When to Start Insulin? Lack of Consensus on Targets

Dennis P. Scanlon, PhD: How about if we think about adding insulin. When is it appropriate to do so?

Robert Gabbay, MD, PhD, FACP: We know that the natural progression of the disease is a continued loss of beta cell function and decreased ability to make insulin. Once you get to the point where oral agents are no longer effective, that’s, in essence, where you’re thinking about starting insulin. At the time of diagnosis, a patient should know that if they live long enough with their diabetes, they’ll ultimately need insulin. But once they’re on 2 treatments, the conversation should be more urgent.“ You’re getting closer to that point, and we may not do it now, but we will do it soon.” And then some people, at diagnosis, if their A1C is particularly high and if they’re symptomatic, would benefit from insulin (even for a short period of time) to get them under control and remove some of the glucose toxicity that might make the oral medications ineffective.

Kenneth Snow, MD, MBA: Unfortunately, we know that there’s a lot of data to support that many providers wait, probably, much too long before initiating both the conversation about insulin and certainly starting the insulin. And part of that clearly reflects on patients’ concern about starting insulin, but part of it also reflects providers’ concern, or at least some providers’ concern, with using insulin in their patients.

Mary Ann Hodorowicz, RDN, MBA, CDE: It’s funny. In the diabetes education world, we call it “insulin resistance” on their part. They’re insulin resistant, and, in starting the insulin, they almost use it, a lot of them, as punishment. They’ll phrase it up as punishment for not adhering to the lifestyle changes. And that’s really unfortunate, because, from my reading, in most cases, especially in type 2, if you live long enough with type 2 and you get beta cell destruction and dysfunction, the need for insulin will be there. But are the providers and the payers going to sync up and allow that to happen?

Zachary Bloomgarden, MD: Another issue, if I could just make 1 more point even though there are so many, is that there’s really a lack of clarity even amongst advisory groups on what our goals should be. There are some groups that have kind of strongly come out and said, “Back off. Glycemic control is not important for this person and this person and that person.” Other groups, such as, for example, the American Association of Clinical Endocrinologists, have weighed in more strongly for aiming for excellent glycemic control in more patients. And the poor practicing clinician, they say, “This group says 8%, and this group says 7%, and this one says 6.5%. I don’t know when I should really push.”

Dennis P. Scanlon, PhD: That’s exactly the point I was going to ask about. With so many different guidelines, and with so many different treatment choices and classes, what’s a clinician to do? How is one to know what the right therapy is to keep up with the evidence? How do you evaluate the quality of the decisions you’re making for your patients?

Robert Gabbay, MD, PhD, FACP: The short answer is, it’s not easy, I guess, particularly for primary care people that deal with diabetes as just 1 of a myriad of different diseases. Although there’s debate on some of this, there is also some consensus around things. And I think the consensus would say, again, going to this idea of being individualized, that people early in the course of the disease should have the tightest control, people much later in the disease with lots of other comorbidities, particularly cardiovascular disease, should have looser recommendations in terms of glycemic control. And then, there are a number of other factors that one uses to help decide where in that range someone is best to be.

In terms of the different medications, unfortunately, there haven’t been lots of randomized trials to be able to say, “Second-line drug, what’s the best choice?” It really ends up being based on logic. There’s concern about hypoglycemia in this patient. Or, do they have a high risk of cardiovascular disease. And some drugs, SGLT2s (sodium-glucose co-transporter 2s) or GLP-1s (glucagon-like peptide-1s) that lower the risk of cardiovascular disease makes that choice better. So, it’s really sort of individualizing based on that patient.

Dennis P. Scanlon, PhD: Dr Snow, with all this variation in heterogeneity, as a payer, how do you address that and handle it?

Kenneth Snow, MD, MBA: It’s quite the challenge to understand what would be the best option for a particular patient. We recognize that the decision of which agent to add, let’s say, second after metformin in those who are metformin candidates, is extraordinarily complex with a lot of nuance. It makes creating any kind of algorithm or recommendation very challenging. What we prefer to look for is whether it makes reasonable sense, and whether folks are achieving good control. Because, ultimately, that’s the most important thing in treating folks with diabetes. What we realize is it probably makes very little difference of what the next step is, if their A1C is 9.1, other than, there needs to be a next step. And, while I agree with Dr Bloomgarden, that there is definitely a lot of confusion at the lower end of what the goal is. I think everyone’s fairly agreed that maybe it’s 6.5%, maybe it’s 7%, maybe it’s 8%. But everyone’s agreed that it should never be 9.5%. And so, that, at least, leads to a consistency saying that in that realm, we really need to do something. And, from a population point of view, those are the folks who are going to have the greatest likelihood of suffering the consequences of poorly controlled diabetes.