.37) | 33 (10.54) | 32 (10.22) | |||||
| Total condition | 1–2 | 16 (10.00) | 21 (7.37) | 0.27 | 16 (10.00) | 17 (10.63) | 0.89 | 18 (5.75) | 23 (5.69) | 0.58 | 18 (5.75) | 17 (5.43) | 0.80 |
| 3–5 | 39 (24.38) | 59 (20.70) | 39 (24.38) | 37 (23.13) | 97 (30.99) | 103 (25.50) | 97 (30.99) | 93 (29.71) | |||||
| 6–10 | 71 (44.38) | 98 (34.39) | 71 (44.38) | 70 (43.75) | 143 (45.69) | 167 (41.34) | 143 (45.69) | 144 (46.01) | |||||
| 11–20 | 32 (20.00) | 83 (29.12) | 32 (20.00) | 31 (19.38) | 51 (16.29) | 99 (24.50) | 51 (16.29) | 58 (18.53) | |||||
| 21–30 | 2 (1.25) | 22 (7.72) | 2 (1.25) | 5 (3.13) | 3 (0.96) | 11 (2.72) | 3 (0.96) | 0 (0) | |||||
| 31–40 | 0 (0) | 2 (0.70) | 0 (0) | 0 (0) | 1 (0.32) | 1 (0.25) | 1 (0.32) | 1 (0.32) | |||||
| Eye | No | 117 (73.13) | 208 (72.98) | 0.84 | 117 (73.13) | 119 (74.38) | 0.86 | 250 (79.87) | 319 (78.96) | 0.39 | 250 (79.87) | 250 (79.87) | 1 |
| Yes | 43 (26.88) | 77 (27.02) | 43 (26.88) | 41 (25.63) | 63 (20.13) | 85 (21.04) | 63 (20.13) | 63 (20.13) | |||||
| Kidney | No | 138 (86.25) | 207 (72.63) | 0.0056*** | 138 (86.25) | 134 (83.75) | 0.63 | 285 (91.05) | 359 (88.86) | 0.48 | 285 (91.05) | 281 (89.78) | 0.69 |
| Yes | 22 (13.75) | 78 (27.37) | 22 (13.75) | 26 (16.25) | 28 (8.95) | 45 (11.14) | 28 (8.95) | 32 (10.22) | |||||
| MI | No | 143 (90.38) | 246 (86.32) | 0.86 | 143 (90.38) | 145 (90.63) | 0.79 | 271 (86.58) | 358 (88.61) | 0.43 | 271 (86.58) | 276 (88.18) | 0.62 |
| Yes | 17 (10.63) | 39 (13.68) | 17 (10.63) | 15 (9.38) | 42 (13.42) | 46 (11.39) | 42 (13.42) | 37 (11.82) | |||||
| Stroke | No | 147 (91.88) | 246 (86.32) | 0.42 | 147 (91.88) | 140 (87.50) | 0.26 | 287 (91.69) | 359 (88.86) | 0.28 | 287 (91.69) | 285 (91.05) | 0.82 |
| Yes | 13 (8.13) | 39 (13.68) | 13 (8.13) | 20 (12.50) | 26 (8.31) | 45 (11.14) | 26 (8.31) | 28 (8.95) | |||||
Notes: *N=sample size after propensity score matching. **Statistical significance. P = 0.0031 (after Bonferroni adjustment). ***Not statistical significance. P = 0.0031 (after Bonferroni adjustment).
Abbreviations: T2DM, type 2 diabetes mellitus; SGLT2 inhibitor, SGLT2 inhibitor; GLP-1 RA, GLP-1 receptor agonist; PCS, physical component summary; MCS, mental component summary; MEPS, Medical Expenditure Panel Survey.
Discussion
This study evaluated total prescription expenditure, OOP costs, MCS, and PCS among patients with T2DM who were on SGLT2 inhibitor or GLP-1 RA, either alone or in combination with metformin. The study found that the OOP for GLP-1 RA was higher than those for SGLT2 inhibitor, either alone or in combination with metformin.
As previous studies have indicated, patients with higher OOP costs might have lower adherence.24 This could further worsen the clinical outcomes including A1C, blood glucose, and kidney function. Quach et al reported higher median annual OOP costs for SGLT2 inhibitor ($430) and for GLP-1 RA ($480) compared to those reported in this study.25 This difference might be because their study used claims data from employer-sponsored pharmacy benefit plans, while this study used nationally representative survey data. Around 40% of this study’s population had public insurance with a lower co-pay. Another reason was that their study did not exclude patients receiving SGLT2 inhibitor or GLP-1 RA for weight loss purposes. However, this study focused on patients using these two medication classes for T2DM treatment. Currently, majority of the employer pharmacy plans exclude coverage for weight loss drugs. While most previous studies focused on the economic perspective between these two medication classes, this study also focused on the HRQoL. Dadwani et al conducted a study using the National Health and Nutrition Examination Survey (NHANES) 2013–2018 database and the United Kingdom Prospective Diabetes Study model to measure HRQoL between SGLT2 inhibitor and GLP-1 RA.26 Their study had similar results as this study. However, the difference was that their study focused on older patients with an average age of 73 years and used life year (LY) and quality-adjusted life year (QALY) to measure HRQoL. Their study indicated that GLP-1 RA (+0.29) has a similar LY compared to SGLT2 inhibitor (+0.26). The QALY also showed similar results (GLP-1 RA vs SGLT2 inhibitor: +0.15 vs 0.13). Torre et al used EQ-5D to compare HRQoL between 2014 and 2015 on patients with T2DM on either SGLT2 inhibitor or GLP-1 RA.27 The baseline scores between SGLT2 inhibitor (0.71) and GLP-1 RA (0.69) groups were similar. After 26 weeks of treatment, there were similar improvements in EQ-5D scores between these two medication classes (SGLT2 inhibitor vs GLP1 RA: 0.76 vs 0.71). Both previous studies and this study demonstrate that patients on SGLT2 inhibitor and GLP-1 RA had similar HRQoL.26,27
When comparing total prescription expenditure between SGLT2 inhibitor and GLP-1 RA, Insiya et al found that patients on GLP-1 RA had significantly higher total prescription expenditure than those on SGLT2 inhibitor.28 This discrepancy was likely because they used Medicare Advantage claim data, resulting in an older population (mean age: 65.1 years (GLP-1 RA), 65.3 years (SGLT2 inhibitor)) than this study (mean age: 59.9 years (GLP-1 RA), 59.2 years (SGLT2 inhibitor)). Another reason for the discrepancy was that the study population in their study was followed for the first 12 months after the patient was started on SGLT2 inhibitor or GLP-1 RA, and the patient needed to have at least two prescription claims.28 In thie current study, total prescription expenditure was evaluated disregarding their diagnosis year and the year they were first prescribed the medication.
This study analyzed health expenditure and HRQoL using MEPS data from 2017 to 2021, which included the COVID-19 pandemic period, during which patients with T2DM reduced their health utilization.29 Parker et al conducted a study using the Cost of Diabetes Economic Model to estimate the difference with and without the COVID-19 pandemic’s impact. Their study found that the direct medical cost of patients with T2DM showed a change of < 5%.1 Because the study by Parker et al used a model to estimate the cost, further real-world study on the COVID pandemic’s impact is needed to be assessed. Additionally, the MEPS data were self-reported survey based. This study cannot rule out recall bias in survey data. MEPS data does not contain any clinical data, therefore SGLT2 inhibitor or GLP-1 RA’s effect on efficacy or side effects in the real world cannot be determined.
Current study has a few limitations. In this study recall bias given MEPS is a self-reported survey database was not omitted.18 Another limitation is, this study cannot compare efficacy or safety outcomes between SGLT2 inhibitor or GLP-1 RA. Thus, further investigation using claims data and electronic medication records to compare health expenditure, efficacy or safety outcomes, and HRQoL between SGLT2 inhibitor and GLP-1 RA with a larger sample size is needed.
Conclusion
This study revealed that GLP-1 RA had significantly higher OOP costs than SGLT2 inhibitor. However, the study results need to be validated with larger patient data sets, since there was no significant difference in PCS or MCS. The findings nevertheless have implications for guiding healthcare providers when they prescribe SGLT2 inhibitor or GLP-1 RA for patients with T2DM. Further investigation using claims data and electronic medical records is needed to compare the economic outcome, clinical outcome, and HRQoL between SGLT2 inhibitor and GLP-1 RA.
Acknowledgment
All the work is conducted by the authors listed. This paper/ The abstract of this paper was presented at the ISPOR 2024 Conference as a poster presentation with interim findings. The poster’s abstract was published in “Poster Abstracts” in Value in Health: https://www.valueinhealthjournal.com/article/S1098-3015(24)00813-1/fulltext.
Funding Statement
The authors have no financial disclosures to report.
Abbreviations
T2DM, type 2 diabetes; ADA, American Diabetes Association; SGLT2 inhibitor, sodium-glucose cotransporter 2 inhibitor; GLP-1 RA, glucagon-like peptide-1 receptor agonist; MACE, major adverse cardiovascular events; GI, gastrointestinal disorders; CORE, Center for Outcomes Research Diabetes Model; MEPS, Medical Expenditure Panel Survey; HRQoL, health-related quality of life; AHRQ, Agency for Healthcare Research and Quality; PCS, physical component summary; MCS, mental component summary; OOP, out-of-pocket; VR 12, veterans rand 12.
Data Sharing Statement
Used data from publicly available files (MEPS). The first author takes responsibility for integrating data, subdata has not been published anywhere. The first author considers sharing with those who are interested.
Ethics Approval and Informed Consent
Under HHS CFR 46.104, this non-interventional study does not require IRB approval, as data obtained in the study, the patients whose information is not identified even to researchers received marketed drugs during routine medical practice and are not assigned to an intervention based on a protocol.
Consent for Publication
No images were copied or used.
Author Contributions
All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis, and interpretation, or in all these areas; took part in drafting, revising, or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.
Disclosure
The authors report no conflicts of interest in this work.
References
- 1.Parker ED, Lin J, Mahoney T, et al. Economic costs of diabetes in the U.S. in 2022. Diabetes Care. 2024;47(1):26–43. doi: 10.2337/dci23-0085 [DOI] [PubMed] [Google Scholar]
- 2.Committee ADAPP. 9. pharmacologic approaches to glycemic treatment: standards of care in diabetes—2024. Diabetes Care. 2023;47(Supplement_1):S158–S178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tian L, Ai S, Zheng H, et al. Cardiovascular and renal outcomes with sodium glucose co-transporter 2 inhibitors in patients with type 2 diabetes mellitus: a system review and network meta-analysis. Front Pharmacol. 2022;13. doi: 10.3389/fphar.2022.986186 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lee YM, Lee SH, Kim TH, Park EJ, Park YA, Jang JS. Cardiovascular outcomes with glucagon-like peptide 1 agonists and sodium-glucose cotransporter 2 inhibitors in patients with type 2 diabetes: a meta-analysis. Cardiol J. 2022;29(3):499–508. doi: 10.5603/CJ.a2020.0140 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Giugliano D, Scappaticcio L, Longo M, et al. GLP-1 receptor agonists and cardiorenal outcomes in type 2 diabetes: an updated meta-analysis of eight CVOTs. Cardiovasc Diabetol. 2021;20(1):1–11. doi: 10.1186/s12933-021-01366-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zou C, Liu X, Sang Y, Wang B, Liang J. Effects of SGLT2 inhibitors on cardiovascular outcomes and mortality in type 2 diabetes: a meta-analysis. Medicine. 2019;98(49):e18245. doi: 10.1097/MD.0000000000018245 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hsia DS, Grove O, Cefalu WT. An update on sodium-glucose co-transporter-2 inhibitors for the treatment of diabetes mellitus. Curr Opin Endocrinol Diabetes Obes. 2017;24(1):73–79. doi: 10.1097/MED.0000000000000311 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Filippatos TD, Panagiotopoulou TV, Elisaf MS. Adverse effects of GLP-1 receptor agonists. Rev Diabet Stud. 2014;11(3–4):202–230. doi: 10.1900/RDS.2014.11.202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hong D, Si L, Jiang M, et al. Cost effectiveness of sodium-glucose cotransporter-2 (SGLT2) inhibitors, glucagon-like peptide-1 (GLP-1) receptor agonists, and dipeptidyl peptidase-4 (DPP-4) inhibitors: a systematic review. Pharmacoeconomics. 2019;37(6):777–818. doi: 10.1007/s40273-019-00774-9 [DOI] [PubMed] [Google Scholar]
- 10.Yoshida Y, Cheng X, Shao H, Fonseca VA, Shi L. A systematic review of cost-effectiveness of sodium-glucose cotransporter inhibitors for type 2 diabetes. Curr Diabet Rep. 2020;20(4). doi: 10.1007/s11892-020-1292-5 [DOI] [PubMed] [Google Scholar]
- 11.Laursen HVB, Jørgensen EP, Vestergaard P, Ehlers LH. A systematic review of cost-effectiveness studies of newer non-insulin antidiabetic drugs: trends in decision-analytical models for modelling of type 2 diabetes mellitus. PharmacoEconomics. 2023;41(11):1469–1514. doi: 10.1007/s40273-023-01268-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ramos M, Ustyugova A, Hau N, Lamotte M. Cost-effectiveness of empagliflozin compared with liraglutide based on cardiovascular outcome trials in type II diabetes. J Comp Eff Res. 2020;9(11):781–794. doi: 10.2217/cer-2020-0071 [DOI] [PubMed] [Google Scholar]
- 13.Lee WC, Conner C, Hammer M. Results of a model analysis of the cost-effectiveness of liraglutide versus exenatide added to metformin, glimepiride, or both for the treatment of type 2 diabetes in the United States. Clin Ther. 2010;32(10):1756–1767. doi: 10.1016/j.clinthera.2010.08.010 [DOI] [PubMed] [Google Scholar]
- 14.Neslusan C, Teschemaker A, Willis M, Johansen P, Vo L. Cost-effectiveness analysis of canagliflozin 300 mg versus dapagliflozin 10 mg added to metformin in patients with type 2 diabetes in the United States. Diabetes Ther. 2018;9(2):565–581. doi: 10.1007/s13300-018-0371-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hunt B, Mocarski M, Valentine WJ, Langer J. Evaluation of the long-term cost-effectiveness of IDegLira versus liraglutide added to basal insulin for patients with type 2 diabetes failing to achieve glycemic control on basal insulin in the USA. J Med Econ. 2017;20(7):663–670. doi: 10.1080/13696998.2017.1301943 [DOI] [PubMed] [Google Scholar]
- 16.Hunt B, McConnachie CC, Gamble C, Dang-Tan T. Evaluating the short-term cost-effectiveness of liraglutide versus lixisenatide in patients with type 2 diabetes in the United States. J Med Econ. 2017;20(11):1117–1120. doi: 10.1080/13696998.2017.1347793 [DOI] [PubMed] [Google Scholar]
- 17.Luo J, Feldman R, Callaway Kim K, et al. Evaluation of out-of-pocket costs and treatment intensification with an SGLT2 inhibitor or GLP-1 RA in patients with type 2 diabetes and cardiovascular disease. JAMA Netw Open. 2023;6(6):e2317886. doi: 10.1001/jamanetworkopen.2023.17886 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Medical expenditure panel survey background. Available from: https://meps.ahrq.gov/mepsweb/about_meps/survey_back.jsp. Accessed March 12, 2023.
- 19.CPI inflation calculator. 2023.
- 20.Medical expenditure panel survey PUF codebook. Available from: https://www.meps.ahrq.gov/mepsweb/data_stats/download_data_files_codebook.jsp?PUFId=H233&varName=TOTEXP21. Accessed February 23, 2025.
- 21.Medical expenditure panel survey PUF codebook. Available from: https://www.meps.ahrq.gov/mepsweb/data_stats/download_data_files_codebook.jsp?PUFId=H229A&varName=RXSF21X. Accessed February 23, 2025.
- 22.Ou H, Mukherjee B, Erickson SR, Piette JD, Bagozzi RP, Balkrishnan R. Comparative performance of comorbidity indices in predicting health care-related behaviors and outcomes among Medicaid enrollees with type 2 diabetes. Popul Health Manag. 2012;15(4):220–229. doi: 10.1089/pop.2011.0037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gasparini A, Williman J, Chia SY, Teo E. Comorbidity: computing comorbidity scores. 2024. Available from: https://cran.r-project.org/web/packages/comorbidity/index.html. Accessed February 16, 2025.
- 24.Essien UR, Singh B, Swabe G, et al. Association of prescription co-payment with adherence to glucagon-like peptide-1 receptor agonist and sodium-glucose cotransporter-2 inhibitor therapies in patients with heart failure and diabetes. JAMA Network Open. 2023;6(6):e2316290. doi: 10.1001/jamanetworkopen.2023.16290 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Quach J, Midlam C, Sell C, Levin-Scherz J. Out-of-pocket costs for diabetes medications in employer-sponsored health insurance plans. Am J Manag Care. 2024;30(3):107–108. [DOI] [PubMed] [Google Scholar]
- 26.Dadwani RS, Wan W, Skandari MR, Huang ES. Expected health benefits of SGLT-2 inhibitors and GLP-1 receptor agonists in older adults. MDM Policy Pract. 2023;8(2). doi: 10.1177/23814683231187566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Torre C, Guerreiro J, Longo P, Raposo JF, Leufkens H, Martins AP. Health-related quality of life in adults with type 2 diabetes mellitus starting with new glucose lowering drugs: an inception cohort study. Primary Care Diabetes. 2019;13(3):221–232. doi: 10.1016/j.pcd.2018.11.009 [DOI] [PubMed] [Google Scholar]
- 28.Poonawalla IB, Bowe AT, Tindal MC, Meah YA, Schwab P. A real-world comparison of cardiovascular, medical and costs outcomes in new users of SGLT2 inhibitors versus GLP-1 agonists. Diabet Res Clin Pract. 2021;175:1. [DOI] [PubMed] [Google Scholar]
- 29.Amsah N, Md Isa Z, Ahmad N, Abdul Manaf MR. Impact of COVID-19 pandemic on healthcare utilization among patients with type 2 diabetes mellitus: a systematic review. Int J Environ Res Public Health. 2023;20(5):4577. doi: 10.3390/ijerph20054577 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Used data from publicly available files (MEPS). The first author takes responsibility for integrating data, subdata has not been published anywhere. The first author considers sharing with those who are interested.