top of page

“Alarming” Number of Patients Stop Taking Second-Line Diabetes Drugs

A large proportion—about 40%—of people with type 2 diabetes who were prescribed second-line medications after taking metformin to manage their condition stopped the drugs within 1 year, according to a retrospective cohort study that involved more than 82 000 insured participants between 2013 and 2017.


Nearly two-thirds modified their treatment in some way, including switching or intensifying their medications. About half of the people who stopped taking their medications had been prescribed glucagon-like peptide-1 (GLP-1) receptor agonists.


The findings “highlight the need for new prescribing approaches and patient supports that can maximize medication adherence and reduce health system waste,” the researchers wrote in the American Journal of Managed Care.

---------------------------------------------------------------------------------------------------------------------

Treatment Modification After Initiating Second-Line Medication for Type 2 Diabetes

December 12, 2023


ABSTRACT

Objectives: To describe changes in antidiabetic medication (ADM) use and characteristics associated with changes in ADM use after initiation of noninsulin second-line therapy.


Study Design: Retrospective cohort study.


Methods: This study analyzed private health plan claims for adults with type 2 diabetes who initiated 1 of 5 index ADM classes: sulfonylureas, dipeptidyl peptidase 4 inhibitors (DPP4is), sodium-glucose cotransporter 2 inhibitors, glucagon-like peptide-1 receptor agonists (GLP-1 RAs), or thiazolidinediones. Analyses evaluated 3 treatment modification outcomes—discontinuation, switching, and intensification—over 12-month follow-up.


Results: Of 82,624 included adults, nearly two-thirds (63.6%) experienced any treatment modification. Discontinuation was the most common modification (38.6%), especially among patients prescribed GLP-1 RAs (50.3%). Switching occurred in 5.2% of patients and intensification in 19.8%. In adjusted analysis, compared with patients prescribed sulfonylureas, discontinuation risk was 7% higher (HR, 1.07; 95% CI, 1.04-1.10) among patients prescribed DPP4is and 28% higher (HR, 1.28; 95% CI, 1.23-1.33) among patients prescribed GLP-1 RAs. Compared with sulfonylureas, all other index ADM classes had higher risks of switching and lower risks of intensification. Younger age group and female sex were both associated with higher risks of all modifications. Compared with index ADM prescription by a family medicine or internal medicine physician, index prescription by an endocrinologist was associated with both lower discontinuation risk and higher intensification risk.


Conclusions: Most patients experienced a treatment modification within 1 year. Results highlight the need for new prescribing approaches and patient supports that can maximize medication adherence and reduce health system waste.


Am J Manag Care. 2023;29(12):661-668. https://doi.org/10.37765/ajmc.2023.89466


_____


Takeaway Points

Among privately insured adults starting noninsulin second-line therapy for type 2 diabetes, nearly two-thirds (63.6%) experienced any of 3 treatment modification outcomes—discontinuation, switching, and intensification—over a 12-month follow-up.

  • Discontinuation was the most common modification (38.6%), followed by intensification (19.8%) and switching (5.2%).

  • Patients prescribed sulfonylureas had the lowest adjusted switching risk and the highest adjusted intensification risk.

  • Younger age group and female sex were associated with higher risks of all modification outcomes.

  • Findings highlight the need to better understand barriers to medication adherence among patients who discontinued treatment to maximize adherence and reduce health system waste.


_____


Recent research findings have identified metformin as the initial medication choice for 77% of patients starting pharmacotherapy for type 2 diabetes,1 reflecting treatment patterns largely in accordance with clinical guidelines. However, to achieve long-term glycemic control, most patients ultimately require combination therapy with other antidiabetic medication (ADM) classes.2 Options for initial second-line ADM after failure of metformin monotherapy are increasing rapidly, with multiple therapeutic classes demonstrating effective glycemic control and cardiovascular and renal benefits.3-7 However, there is no clinical consensus about which ADM should be used as initial second-line pharmacotherapy in combination with metformin.


Currently, the following 5 noninsulin ADM classes are available when metformin monotherapy is no longer sufficient for glycemic control: sulfonylureas/meglitinides (henceforth, sulfonylureas), dipeptidyl peptidase 4 inhibitors (DPP4is), sodium-glucose cotransporter 2 inhibitors (SGLT2is), glucagon-like peptide-1 receptor agonists (GLP-1 RAs), and thiazolidinediones (TZDs). Practice guidelines direct clinicians to consider multiple factors when selecting a second-line ADM, including the drug’s effects on glycemic control,8 cardiovascular events,4,9 renal disease progression,10 and body weight.11 Patients’ preferences based on potential benefits, adverse effects, and medication costs should also be considered.


There are limited data describing patterns of ADM use after failure of metformin monotherapy. One recent national study of adults with type 2 diabetes found that approximately half of patients receiving metformin monotherapy required second-line treatment within 3.4 years, with sulfonylureas being the most common choice.12 National data on US adults with type 2 diabetes demonstrate the following consistent trends in second-line ADM use over the past 20 years: (1) a large proportional decrease in TZD use, (2) a small reduction in sulfonylurea use, and (3) increases in use of newer medication classes (ie, DPP4is, GLP-1 RAs, and SGLT2is) after introduction to market.12-14 Although prior research has examined some ADM changes after initiation of second-line therapy,1,12,15-17 this literature has not examined all types of medication changes from patients’ and clinicians’ perspectives, namely discontinuing the second-line ADM (at the behest of either patients or clinicians), switching to a new ADM (after a prescribing clinician’s decision to change the pharmacotherapy regimen because of factors like adverse effects or inadequate glycemic control), and intensifying the ADM regimen (presumably due to inadequate glycemic control). Additionally, prior research was not designed to examine predictors of these medication changes. Finally, earlier studies were largely confined to the years before the introduction of newer ADM classes such as SGLT2is.


The current study evaluated data from the period after SGLT2i introduction to address these gaps in the evidence base. Our aims were to (1) describe changes in medication use experienced by patients who initiated different classes of noninsulin second-line ADM therapy and (2) identify patient and provider characteristics that are independently associated with changes in medication use after initiation of second-line therapy.


METHODS

Study Design and Data

This was a retrospective cohort study comparing modifications in diabetes treatment regimen among adults with type 2 diabetes who were initially prescribed 1 of the 5 classes of noninsulin second-line ADMs. As done previously,13,18,19 our study team used nationwide sources of administrative claims data for privately insured US adults with type 2 diabetes. Data sources included health plan enrollment files, medical inpatient and ambulatory claims, and pharmacy claims between July 2013 and December 2017. Our use of coded, nonidentifiable data was deemed non–human subjects research by Northwestern University’s institutional review board.


Study Setting and Inclusion Criteria

The study included adults (18 years or older) who were enrolled in a commercial or Medicare Advantage health plan and initiated an index, second-line ADM prescription between January 1, 2014, and June 30, 2017. Included patients were required to have continuous health plan enrollment for at least 6 months before and 6 months after the date of index medication initiation (ie, the index date). During this 12-month period of continuous enrollment, patients were required to have no change in health plan insurance segment (commercial vs Medicare Advantage) or health plan structure (health maintenance organization [HMO], preferred provider organization [PPO], etc). Each patient’s follow-up period lasted up to 12 months after the index date. The eAppendix Figure (eAppendix available at ajmc.com) provides additional details on dates of data collection and identification of study periods.


Included patients had preexisting type 2 diabetes, defined as 2 or more outpatient diagnosis codes or 1 or more inpatient diagnosis codes for type 2 diabetes at any point before the index date (extending back as far as January 1, 2011). Included patients had at least 1 pharmacy claim for metformin during the 6 months before the index date and a pharmacy claim on the index date for at least a 28-day supply of 1 included index medication.

Patients were excluded if they had 2 or more diagnosis codes for type 1 diabetes, any claims for a nonmetformin ADM before the index date (extending back as far as January 1, 2011), or a pharmacy claim for 2 or more nonmetformin ADMs on the index date. Patients missing data for any variables used in regression analysis were also excluded.


Treatment Modification Outcomes

Study outcomes were defined as time-to-event data to be evaluated in Cox proportional hazards regression models. All included patients were followed after their index dates until experiencing 1 of 4 diabetes treatment modification outcomes: no treatment modification for up to 12 months; discontinuation of the index ADM without replacement with another ADM; switching (discontinuation of the index ADM with replacement by another ADM class); or intensification. We only evaluated each patient’s first observed treatment modification; any subsequent modifications were not considered during analysis.


Patients were classified as having no treatment modification if they consistently obtained prescription refills from the index medication class and did not initiate medications from any other ADM classes throughout the follow-up period. As done previously,20,21 we defined a prescription refill as a pharmacy claim for a subsequent fill in the same ADM class within 60 days after the prior fill was exhausted (based on calculated days’ supply). Discontinuation was defined as times when a prescription was exhausted and the patient did not refill a medication from the index medication class or initiate a new medication from another ADM class within 60 days. Switching was defined as discontinuation of the index medication and—within 60 days of exhausting the final index medication fill—initiation of a medication from another index medication class.


Patients met the definition of treatment intensification in 3 ways: (1) doubling (or more) the dose of their index medication from one fill to the next; only patients taking the 4 index medication classes besides GLP-1 RAs—for which patients often titrate upward to the maximum tolerable dose22—were eligible for this form of intensification; (2) starting another index medication class while continuing to refill the index medication; or (3) initiating basal insulin (regardless of whether the index medication continued to be refilled).


Predictor and Covariate Measures

We collected several predictor and covariate measures for each patient at the time of the index date. The primary predictor was a 5-category measure of index medication class (sulfonylureas were the referent group in regression analyses). We also collected measures on patient demographics including age, sex, and US Census region. Because provider characteristics can influence the choice of index medication,13 we collected data on the type of clinician who prescribed the index medication. Measures of health plan characteristics included insurance segment, consumer-driven health plan (CDHP) enrollment (ie, having a commercial plan with a health reimbursement arrangement or health savings account), and health plan structure. Diabetes-related comorbidity was measured using the Diabetes Complications Severity Index (DCSI) score.23 For patients with available hemoglobin A1c (HbA1c) data, we evaluated glycemic control prior to the index date using the most recent HbA1c result from the 6-month period preceding the index date (the insurer whose claims were evaluated obtained HbA1c results through data sharing agreements with several individual laboratory vendors; however, HbA1c results were not available for patients whose HbA1c testing was done at laboratories that did not share data with the insurer). To adjust for temporal patterns over time, we also collected the year of the index date.


Statistical Analysis

Statistical analyses were conducted using SAS Viya software (SAS Institute Inc) with α = .05. We computed descriptive statistics for baseline characteristics and unadjusted rates of each treatment modification outcome during follow-up.


In adjusted analysis to evaluate differences in the time to each outcome, we estimated multivariable Cox proportional hazards models that account for competing risks. We estimated 3 regression models, 1 for each of the 3 treatment modification outcomes. Each model adjusted for all covariates outlined earlier (with a composite indicator for insurance segment and CDHP enrollment). Our censoring approach followed the cause-specific proportional hazards approach to competing risks analysis.24,25 Patients were censored at the time of any observed treatment modification, with modifications besides each modeled outcome treated as competing risks (eg, in the model for the discontinuation outcome, patients were censored at the time of switching or intensification). Patients were also censored at the time of health plan disenrollment, any change in insurance segment or health plan structure, or the end of 12-month follow-up.


When reviewing the timing of observed treatment modifications, we identified 1192 patients (1.4% of all included patients) who switched or intensified treatment within 1 week of their index date. Because administrative claims data have inherent limitations that precluded us from identifying the reasons for these rapid treatment modifications, we also conducted a sensitivity analysis omitting these 1192 patients.


RESULTS

A total of 82,624 adults with type 2 diabetes met all inclusion criteria (the eAppendix Table presents detailed data on sample size after individual inclusion and exclusion criteria were applied). As shown in Table 1, approximately half of included patients received a sulfonylurea as their index medication (42,118; 51.0%); the next most common index medication classes were DPP4is (19,830; 24.0%), SGLT2is (9624; 11.6%), GLP-1 RAs (6707; 8.1%), and TZDs (4345; 5.3%). Most of the included patients were men (54.0%), had a non-Medicare commercial insurance plan (57.0%), and had no diabetes complications (61.5%). Pluralities of patients had their index medication prescribed by a family medicine physician (42.1%), had a point-of-service health plan (which combines elements of HMO and PPO plans) (40.7%), or were aged 65 to 74 years (28.2%). HbA1c data were available for a total of 43,457 patients (52.6%).


The Figure presents data on treatment modification outcomes, defined as each patient’s first observed treatment modification. As shown in the figure, 63.6% of all patients experienced a treatment modification during follow-up. Discontinuation was by far the most common modification (38.6%). Approximately 1 in 20 (5.2%) patients switched their ADM class, and approximately 1 in 5 (19.8%) intensified treatment. Only 36.4% of included patients did not experience any treatment modifications. By a wide margin, discontinuation was most common among patients initially prescribed GLP-1 RAs (50.3%); for other index medication classes, the proportion of patients who discontinued ranged from 34.2% (TZDs) to 39.5% (DPP4is). Switching was most common among patients initially prescribed DPP4is (8.7%) and least common among patients initially prescribed sulfonylureas (3.0%). Intensification was most common among patients initially prescribed sulfonylureas (23.0%).


Table 2 presents detailed unadjusted results on treatment switching and intensification. Among the 4342 patients who switched ADM classes, switching to a sulfonylurea (1429; 32.9%) was most common, particularly among the 1729 patients who switched after an initial DPP4i prescription (978; 56.6%). Switching patterns varied across other classes; for example, patients with an SGLT2i index medication who switched ADM classes (669) most frequently switched to DPP4is (267; 39.9%). Among the 16,327 patients who experienced 1 of the 3 forms of treatment intensification, the majority intensified by increasing their index medication dose (8559; 52.4%); this form of intensification was especially common among the 9679 participants initially prescribed sulfonylureas who intensified treatment (6602; 68.2%). There were 2112 patients who intensified treatment by initiating basal insulin (12.9%); this form of intensification was proportionally least common among the 1924 patients initially prescribed SGLT2is who intensified treatment (177; 9.2%).

In multivariable regression analysis using multivariable Cox proportional hazards models, some index medication classes had a higher adjusted risk of treatment discontinuation than sulfonylureas (Table 3). Compared with patients initially prescribed sulfonylureas, the adjusted risk of discontinuation was 7% higher (HR, 1.07; 95% CI, 1.04-1.10) among patients initially prescribed DPP4is and 28% higher (HR, 1.28; 95% CI, 1.23-1.33) among patients initially prescribed GLP-1 RAs. There was no adjusted difference in discontinuation between those initially prescribed sulfonylureas, SGLT2is, or TZDs.


Patients initially prescribed sulfonylureas had the lowest adjusted risk of switching and the highest adjusted risk of intensification. Among other index medication classes, the adjusted risk of switching ranged from being 73% greater for GLP-1 RAs (HR, 1.73; 95% CI, 1.54-1.95) to 188% greater for DPP4is (HR, 2.88; 95% CI, 2.68-3.10) compared with sulfonylureas. Conversely, the adjusted risk of intensification ranged from being 8% lower for TZDs (HR, 0.92; 95% CI, 0.86-0.99) to 60% lower for GLP-1 RAs (HR, 0.40; 95% CI, 0.37-0.44) compared with sulfonylureas. In sensitivity analyses omitting patients who switched or intensified within 1 week of the index date, inference for each index medication class was identical (results available from authors upon request).


After multivariable adjustment, several covariates were independently associated with treatment modification (Table 3). Compared with all older age groups, patients aged 18 to 44 years were at significantly higher risk of discontinuation, switching, and intensification. Men were at significantly lower risk of each modification than women. The type of prescribing clinician was associated with multiple modifications; for example, compared with having the index medication prescribed by either a family medicine physician or internal medicine physician, prescription by an endocrinologist was associated with both lower discontinuation risk and higher intensification risk. Higher DCSI scores were associated with incrementally greater intensification risk, and higher recent HbA1c levels were associated with incrementally higher risks of both switching and intensification.


DISCUSSION

In this retrospective cohort study of more than 80,000 privately insured patients who initiated a second-line ADM, nearly two-thirds of included patients modified treatment within 1 year of their index medication fill. Discontinuation was by far the most common modification (38.6%), followed by intensification (19.8%) and switching (5.2%). Patients initially prescribed GLP-1 RAs had the highest observed risk of discontinuation. Patients initially prescribed sulfonylureas had the lowest adjusted risk of switching and the highest adjusted risk of intensification (largely via increased dose of their sulfonylurea). Several variables were independently associated with treatment modifications; for example, being in a younger age group and female were both associated with higher risks of all 3 modification outcomes. Also, compared with index medication prescription by a family medicine or internal medicine physician, prescription by an endocrinologist was associated with lower discontinuation risk and higher intensification risk.


Consistent with the literature, discontinuation of second-line ADM treatment was relatively common in patients with type 2 diabetes. The particularly high discontinuation rate for GLP-1 RAs was likely due to previously observed factors such as gastrointestinal adverse effects.26 Although our observed rates of discontinuation exceed those from a national study of outpatient electronic health record data between 1995 and 2016—in which discontinuation ranged between 10% for patients prescribed sulfonylureas and 21% for patients prescribed GLP-1 RAs and SGLT2is12—they are similar to those of a study of patients in managed care plans, 35% of whom discontinued ADM therapy between 2006 and 2009.27


The observed persistence of sulfonylurea therapy—patients with a sulfonylurea index medication rarely switched to a different ADM class or initiated a second index medication class—was notable. This finding may be indicative of elevated price sensitivity (for co-pays) among patients with a sulfonylurea index medication, who were somewhat more likely than the overall study population to be 65 years or older or to have a Medicare Advantage plan. Price sensitivity among these patients would generally align with findings from a prior study that found the high out-of-pocket costs of brand-name ADMs were associated with medication nonadherence and discontinuation in Medicare patients.20


From the perspective of health plan design and administration, this study’s outcomes highlight the need to better understand barriers to ADM adherence in the alarmingly large number of patients who discontinued second-line therapy. When patients discontinue a second-line ADM soon after treatment initiation, both resources (among health plans and patients) and time (among patients and clinicians) are wasted. Additionally, medication nonadherence in diabetes increases the likelihood of subsequent inpatient admissions28-30 and avoidable health care costs,28-31 underscoring the need for interventions that promote informed prescribing practices and postprescription patient supports that maximize medication adherence after initiation of second-line ADM therapy. Research is especially needed to examine patients’ reasons for discontinuation and the extent to which patients’ discontinuation decisions were made without—or in conflict with—physician feedback. Research and intervention development efforts should prioritize groups shown here to be at high risk of discontinuation, including patients prescribed GLP-1 RAs, relatively younger patients, women, and patients with CDHPs.


This study had several limitations. Patients who switched medications within the same index medication class or gradually increased their index medication dose over time (which could eventually lead to doubling of their initial dose) were classified as having no treatment modifications. Claims data do not provide contextual information on patients’ or clinicians’ decision-making processes, thus precluding investigation of underlying reasons for observed treatment modifications. Also, the study did not include patients who were prescribed basal insulin as a second-line ADM.32 Because it is often not possible to properly estimate days’ supply of insulin,33 we were unable to accurately calculate treatment modification outcomes for patients taking basal insulin. Also, prior research by our team identified distinct characteristics among patients prescribed basal insulin, including elevated HbA1c and high health care costs,13,19 compared with patients prescribed noninsulin second-line ADMs; assuming some unique characteristics of these patients remain unmeasured, including them in the current study would introduce selection bias into relevant statistical comparisons. Finally, our use of a 60-day threshold—after the prior fill was exhausted—for defining treatment modification outcomes represented a middle ground between a more stringent 30-day threshold34 and a more permissive 90-day threshold35 that have been used in prior medication adherence studies. A threshold of a different length would have led to different observed proportions of each treatment modification outcome.


This study also had notable strengths. To our knowledge, this was the first large study of treatment modifications for second-line ADM therapy that solely focused on the years after the 2013 introduction of SGLT2is. The large data set provided evidence on nationwide patterns of treatment modifications in privately insured populations, facilitating comprehensive comparisons across multiple classes of second-line ADM therapy. Detailed pharmacy claims data allowed for construction of valid outcome measures, with clinical and sociodemographic data leveraged to identify patient and provider characteristics that were associated with the treatment modifications under study.


CONCLUSIONS

Most included patients modified their type 2 diabetes medication regimen within 1 year of initiating a noninsulin second-line ADM, most frequently via discontinuation of their index medication. Future research is needed to improve our understanding of barriers to ADM adherence among patients who discontinue second-line therapy and to test new prescribing approaches and patient supports to maximize medication adherence and reduce health system waste among patients who initiate second-line ADM therapy.


Acknowledgments

The authors thank Sanya Pasricha, UnitedHealth Group, for help with preparation of manuscript tables and figures. The authors thank Sterling Harris, Northwestern University, for project management support. Interim study findings were presented virtually at the American Diabetes Association Scientific Sessions on June 13, 2020.

Funding support and data access for this work were provided through a grant to Northwestern University from UnitedHealth Group, the company whose insurance arm provided coverage to the members under study.


Author Affiliations: Division of General Internal Medicine (DTL, MC, MJO, CA, AJC, RTA), Institute for Public Health and Medicine (DTL, MJO, RHK, CA, AW, RTA), and Division of Endocrinology, Metabolism, and Molecular Medicine (AW, RTA), Northwestern University Feinberg School of Medicine, Chicago, IL; Applied Research, UnitedHealth Group (EK, EDP), Minneapolis, MN.


Source of Funding: Funding support for this work was provided through a grant to Northwestern University from UnitedHealth Group.


Author Disclosures: Dr Liss’ work on the current manuscript was supported through a grant from UnitedHealth Group. UnitedHealth Group includes the insurer that paid the health care claims of patients in this study and, as such, has a financial interest in the article’s subject matter. Dr Wallia has received grants from Novo Nordisk and Eli Lilly and Company and research support from UnitedHealth Group. Ms Koep and Dr Parker are employed by and own stock in UnitedHealth Group, which provided Northwestern University with access to the data for this research. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.


Authorship Information: Concept and design (DTL, MJO, RHK, AW, EDP, RTA); acquisition of data (AJC, CA, EK, EDP, RTA); analysis and interpretation of data (DTL, MC, MJO, RHK, AW, AJC, EK, EDP, RTA); drafting of the manuscript (DTL, MC, EDP); critical revision of the manuscript for important intellectual content (DTL, MJO, RHK, AW, EDP, RTA); statistical analysis (DTL, MC, RHK, AJC); obtaining funding (CA); administrative, technical, or logistic support (AJC, CA, EK); and supervision (DTL).


Address Correspondence to: David T. Liss, PhD, Division of General Internal Medicine, Northwestern University Feinberg School of Medicine, 750 N Lake Shore Dr, 10th Floor, Chicago, IL 60611. Email: david.liss@northwestern.edu.


REFERENCES

1. Bonafede M, Chandran A, DiMario S, Saltiel-Berzin R, Saliu D. Medication usage, treatment intensification, and medical cost in patients with type 2 diabetes: a retrospective database study. BMJ Open Diabetes Res Care. 2016;4(1):e000189. doi:10.1136/bmjdrc-2015-000189

2. Bailey T. Options for combination therapy in type 2 diabetes: comparison of the ADA/EASD position statement and AACE/ACE algorithm. Am J Med. 2013;126(9 suppl 1):S10-S20. doi:10.1016/j.amjmed.2013.06.009

3. Zinman B, Wanner C, Lachin JM, et al; EMPA-REG OUTCOME Investigators. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015;373(22):2117-2128. doi:10.1056/NEJMoa1504720

4. Marso SP, Bain SC, Consoli A, et al; SUSTAIN-6 Investigators. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375(19):1834-1844. doi:10.1056/NEJMoa1607141

5. 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

6. Husain M, Birkenfeld AL, Donsmark M, et al; PIONEER 6 Investigators. Oral semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2019;381(9):841-851. doi:10.1056/NEJMoa1901118

7. Gerstein HC, Colhoun HM, Dagenais GR, et al; REWIND Investigators. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet. 2019;394(10193):121-130. doi:10.1016/S0140-6736(19)31149-3

8. Qaseem A, Barry MJ, Humphrey LL, et al; Clinical Guidelines Committee of the American College of Physicians. Oral pharmacologic treatment of type 2 diabetes mellitus: a clinical practice guideline update from the American College of Physicians. Ann Intern Med. 2017;166(4):279-290. doi:10.7326/M16-1860

9. Fitchett D, Inzucchi SE, Cannon CP, et al. Empagliflozin reduced mortality and hospitalization for heart failure across the spectrum of cardiovascular risk in the EMPA-REG OUTCOME trial. Circulation. 2019;139(11):1384-1395. doi:10.1161/CIRCULATIONAHA.118.037778

10. Coppolino G, Leporini C, Rivoli L, et al. Exploring the effects of DPP-4 inhibitors on the kidney from the bench to clinical trials. Pharmacol Res. 2018;129:274-294. doi:10.1016/j.phrs.2017.12.001

11. Pratley RE, Aroda VR, Lingvay I, et al; SUSTAIN 7 Investigators. Semaglutide versus dulaglutide once weekly in patients with type 2 diabetes (SUSTAIN 7): a randomised, open-label, phase 3b trial. Lancet Diabetes Endocrinol. 2018;6(4):275-286. doi:10.1016/S2213-8587(18)30024-X

12. Montvida O, Shaw J, Atherton JJ, Stringer F, Paul SK. Long-term trends in antidiabetes drug usage in the U.S.: real-world evidence in patients newly diagnosed with type 2 diabetes. Diabetes Care. 2018;41(1):69-78. doi:10.2337/dc17-1414

13. 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

14. Le P, Chaitoff A, Misra-Hebert AD, Ye W, Herman WH, Rothberg MB. Use of antihyperglycemic medications in U.S. adults: an analysis of the National Health and Nutrition Examination Survey. Diabetes Care. 2020;43(6):1227-1233. doi:10.2337/dc19-2424

15. Montvida O, Shaw JE, Blonde L, Paul SK. Long-term sustainability of glycaemic achievements with second-line antidiabetic therapies in patients with type 2 diabetes: a real-world study. Diabetes Obes Metab. 2018;20(7):1722-1731. doi:10.1111/dom.13288

16. Koye DN, Montvida O, Paul SK. Third-line antidiabetic therapy intensification patterns and glycaemic control in patients with type 2 diabetes in the USA: a real-world study. Drugs. 2020;80(5):477-487. doi:10.1007/s40265-020-01279-y

17. Flory JH, Li J, Dawwas GK, Leonard CE. Compared to commercially insured patients, Medicare Advantage patients adopt newer diabetes drugs more slowly and adhere to them less. Endocrinol Diabetes Metab. 2021;4(3):e00245. doi:10.1002/edm2.245

18. O’Brien MJ, Karam SL, Wallia A, et al. Association of second-line antidiabetic medications with cardiovascular events among insured adults with type 2 diabetes. JAMA Netw Open. 2018;1(8):e186125. doi:10.1001/jamanetworkopen.2018.6125

19. Liss DT, Kang RH, Lancki N, et al. Costs for commercially insured adults prescribed second-line diabetes medications. Am J Manag Care. 2021;27(3):e72-e79. doi:10.37765/ajmc.2021.88601

20. Pawaskar MD, Xu L, Tang Y, Puckrein GA, Rajpathak SN, Stuart B. Effect of medication copayment on adherence and discontinuation in Medicare beneficiaries with type 2 diabetes: a retrospective administrative claims database analysis. Diabetes Ther. 2018;9(5):1979-1993. doi:10.1007/s13300-018-0489-y

21. Peterson AM, Nau DP, Cramer JA, Benner J, Gwadry-Sridhar F, Nichol M. A checklist for medication compliance and persistence studies using retrospective databases. Value Health. 2007;10(1):3-12. doi:10.1111/j.1524-4733.2006.00139.x

22. Prasad-Reddy L, Isaacs D. A clinical review of GLP-1 receptor agonists: efficacy and safety in diabetes and beyond. Drugs Context. 2015;4:212283. doi:10.7573/dic.212283

23. 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

24. Lau B, Cole SR, Gange SJ. Competing risk regression models for epidemiologic data. Am J Epidemiol. 2009;170(2):244-256. doi:10.1093/aje/kwp107

25. Austin PC, Lee DS, Fine JP. Introduction to the analysis of survival data in the presence of competing risks. Circulation. 2016;133(6):601-609. doi:10.1161/CIRCULATIONAHA.115.017719

26. Sikirica MV, Martin AA, Wood R, Leith A, Piercy J, Higgins V. Reasons for discontinuation of GLP1 receptor agonists: data from a real-world cross-sectional survey of physicians and their patients with type 2 diabetes. Diabetes Metab Syndr Obes. 2017;10:403-412. doi:10.2147/DMSO.S141235

27. Pawaskar M, Bonafede M, Johnson B, Fowler R, Lenhart G, Hoogwerf B. Medication utilization patterns among type 2 diabetes patients initiating exenatide BID or insulin glargine: a retrospective database study. BMC Endocr Disord. 2013;13:20. doi:10.1186/1472-6823-13-20

28. Sokol MC, McGuigan KA, Verbrugge RR, Epstein RS. Impact of medication adherence on hospitalization risk and healthcare cost. Med Care. 2005;43(6):521-530. doi:10.1097/01.mlr.0000163641.86870.af

29. Curtis SE, Boye KS, Lage MJ, Garcia-Perez LE. Medication adherence and improved outcomes among patients with type 2 diabetes. Am J Manag Care. 2017;23(7):e208-e214.

30. Jha AK, Aubert RE, Yao J, Teagarden JR, Epstein RS. Greater adherence to diabetes drugs is linked to less hospital use and could save nearly $5 billion annually. Health Aff (Millwood). 2012;31(8):1836-1846. doi:10.1377/hlthaff.2011.1198

31. Gentil L, Vasiliadis HM, Préville M, Berbiche D. Adherence to oral antihyperglycemic agents among older adults with mental disorders and its effect on health care costs, Quebec, Canada, 2005-2008. Prev Chronic Dis. 2015;12:E230. doi:10.5888/pcd12.150412

32. Nathan DM, Buse JB, Kahn SE, et al; GRADE Study Research Group. Rationale and design of the glycemia reduction approaches in diabetes: a comparative effectiveness study (GRADE). Diabetes Care. 2013;36(8):2254-2261. doi:10.2337/dc13-0356

33. Stolpe S, Kroes MA, Webb N, Wisniewski T. A systematic review of insulin adherence measures in patients with diabetes. J Manag Care Spec Pharm. 2016;22(11):1224-1246. doi:10.18553/jmcp.2016.22.11.1224

34. Luo J, Feldman R, Rothenberger S, Korytkowski M, Fischer MA, Gellad WF. Incidence and predictors of primary nonadherence to sodium glucose co-transporter 2 inhibitors and glucagon-like peptide 1 agonists in a large integrated healthcare system. J Gen Intern Med. 2022;37(14):3562-3569. doi:10.1007/s11606-021-07331-1

35. Karter AJ, Parker MM, Solomon MD, et al. Effect of out-of-pocket cost on medication initiation, adherence, and persistence among patients with type 2 diabetes: the Diabetes Study of Northern California (DISTANCE). Health Serv Res. 2018;53(2):1227-1247. doi:10.1111/1475-6773.12700


Article Information

Published Online: January 3, 2024. doi:10.1001/jama.2023.26110

1 view

Commentaires


bottom of page