Costs Associated With Long-Acting Insulin Analogues in Patients With Diabetes

July 11, 2018
Berhanu Alemayehu, DrPH, MS

,
Jessica Speiser, MPH

,
Lisa Bloudek, PharmD, MS

,
Evelyn Sarnes, PharmD, MPH

Volume 24, Issue SP 8

Findings from a literature review indicate that overall costs of long-acting insulin analogues are not significantly different from those of intermediate-acting human insulin and oral antidiabetic agents.

ABSTRACTObjectives: The objective of this literature review was to evaluate the costs associated with the use of long-acting insulin analogues (LAIAs) compared with non-LAIA agents, including human insulin, oral antidiabetic drugs, and other injectable therapies, in the treatment of patients with type 1 diabetes (T1D) or type 2 diabetes (T2D).

Study Design: A systematic review of the medical literature (MEDLINE, EMBASE, Cochrane, EconLit) conducted from 2004 to 2016.

Methods: The review protocol was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Inclusion criteria for studies were: patients with T1D and/or T2D, LAIA intervention, and comparators, including oral antidiabetics (OADs) or neutral protamine Hagedorn (NPH). Outcomes of interest were adherence measures; economic outcomes, including total costs, cost savings, and willingness-to-pay; and cost-effectiveness or incremental cost-effectiveness analyses. Real-world costs of individual LAIAs were also evaluated and are often compared against those of other LAIAs in the economic analyses.

Results: We identified and included 117 relevant studies. Patients using LAIAs had higher drug costs than those using OADs and NPH but had neutral or reduced total and diabetes-related costs compared with patients using non-LAIAs. Use of LAIA pen-delivery systems may lead to improved adherence and reduction in costs. Patients receiving insulin glargine demonstrated higher adherence and persistence than patients on insulin detemir. Economic models suggest that LAIAs are more cost-effective than NPH for T1D; for T2D, insulin glargine is more costly than NPH but less so than insulin detemir.

Conclusions: Despite higher drug costs, the real-world overall medical costs of LAIAs are not significantly different from those of NPH in patients with diabetes. The findings may be helpful for formulary decision making for patients with diabetes in a cost-constrained environment.

Am J Manag Care. 2018;24(Spec Issue No. 8):SP265-SP272Takeaway Points

Long-acting insulin analogues (LAIAs) are a potentially cost-effective treatment choice for patients with type 1 or type 2 diabetes compared with neutral protamine Hagedorn (NPH) and oral antidiabetic agents.

  • Despite higher drug acquisition costs, use of LAIAs is associated with neutral or reduced overall diabetes-related direct healthcare costs compared with alternative intermediate-acting human insulin, NPH.
  • Limited real-world evidence suggests that use of LAIA pen-delivery systems may improve patient adherence and reduce the frequency of hypoglycemic events.
  • Formulary decision makers may use these findings to help guide the management of patients with diabetes in a cost-constrained environment.

Diabetes is among the most common chronic medical conditions worldwide, with 422 million people affected.1 Early treatment options for the replacement of basal insulin in patients have included intermediate-acting insulins, such as neutral protamine Hagedorn (NPH). Although they are effective in the management of hyperglycemia, these agents require 2 or more daily injections, often in combination with other therapies, including mealtime insulin. Since their introduction in the early 2000s, long-acting insulin analogues (LAIAs) have become a mainstay in the long-term management of patients with type 1 diabetes (T1D) or type 2 diabetes (T2D). In contrast to intermediate-acting insulins, which generally reach the bloodstream approximately 1 to 2 hours after injection, peak 4 to 12 hours later, and are effective for about 12 to 18 hours, LAIAs reach the bloodstream up to 2 hours after injection and tend to lower glucose levels fairly evenly over a 24-hour period.2 This combination of delayed absorption and fewer peaks reduces the potential for hypoglycemic episodes3; the improved pharmacokinetic properties of the LAIAs also provide greater flexibility with regard to meals and exercise and allow for less frequent administration.3 As such, LAIAs may result in improved glycemic control and higher quality of life and treatment satisfaction, with fewer diabetic complications than with NPH, non-LAIA oral diabetic agents, and short-acting insulin.4-6 Currently, the LAIAs approved in the United States are insulin glargine (eg, Lantus, Toujeo, Abasaglar), insulin detemir (Levemir), and insulin degludec (Tresiba). LAIAs are available in multiple formulations, including prefilled pen devices and vials.

As the treatment of diabetes has become more intensive and complex, drug costs have become a substantial and increasing part of total healthcare costs. From a US payer perspective, a challenge with the increasing utilization of LAIAs within the diabetic population is the higher drug acquisition cost compared with other insulin products; analyses of retrospective data from different countries demonstrate consistently higher annual drug costs for LAIAs compared with oral antidiabetics (OADs) and with shorter-acting insulins, such as NPH.7,8 However, increased drug acquisition cost is only 1 aspect of the overall cost of care for patients treated with LAIAs. The total costs of LAIAs must also be considered when comparing them with alternative regimens; for example, the use of these may include the impact of LAIAs on overall diabetic treatment costs, direct medical costs, and consequent health outcomes, such as improvements in goal attainment, risk of chronic complications, and associated ambulatory visits and hospitalizations.

Currently available LAIAs have been studied extensively over the last 10 years, resulting in a substantial body of clinical and health economic research. However, the real-world costs of LAIAs have not been definitively examined; although a large number of studies have been undertaken, there is considerable variation in their findings. Information on the real-world costs of LAIAs, which are intertwined with clinical effectiveness, are important to healthcare providers and drug formulary decision makers in evaluating the appropriate treatment strategy for each patient with T1D and T2D.

Therefore, the present investigation evaluated the real-world costs of LAIAs in patients with diabetes through a systematic medical literature review of published interventional and observational studies. The key research questions the review sought to answer were: What economic end points have been included in randomized controlled trials with LAIAs? What economic evaluations and associated economic value have been published for LAIAs? What are the real-world health outcomes associated with LAIAs? What are the real-world adherence rates associated with LAIAs?

METHODS

A systematic review of the literature evaluating the use of LAIAs in patients with T1D and T2D was conducted from January 2004 to November 2014, with a subsequent update completed in 2016 to include literature published between August 2014 and October 2016. Database sources included MEDLINE (via PubMed), EMBASE, the Cochrane Library, and EconLit. The research questions for this study were part of a broader systematic review. Publications were initially identified for this broader review using a search protocol following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. From these, the PICOS (Population, Intervention, Comparators, Outcomes, and Study design) criteria9 were used to identify studies that were relevant to the research questions in this review; a brief summary of the search strategy is shown in Table 1. Patient populations included adults with T2D and patients 6 years and older diagnosed with T1D treated with 1 or more of the following LAIAs: insulin glargine 100 units/mL (U100) (Lantus), insulin glargine 300 units/mL (U300) (Toujeo), biosimilar insulin glargine (Basaglar, Abasria/Abasaglar), insulin detemir (Levemir), and insulin degludec (Tresiba). Suitable comparators for inclusion in the review included a non-LAIA treatment for diabetes, such as an OAD or NPH, or a different LAIA as comparator. Outcomes of interest included economic outcomes, including total costs, event costs, costs avoided, cost savings, willingness-to-pay assessments, cost-effectiveness, or incremental cost-effectiveness (including cost per quality-adjusted life-year [QALY]), as well as adherence outcomes, such as medication possession ratio (MPR), persistence, or discontinuation rate. Study designs of interest were controlled clinical trials, observational studies, retrospective analyses, economic models, health technology assessments, and systematic review articles reporting adherence to or economic profiles of LAIAs.

Only English-language studies were included. Unpublished studies (including conference abstracts), nonsystematic reviews, editorials, commentaries, and letters were excluded. Nonhuman studies, theses, dissertations, and pharmacokinetic/pharmaco­dynamic studies were also excluded. Additional exclusion criteria are described in Table 1.

RESULTS

The number of unique, potentially relevant publications identified and screened for retrieval as part of the broader systematic review was 1281, of which 230 met the eligibility for extraction. A further 23 studies were added following post hoc searching and validation, producing a total of 253 articles. From these 253, 105 studies were identified using PICOS criteria as relevant to the research questions of interest in this study (eAppendix Figure [eAppendix available at ajmc.com]). This search protocol was rerun in October 2016, and after applying the same criteria, an additional 17 studies were identified and included, for a total of 122 relevant studies (Figure). The majority of these studies were conducted in the T2D population (99 studies); 18 studies were conducted in a T1D population, and 5 were conducted in a mixed T1D and T2D population. The remainder of this section focuses on the studies conducted in either a T1D or T2D population (117 studies).

Economic Impact of LAIAs in T1D

Economic outcomes. Although the evidence base is limited, LAIAs were associated with decreased medical costs in patients with T1D, particularly in the costs of diabetic testing supplies. Three studies evaluated healthcare costs and utilization in patients with T1D initiating LAIAs. Compared with NPH, insulin glargine was associated with lower treatment costs, including costs of insulin and diabetic testing supplies (test strips and lancets), but the difference was not statistically significant.10 Insulin glargine was associated with significantly lower medical costs compared with insulin detemir (£1198 vs £1330, respectively; P <.001), driven by reduced costs for insulin, reagents, hypoglycemia rescue medication, sharps, and pen devices.11 Insulin glargine administered via vial/syringe was associated with a yearly total cost of $4900 compared with $9373 when administered via an insulin pump.12

Economic evaluations of LAIAs. When compared in traditional health economic analyses, the available evidence consistently supported that LAIAs were more cost-effective and should be used in lieu of standard insulin therapy. However, there is uncertainty with respect to selection of one LAIA over another, as small differences in benefits were imbalanced with substantial difference in costs.

Fifteen economic models were identified that evaluated the economic impact of introducing LAIAs into the T1D treatment paradigm (key findings are summarized in Table 2). Although data to support a specific LAIA for T1D are lacking, there was reasonably consistent evidence to support that LAIAs should be used in lieu of NPH because, despite greater drug costs, there was an improved safety profile relative to NPH, especially in patients with a high risk of hypoglycemia. Insulin glargine or insulin detemir were found to be the dominant treatment strategies in 3 of the 10 studies comparing insulin glargine or insulin detemir with NPH, but they also demonstrated incremental cost-effectiveness ratios of $73,266 per QALY and $323,062 per QALY, respectively, in other studies, indicative of high variability.

Adherence and persistence with LAIAs. The literature review yielded no studies that assessed adherence and persistence to therapy among patients with T1D.

Economic Impact of LAIAs in T2D

Economic outcomes. Although LAIAs may be associated with higher drug costs compared with alternative pharmacotherapies to treat T2D, the literature supports that the overall cost of care for patients using LAIAs is not significantly different compared with the overall cost of care for patients not using LAIAs.

Thirty-three studies evaluated healthcare costs and utilization in patients with T2D utilizing LAIAs. Table 3 summarizes the impact of treatment with insulin glargine compared with non-LAIAs in patients with T2D on diabetes-related costs and total healthcare costs, including annual pharmacy costs and medical costs (eg, hypoglycemic episodes, hospitalizations, outpatient costs, emergency department [ED] costs, and other direct costs), as identified in the literature review.7,13-20 As shown in Table 3, the results are not consistent; 3 studies demonstrated reduced total healthcare costs, and 2 showed higher total healthcare costs, with insulin glargine compared with non-LAIAs. Among studies comparing insulin glargine with alternate therapies, including OADs, exenatide, NPH, and other insulins, the impact on healthcare costs and utilization varied.7,13-26 In analyses ranging from 6 months to 2 years that evaluated 150 to 6300 patients, overall total healthcare costs and diabetes-related costs were generally higher for individuals receiving insulin glargine compared with those not receiving LAIAs. However, findings often did not reach statistical significance, with similar findings reported for measures of healthcare utilization. No obvious population or geographical bias was identified to explain the variable findings among studies. For example, analyses by Misurski et al17 and Kleinman et al15 were conducted among US adults with T2D who had initiated therapy with insulin glargine or another antidiabetic agent. In the analysis by Misurski et al, the annual total medical cost was $23,782 for patients who initiated insulin glargine (2007 US$), which was significantly higher than for patients who initiated exenatide ($19,293; P <.0001).17 In contrast, Kleinman et al found that in adults who had filled at least 1 insulin prescription, insulin glargine was associated with better adherence and lower total annual medical and pharmacy costs than patients initiating all other insulins ($6771 vs $7969, respectively; P = .0046).15 It is not clear why the annual total costs were so dramatically different among groups.

A small group of studies evaluated economic outcomes in patients with T2D receiving mixed basal insulin.8,27-30 In an analysis of patients receiving any basal insulin compared with an insulin mixture, individuals with documented persistence to basal insulin consistently had fewer diabetes-related events (including inpatient admissions and use of the ED) and cardiovascular-related events.27 Overall annual total costs were lower in patients receiving an LAIA compared with NPH insulin ($5858 vs $6607, respectively; P = .30).8 Although initiating therapy with an LAIA may result in higher up-front drug costs, a UK analysis found that cost differences among insulin glargine, insulin detemir, and NPH insulin were negligible after 2 or 3 years.30

An additional 6 studies directly compared economic outcomes with the use of insulin glargine versus insulin detemir.11,30-34 Results were inconsistent, with 1 British study in patients naïve to basal insulin reporting insulin glargine to have significantly lower costs of treatment, while 1 US study found insulin detemir to have lower costs of diabetes-related healthcare and the results of another US study indicated that insulin detemir had lower insulin costs. The 3 remaining studies (also from the United Kingdom and United States) showed no significant differences. Again, there were no obvious differences in the patient populations or demographics, or clear bias, to explain the variability in results.

Six studies assessed the economic impact of using a pen versus a vial/syringe for LAIA administration in T2D. These studies found that although medication costs were notably increased for patients using pens, overall costs were less or not significantly different.35-40

Economic evaluations of LAIAs. Forty-four economic models were identified that assessed the economic implications of introducing LAIAs into the T2D treatment paradigm, but with seemingly conflicting results (Table 4). Although the overall findings show variable cost savings associated with insulin glargine compared with alternative insulin products, the key benefits associated with LAIAs included convenient dosing with a reduced risk of hypoglycemia.

For patients with T2D, insulin glargine was more costly than NPH but less costly than other LAIAs; as such, the proposed benefits of LAIAs in T2D depend largely on a more convenient treatment paradigm coupled with a reduced risk of hypoglycemia. Although the majority of the models leveraged the existing IQVIA CORE Diabetes Model, helping to establish consistency in the types of costs and outcomes that were analyzed, a great degree of variability in findings was reported, making it challenging to draw clear conclusions. For example, when comparing insulin glargine with NPH in patients with T2D, of 6 studies included, 2 found that insulin glargine was the dominant treatment strategy only among patients with baseline glycated hemoglobin (A1C) of 10% or greater; cost-effectiveness decreased with lower baseline A1C levels. Of 5 studies that conducted analyses comparing insulin glargine with NPH for the overall T2D patient populations, 4 found that insulin glargine was cost-effective, with incremental cost-effectiveness ratios ranging from $6587/QALY to $26,179/QALY, and 1 study found that insulin glargine was not cost-effective at $536,745/QALY (2017 US$). Of these 6 studies overall, only 2 were based on the IMS CORE model, with data coming from 4 countries, and 4 were from payer perspectives. The variability may be explained by differing patient input populations.

Adherence and persistence with LAIAs. Although a substantial number of studies evaluated the impact of LAIAs on adherence and persistence, the findings were mixed. The general trend in the literature suggests that adherence and persistence are improved in patients treated with insulin glargine compared with insulin detemir.

Twenty-two studies were identified in patients with T2D evaluating the impact of LAIAs on medication adherence and persistence over a study period of up to 3 years. Study comparisons included LAIAs versus insulin mixtures or alternate therapies (eg, NPH, exenatide; n = 8); direct comparisons of insulin glargine and insulin detemir (n = 4); a retrospective 4-way comparison among insulin glargine, insulin detemir, NPH insulin, and premixed insulin (n = 1); and a retrospective 3-way comparison among insulin glargine, insulin detemir, and NPH (n = 1). Other studies included a cohort of patients using insulin glargine or insulin detemir (n = 1), patients initiated on combination therapy (exenatide and glargine; n = 2), and patients utilizing different devices with varying basal insulin products (n = 1).

When comparing newly initiated LAIA therapies with non-LAIA therapies in US patients, the reported 1-year adherence (as measured by mean or adjusted MPR) ranged from 0.50 to 0.67 (for reference, an MPR of 1 indicates 100% adherence).7,13,41,42 Persistence (defined as the proportion of patients remaining on the study drug during the follow-up period, without discontinuation or switching after initiation) over 1 year was 54.5% to 55.9% for insulin glargine compared with 43.8% to 45.4% for non-LAIAs,7,13 and the mean number of persistent days for patients newly initiated on LAIAs ranged from 167 (insulin detemir)41 to 284 days (insulin glargine).7

Direct comparisons of insulin glargine and insulin detemir revealed significant differences favoring insulin glargine. In a retrospective analysis of nearly 6000 US patients, those continuing therapy with insulin glargine reported significantly higher treatment persistence over 1 year (53% to 56%) compared with patients who switched to insulin detemir therapy (41% to 45%; P <.001).31 Significantly higher adherence (defined as adjusted MPR ≥0.80) was demonstrated in patients continuing with insulin glargine (59% to 61%) compared with those switching to insulin detemir (38% to 45%; P <.0001).31 It is important to note that although propensity score matching was used to balance the 2 groups, selection bias still may have remained that influenced the adherence findings. Information on why patients switched was not provided in the study. In a separate analysis of retrospective data for almost 1700 patients, significantly more patients using insulin glargine were persistent at 12 months (65%) compared with patients with insulin detemir therapy (51%; P <.001).32 In a retrospective analysis of approximately 8000 UK patients, individuals receiving insulin glargine therapy had better persistence at 12 months (83% vs 78%, respectively) and at 24 months (75% vs 68%, respectively) compared with insulin detemir (P value not reported).43 Finally, in a retrospective analysis of more than 2500 German patients on a basal insulin and OAD regimen, persistence at 2 years was significantly higher for users of insulin glargine (65%) compared with insulin detemir (53%) and NPH (59%) (P <.001).44 In the same analysis, contrasting results were reported for a different cohort of nearly 3200 patients who were on various basal-bolus insulin regimens. No significant differences in 2-year persistence were reported among insulin glargine, insulin detemir, and NPH in this cohort (84%, 85%, and 86%, respectively; P = .536).44

Administration of LAIAs using a pen-delivery system compared with vial/syringe yielded mixed results on adherence and persistence. Baser et al reported a significant difference in adherence over 1 year favoring vial/syringe application (adjusted MPR, 0.60 vs 0.55; P <.001) compared with FlexPen. However, change from baseline (improvement) in adherence over 1 year favored FlexPen (0.22 vs 0.13; P = .0011).45 Two retrospective analyses demonstrated statistically significantly greater adherence and persistence among patients with T2D initiating insulin glargine via the SoloSTAR pen compared with vial/syringe, supporting an adherence advantage associated with pen devices.35,36

DISCUSSION

LAIAs are an integral aspect of T1D and T2D management. Although the use of LAIAs carries a theoretical clinical benefit due to the less frequent administration and basal insulin action compared with non-LAIA treatment options, drug acquisition costs are relatively high and there is uncertainty as to whether the clinical benefits translate to real-world cost savings. The findings of this review suggest that patients using LAIAs often have neutral or reduced overall and diabetes-related healthcare costs, compared with patients not using LAIAs, and that adherence and persistence are improved with insulin glargine compared with insulin detemir. Some limited evidence also suggests that use of LAIA pen-delivery systems may lead to improved adherence and/or reduction in the frequency of hypoglycemic events, with associated reduction in costs, but a paucity of data prevents definitive assessment.

The relationship among dependent variables such as adherence, persistence, and economic cost remains unclear. It is difficult to ascertain whether adherence and persistence have an effect on subsequent economic outcomes or whether it is the choice of LAIA versus non-LAIA or mode of delivery alone affecting the increased adherence/persistence and cost outcomes. Certainly, it is intuitive that greater adherence would lead to improved clinical outcomes and therefore cost savings.

Limitations

Certain limitations should be taken into account with regard to this systematic review. The majority of studies reviewed were sponsored by pharmaceutical manufacturers, evoking potential bias; there is a general need for more industry-independent research to assess the cost-effectiveness of LAIAs relative to standard insulin regimens and to one another. A common feature reported in this analysis was the inconsistency of findings between different studies with ostensibly the same comparisons and populations, rendering firm conclusions difficult. There is considerable variation in findings across different studies; the disparate geographic locations and differing populations of many of these studies mean that each has limited representativeness or generalizability. Diverse patient populations were evaluated, variable data sources were used to inform efficacy and safety parameters, and different assumptions were made with regard to key inputs, such as the impact of hypoglycemia. Consequently, these differences significantly influenced model results. Furthermore, some studies analyzed in the present review were not prospectively designed to assess economic or adherence/persistence outcomes, and it is impossible to fully adjust for behavioral characteristics that may influence choice of device or treatment. Establishment of robust data sources to inform efficacy and safety inputs with greatest influence on economic model results (eg, rates of hypoglycemia, utility/disutility estimates) would therefore be a future goal.

CONCLUSIONS

Because much ambiguity remains, additional future research quantifying the economic burden of LAIAs in diabetes, and the resulting implications for payers (in terms of costs) and providers (in terms of quality outcome measures), is warranted. A better understanding of long-term microvascular and macrovascular outcomes with LAIAs relative to other diabetes treatment regimens would further aid in determining overall cost-effectiveness.&ensp;

Acknowledgments

The authors would like to acknowledge the following individuals for their support in the concept and design of this research and assistance in the manuscript development: Shengsheng Yu, PhD (currently at Abbott); Jennifer Colby, PharmD (Xcenda); Minh Luu, MBBS (Xcenda); and Tracy Liu, MD (Xcenda).Author Affiliations: Merck & Co, Inc (BA, JS), Kenilworth, NJ; Xcenda, LLC (LB, ES), Palm Harbor, FL.

Source of Funding: This work was funded by Merck & Co, Inc, Kenilworth, NJ.

Author Disclosures: Dr Alemayehu reports employment and stock ownership with Merck & Co. Ms Speiser reports employment with Merck & Co. Dr Bloudek and Dr Sarnes report employment with Xcenda, LLC, which received funding for this research.

Authorship Information: Concept and design (BA, JS, LB, ES); acquisition of data (BA, LB, ES); analysis and interpretation of data (BA, LB, ES); drafting of the manuscript (BA, JS, ES); critical revision of the manuscript for important intellectual content (BA, JS, LB, ES); provision of patients or study materials (JS); obtaining funding (BA); administrative, technical, or logistic support (JS); and supervision (BA, ES).

Address Correspondence to: Berhanu Alemayehu, DrPH, MS, Center for Observational and Real-World Evidence, Merck & Co, Inc, 351 N Sumneytown Pike, North Wales, PA 19454. Email: berhanu.alemayehu@merck.com.REFERENCES

1. World Health Organization. Global Report on Diabetes. Geneva, Switzerland; World Health Organization; 2016. who.int/diabetes/publications/grd-2016/en. Accessed December 1, 2016.

2. Types of insulin. University of California, San Francisco Diabetes Education Online website. ucsf.edu/types-of-diabetes/type2/treatment-of-type-2-diabetes/medications-and-therapies/type-2-insulin-rx/types-of-insulin/#insulinaction. Accessed May 17, 2018.

3. Barnett AH. Insulin glargine in the treatment of type 1 and type 2 diabetes. Vasc Health Risk Manag. 2006;2(1):59-67.

4. Tricco AC, Ashoor HM, Antony J, et al. Safety, effectiveness, and cost effectiveness of long acting versus intermediate acting insulin for patients with type 1 diabetes: systematic review and network meta-analysis. BMJ. 2014;349:g5459. doi: 10.1136/bmj.g5459.

5. Meneghini LF, Traylor L, Schwartz SL. Improved glycemic control with insulin glargine versus pioglitazone as add-on therapy to sulfonylurea or metformin in patients with uncontrolled type 2 diabetes mellitus. Endocr Pract. 2010;16(4):588-599. doi: 10.4158/EP09281.OR.

6. Bretzel RG, Nuber U, Landgraf W, Owens DR, Bradley C, Linn T. Once-daily basal insulin glargine versus thrice-daily prandial insulin lispro in people with type 2 diabetes on oral hypoglycaemic agents (APOLLO): an open randomised controlled trial. Lancet. 2008;371(9618):1073-1084. doi: 10.1016/S0140-6736(08)60485-7.

7. Wang L, Wei W, Miao R, Xie L, Baser O. Real-world outcomes of US employees with type 2 diabetes mellitus treated with insulin glargine or neutral protamine Hagedorn insulin: a comparative retrospective database study. BMJ Open. 2013;3(4):e002348. doi: 10.1136/bmjopen-2012-002348.

8. Gundgaard J, Christensen TE, Thomsen TL. Direct healthcare costs of patients with type 2 diabetes using long-acting insulin analogues or NPH insulin in a basal insulin-only regimen. Prim Care Diabetes. 2010;4(3):165-172. doi: 10.1016/j.pcd.2010.04.004.

9. Methley AM, Campbell S, Chew-Graham C, McNally R, Cheraghi-Sohi S. PICO, PICOS and SPIDER: a comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews. BMC Health Serv Res. 2014;14:579. doi: 10.1186/s12913-014-0579-0.

10. Kostev K, Dippel FW, Bierwirth R. Resource consumption and costs of treatment in patients with type 1 diabetes under intensified conventional therapy under German real-life conditions. J Diabetes Sci Technol. 2013;7(3):736-742. doi: 10.1177/193229681300700319.

11. Poole CD, Tetlow T, McEwan P, Holmes P, Currie CJ. The prescription cost of managing people with type 1 and type 2 diabetes following initiation of treatment with either insulin glargine or insulin detemir in routine general practice in the UK: a retrospective database analysis. Curr Med Res Opin. 2007;23(suppl 1):S41-S48. doi: 10.1185/030079907X167589.

12. Garg SK, Walker AJ, Hoff HK, D’Souza AO, Gottlieb PA, Chase HP. Glycemic parameters with multiple daily injections using insulin glargine versus insulin pump. Diabetes Technol Ther. 2004;6(1):9-15. doi: 10.1089/152091504322783350.

13. Baser O, Tangirala K, Wei W, Xie L. Real-world outcomes of initiating insulin glargine-based treatment versus premixed analog insulins among US patients with type 2 diabetes failing oral antidiabetic drugs. Clinicoecon Outcomes Res. 2013;5:497-505. doi: 10.2147/CEOR.S49279.

14. Baser O, Wei W, Baser E, Xie L. Clinical and economic outcomes in patients with type 2 diabetes initiating insulin glargine disposable pen versus exenatide BID. J Med Econ. 2011;14(6):673-680. doi: 10.3111/13696998.2011.605818.

15. Kleinman NL, Schaneman JL, Lynch WD. The association of insulin medication possession ratio, use of insulin glargine, and health benefit costs in employees and spouses with type 2 diabetes. J Occup Environ Med. 2008;50(12):1386-1393. doi: 10.1097/JOM.0b013e3181875e9b.

16. Lee LJ, Yu AP, Johnson SJ, et al. Direct costs associated with initiating NPH insulin versus glargine in patients with type 2 diabetes: a retrospective database analysis. Diabetes Res Clin Pract. 2010;87(1):108-116. doi: 10.1016/j.diabres.2009.09.023.

17. Misurski D, Lage MJ, Fabunmi R, Boye KS. A comparison of costs among patients with type 2 diabetes mellitus who initiated therapy with exenatide or insulin glargine. Appl Health Econ Health Policy. 2009;7(4):245-254. doi: 10.2165/11318730-000000000-00000.

18. Pawaskar M, Zagar A, Sugihara T, Shi L. Healthcare resource utilization and costs assessment of type 2 diabetes patients initiating exenatide BID or glargine: a retrospective database analysis. J Med Econ. 2011;14(1):16-27. doi: 10.3111/13696998.2010.544797.

19. Rhoads GG, Dain MP, Zhang Q, Kennedy L. Two-year glycaemic control and healthcare expenditures following initiation of insulin glargine versus neutral protamine Hagedorn insulin in type 2 diabetes. Diabetes Obes Metab. 2011;13(8):711-717. doi: 10.1111/j.1463-1326.2011.01394.x.

20. Schöffski O, Breitscheidel L, Benter U, et al. Resource utilisation and costs in patients with type 2 diabetes mellitus treated with insulin glargine or conventional basal insulin under real-world conditions in Germany: LIVE-SPP study. J Med Econ. 2008;11(4):695-712. doi: 10.3111/13696990802645726.

21. Anderten H, Dippel FW, Kostev K. Early discontinuation and related treatment costs after initiation of basal insulin in type 2 diabetes patients: a German primary care database analysis. J Diabetes Sci Technol. 2015;9(3):644-650. doi: 10.1177/1932296814566232.

22. Bullano MF, Fisher MD, Grochulski WD, Menditto L, Willey VJ. Hypoglycemic events and glycosylated hemoglobin values in patients with type 2 diabetes mellitus newly initiated on insulin glargine or premixed insulin combination products. Am J Health Syst Pharm. 2006;63(24):2473-2482. doi: 10.2146/ajhp050552.

23. Lechleitner M, Roden M, Haehling E, Mueller M. Insulin glargine in combination with oral antidiabetic drugs as a cost-equivalent alternative to conventional insulin therapy in type 2 diabetes mellitus. Wien Klin Wochenschr. 2005;117(17):593-598.

24. Levin PA, Zhang Q, Mersey JH, et al. Glycemic control with insulin glargine plus insulin glulisine versus premixed insulin analogues in real-world practices: a cost-effectiveness study with a randomized pragmatic trial design. Clin Ther. 2011;33(7):841-850. doi: 10.1016/j.clinthera.2011.05.091.

25. Miller DR, Gardner JA, Hendricks AM, Zhang Q, Fincke BG. Health care resource utilization and expenditures associated with the use of insulin glargine. Clin Ther. 2007;29(3):478-487. doi: 10.1016/S0149-2918(07)80086-5.

26. Pasquel FJ, Powell W, Peng L, et al. A randomized controlled trial comparing treatment with oral agents and basal insulin in elderly patients with type 2 diabetes in long-term care facilities. BMJ Open Diabetes Res Care. 2015;3(1):e000104. doi: 10.1136/bmjdrc-2015-000104.

27. Bonafede MM, Kalsekar A, Pawaskar M, et al. A retrospective database analysis of insulin use patterns in insulin-naïve patients with type 2 diabetes initiating basal insulin or mixtures. Patient Prefer Adherence. 2010;4:147-156. doi: 10.2147/PPA.S10467.

28. Dalal MR, Xie L, Baser O, DiGenio A. Adding rapid-acting insulin or GLP-1 receptor agonist to basal insulin: outcomes in a community setting. Endocr Pract. 2015;21(1):68-76. doi: 10.4158/EP14290.OR.

29. Ganz ML, Wintfeld NS, Li Q, Lee YC, Gatt E, Huang JC. Severe hypoglycemia rates and associated costs among type 2 diabetics starting basal insulin therapy in the United States. Curr Med Res Opin. 2014;30(10):1991-2000. doi: 10.1185/03007995.2014.936930.

30. Idris I, Gordon J, Tilling C, Vora J. A cost comparison of long-acting insulin analogs vs NPH insulin-based treatment in patients with type 2 diabetes using routinely collected primary care data from the UK. J Med Econ. 2015;18(4):273-282. doi: 10.3111/13696998.2014.991788.

31. Wei W, Zhou S, Miao R, et al. Much ado about nothing? a real-world study of patients with type 2 diabetes switching basal insulin analogs. Adv Ther. 2014;31(5):539-560. doi: 10.1007/s12325-014-0120-1.

32. Xie L, Wei W, Pan C, Du J, Baser O. A real-world study of patients with type 2 diabetes initiating basal insulins via disposable pens. Adv Ther. 2011;28(11):1000-1011. doi: 10.1007/s12325-011-0074-5.

33. Asias BD, Stock EM, Small NL, et al. Clinical and financial outcomes of switching insulin glargine to insulin detemir in a veteran population with type 2 diabetes. J Diabetes Metab Disord. 2015;14:53. doi: 10.1186/s40200-015-0180-z.

34. Borah BJ, Darkow T, Bouchard J, Aagren M, Forma F, Alemayehu B. A comparison of insulin use, glycemic control, and health care costs with insulin detemir and insulin glargine in insulin-naive patients with type 2 diabetes. Clin Ther. 2009;31(3):623-631. doi: 10.1016/j.clinthera.2009.03.005.

35. Fischer JS, McLaughlin T, Loza L, Beauchamp R, Schwartz S, Kipnes M. The impact of insulin glargine on clinical and humanistic outcomes in patients uncontrolled on other insulin and oral agents: an office-based naturalistic study. Curr Med Res Opin. 2004;20(11):1703-1710. doi: 10.1185/030079904X5526.

36. Grabner M, Chu J, Raparla S, Quimbo R, Zhou S, Conoshenti J. Clinical and economic outcomes among patients with diabetes mellitus initiating insulin glargine pen versus vial. Postgrad Med. 2013;125(3):204-213. doi: 10.3810/pgm.2013.05.2656.

37. Pawaskar MD, Camacho FT, Anderson RT, Cobden D, Joshi AV, Balkrishnan R. Health care costs and medication adherence associated with initiation of insulin pen therapy in Medicaid-enrolled patients with type 2 diabetes: a retrospective database analysis. Clin Ther. 2007;29(6 pt 1):1294-1305. doi: 10.1016/j.clinthera.2007.07.007.

38. Xie L, Wei W, Pan C, Baser O. Real-world rates, predictors, and associated costs of hypoglycemia among patients with type 2 diabetes mellitus treated with insulin glargine: results of a pooled analysis of six retrospective observational studies. J Med Econ. 2013;16(9):1137-1145. doi: 10.3111/13696998.2013.824458.

39. Xie L, Zhou S, Wei W, Gill J, Pan C, Baser O. Does pen help? a real-world outcomes study of switching from vial to disposable pen among insulin glargine-treated patients with type 2 diabetes mellitus. Diabetes Technol Ther. 2013;15(3):230-236. doi: 10.1089/dia.2012.0253.

40. Xie L, Zhou S, Pinsky BW, Buysman EK, Baser O. Impact of initiating insulin glargine disposable pen versus vial/syringe on real-world glycemic outcomes and persistence among patients with type 2 diabetes mellitus in a large managed care plan: a claims database analysis. Diabetes Technol Ther. 2014;16(9):567-575. doi: 10.1089/dia.2013.0312.

41. Buysman E, Conner C, Aagren M, Bouchard J, Liu F. Adherence and persistence to a regimen of basal insulin in a pre-filled pen compared to vial/syringe in insulin-naïve patients with type 2 diabetes. Curr Med Res Opin. 2011;27(9):1709-1717. doi: 10.1185/03007995.2011.598500.

42. Fabunmi R, Nielsen LL, Quimbo R, et al. Patient characteristics, drug adherence patterns, and hypoglycemia costs for patients with type 2 diabetes mellitus newly initiated on exenatide or insulin glargine. Curr Med Res Opin. 2009;25(3):777-786. doi: 10.1185/03007990802715199.

43. Gordon J, Pockett RD, Tetlow AP, McEwan P, Home PD. A comparison of intermediate and long-acting insulins in people with type 2 diabetes starting insulin: an observational database study. Int J Clin Pract. 2010;64(12):1609-1618. doi: 10.1111/j.1742-1241.2010.02520.x.

44. Pscherer S, Chou E, Dippel FW, Rathmann W, Kostev K. Treatment persistence after initiating basal insulin in type 2 diabetes patients: a primary care database analysis. Prim Care Diabetes. 2015;9(5):377-384. doi: 10.1016/j.pcd.2015.01.011.

45. Baser O, Bouchard J, DeLuzio T, Henk H, Aagren M. Assessment of adherence and healthcare costs of insulin device (FlexPen) versus conventional vial/syringe. Adv Ther. 2010;27(2):94-104. doi: 10.1007/s12325-010-0009-6.