Memorandum

NEJM article on achievement of goals in US diabetes care from 1999 to 2010; targets for diabetes care not met by nearly half of patients – April 25, 2013

Executive Highlights

  • Today, the NEJM published a special article online tracking national progress in diabetes care in the US from 1999-2002, 2003-2006, and 2007-2010.
  • Almost half of US adults with diabetes did not meet the recommended goals for diabetes care. Only 14% of patients are hitting all three goals for A1c, blood pressure, LDL cholesterol, and nonsmoking.
  • Many of the improvements in meeting goals came between periods one and two, with much less improvement in the latest time period assessed (2007-2010). The accompanying editorial concludes that “we have reached a plateau.”

In the April 25 edition of the New England Journal of Medicine, Dr. Mohammed Ali and colleagues published a special article on the achievement of goals in US diabetes care from 1999 to 2010. The paper analyzes data from over 100,000 adults with self-reported diabetes from the National Health and Nutrition Examination Survey (NHANES) and the Behavioral Risk Factor Surveillance System (BRFSS; the same telephone-based study used to compile the obesity maps of the US). The researchers specifically examined risk-factor control (A1c, blood pressure, cholesterol, tobacco), preventive practices (screenings, glucose monitoring, and vaccinations), and risk scores for coronary heart disease from 1999-2002, 2003-2006, and 2007-2010.

Broadly speaking, the most recent results are fairly disheartening for diabetes control. While there were big improvements in A1c between 1999-2002 and 2003-2006, the gains in glycemic control seem to have plateaued or been chipped away – the percentage of patients with an A1c <7% has actually declined from 57% (2003-2006) to 52% (2007-2010). Additionally, the number of patients with very poor glycemic control (A1c >9%) has stayed roughly flat (13% vs. 12.6%) from 2003-2006 and 2007- 2010. We were also very surprised to see that an astonishing 22% of people with diabetes are current smokers, a number that has only declined two percentage points from 1999-2002.

Blood pressure (<130/80 mm Hg) and LDL cholesterol were certainly bright spots in the analysis, with improvements continuing on a positive linear trend in recent years. Additionally, the percentage of patients meeting all goals for A1c, blood pressure, LDL cholesterol, and nonsmoking status notably improved as well – still, just 14% are hitting all four metrics, suggesting there is a lot to do on this front. Additionally, given the lack of improvement in diabetes control and smoking, we suspect the gain in this catch-all metric stems from improvements in lipids and blood pressure.

We thought the first sentence of the accompanying editorial from Drs. Graham McMahon and Robert Dluhy summed it up well: “…the recent report card on our national performance that appears in this issue of the Journal suggests we have reached a plateau…there’s a long way to go to deliver the quality of diabetes care that truly meets our patients’ needs.” Drs. McMahon and Dluhy believe the “chronic care model offers some hope.” They call for greater use of team care, changes in medical education and training, and quality improvement metrics that are tracked and rewarded. We couldn’t agree more, especially on the latter. Diabetes is truly a metrics-driven disease, though to date, we believe landmark trials have perhaps encouraged overreliance on the crude metric of A1c. We hope providers are rewarded for one day for keeping patients “in-zone” more often and avoiding hypoglycemia.

  • Those interested can read the study (subscription needed) at http://www.nejm.org/doi/full/10.1056/NEJMsa1213829, while the editorial is posted at http://www.nejm.org/doi/full/10.1056/NEJMe1302610.
  • The authors of the study (as well as broader media attention) appear to interpret the paper as a sign that diabetes care is improving – we believe the interpretation is more nuanced. Certainly, comparing 1999-2002 with 2007-2010 indeed suggests we’re doing much better. However, a careful look at the data (Table 3) reveals that most of the gains came between 1999-2002 and 2003-2006. Between 2003-2006 and 2007-2010, improvements seem to have plateaued in diabetes. In our view, it’s critical to look at the time-point by time-point trend over the last decade, not just the overall change.
  • To us, the paper speaks to the greater challenges of managing diabetes over blood pressure and cholesterol: 1) much more patient self-management; 2) a greater need for behavior change; 3) therapies that are harder to take (certainly the conditions’ oral meds are comparable, though many patients with diabetes end up on injectables); 3) less clear metrics and targets in diabetes (especially with the recent focus on individualization); 4) the need to balance the acute risk of intensive control (e.g., severe hypoglycemia) with its benefits (long-term avoidance of diabetes complications). Blood pressure, cholesterol, and diabetes certainly do share some characteristics – no immediate pain when not a goal, required motivation to avoid long- term problems, many different therapies available – though we think it’s instructive to consider these underlying differences when seeking to explain the discrepancies.
  • It will be fascinating to see how things change from 2007-2010 and the next update. Incretins have continued to grow significantly since 2010, and we wonder how the commercial success of both DPP-4s and GLP-1s will impact care. The recent addition of SGLT-2s, as well as upcoming fixed-dose combinations and easier-to-take therapies (once-weekly) should also play an interesting role. Our hope is that things will improve markedly, though in diabetes, we need better therapies as much as we need drastic changes in how we care for people with diabetes: better metrics and incentives for providers, better reimbursement, more team care, better understanding of behavior and how to motivate patients, improved education, etc.
  • Preventive practices largely stayed flat between 2003-2006 and 2007-2010. Glucose monitoring at least once per day saw a small gain (67% to 71%), though the magnitude was much smaller than the nine percentage-point gain between 1999-2002 and 2003-2006. Diabetes education has remained fairly flat in recent years; unfortunately, just over half (55%) of patients received it in the 2007-2010 period, suggesting there is ways to go on this front. Similar to A1c, risk of complications improved markedly from 1999-2002 and 2003-2006, though leveled off from 2007-2010.
  • When the authors broke down A1c goals on an individualized basis (age and complications), the trends were slightly more nuanced. (The paper uses individualized ADA/EASD A1c targets depending on age and complications status, ranging from a low of <6.5% for 18-44 years old and no complications to a high of <8% for patients 45-64 years old with complications.) Notably, many more of those 18-44 years old without complications met an A1c<6.5% over time – 37% in 1999-2002, 35% in 2003-2006, and 55% in 2007-2010. Yet, in thosewith complications in the same age group, the trend was more disappointing: 39% met the <7% goal in 1999-2002, rising to 53% in 2003-2006, and dropping substantially to just 28% in 2007- 2010. In those over 65 years old with complications, the trend is on a slow positive climb from avery low base: 3.4% met the <8% goal in 1999-2002, rising to 3.7% in 2003-2006 and 4.3% in2007-2010.
  • The researchers note that younger adults were less likely than older adults to meet goals for treatment and preventive practices. Younger adults had no significant improvements in these areas over time. The authors note, “It is unclear whether physician inattentiveness, poor access to health care, or other factors account for these gaps.” We would guess it’s the middle one, as younger people might be less likely to see a doctor to begin with.
  • We note that FDA EMDAC Advisory Committee member Dr. Edward Gregg (Centers for Disease Control and Prevention, Atlanta GA) was the final author of the study. Dr. Gregg is voted against approval of canagliflozin due to the absence of two-year data and the concerning stroke data. He admitted that canagliflozin had a diverse set of benefits, but stated that there were many lingering questions (bone density, fractures, renal function, volume depletion) that could not be answered without long-term data. He also would have liked to see quality of life data. In Dr. Gregg’s view, canagliflozin’s benefits were also clouded by the lower efficacy in those with renal impairment, a target population. He was not concerned about the cardiovascular risk in the first 30 days, a finding he attributed to chance. Dr. Gregg voted against the approval of dapagliflozin on the basis of uncertainty surrounding its breast and bladder cancer risk; against approval at the first Advisory Committee meeting for Belviq (Lorqess) and in favor at the second; in favor of approval of Qnexa (Qsymia) in 2012; and “yes” at the Advisory Committee meeting on CV risk assessment for obesity drugs (he was one of the few “yes” voters who expressed concern about a two-tiered approach to CV risk evaluation, noting that it may not provide clear guidance to sponsors).

-- by Adam Brown and Kelly Close