Publication
Article
Author(s):
Real-world treatment of diabetic kidney disease in the United States, based on national-level health care claims and electronic health records data, is inconsistent with the current guidelines.
ABSTRACT
Objectives: Recent trials of glucose-lowering drugs (GLDs) have drawn attention to renal outcomes. Our goal was to understand how patients with diabetic kidney disease (DKD) are treated in general practices in the United States.
Study Design: Retrospective cohort study using a national-level claims data set and electronic health records.
Methods: Patients (≥ 18 years) with type 2 diabetes, whose estimated glomerular filtration rates (eGFRs) were between 15 and 89 mL/min/1.73 m2 between 2016 and 2018, were selected. Use of different GLDs during a 12-month period was examined across all eGFR levels.
Results: Of the 25,486 sample patients, 69.2%, 18.9%, 9.6%, and 2.3% had an eGFR in the ranges of 60 to 89, 45 to 59, 30 to 44, and 15 to 29 mL/min/1.73 m2, respectively. Metformin was used by nearly 33% of patients with an eGFR of 30 to 44 mL/min/1.73 m2 and by 10% of patients with an eGFR less than 30 mL/min/1.73 m2. Less than 10% (across all eGFR levels) of patients used glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors. Use of insulin was more frequent among patients with a lower eGFR (P < .05). The findings were similar in subgroups with different hemoglobin A1c levels (< 7% and ≥ 7%).
Conclusions: Real-world treatment of DKD in the United States is suboptimal. Inappropriate use of some GLD classes, especially in advanced DKD stages, was found along with lower than expected use of modern agents that are considered safe and effective to treat glycemic outcomes. Efforts may be needed to improve understanding of safety, glycemic efficacy, and overall clinical value of GLDs across DKD stages.
Am J Manag Care. 2022;28(8):e301-e307. https://doi.org/10.37765/ajmc.2022.89152
Takeaway Points
Our data demonstrate that real-world treatment of diabetic kidney disease (DKD) in the United States is inconsistent with the current guidelines, which may result in suboptimal outcomes.
Type 2 diabetes (T2D) is the leading cause of chronic kidney disease (CKD) in the United States,1 and approximately 40% of patients with T2D have CKD.2 CKD directly increases cardiovascular disease risk, the leading cause of morbidity and mortality in the T2D population.3 CKD attributed to diabetes, referred to as diabetic kidney disease (DKD), is the most common cause of end-stage renal disease.3
Poor glycemic control has been associated with DKD progression.4 Optimizing glucose control while slowing the progression of CKD has been a long-established goal in treating patients with DKD.3 However, many glucose-lowering drugs (GLDs) either are contraindicated or require careful dose adjustment in patients with DKD because the accumulation of active metabolites of GLDs may cause severe adverse effects (AEs) or hypoglycemia in patients with impaired renal function because of their altered metabolism and kidney clearance.5,6 In addition, accumulation of uremic toxins, inflammatory factors, and oxidative stress associated with CKD progression can induce insulin resistance.7
The impact of GLDs on renal function among patients with T2D and CKD have been designated as primary outcomes in recent clinical trials, including the CREDENCE, FLOW, and DAPA-CKD trials.8-11 Renal outcomes as secondary end points have been reported in many completed clinical trials of GLDs.8,12-16 The results from these trials aim to provide insights on using GLDs in patients with DKD.
Furthermore, researchers have also generated real-world evidence of the impact of GLDs on renal outcomes, in addition to their impact on glycemic outcomes.17-21 Based on an electronic health record database in the United States, Boye et al found that initiating glucagon-like peptide-1 receptor agonists (GLP-1 RAs), compared with other GLDs, was associated with significantly smaller declines in estimated glomerular filtration rates (eGFRs) and larger reductions in hemoglobin A1c (HbA1c) over 1 year.19 The authors also reported similar findings in a later study that compared initiating dulaglutide, a GLP-1 RA agent, with insulin glargine.20 Evidence from health systems outside of the United States have also demonstrated the beneficial impacts on renal outcomes from the use of sodium-glucose cotransporter-2 (SGLT2) inhibitors.17,18,21-23
However, data on GLD treatment patterns among patients with DKD in the United States are scarce. Based on national survey data from 2007 to 2012, Wu et al reported that GLD use was consistent with treatment guidelines published in 2012 by the National Kidney Foundation.24 On the other hand, using a large claims data set, Rhee et al found that GLD use in 2014 was not consistent with recent treatment guidelines regardless of CKD severity; they did not, however, examine the use of SGLT2 inhibitors.25 Use of some GLD classes, including metformin, dipeptidyl peptidase-4 (DPP4) inhibitors, GLP-1 RAs, and SGLT2 inhibitors, increased noticeably from 2007 to 2016 among Medicare beneficiaries who had CKD and T2D,26 and a steadily increasing trend of initiating SGLT2 inhibitors and GLP-1 RAs was observed between January 2013 and March 2020 among patients with DKD.27 However, utilization of new agents remains low.26
More evidence on GLD use among patients with DKD is needed to investigate how the recent developments in treatments are reflected in prescription patterns of general practitioners in the United States. Thus, the purpose of this study was to describe real-world use of GLDs in patients with T2D and renal impairment in the United States and to understand how GLD use varies by severity of renal impairment and by HbA1c levels.
METHODS
Data Source
The IBMMarketScan Explorys Claims–Electronic Medical Record (EMR) data were used in this study. This data set links patients’ longitudinal treatment and medical claims from the MarketScan Commercial and Medicare Supplemental databases to the same patients’ clinical records from the Explorys EMR data set. The combined data set provides clinical, claims, and financial data to support health economics and outcomes research, market research, and epidemiologic analyses. All databases are compliant with the Health Insurance Portability and Accountability Act; institutional review board approval was not required for this study.
Study Design and Sample Selection
Patients with T2D were identified using all available data from January 1, 2015, through December 31, 2019. Patients were classified as having T2D if they had more claims with T2D diagnosis codes (International Classification of Diseases, Ninth Revision [ICD-9] codes: 250.*0, 250.*2; International Classification of Diseases, Tenth Revision [ICD-10] code: E11) than claims with type 1 diabetes diagnosis codes (ICD-9 codes: 250.*1, 250.*3; ICD-10 code: E10). We estimated eGFR values for all dates between January 1, 2016, and December 31, 2018, from data on serum creatinine, age, race, and sex using the Chronic Kidney Disease–Epidemiology Collaboration equation.28 Due to potential multiple measures of eGFR for a patient, a randomization process aligned with a rolling cross-sectional study design was used to identify the index eGFR (eAppendix Figure 1 [eAppendix available at ajmc.com]).29 Each patient’s index date is defined by the date corresponding to the index eGFR.
The study focused on patients with index eGFR values in the range of 15 to 89 mL/min/1.73 m2 (ie, CKD stage 2 to 4) who were categorized into 4 index eGFR levels: 60 to 89, 45 to 59, 30 to 44, and 15 to 29 (the unit of eGFR is mL/min/1.73 m2 if not stated otherwise). The 6-month period before each patient’s index date serves as their baseline period; all patients were followed for 12 months from their index dates.
Patients were further required to have a T2D diagnosis code before July 1, 2015, to ensure that they developed CKD after having some experience managing T2D. Patients were also required to be 18 years or older as of January 1, 2016, and to have been continuously enrolled in their health and pharmacy plans during their baseline and follow-up periods. Patients were excluded if they had (1) evidence of dialysis or an eGFR less than 15 before the index date or (2) a diagnosis of gestational diabetes, pregnancy, or acute kidney injury in their baseline or follow-up periods.
Study Measures
Patient characteristics were assessed during the baseline period or as of the index date. Demographic characteristics included age, sex, race, health plan type, and index year. Clinical characteristics included the Charlson Comorbidity Index (CCI) score,30 the Diabetes Complications Severity Index (DCSI) score,31 and HbA1c level. The baseline HbA1c value was based on the mean laboratory result during the baseline period and index date.
The percentage of patients using the following classes of GLDs (ie, the prevalence of use) was assessed during the follow-up period: metformin, sulfonylureas (SUs), thiazolidinediones (TZDs), DPP4 inhibitors, GLP-1 RAs, SGLT2 inhibitors, alpha-glucosidase inhibitors (AGIs), meglitinides (MEGs), amylin agonists, and insulins (including basal insulin, bolus insulin, or premixed insulin). Prevalence was also assessed for the following regimens: oral GLDs only (as well as 1 class, 2 classes, or ≥ 3 classes of oral GLDs ever being used) and insulin without a GLP-1 RA prescription (with or without oral GLDs as a combination therapy), where combination therapy was defined by medications that overlapped for at least 60 days.
Statistical Analysis
Summary statistics (means and SDs for continuous measures, and counts and proportions for categorical measures) were computed and reported for all study measures stratified by their index eGFR levels. The Student t and χ2 tests were used to compare continuous and categorical study measures, respectively, between patients in one index eGFR level and those in the next-worse index eGFR level (ie, eGFR 60-89 vs 45-59, eGFR 45-59 vs 30-44, and eGFR 30-44 vs 15-29). The Cochran-Armitage test for trend was also used to assess the homogeneity of treatment patterns across index eGFR levels. These analyses were repeated in subgroups defined by baseline HbA1c (< 7% and ≥ 7%). All analyses were conducted using SAS version 9.4 (SAS Institute).
RESULTS
Patient Characteristics
The study sample included 25,486 patients who met the sample selection criteria (Figure 1). Most patients (n = 17,645; 69.2%) had index eGFR values of 60 to 89; much smaller proportions of patients had eGFR values of 45 to 59 (n = 4815; 18.9%), 30 to 44 (n = 2434; 9.6%), and 15 to 29 (n = 592; 2.3%). Patient characteristics, summarized across index eGFR levels, are presented in the Table. Patients with eGFRs in the range of 45 to 59 were older than those in the range of 60 to 89 (mean age, 72 vs 65 years; P < .05). Compared with patients with eGFRs of 45 to 59, patients with eGFRs of 30 to 44 were also older and less likely to be male. No statistically significant differences in demographic characteristics were observed between patients with eGFRs of 30 to 44 and those with eGFRs of 15 to 29. Patients’ mean baseline CCI and DCSI scores were higher for patients with more severe renal impairment (all P < .05). The mean baseline HbA1c value was 7.2 to 7.3 across all eGFR levels.
Treatment Patterns
As illustrated in Figure 2, around 78% to 80% of patients across all eGFR levels were treated with GLDs. Metformin, SUs, DPP4 inhibitors, and insulin were the most used agents (> 10% across all eGFR levels); less than 10% of patients used TZDs, GLP-1 RAs, SGLT2 inhibitors, or MEGs. Because so few (< 1%) patients used AGIs or amylin agonists, use of these 2 agents was not reported.
Patients with higher index eGFRs were more likely to have used metformin than patients with lower eGFR values (66% among patients with eGFR values of 60 to 89 vs 10% among those with eGFR values of 15 to 29; P < .05). Likewise, patients with higher index eGFRs were more likely to have used GLP-1 RAs and SGLT2 inhibitors (9% and 9% among patients with eGFR values of 60 to 89 vs 6% and 1% among those with eGFR values of 15 to 29; all P < .05). On the other hand, patients with higher index eGFRs were less likely to have used insulin (23% among patients with eGFR values of 60 to 89 vs 45% among those with eGFR values of 15 to 29; P < .05). The most commonly used insulin was basal insulin, followed by bolus and premixed insulin. Also, patients with higher index eGFRs were less likely to use SUs, DPP4 inhibitors, and MEGs (from eGFR values of 60 to 89 to eGFR values of 15 to 29, SUs: 28% to 37%; DPP4 inhibitors: 17% to 18%; MEGs: 1% to 2%; all P < .05).
A large proportion of patients only used oral GLDs despite an index eGFR of less than 45 (40% for eGFR values of 30 to 44 and 32% for eGFR values of 15 to 29; P < .05). The exclusive use of oral GLDs decreased with lower eGFR values, and a similar trend was observed when different numbers of oral GLDs (1, 2, and ≥ 3 classes of oral GLDs) were examined. Further investigation showed that proportionately fewer patients with index eGFR values of 60 to 89 used insulin without a concurrent injectable GLD (ie, GLP-1 RA) than those with index eGFR values of 15 to 29 (19% vs 40%; P < .05). This difference was mostly driven by the use of insulin monotherapy without GLP-1 RAs or oral GLDs (7% among patients with index eGFR values of 60 to 89 vs 27% among those with index eGFR values of 15 to 29; P < .05).
Subgroups Defined by Baseline HbA1c
The subgroups included 9392 patients with a baseline HbA1c less than 7% and 8782 patients with a baseline HbA1c of 7% or greater. The distributions of index eGFR levels were similar in the 2 subgroups: around 70%, 19%, 9%, and 2% had index eGFRs in the ranges of 60 to 89, 45 to 59, 30 to 44, and 15 to 29, respectively. The differences in baseline characteristics across all eGFR levels observed in the overall sample were similar in each of the HbA1c subgroups (eAppendix Table).
Treatment patterns across all eGFR levels among patients with a baseline HbA1c less than 7% were similar to those for the overall sample: Metformin, GLP-1 RAs, and SGLT2 inhibitors were less frequently used by patients with higher index eGFRs than by patients with lower index eGFRs, whereas the opposite was true for insulin, SUs, TZDs, DPP4 inhibitors, and MEGs (Figure 3). Proportionately fewer patients with higher index eGFRs exclusively used oral GLDs (P < .05). Use of insulin in this subgroup was noticeably less common than in the overall sample, whereas the exclusive use of oral GLDs, especially use of only 1 oral GLD class, was much more frequent.
More than 95% of patients with a baseline HbA1c of 7% or greater were treated with GLDs regardless of index eGFR level (Figure 4). The proportions of patients using TZDs and DPP4 inhibitors were not statistically significantly different across all eGFR levels, whereas the relationships between use of other treatments and index eGFR values were the same as those in the overall sample. Despite having a baseline HbA1c of 7% or greater, a large proportion of patients with an index eGFR less than 45 used only oral GLDs (31% and 26% of patients with index eGFR values of 30 to 44 and 15 to 29, respectively).
DISCUSSION
This study provides important evidence on how well the use of real-world GLD treatments for patients with DKD in the United States compare with the most updated guidelines published by the American Diabetes Association.3 We found that treatments for glycemic control varied by renal impairment severity. Metformin, GLP-1 RAs, and SGLT2 inhibitors were less commonly used by patients with more severe renal impairment, whereas SUs, TZDs, DPP4 inhibitors, MEGs, and insulins were more commonly used by patients with more severe renal impairment. Metformin, SUs, DPP4 inhibitors, and insulin were the most commonly used agents across all eGFR levels. These findings are consistent with evidence using earlier data.25
Despite the evidence of the cardiorenal benefits of newer agents, such as GLP-1 RAs and SGLT2 inhibitors, that has been confirmed by various trials conducted since 2016,32-36 the utilization of these newer agents was quite low across all eGFR levels in our study—for example, less than 16% of patients with a high baseline HbA1c used these agents. The suboptimal use of these newer agents may be affected by providers’ established prescribing behaviors, medication costs, and/or formulary restrictions.37,38 Treatment decision algorithms specific to patients with DKD may be needed for health care providers, especially primary care providers39; only 59% of patients in our study with eGFR values of 15 to 29 had outpatient visits to a specialist (cardiologist, endocrinologist, or nephrologist) during their follow-up periods (eAppendix Figure 2). Moreover, efforts should be made to provide easier availability with better pricing of the newer agents.
A previous study, based on a 5% random sample of Medicare enrollees with CKD and T2D from 2007 to 2016, showed that the use of metformin, DPP4 inhibitors, GLP-1 RAs, and SGLT2 inhibitors increased significantly over that time period.26 A similar trend was observed in an ad hoc analysis of this study, which showed that use of GLP-1 RAs increased annually from 2016 to 2019 regardless of patients’ eGFR levels (eAppendix Figure 3). This increase was noticeable for patients with low eGFR levels (from 6% in 2016 to 21% in 2019 among patients with a baseline eGFR of 15 to 29; P < .05). The use of SGLT2 inhibitors also increased annually, except for patients with baseline eGFR values of 15 to 29. Although the use of GLP-1 RAs and SGLT2 inhibitors increased over time, the proportion of patients using these medications remained small. Efforts are still needed to further promote using the agents that have evidence of renal and cardiovascular benefits in patients with T2D.
Although insulin has been the mainstream treatment for patients with T2D and moderate to severe CKD (eGFR values of 15-45), especially for patients with uncontrolled glycemic levels, evidence has accumulated that the AEs of intensive glycemic control (hypoglycemia and mortality) have increased among patients with CKD.40 Use of insulin was noticeably more frequent among patients with lower eGFR levels in our study. This association was most obvious for insulin monotherapy (ie, insulin without GLP-1 RAs or oral GLDs). Although insulin can be used across all eGFR levels, such therapy must be monitored and carefully titrated for patients with low eGFR levels because of increasing hypoglycemia risk. The higher risk of AEs associated with insulin use may explain the observed large proportion of patients exclusively using oral GLDs, even among patients with a baseline HbA1c of 7% or greater and an eGFR less than 45.
Although metformin is the first-line treatment for T2D, it is contraindicated for patients with an eGFR less than 30 and is not recommended for patients with eGFRs between 30 and 45; increased renal function monitoring is recommended if metformin is continued in patients with an eGFR less than 45.41 Nearly one-third of patients with eGFR values of 30 to 44 used metformin in our study; more concerning is that 10% of patients with an eGFR less than 30 used this agent, which is clearly contraindicated.
SUs and DPP4 inhibitors were the most commonly used second-line noninsulin treatments in the study (> 15% of the patients used these agents across all eGFR levels). However, SUs are associated with high risk of hypoglycemia, especially among those with renal impairment, and they are also associated with weight gain; DPP4 inhibitors, however, are modestly effective at controlling blood glucose levels without evidence of reducing cardiovascular disease or CKD progression risk.42 These findings, along with the high use of insulin and contraindicated use of metformin among patients with low eGFR levels, suggest that GLD treatment options for patients with advanced DKD may be limited and it is challenging to achieve glycemic control in these patients.
Limitations
Our study is subject to limitations that commonly apply to claims- and EMR-based studies, such as possible coding errors or missing data and differences in patient characteristics due to changes in data contributors. Furthermore, the database contained patients covered by commercial insurance, which may not be generalizable to the entire US population of patients with DKD. Lastly, due to the lack of albumin-to-creatinine ratio values (available in about 10% of the sample), patients’ renal function status was measured only by eGFR, which may result in misclassification of CKD severity. However, these limitations should have minimal impacts on our findings as the distribution of CKD stage defined by the index eGFR in our study sample was aligned with that observed in nationally representative data.24
CONCLUSIONS
Real-world treatment of DKD in the United States is suboptimal for lowering risk of cardiorenal events, as evidenced by the low usage of noninsulin GLDs that not only effectively control blood glucose but have proven beneficial effects on renal and cardiovascular outcomes even among patients with advanced renal function decline (eGFR extending down to 15). Treatment of hyperglycemia in patients with DKD was not consistent with current guidelines, as evidenced by metformin use by patients with an eGFR less than 30. This study suggests that, given the limited pharmacologic options with which to control glucose and reduce cardiorenal risk among patients with advanced DKD, efforts may be needed to improve understanding of safety, glycemic control effectiveness, and overall clinical value of GLDs across DKD stages.
Acknowledgments
Robert Donaldson, PhD, Real-world Evidence, Evidera, England, helped in the programming of this study.
Author Affiliations: Novo Nordisk Inc (NNI, TDT, CG, SM), Plainsboro, NJ; Real-world Evidence, Evidera (QL, SS, MLG), Waltham, MA; The University of Chicago Medicine (GB), Chicago, IL.
Source of Funding: This study was sponsored by Novo Nordisk Inc.
Author Disclosures: Dr Iyer, Dr Dang-Tan, Dr Gamble, and Dr Mehanna are employed by Novo Nordisk Inc; Dr Iyer and Dr Dang-Tan also own employee stock. Dr Li, Dr Shah, and Dr Ganz are employed by Evidera, a health economics and outcomes research consultancy that was hired by Novo Nordisk to conduct the study and prepare the manuscript. Dr Bakris served as a consultant or paid advisor on new product development not in this paper.
Authorship Information: Concept and design (NNI, QL, SS, CG, GB); acquisition of data (NNI); analysis and interpretation of data (NNI, QL, SS, MLG, TDT, CG, SM, GB); drafting of the manuscript (NNI, QL, SS, TDT, SM, GB); critical revision of the manuscript for important intellectual content (NNI, QL, MLG, TDT, CG, SM); statistical analysis (NNI, QL); administrative, technical, or logistic support (NNI); and supervision (NNI, MLG).
Address Correspondence to: Neeraj N. Iyer, PhD, Novo Nordisk Inc, 800 Scudders Mill Rd, Plainsboro, NJ 08536. Email: nriy@novonordisk.com.
REFERENCES
1. Bailey RA, Wang Y, Zhu V, Rupnow MFT. Chronic kidney disease in US adults with type 2 diabetes: an updated national estimate of prevalence based on Kidney Disease: Improving Global Outcomes (KDIGO) staging. BMC Res Notes. 2014;7:415. doi:10.1186/1756-0500-7-415
2. Koro CE, Lee BH, Bowlin SJ. Antidiabetic medication use and prevalence of chronic kidney disease among patients with type 2 diabetes mellitus in the United States. Clin Ther. 2009;31(11):2608-2617. doi:10.1016/j.clinthera.2009.10.020
3. American Diabetes Association. 10. Cardiovascular disease and risk management: Standards of Medical Care in Diabetes – 2020. Diabetes Care. 2020;43(suppl 1):S111-S134. doi:10.2337/dc20-S010
4. Pecoits-Filho R, Abensur H, Betonico CC, et al. Interactions between kidney disease and diabetes: dangerous liaisons. Diabetol Metab Syndr. 2016;8:50. doi:10.1186/s13098-016-0159-z
5. Dreisbach AW, Lertora JJL. The effect of chronic renal failure on drug metabolism and transport. Expert Opin Drug Metab Toxicol. 2008;4(8):1065-1074. doi:10.1517/17425255.4.8.1065
6. Neumiller JJ, Alicic RZ, Tuttle KR. Therapeutic considerations for antihyperglycemic agents in diabetic kidney disease. J Am Soc Nephrol. 2017;28(8):2263-2274. doi:10.1681/ASN.2016121372
7. Smith D, DeFronzo RA. Insulin resistance in uremia mediated by postbinding defects. Kidney Int. 1982;22(1):54-62. doi:10.1038/ki.1982.132
8. Evaluation of the effects of canagliflozin on renal and cardiovascular outcomes in participants with diabetic nephropathy (CREDENCE). ClinicalTrials.gov. Updated December 5, 2019. Accessed May 15, 2021. https://clinicaltrials.gov/ct2/show/NCT02065791
9. A research study to see how semaglutide works compared to placebo in people with type 2 diabetes and chronic kidney disease (FLOW). ClinicalTrials.gov. Updated January 27, 2022. Accessed May 5, 2022. https://clinicaltrials.gov/ct2/show/NCT03819153
10. Researching cardiovascular Events with a Weekly INcretin in Diabetes (REWIND). ClinicalTrials.gov. Updated October 8, 2019. Accessed May 15, 2021. https://clinicaltrials.gov/ct2/show/NCT01394952
11. Heerspink HJL, Stefánsson BV, Correa-Rotter R, et al; DAPA-CKD Trial Committees and Investigators. Dapagliflozin in patients with chronic kidney disease. N Engl J Med. 2020;383(15):1436-1446. doi:10.1056/NEJMoa2024816
12. Mosenzon O, Blicher TM, Rosenlund S, et al; PIONEER 5 Investigators. Efficacy and safety of oral semaglutide in patients with type 2 diabetes and moderate renal impairment (PIONEER 5): a placebo-controlled, randomised, phase 3a trial. Lancet Diabetes Endocrinol. 2019;7(7):515-527. doi:10.1016/S2213-8587(19)30192-5
13. Davies MJ, Bain SC, Atkin SL, et al. Efficacy and safety of liraglutide versus placebo as add-on to glucose-lowering therapy in patients with type 2 diabetes and moderate renal impairment (LIRA-RENAL): a randomized clinical trial. Diabetes Care. 2016;39(2):222-230. doi:10.2337/dc14-2883
14. Cardiovascular and renal microvascular outcome study with linagliptin in patients with type 2 diabetes mellitus (CARMELINA). ClinicalTrials.gov. Updated April 4, 2019. Accessed May 15, 2021. https://clinicaltrials.gov/ct2/show/NCT01897532
15. A study comparing dulaglutide with insulin glargine on glycemic control in participants with type 2 diabetes (T2D) and moderate or severe chronic kidney disease (CKD) (AWARD-7). ClinicalTrials.gov. Updated September 19, 2019. Accessed May 15, 2021. https://clinicaltrials.gov/ct2/show/NCT01621178
16. Neal B, Perkovic V, Mahaffey KW, et al; CANVAS Program Collaborative Group. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med. 2017;377(7):644-657. doi:10.1056/NEJMoa1611925
17. Cahn A, Melzer-Cohen C, Pollack R, Chodick G, Shalev V. Acute renal outcomes with sodium-glucose co-transporter-2 inhibitors: real-world data analysis. Diabetes Obes Metab. 2019;21(2):340-348. doi:10.1111/dom.13532
18. Lin YH, Huang YY, Hsieh SH, Sun JH, Chen ST, Lin CH. Renal and glucose-lowering effects of empagliflozin and dapagliflozin in different chronic kidney disease stages. Front Endocrinol (Lausanne). 2019;10:820. doi:10.3389/fendo.2019.00820
19. Boye KS, Botros FT, Haupt A, Woodward B, Lage MJ. Glucagon-like peptide-1 receptor agonist use and renal impairment: a retrospective analysis of an electronic health records database in the U.S. population. Diabetes Ther. 2018;9(2):637-650. doi:10.1007/s13300-018-0377-5
20. Boye KS, Mody R, Wu J, Lage MJ, Botros FT, Woodward B. Effects of dulaglutide and insulin glargine on estimated glomerular filtration rate in a real-world setting. Clin Ther. 2018;40(8):1396-1407. doi:10.1016/j.clinthera.2018.07.002
21. Kobayashi K, Toyoda M, Kimura M, et al. Retrospective analysis of effects of sodium-glucose co-transporter 2 inhibitor in Japanese type 2 diabetes mellitus patients with chronic kidney disease. Diab Vasc Dis Res. 2019;16(1):103-107. doi:10.1177/1479164118802759
22. Smith SM. SGLT2 inhibitors and kidney outcomes in the real world. BMJ. 2020;369:m1584. doi:10.1136/bmj.m1584
23. Heerspink HJL, Karasik A, Thuresson M, et al. Kidney outcomes associated with use of SGLT2 inhibitors in real-world clinical practice (CVD-REAL 3): a multinational observational cohort study. Lancet Diabetes Endocrinol. 2020;8(1):27-35. doi:10.1016/S2213-8587(19)30384-5
24. Wu B, Bell K, Stanford A, et al. Understanding CKD among patients with T2DM: prevalence, temporal trends, and treatment patterns-NHANES 2007-2012. BMJ Open Diabetes Res Care. 2016;4(1):e000154. doi:10.1136/bmjdrc-2015-000154
25. Rhee JJ, Han J, Montez-Rath ME, et al. Antidiabetic medication use in patients with type 2 diabetes and chronic kidney disease. J Diabetes Complications. 2019;33(11):107423. doi:10.1016/j.jdiacomp.2019.107423
26. Zhao JZ, Weinhandl ED, Carlson AM, St Peter WL. Glucose-lowering medication use in CKD: analysis of US Medicare beneficiaries between 2007 and 2016. Kidney Med. 2021;3(2):173-182.e1. doi:10.1016/j.xkme.2020.09.016
27. Harris ST, Patorno E, Zhuo M, Kim SC, Paik JM. Prescribing trends of antidiabetes medications in patients with type 2 diabetes and diabetic kidney disease, a cohort study. Diabetes Care. 2021;44(10):2293-2301. doi:10.2337/dc21-0529
28. Glomerular filtration rate (GFR) calculators. National Institute of Diabetes and Digestive and Kidney Diseases. Accessed May 15, 2021. https://www.niddk.nih.gov/health-information/communication-programs/nkdep/laboratory-evaluation/glomerular-filtration-rate-calculators
29. Brady HE, Johnston R. The rolling cross-section and causal attribution. In: Brady HE, Johnston R, eds. Capturing Campaign Effects. University of Michigan Press; 2006:164-195.
30. Charlson ME, Charlson RE, Peterson JC, Marinopoulos SS, Briggs WM, Hollenberg JP. The Charlson comorbidity index is adapted to predict costs of chronic disease in primary care patients. J Clin Epidemiol. 2008;61(12):1234-1240. doi:10.1016/j.jclinepi.2008.01.006
31. Glasheen WP, Renda A, Dong Y. Diabetes Complications Severity Index (DCSI)—update and ICD-10 translation. J Diabetes Complications. 2017;31(6):1007-1013. doi:10.1016/j.jdiacomp.2017.02.018
32. Davidson JA. SGLT2 inhibitors in patients with type 2 diabetes and renal disease: overview of current evidence. Postgrad Med. 2019;131(4):251-260. doi:10.1080/00325481.2019.1601404
33. Perkovic V, Jardine MJ, Neal B, et al; CREDENCE Trial Investigators. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med. 2019;380(24):2295-2306. doi:10.1056/NEJMoa1811744
34. Tuttle KR, Lakshmanan MC, Rayner B, et al. Dulaglutide versus insulin glargine in patients with type 2 diabetes and moderate-to-severe chronic kidney disease (AWARD-7): a multicentre, open-label, randomised trial. Lancet Diabetes Endocrinol. 2018;6(8):605-617. doi:10.1016/S2213-8587(18)30104-9
35. Kelly MS, Lewis J, Huntsberry AM, Dea L, Portillo I. Efficacy and renal outcomes of SGLT2 inhibitors in patients with type 2 diabetes and chronic kidney disease. Postgrad Med. 2019;131(1):31-42. doi:10.1080/00325481.2019.1549459
36. Marso SP, Daniels GH, Brown-Frandsen K, et al; LEADER Steering Committee; LEADER Trial Investigators. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375(4):311-322. doi:10.1056/NEJMoa1603827
37. Ackermann RT, Wallia A, O’Brien MJ, et al. Correlates of second-line type 2 diabetes medication selection in the USA. BMJ Open Diabetes Res Care. 2017;5(1):e000421. doi:10.1136/bmjdrc-2017-000421
38. Grant RW, Wexler DJ, Watson AJ, et al. How doctors choose medications to treat type 2 diabetes: a national survey of specialists and academic generalists. Diabetes Care. 2007;30(6):1448-1453. doi:10.2337/dc06-2499
39. Li J, Albajrami O, Zhuo M, Hawley CE, Paik JM. Decision algorithm for prescribing SGLT2 inhibitors and GLP-1 receptor agonists for diabetic kidney disease. Clin J Am Soc Nephrol. 2020;15(11):1678-1688. doi:10.2215/CJN.02690320
40. Williams ME, Garg R. Glycemic management in ESRD and earlier stages of CKD. Am J Kidney Dis. 2014;63(2 suppl 2):S22-S38. doi:10.1053/j.ajkd.2013.10.049
41. FDA drug safety communication: FDA revises warnings regarding use of the diabetes medicine metformin in certain patients with reduced kidney function. FDA. Updated November 14, 2017. Accessed May 15, 2021. https://www.fda.gov/drugs/drug-safety-and-availability/fda-drug-safety-communication-fda-revises-warnings-regarding-use-diabetes-medicine-metformin-certain
42. Bennett WL, Balfe LM, Faysal JM. AHRQ’s comparative effectiveness research on oral medications for type 2 diabetes: a summary of the key findings. J Manag Care Pharm. 2012;18(1 suppl A):1-22. doi:10.18553/jmcp.2012.18.S1-A.1