Deintensification Discourse: Why are we still hesitant to pump the breaks on diabetes management?

It is well-established that the efficacy of diabetes therapy is frequently limited by clinical inertia, whether it be attributed to the patient, the clinician, or the health care system. The need for more frequent follow-up and education in the management of diabetes mellitus has created an area of opportunity for pharmacists, as we are more available to meet with patients and facilitate the initiation and acceleration of different therapies. Yet, we are often so concerned with adding on antihyperglycemic agents to getting our patients to goal that we often forget to pause and reassess if the patient really needs all of the medication they are prescribed.

In 2012, the American Diabetes Association (ADA) and European Association for the Study of Diabetes dismissed the one-size-fits-all-approach for glycemic targets and released a position statement describing the importance of individualizing treatment strategies, with an emphasis on patient-centered care and shared decision-making in the management of patients with type 2 diabetes.(1) This was based on literature that demonstrated intensive glycemic control (A1c <6–6.5%) may improve surrogate outcomes for microvascular complications and reduce cardiovascular events at the possible expense of all-cause mortality.(2-5) Factors such as disease duration, life expectancy, comorbidities, established vascular complications, patient preferences, available resources and support, along with risk for hypoglycemia and other adverse effects should be taken into consideration when establishing patient-specific glycemic targets.(6) As clinicians, we frequently assess these factors at diagnosis, but do we do the same for our aging patients with long-standing diabetes?

While we may think this isn’t a common problem, multiple studies have demonstrated possible overtreatment in vulnerable populations (7) and poor rates of treatment deintensification among patients with low HbA1c values or those at risk for hypoglycemia.(8, 9) A retrospective cohort study of data from the Veterans Health Administration examined rates of treatment deintensification among patients that were aged 70 years or older on active treatment for diabetes.(8) Of the 23,769 patients with moderately low HbA1c levels (6.0–6.4%) and the 12,917 patients with very low HbA1c levels (<6%), treatment deintensification was initiated in 20.9 and 27.0% of patients, respectively. Another retrospective analysis using OptumInsight data from 2004 through 2010 determined that antihyperglycemic therapy was deintensified in 18.3% of all patients with recently diagnosed type 2 diabetes mellitus.(9) Furthermore, treatment deintensification occurred in only 19.4% of patients with multiple comorbidities and 21.2% of those that met authors’ definition for frailty, regardless of glycemic control at baseline. This demonstrates the need for us as clinicians to step back and think, “does my patient actually need all of this medication?”

The widespread lack of treatment deintensification described by these studies demonstrates the need for further education among health care professionals and stakeholders. Patients that have not yet experienced a hypoglycemic event and are comfortable with their diabetes treatment are often empowered by their glycemic control, and therefore, patient preference serves as a barrier to deintensification. Providers may struggle identifying the need for and prioritizing diabetes treatment deintensification because what appears to be well-controlled diabetes may seem like the smallest obstacle in the road to comprehensive disease management, especially when patients present with a list of more active problems that require immediate attention. So, how can we better recognize the patient-specific transition of well-controlled to overcontrolled diabetes? How often do you assess an A1c goal in a well-controlled patient and consider their risk of hypoglycemia and falls along with their comorbid conditions to incorporate those factors into their treatment plan?

Just as pharmacists have demonstrated value in the implementation and optimization of antihyperglycemic therapy, their role in identifying those that would benefit from treatment deintensification and facilitating that process is equally important. That being said, what are some strategies you have used to identify the need for and to initiate deintensification of therapy in clinical practice?

Authored by Liz Van Dril, PharmD, PGY-1 Resident, Midwestern University

  1. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care 2012;35:1364–79.
  2. Zoungas S, Chalmers J, Neal B, et al. Follow-up of blood-pressure lowering and glucose control in type 2 diabetes. N Engl J Med. 2014;371:1392–406.
  3. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545–59.
  4. Gerstein HC, Miller ME, Genuth S, et al. Long-term effects of intensive glucose lowering on cardiovascular outcomes. N Engl J Med. 2011;364:818–28.
  5. Hayward RA, Reaven PD, Wiitala WL, et al. Follow-up of glycemic control and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2015;372:2197–206.
  6. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38:140–149.
  7. Lipska KJ, Ross JS, Miao Y, et al. Potential overtreatment of diabetes mellitus in older adults with tight glycemic control. JAMA Intern Med. 2015;175:356–362.
  8. Sussman JB, Kerr EA, Saini SD, et al. Rates of deintensification of blood pressure and glycemic medication treatment based on levels of control and life expectancy in older patients with diabetes mellitus. JAMA Intern Med. 2015;175:1942–1949.
  9. McAlister FA, Youngson E, Eurich DT. Treatment deintensification is uncommon in adults with type 2 diabetes mellitus: a retrospective cohort study. Circ Cardiovasc Qual Outcomes. 2017;10(4).
  10. National Committee for Quality Assurance. Healthcare effectiveness data and information set: comprehensive diabetes care. (accessed 2017 June 11).



A pharmacy business model that’s bad for pharmacy and bad for some patients

We’ve all heard about and experienced the impact of high medication costs in recent years ( While the ACA has increased the number of people who can obtain insurance, it has done nothing to help patients actually afford their medications. While there is no one person to blame, there is one piece of the puzzle that has a domino effect on medication costs.  The Pharmacy Benefit Manager (PBM) is a 4thparty in the mix of insurance that directly manages the prescription benefit for insurance companies.  It is the PBM who negotiates with manufacturers, insurance companies, and pharmacies. It is the PBM that offers different ‘formularies’ to health plans, in many cases disenfranchising patients who may need medications not on the ‘formulary’. A perfect example is the insulin marketplace.  (  Here is a good visual and explanation of how PBM’s work: )  I think they have created a perfect storm of healthcare inefficiency that costs patients access to medications and increases the overall cost of healthcare. Why?  I could list a bunch of reasons, but what do you think?

Metformin: Stepchild of Diabetes Care

The Merriam Webster definition of Stepchild is: “one that fails to receive proper care or attention”. We all know that metformin is pretty much a ‘given’ when we initiate care for diabetes based on nearly all type 2 diabetes treatment guidelines. We also know that metformin has been used effectively in diabetes prevention in patients with prediabetes with long-term effectiveness (1) and that this effect was dose and adherence related (2). Yet despite this, one survey suggested that only 36% of primary care providers prescribe metformin for patients with prediabetes at all (3). Combine this with the fact that we that there is a relationship of dose to the intensity of the hypoglycemic effect in diabetes in clinical trials (4) and that there is an association of glucose control in early treatment with ‘regression’ of pre-diabetes (5).

Guidelines are clear on the need to aggressively intensify therapy in patients with newly diagnosed type 2 diabetes to get to the A1C goal, but patients and physicians may feel less urgency early in the course of the disease, and ‘clinical inertia’ or lack of appreciation of the benefits of early control mean that many patients will be at increased risk of later cardiovascular complications from inadequate intensity of metformin dosing even in the few years subsequent to diagnosis (6). With more recent publications suggesting more flexibility in dosing metformin in patients with impaired renal function, we have even more reason to be comfortable with dosing this important medication (7). When was the last time you recommended metformin for a patient with prediabetes? When was the last time your reviewed your type 2 diabetes patients’ doses of metformin and pushed the envelope on dosing? Sounds like there’s a need to me!

1. Diabetes Prevention Program Research Group. 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study Lancet. 2009 November 14; 374(9702): 1677

2. The Diabetes Prevention Program Research Group. Long-Term Safety, Tolerability, and Weight Loss Associated With Metformin in the Diabetes Prevention Program Outcomes Study. Diabetes Care 2012;35:731

3. Mainous AG, et al. Prediabetes Screening and Treatment in Diabetes
Prevention: The Impact of Physician Attitudes. J Am Board Fam Med 2016;29:663

4. Hirst JA, et al. Quantifying the Effect of Metformin Treatment and Dose on Glycemic Control. Diabetes Care 2012;35:446

5. Perreault L, et al. Effect of regression from prediabetes to normal glucose regulation on long-term reduction in diabetes risk: results from the Diabetes Prevention Program Outcomes Study. Lancet 2012;379:2243

6. Svensson E, et al. Early Glycemic Control and Magnitude of HbA1c Reduction
Predict Cardiovascular Events and Mortality: Population-Based Cohort Study of 24,752 Metformin Initiators. Diabetes Care 2017;40_online April 17

7. Inzucchi, SE, et al. Metformin in Patients With Type 2 Diabetes and Kidney Disease: A Systematic Review. JAMA. 2014;312:2668

Laugh and the World Laughs With You… …Snore and You Sleep Alone!


Of course, all snoring isn’t necessarily bad (unless you are a sleep partner), but most of us have heard about Obstructive Sleep Apnea (OSA) and have pre-conceived notions about risk factors, prevalence and outcomes. It is safe to say, however, that obstructive sleep apnea is the  ‘elephant in the room’ of cardiovascular risk.  It is true that most individuals who have OSA are overweight or obese, as that is the most common risk factor.  We know that the upper airway collapses in some people who have more tissue in the upper airway (tonsils, adenoids, tongue, palate and uvula).  When the OSA sleeper lays in his/her back, gravity is not in their favor as muscles relax during sleep.  When the sleeper attempts to breath in, the tissue obstructs resulting in snoring or gasping sounds and either a decrease in flow (hypopnea) or a cessation of airflow into the lungs (apnea).  The more often this happens the more likely the sleeper is to develop the consequences of chronic intermittent hypoxia.  OSA affects men about 2-3 times more commonly than women. An older study suggested that roughly 26% of a primary care population was potentially at risk of OSA, yet screen was very rare (1,2). If we look at one specific population, among an estimated 14 million US commercial drivers, 17–28% or 2.4 to 3.9 million are expected to have OSA (3).  In this population the effects of sleep deprivation secondary to OSA can be disastrous.

A much more insidious and common occurrence of OSA is in people with diabetes.  If we look at the type 2 diabetes patient population, the proportion at risk is significantly higher, due to the higher prevalence of older patients, and those with obesity. A recent review suggested the overall prevalence of diagnosed OSA in diabetic patients is approximately 71% based on the average data from five studies including a total number of nearly 1200 patients with type 2 diabetes (4). In 2008, the IDF Taskforce on Epidemiology and Prevention released a consensus statement that recommended a targeted approach to “screen individuals with type 2 diabetes and obesity for sleep disordered breathing (SDB)”. Briefly, the IDF recommended that healthcare professionals should consider the possibility of OSA in patients with type 2 diabetes and work in tandem with the local ‘sleep service’ to provide a clinically appropriate process of assessment, referral and intervention (5).

Several screening tests are available and include the Berlin Questionnaire, the STOP-BANG Questionnaire, and the Epworth Sleepiness Scale.  This last screener may be less effective than the others primarily due to the fact that daytime sleepiness, while common, is not universal, and appears less often in women and in individuals with heart failure. I prefer the “Canadian modifications” of the STOP-BANG screening tool (6).  Further diagnostic tests include the ‘gold standard’ sleep study (polysomnography), and within the last several years more and more products have been introduced that can be used as home diagnostic tests (7).

Studies have shown that cardiac remodeling occurs In OSA patients and that the changes are similar to predisposing changes for heart failure.  There is a significant increase in cardiovascular risk from the downstream consequences of chronic intermittent hypoxia from repeated episodes of apnea or hypopnea during sleep: atherosclerosis, cardiovascular disease including conditions such as myocardial infarction, congestive heart failure, cerebrovascular accident, resistant hypertension, and cardiac arrhythmia, as well as cognitive dysfunction, depression, poor glucose control in diabetes and motor vehicle accidents to name just some of them.

So, the prevalence in people with diabetes is high, and the outcomes of cardiovascular morbidity and mortality are well described.  Yet, the screening rate is abysmally low (in one study around 5%).  Routine screening of diabetes patients should lead many more people to a diagnostic procedure and to CPAP as the most effective treatment.  An old saying about how to eat an elephant is “one bite at a time”.  In the case of OSA, I would submit that some of these bites are up to you.  Pharmacists, involved in screening for OSA you ask (?) Of course! In this era of patient-centered care, how could a credible “diabetes practitioner” [yes, that’s you…] NOT screen patients for Obstructive Sleep Apnea!

  1. Hiestand DM, Britz P, Goldman M, Phillips B. Prevalence of symptoms and risk of sleep apnea in the US population: results from the National Sleep Foundation sleep in America 2005 poll. Chest 2006; 130:780 – 6.
  2. Grover M, et al.  Identifying Patients at Risk for Obstructive Sleep Apnea in a Primary Care Practice. J Am Board Fam Med 2011;24:152–160
  3. Kales S, and Straubel,M. Obstructive Sleep Apnea in North American Commercial Drivers. Industrial Health 2014, 52, 13–24
  4. Pamidi S and Tasali E . Obstructive Sleep Apnea And Diabetes-IsThereALink_Pamidi FrontNeurol_2012_v3_Article128
  5. Seetho I, et al. Obstructive sleep apnoea in diabetes – assessment and awareness. British Journal of Diabetes  2014(3):105-108

New Year and New Guidelines: Incorporating Cardiovascular Outcome Data

After a long year of being on the look-out for new cardiovascular outcome trial (CVOT) data, the American Diabetes Association (ADA) has released the much anticipated 2017 Standards of Medical Care in Diabetes update. The overall value of the CVOT data has been called into question and clinicians have varied predictions regarding how the ADA may incorporate those findings into their recommendations.  So let’s take a look:

The ADA has incorporated their recommendation regarding the positive outcomes seen with the use of empagliflozin in the EMPA-REG OUTCOME trial and liraglutide in the LEADER trial into the section on the pharmacologic approaches to glycemic treatment in type 2 diabetes mellitus (T2DM). Regretfully, the data was not incorporated into the general treatment algorithm, which continues to focus on patient centered characteristics for the selection of second line agents. Instead, the level B ADA recommendation comes with some caveats. The recommendation outlines the use of empagliflozin and liraglutide in patients with long-standing, poorly controlled T2DM and established atherosclerotic cardiovascular disease (ASCVD) in addition to the existing standards of care. (ref 1)  These guideline updates are on the heels of the new indication approval by the FDA for empagliflozin. This new indication is to reduce the risk of cardiovascular death in adults with T2DM and cardiovascular disease.(ref 2)

While this blanket statement for established ASCVD may by easier for clinicians to apply in practice, it is important to recognize the differences in patient population that were included in both the EMPA-REG OUTCOME and the LEADER trials.(ref 3,4) The LEADER trial exemplified a higher level of external validity, because of the more inclusive study design and expansive definition of high cardiovascular risk. (ref 4) While the EMPA-REG OUTCOME study included patients with only established cardiovascular disease, the LEADER trial also included patients with at least one cardiovascular disease risk factor and aged > 60 years. (ref 3,4) The LEADER trial also considered CKD stage 3 or greater or CHF of NYHA class II or III in their definition of coexisting cardiovascular condition, which only had an inclusion age cut off of 50 years.(ref 4) I feel that in my practice, while established ASCVD is not rare, the positive CVOT data can be applied more widely for liraglutide than for empagliflozin due to the broader inclusion criteria seen in the LEADER study.

How do the differences in study population affect how other clinicians apply this data? I am eager to hear other opinions in the clinical community about how the ADA has classified this data and if it will change your practice. Another clinical discussion starter can also be if this data fulfils the original purpose of CVOT and if they are worth the large cost and limited external  . Lastly, does anyone think that metformin with be over-thrown from the high castle as the backbone of therapy with any of this CVOT data? If you aren’t convinced just yet, what do you think it would take to convince you?

  1. American Diabetes Association (ADA). Standards of medical care in diabetes 2017. Diabetes Care 2017 Jan; 40 (Supplement 1): S64-S74.
  2. FDA approves Jardiance to reduce cardiovascular death in adults with type 2 diabetes. Available at: [Accessed January 10, 2017]
  3. Zinman B, Wanner C, Lachine JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med 2015;373:2117-28.
  4. Marso SP, Daniels, GH, Brown-Frandsen K, et al. Liraglutide and cardiovascular outcomes in Type 2 diabetes. N Eng J Med 2016;375(4):311-22


Low hanging fruit

Hopefully everyone had a good holiday break, and maybe even the chance to curl up with a good book…or better, with the Federal Register!  Yes, you heard that right, the Federal appleRegister from 2016 has a number of positive developments in areas that are or will soon be ‘ripe’ for pharmacist intervention.  Notices and rulemaking for Medicare are published by CMS in the Federal Register. These notices are often accompanied by CMS’ responses to comments received on proposed rulemaking as well as some interesting background data.  So let’s look at what CMS considers important (i.e. what they are willing to pay for).  First, they recognize that medication misadventures often result in costly adverse effects including ED visits and hospitalizations.  They recognized MTM as needed (Ref 1), and when it was clear that it wasn’t being used as often as they wanted, they expanded the criteria to qualify more patients for MTM.  Realizing that medication misadventures were more likely during care transitions, CMS decided to reimburse for Transitional Care Management (CPT 99495, 99496) that included Medication Reconciliation. (Ref 2) With CMS’ announced aggressive plan to move to a system of more value-based reimbursement, new payment models were recently release as MACRA-MIPS (Ref 3), and medication reconciliation was a key component for the Merit-based Incentive Payment System (Ref 3, see pgs 77225 and 77230).

Switching gears to a key condition recognized by CMS, diabetes, it is clear that they value Diabetes Self Management Education (CPT G0108, G0109), and that they have increased the reimbursement for provision of that service. (4) Recently, CMS has announced that it will reimburse diabetes prevention in the proposed Medicare Diabetes Prevention Program (Ref 5, see section III.J)  The preliminary structure proposal is in Ref 5, and establishes how you can be prepared for the final rulemaking in 2017, and implementation in 2018.  In this document you can find links to the proposed standards and to the proposed curriculum developed by CDC. This is a must read if you want to be prepared to offer this service!

Finally, CMS has not only established reimbursement for Chronic Care Management (CPT 99490), but updated their rules with new codes for more complex patients who are involved in CCM (CPT 99487-99489) so that the reimbursement could better reflect the amount of work involved with CCM in highly complex patients (Ref 5, see section E.4 and Table 11).  As most of you already know, CCM can be provided to patients with 2 or more chronic diseases, and thus nearly all your patients with diabetes would qualify.

It is clear that medications, diabetes and chronic care are not only on the CMS’ radar screen, but they are addressing concerns related to expansion of these services as well as augmenting reimbursement.  Members of the ACCP Endocrine & Metabolism PRN are particularly well positioned to take advantage of many of these services!

At the 2016 ACCP Annual Meeting in October, a session discussing TCM and CCM was poorly attended, yet just down the hall a session on PCSK-9 inhibitors was packed.  While being the local guru on pharmacodynamics of PCSK-9 inhibitors may bring personal satisfaction, providing services such as TCM, CCM, DSMT, and soon MDPP brings revenue.  We are fighting for recognition as ‘providers’, and in any fight, you are lucky when the other guy telegraphs his moves.  CMS is not telegraphing, they are shouting it from the rooftops, yet only a few ACCP members are doing any of these services.  My bias is obvious, if you want to be recognized as ‘providers’, then start providing what your customer wants!

  1. The Medicare Prescription Drug, Improvement, and Modernization Act of 2003. Public Law 108-173. December 8, 2003. Available at: [accessed 12/27/2016]
  2. [accessed 12/27/2016]
  3. Medicare Program; Merit-Based Incentive Payment System (MIPS) and Alternative Payment Model (APM) Incentive Under the Physician Fee Schedule, and Criteria for Physician- Focused Payment Models. Federal Register / Vol. 81, No. 214 / Friday, November 4, 2016 / Rules and Regulations. [accessed 12/27/2016]
  4. Medicare Benefit Policy Manual Chapter 15 – Covered Medical and Other Health Services [accessed 12/27/2016]
  5. Medicare Program; Revisions to Payment Policies Under the Physician Fee Schedule and Other Revisions to Part B for CY 2017; Medicare Advantage Bid Pricing Data Release; Medicare Advantage and Part D Medical Loss Ratio Data Release; Medicare Advantage Provider Network Requirements; Expansion of Medicare Diabetes Prevention Program Model; Medicare Shared Savings Program Requirements. [accessed 12/27/2016]

Co-Pays and Policy

An recent editorial in New England Journal of Medicine (N Engl J Med 2016; 375:2013-2015) from Dr. Leemore Dafny and others spoke of the U.S. DHHS announced goal of linking at least 50% of Medicare spending to value-based payment models. They noted that “health care providers are now scrambling to reorganize in a way that delivers value while preserving or enhancing ‘commercial success’ […of their health care businesses]…For years, insurers and pharmacy benefits managers have steered consumers toward generic and other ‘high-value’ drugs by categorizing drugs into ‘tiers’ and requiring lower copayments for ‘preferred’ drugs.” High value and preferred decisions are being made by the insurer/PBM based perhaps on criteria other than what a clinical pharmacist might consider high value… The authors note that “under tiering, insurers offer manufacturers favorable tier placement in exchange for better discounts” a clear benefit to the insurer or PBM. “Placement on a preferred-brand tier, with a [low] co-pay, will [arguably] yield higher sales than placement on a non-preferred-brand tier with a typical copayment of more than $50. Insurers can also negotiate lower prices for drugs that have therapeutic substitutes or questionable benefits by threatening to exclude them from their formularies entirely.”  The philosophy espoused by the authors of the article is that this is all good!  However, they note that the pharmaceutical industry “counterattacked” (…interesting choice of emotional vocabulary) “by offering ‘copayment coupons’. These coupons or discount cards — distributed by physicians’ offices, through the mail, and online — enable the manufacturer to pay some or all of a consumer’s copayment for a prescription. By severing the link between cost sharing and the ‘value’ generated by a drug, copayment coupons can undo the perceived beneficial effects of tiering.”  They estimate that “coupons increase the percentage of prescriptions filled with brand-name formulations by more than 60%” (!)

OK, I get it! The tiering process is one way to control costs, and healthcare policymakers like that!  I wanted to present a flip side view from the patient care perspective. It is well known that medication adherence is related in part to the size of co-pay assessments in an attempt at ‘cost sharing’ (1-2) both generally and for those entering the Medicare ‘donut hole’.(3)  The use of ‘high value’ [my definition is not necessarily based on medication cost…] drugs such as those that are still branded, may offer improved outcomes, but they are often made second or third tier by insurers to limit their own costs.   Adherence with a number of new medications, especially those with minimal co-pays suggests that there are adherence benefits to be gleaned from the discount cards or coupons.  Add to that the evidence that some of the older diabetes medications, available with little or no co-pays may have a higher rate of adverse effects. (4-7)

So ask yourself, if your mother or father were diagnosed with type 2 diabetes, would you want early aggressive therapy with agents that may have significantly more benefits (8-9), not to mention the benefit of early glucose control, or would you want them to start older agents (some with potential CV risk) and then stay with them for years in the clinical inertia that plagues diabetes care. I know what the policy makers prefer, and I know what my preference is for making new and better drugs more available regardless of ability to handle the co-pay…How about you?

  1. Effect of prescription copayments on adherence and treatment failure with oral antidiabetic medications. Barron J, Wahl P, Fisher M, Plauschinat C. P.T. 2008 Sep;33(9):532-53
  2. How patient cost-sharing trends affect adherence and outcomes: a literature review.Eaddy MT, Cook CL, O’Day K, Burch SP, Cantrell CR. P T. 2012 Jan;37(1):45-55
  3. Part D coverage gap and adherence to diabetes medications. Gu Q1, Zeng F, Patel BV, Tripoli LC. Am J Manag Care. 2010;16(12):911-8
  4. Sulphonylureas and risk of cardiovascular disease: systematic review and meta-analysis. Phung OJ1, Schwartzman E, Allen RW, Engel SS, Rajpathak SN. Diabet Med. 2013 Oct;30(10):1160-71
  5. Hyperinsulinemia and sulfonylurea use are independently associated with left ventricular diastolic dysfunction in patients with type 2 diabetes mellitus with suboptimal blood glucose control. Inoue T, Maeda Y, Sonoda N, Sasaki S, Kabemura T, Kobayashi K, Inoguchi T. BMJ Open Diabetes Res Care. 2016 Aug 18;4(1):e000223
  6. Cardiovascular risk associated with the use of glitazones, metformin and sufonylureas: meta-analysis of published observational studies.Pladevall M, Riera-Guardia N, Margulis AV, Varas-Lorenzo C, Calingaert B, Perez-Gutthann S. BMC Cardiovasc Disord. 2016 Jan 15;16:14
  7. Mortality risk among sulfonylureas: a systematic review and network meta-analysis.Simpson SH, Lee J, Choi S, Vandermeer B, Abdelmoneim AS, Featherstone TR. Lancet Diabetes Endocrinol. 2015 Jan;3(1):43-51
  8. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. Zinman B, Wanner C, Lachin JM, Fitchett D, Bluhmki E, Hantel S, Mattheus M, Devins T, Johansen OE, Woerle HJ, Broedl UC, Inzucchi SE; EMPA-REG OUTCOME Investigators. N Engl J Med. 2015 Nov 26;373(22):2117-28
  9. Empagliflozin and Progression of Kidney Disease in Type 2 Diabetes. Wanner C, Inzucchi SE, Lachin JM, Fitchett D, von Eynatten M1, Mattheus M, Johansen OE, Woerle HJ, Broedl UC, Zinman B; EMPA-REG OUTCOME Investigators. N Engl J Med. 2016 Jul 28;375(4):323-34