Type 2 Diabetes Podcast

How to Care for Older Adults With Type 2 Diabetes

Carol Wysham, MD; Medha N. Munshi, MD


February 01, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Carol Wysham, MD: Hello. I'm Dr Carol Wysham. Welcome to the second season of Medscape's InDiscussion series on type 2 diabetes. Today, we'll be discussing the treatment of older adults with diabetes. First, let me introduce my guest, Dr Medha Munshi. Dr Munshi is the director of the Joslin Geriatric Diabetes Program and staff geriatrician at Beth Israel Deaconess Medical Center in Boston. She is also associate professor of medicine at Harvard Medical School. Welcome to InDiscussion, Medha.

Medha N. Munshi, MD: Thank you very much, Carol. It's a pleasure.

Wysham: I always like to ask each of my guests an icebreaker question. My question to you is this: Given that this field did not exist when you started your practice, how was it that you became interested in taking care of older patients with diabetes?

Munshi: A short answer is that life happened. It was purely by accident that I did my endocrine fellowship first, and for variety of reasons, I ended up doing geriatrics. I really didn't have a great understanding of what geriatric medicine was when I did that in 1999. Once I started taking care of older adults, a lot of my colleagues sent me their patients with diabetes because they knew I had an endocrine background. As I was seeing more and more older adults, I just loved it. I completely enjoy taking care of and talking to these patients. I also realized that there was a difference with how my geriatric patients reacted or responded to what I was doing, and I started looking more deeply into what else was going on. The basic principle of geriatrics is to treat the whole person — not the disease. When I moved to Boston in 2001, Joslin Clinic did not have a geriatrics program. I talked to them and asked if I could work with one of their educators and their patients because I knew this population required a different approach. And their answer was, "We don't have any educators, but why don't you start?" That was the step I needed to get into this area.

Wysham: Very good. Before we discuss treating older patients with diabetes, let's define an older adult. How do you define it?

Munshi: There are so many definitions that have been tried, and I think the easiest one came from Medicare. In the United States, Medicare starts at age 65, and so most people think that 65 is a good age to use. What I've seen, though, is that age as a number is not a good way of defining what aging does, especially in the context of diabetes. When the American Diabetes Association developed the consensus report in 2012, there was a lot of discussion about how we should define the older adults. The thinking was that older adults are like the pediatric age group, right? If you think about people between 11 and 14 years of age, they are so different in how their body is — their physical and mental health. Similarly, in older adults, it is not the number of their age, but how they are. Patient goals and how their diabetes is managed are made more in the context of what else is going on, what their cognitive capacity is, what their physical ability is, and what the other comorbid conditions are that they have. As a geriatric diabetologist, when I am asked when I think people need to be careful with aging, I say that in Western countries and in the United States, it's probably somewhere after 75. That's when I would start looking for any age-related issues, even if the person I'm talking to looks absolutely fantastic.

Wysham: That's very interesting and leads quite well into the next question. What are the challenges that you and other healthcare providers face when dealing with older adults? And specifically, how do we use frailty and life expectancy in how we define our targets for these patients?

Munshi: I think the first obstacle is convincing clinicians that they need to look at the aging population differently. In the world of diabetes, we are very aware that we don't treat a kid like a 40-year-old person. We know that the goals are different and that how a body reacts to medications is different. For a long time, the problem was trying to convince people that you don't treat a 90-year-old like a 40-year-old. The biggest part of this is thinking back to why we treat diabetes. The major focus of managing diabetes is that it causes complications over the long term. When you talk to a 20- or 40-year-old, we are looking at a different time frame vs somebody who is in their eighties or nineties. How we worry about them developing complications is different. The other way diabetes is very different from other chronic conditions is that the safety window is very tight. Think about high blood pressure or cholesterol levels as an example. When you put somebody on medications to lower their numbers, we know that too low of a blood pressure or too low of a cholesterol level can be difficult, but it is not as big of a problem as hypoglycemia. Overtreatment of diabetes then can lead to too low of a blood sugar level, which we have realized in this population can be a much bigger problem because the consequences of hypoglycemia are much worse. I tell my patients that when you are 40 years old and you have hypoglycemia, if you lose your step and fall, you get up and go on with life. For my 85- or 90-year-old patient with hypoglycemia, if they miss a step and fall and get a hip fracture, we have defeated the purpose of managing their diabetes. That's a major problem.

Wysham: I think it's a good time to discuss what you think the targets for glucose control should be in older adults and how you define different levels of treatment targets by their frailty or life expectancy.

Munshi: Again, we have to talk about how hard it is to understand or manage this population because one side of the coin is how variable the patient's clinical picture is. People who have frailty, as you mention — people with cognitive impairment and people who have multiple comorbid conditions that limit their life expectancy on the one side — this makes it difficult for them to do the self-care behaviors they need for diabetes management on the other side. Checking fingerstick readings on a regular basis or giving injections on a regular basis — that is only one side of the story. The other side of the story is a very heterogeneous population in the sense of where you find them. As we think about goal setting, we assess if a patient is living in their own home in a community or in an assisted living facility, and if some supervision is provided where they live. Are they in the hospital, or they have just gone to the rehabilitation facility? Are they in a nursing home? We need to pull all that information together and synthesize it to determine the goals for their diabetes management. And again, the Endocrine Society recently came out with guidelines for setting up these goals. We understand that people who are intact in regard to cognition have the ability to take care of their physical function, and people with fewer comorbidities should have a little tighter control of diabetes because their life expectancy is longer. In that case, we want to prevent complications of long-term diabetes vs the way we manage somebody who has moderate to severe cognitive impairment. If they require a lot of help with their day-to-day activities or they have severe comorbidities, they should have liberal goals. At that level, we do assign numbers and say the A1c should be less than 7.5 in the healthy group, less than 8 in the intermediate group, and less than 8.5 in the poor-health group. I would like the majority of clinicians to think about it in that way, and then shoot for the best goal we can achieve without putting patients at risk for hypoglycemia. That would be my way of thinking.

Wysham: Actually, this brings up another question, and that is the question about whether there should be a floor. If you have an A1c, for instance, in your intermediate group, that is less than 7 — would that mean we should be backing off on treatment?

Munshi: That is a very good point. I know the Endocrine Society tried hard to do that. And it's a good idea, no question about it, but my worry is that there can be too much of a good thing. The problem is that when we assign a higher and lower number, it becomes very difficult to land in that one particular good place. The other issue I really struggle with — and I know it's not an easy thing to talk about — is the accuracy of A1c in the older population. I am really worried about that for a variety of reasons, including other conditions this older population suffers with, like anemia, chronic disease, multiple infections that occur off and on, and even flu or a urinary tract infection that cause hypoglycemia. And then we ask, "Was the A1c too high, or was it anemia or a blood transfusion?" Many conditions that change the red blood cell life-span or turnover can impact the A1c. A1c in a really frail population is not something I am very comfortable putting my money on. The excursions are not identified by A1c either. We have shown in many of our studies over and over again that even people with high A1c, over 8, have had multiple hypoglycemic episodes. These are some of the reasons I worry about making the range of A1c goal very tight from the upper and lower level. The other way of thinking about it, then, is that if someone is on medications that have a low risk of hypoglycemia, their A1c might be easier to manage. If they are on medications that have a high risk of hypoglycemia, then even high A1c doesn't actually take away the risk of hypoglycemia. So, my message would be to look for excursions and to look for a risk of hypoglycemia rather than identifying the tight A1c range.

Wysham: That's very good. I know you use a lot of continuous glucose monitoring (CGM) in your patients. What about using it in patients who are on sulfonylurea or just on basal insulin? Do you use the professional CGM in that setting if you can't get approval for the personal use?

Munshi: I do. In fact, I like professional CGM for almost all my older patients when I'm not sure about their ability to do fingerstick readings or eat consistently, if they might be missing meals, or they are on, as you mentioned, medications that have a high risk of hypoglycemia. The population I see in my geriatric diabetes clinic is more extreme than what you might normally find. In my mind, it's more about people who are much frailer and older. And even if they are on multiple insulin injections, many times they are not able to use real-time CGM accurately or effectively. Having professional CGM gives me an understanding of what their excursions are, what their pattern is, and how I should change my regimen to not only simplify their regimen so they don't have as much burden of therapy, but also identify where there is a risk of hypoglycemia. If somebody is on sulfonylurea and their excursions are high, I might want to change it. If they don't have excursions, then then a lot of times I leave it alone for the time being.

Wysham: Besides being on insulin or sulfonylureas, what are the other risk factors you look for in patients who might be at risk for hypoglycemia or for us overtreating their diabetes?

Munshi: The biggest concern I see as a geriatrician in older adults with type 2 diabetes is cognitive function. The frontal lobe and frontal lobe function are most commonly affected. The frontal lobe controls behaviors such stopping old behaviors, starting new behaviors, and problem-solving. Many times, these patients have very intact memory, which makes it very difficult to identify they're having problems when you sit in front of them. For instance, you can teach them something and ask them to repeat it back. They can completely repeat it back, but then they go home and do not change anything. When they come back, you say, "Didn't we talk about this last time?" and they say, "Oh, yeah, you did. And you know, you told me this, I'm going to do that." They go back home and come back again without any change. That scenario is a big red flag for me because the best part of managing older population is that they actually do what you ask them to do. They understand they are not immortal. When somebody starts not following things you know they would want to, it is a big red flag. Someone who says they occasionally miss their medication is a big red flag for me because that's what they remember. That is how many times they forgot something they don't remember. I worry about someone who has suddenly changed their glycemic control, whether it's too tight or too loose, I really worry about people who live alone in general because I have repeatedly found in my primary care practice that when someone goes to do a home visit, they realize how bad things get for a patient when nobody else knows what is happening. These are some of the scenarios where I would want to do professional CGM to see what's actually going on.

Wysham: That's very helpful. Before we end, I'd like to have a brief conversation about how you change your approach to cardiovascular risk factor management, which is something we've discussed recently. What's your view on that?

Munshi: Any time a new set of medications come in, I'm a little bit of worried about it because, as you know, a majority of the studies still do not take older adults as part of the overall cohort. They're sort of a subpopulation with a different type of analysis. I am a little bit worried about medications like SGLT2 and SLGT1 receptor agonists going backward and if GLP-1 receptor agonists and SGLT2 inhibitors are going to cause too much weight loss. That's another difference in how we worry about older adults — it's not so much about obesity but too much weight loss, urinary tract infections, genital infection, and so forth. What I have seen over past years is that these medications are very well tolerated. And also the data have shown that the impact of using these medications in people with cardiovascular risk or congestive heart failure has a tremendous benefit. I do like to use them for people who are in the right cohort. Again, not having hypoglycemia is a huge benefit. There are some medications, especially SGLT2 inhibitors, where I have had experience with euglycemic diabetic ketoacidosis, and that's very frightening in older adults and in those with type 2 diabetes. So, we need more data. However, I do think that the data as far as cardiovascular risk factors are tremendous, and I'm really excited about that.

Wysham: That's great. We obviously need to be careful about side effects of our other medications, blood pressure lowering, and statins as well, since they can have their own challenges in older folks. I want to thank you for your very good conversation that we've had about this burgeoning area we all have developed interest in, and that is caring for our older adults with diabetes. You've really helped us understand how to look at our older patients as a heterogeneous group, which is going to be helpful in setting specific targets, whether it be for blood pressure or for cardiovascular risk reduction. I want to thank the audience for joining us today. This is Dr Carol Wysham for InDiscussion.

Listen to additional seasons of this podcast.


Type 2 Diabetes Mellitus

Diabetes in Older Adults: A Consensus Report



Treatment of Diabetes in Older Adults: An Endocrine Society Clinical Practice Guideline

Sodium-Glucose Cotransporter 2 Inhibitors: An Overview

Euglycemic Diabetic Ketoacidosis

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.