In my last blog, I explained why I’m not a fan of the highly awaited ADA/EASD position statement on management of type 2 diabetes, which was published online April 19th, 2012. Overall, I expressed disappointment in the document’s lack of clinical usefulness and specific clinical recommendations, despite its proclaimed patient-centered approach. Here, as I elaborate a bit more on my feelings about specific statements and recommendations within the document, I will pay homage to Clint Eastwood's American classic The Good, the Bad, and the Ugly.
I was prepared to dislike the section Advancing to triple combination therapy. Being an endocrinologist, I have too many patients with uncontrolled type 2 diabetes referred to me who are on three or four noninsulin therapies even though their A1c levels had been so high when the 3rd or 4th agent was started that it had no chance of success. A strong message from the diabetes specialty world for years has been that many patients need insulin long before it’s prescribed. So a section on triple therapy looked unsettling. Fortunately, though, this section mostly advocates insulin after two drugs, especially if the A1c is ≥8.5%. Good.
I’m also OK with the section Transitions to and titrations of insulin. I wanted specific instructions on insulin doses and adjustment algorithms akin to the earlier ADA/EASD consensus statements. That information is here—starting U/kg doses, the same titration algorithm as the 2006 and 2009 ADA/EASD consensus statements, and U/kg doses for starting prandial insulin. And it’s done in an up-to-date fashion.
The authors also get high marks for the Other considerationssection. There’s a lot of useful information here.
I’m unhappy with the Initial drug therapy section. It says metformin. Fine—no surprise. But I was caught off guard with the statement that it is now okay to wait on starting metformin at diagnosis and to try lifestyle efforts first with an A1c up to 7.5%. Really? I know that the rules that governed what went into the Position statement required published evidence for a recommendation, and I agree data for and against immediate metformin use with lowish A1c levels are lacking. Still, the argument made in 2006 and 2009 for starting metformin at diagnosis at any A1c level was the well-known, long-term failure of lifestyle efforts in this disease, or of monotherapy, quite frankly. Going back to waiting for lifestyle efforts to fail before adding metformin with modest A1c increases, I’m afraid, will promote therapeutic inertia. Not my favorite recommendation.
I also am a bit shocked how the authors handle the discussion of stratification of initial therapy for the starting A1c level—a reasonably hot diabetes topic. They recommend combination therapy when the A1c exceeds 9%. Specific wording is “Patients with….HbA1c ≥9% have a low probability of achieving a near-normal target with monotherapy. It may be….justified to start directly with a combination of two noninsulin agents or with insulin itself.” Wow. That feels really conservative. Plus the AACE roadmap starts one agent with an A1c below 7.5%, two up to 9%, and three or more with insulin above that. Interesting but a little disheartening how divergent these recommendations are. Given the global visibility of the ADA/EASD guidelines, I find the super-conservative approach to be unfortunate.
So far I’ve voiced a few likes and dislikes. Nothing horrible. But in reality my overall feeling is considerably more negative: the whole document feels inadequate, and I am really disappointed. Why? It finally dawned on me. One of the major take-home messages of the prior ADA/EASD consensus statements was the importance of the figure(s) for communicating the guidelines (See my comments in my first blog on this subject). That’s the only thing the vast majority of people will ever see. And that’s what is used to communicate the information. So from the get go, my belief while waiting was the success or failure of the new guidelines would hinge on the figures. I expected a spectacular algorithm figure that was clear, powerful, informative, and dictatorial—one that would tell providers how to be better diabetes clinicians. Do this! Do that! Maybe don’t do that! Meet this goal! Do it in this time frame! Go quickly to insulin if you need to—that’s okay because many patients need it.
It doesn’t. Not even close to my eyes. Figure 2 is visually poor—too much information with no real guidance. Effectively it says that a provider can do anything: one, two, or three drugs, or add insulin whenever. Regarding insulin, the figure most strongly visually suggests starting after three drugs, although admittedly there is the right-hand column (in grey, the most muted color, so it’s easy to miss).
And then there’s the insulin figure (Fig 3.). One of the parts of the 2006 and 2009 consensus statements that I liked best was the separate insulin figure, the same in both versions, that included specific details on how to do insulin—starting doses, blood glucose and A1c goals, and how long to wait before intensification. They were highly effective teaching tools so that specifics on insulin use were an important part of presenting the document’s recommendations. That’s gone in the current version. And I miss it. Figure 3 in this new position statement is very general with no specifics. Not a “how to” slide. A major omission from my perspective.
In deference and respect to the smart people who worked long and hard on the position statement, forgive me for being so negative. Truthfully, I have no idea what I would have done differently except probably to have pushed for more specifics and better figures. Regarding the latter, I wonder how the existing figures will be presented to an audience, especially Figure 2. What does it say? Is the main message to do whatever you and the patient want, whenever you want? And as a general question, will this document improve diabetes care? I genuinely hope so, but I’m not optimistic. I hope I’m wrong.
Tell me your opinion
Send me your comments about the document and/or my review. This seems likes a timely important topic for an active discussion.