Skip to content


What the heck is it? Why does it matter? How do we recognize it and if we do recognize it, is there anything we can do about it? On today’s podcast we talk to Linda Fried, Dean of Columbia University’s Mailman School of Public Health and world renown frailty researcher about all things frailty.

We talk to Dr. Fried about how she first got interested in frailty, how we define it including the difference between phenotypic frailty and a “deficit accumulation model’ frailty index”, and how we should think about assessing frailty and managing it.

We also talk with Dr. Fried about how she thinks about resiliency and the analogy put forth by George A. Kuchel in a wonderful article published in the Journal of American Geriatrics Society (JAGS)that uses the Golden Gate Bridge to explain different definitions of frailty as shown below:

Eric: Welcome to the GeriPal Podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: Alex, I am so excited for today. We have a super very special guest with us.

Alex: Super very special guest. We have Linda Fried, who was a geriatrician, a famous geriatrician, one of our celebrities. She formerly was a chair at Hopkins and is now at Columbia, where she’s director of the Robert Butler Aging Center and dean of the School of Public Health. Welcome to the GeriPal Podcast.

Linda: I’m so happy to be here.

Eric: I am excited to talk about frailty. It is something, as a geriatrician, I’m still trying to get my head around, which I feel like I shouldn’t be acknowledging that as a geriatrician. But it is still something I’m trying to get my head around. But before we go into the topic of frailty, Linda, do you have a song request for Alex?

Linda: So I think the song du jour. Certainly the song of the year. It goes back to Bob Dylan, because The Times Are A-Changing.

Alex: (singing)

Linda: Thank you.

Eric: That was awesome, Alex. Nicely done, Alex.

Eric: So, Linda, I’d love to take a big step in time and just think about how you are a guru, like the ‘it’ person when it comes to frailty. How did you get interested in this as a subject?

Linda: I got interested in it because I became a geriatrician. As a newly trained geriatrician, I was told that frailty was the raison d’etre of geriatric medicine. That seemed important. I started reading what there was because I was looking at people who seemed frail to me, and the literature said that I would know it when I saw it, and then everybody else knew it when they saw it. Well, we couldn’t put really a name to it beyond that.

Linda: In fact, the more I saw patients, the more I wasn’t really comfortable that the definitions I saw matched my patients. So there was a lot of … At the time, this was in the ’80s, if you looked at the publications of the National Institute on Aging, they said frailty was the same as ADL disability. Well, I saw people who seemed frail to me, but were not ADL disabled. So that didn’t work for me.

Linda: There were a lot of really innovative work going on to establish geriatric assessment as the leading modality of geriatric care evaluation and care. In fact, I set up and directed the Geriatric Assessment Center of Hopkins, and we’re seeing a lot of very complicated patients, and was deeply interested in world literature that was coming out from David Reuben and many others about evaluating geriatric assessment clinics [inaudible 00:04:22].

Linda: The puzzling thing to me, as I eagerly looked forward to their reporting the results of their randomized trials in geriatric assessment, the thing that was really disturbing was that I knew what we were doing made a difference. I led a large multidisciplinary team. We were seeing highly complicated patients. I thought that our evaluations really mattered in terms of figuring out what was going on with those patients and how to intervene, and the clinical trials were repeatedly coming up negative, negative meaning they said that all that investment of brainpower and resources didn’t matter.

Linda: That concerned me because, again, it didn’t match what I saw clinically. And so, I probably went down a deep rabbit hole of reading everything I could to try and figure out why this was.

Linda: What really emerged for me was that the way people were being selected into geriatric assessment, the way the criteria for who was frail didn’t match from one trial to another, from one center to another, and in fact was a grab bag of everything that could possibly occur with aging.

Linda: So, in general, the criteria for “frailty” was all the things that geriatricians are so important for, multimorbidity, the very old old, disability along a whole continuum. All of those and more were in the grab bag of who was being characterized as frail, and the trials were showing no difference.

Linda: So it doesn’t take a genius to look at that and say, well, maybe we have a targeting problem. If we’re showing no difference then and we’re just using this modality for everybody, maybe we have to learn how to be more sophisticated about saying who this expensive clinical investment really benefits.

Linda: And so, that was my starting point in thinking about frailty, trying to say, well, we need to understand who is frail better so that we can really be more meticulous about matching the clinical intervention to the patient.

Eric: Yeah. First of all, the frailty, I feel like nowadays I see a new paper about frailty every time I open up a JAGS publication. But even going back to … I just pulled up the PubMed numbers, there is less than 100 publications around 2003 around frailty. Now we have like 1500 a year on frailty. So really a hot topic right now.

Eric: But I feel like there’s still this idea … When we’re defining frailty, it’s kind of like the pornography definition, like you know it when you see it. I know there’s a lot of other … Including you, have defined frailty. I wonder if you can give us a big step back. How should we be thinking currently about how we define frailty?

Linda: Well, maybe I can start out by saying how I tried to explore this, because I went in agnostic in the sense that I didn’t know the definition. I just knew that I was seeing patients who met what the late 1980s literature said the concept of frailty was about, which is that they were people who were very vulnerable in the face of stressors, at very high risk of things going wrong.

Linda: Sometimes I looked at them and they passed the visual sniff test, if you will, that you just looked at them and you just knew these were people who were vulnerable, and sometimes you only knew it when a stressor happened like an accident or the flu and everything fell apart. Much more subtle than we would have expected in another patient.

Linda: So I started out without a predisposition as to what the answer on this was. I just knew that there were people who were frail, and I didn’t know why, or whether there was a way to identify it. I went through years of trying to figure out how to get there.

Linda: I used to talk to … Like any geriatrician, you get invited to give lots of talks to people in assisted living facilities and nursing homes. And so, I would always say to them, “Do you know anybody who’s frail?” Everybody always raised their hand in the audience and said, yes, they knew someone who was frail. It was never them. It was somebody else that they knew.

Linda: Then I say, “Well, how do you know they’re frail?” Interestingly, they always said the same things. It wasn’t one thing, but there was a repertoire of answers that older adults in assisted living facilities and nursing homes told them someone was frail. It was that they were thin, they were a weak, they were slow, they were tired, and perhaps they lost weight.

Linda: Now no person would say all of those, but if I was talking to an audience of 50 people, those were the things I would hear, and I would hear it from these older adults who were themselves at risk of frailty over and over and over again, probably hundreds of times, in all the different talks I gave.

Linda: Then we did studies. Some of them I partnered with my fellow Jeff Williamson, and we did a series of studies asking geriatricians across many medical centers what they fought frailty was. Interestingly, what we learned, some of which is published and some of which nobody would ever publish, was that geriatricians didn’t think frailty was the same as disability, even though that was the assumption, that they thought frailty was a distinct entity, that they didn’t think frailty was the same as disease, in fact, and there was no disease that was the same as frailty, but that there were some shared sequelae of some diseases which told a clinician they were frail.

Linda: Interestingly, the sequelae that they identified as giving clues of frailty were consistent with what the people in assisted living facilities said they recognize as frailty.

Linda: So that’s how I started out trying to understand this, was … And then, of course, studying my own patients and trying to see if I could glean some consistent principle. So my consistent principles came out to be pretty aligned with what the older adults were saying in terms of my own recognition.

Linda: So that was where I started. Then I spent, honestly, eight years puzzling over the list that older adults gave me through all those talks, that they kept saying the same five things over and over again, that told one or another that somebody was frail. I kept thinking, “Why on earth are these five things ending up on the same list?”

Linda: I spent one day a week for eight years trying to figure this out. I read everything I could possibly read. Then one day I saw at least what for me started to put it together, which was actually those five things which were isolated things on the list actually have a physiologic explanation of fitting into a clinical vicious cycle, because at that time we knew there were [inaudible 00:13:55] connections between each of those things on that list.

Linda: For example, if people … First of all, we knew that sarcopenia was a core anchoring component of both aging and frailty, that somehow this thing that was first described by Shakespeare of a shrunk shank associated with getting older, that that thing which we now call sarcopenia of aging was a defining attribute of being frail.

Linda: I mean it’s beyond common knowledge to say when people get sarcopenia, they get muscle weakness and they become weaker. When they’re weaker, they actually slow down. And loss of muscle mass also diminishes exercise tolerance and fitness. All of those things actually together predict disability. They’re not the same as disability, but they predict who’s going to become disabled.

Linda: All together, they also predict people cutting back on their physical activity. We actually, even in the late ’80s, there was really important work going on, particularly at Tufts, that show that in the subset of older adults who may well be frail, for some reason, in that situation, people don’t regulate tightly the balance between how much energy they’re putting out in physical activity and how much they’re taking in in their diet. It gets dysregulated.

Linda: We count on that to be homeostatically balanced that we just will automatically not be fluctuating wildly in our weight every day from day to day to day because we tightly regulate this. But in a subset of older adults, the dysregulation gets profound, and even when people have very little physical activity, they’re taking in even less nutritionally.

Eric: And these five key components, so weakness, slow walking speed, low physical activity, low energy or exhaustion, and weight loss, is this what people usually describe as like phenotypic frailty?

Linda: Oh, that’s the conclusion I came to was that not only did geriatricians recognize these elements, one or another of them, as marking people who might fail, but that physiologically they actually are linked. Clinically, it seems like when they had a lot of those things, they were worse off than when they had only one or two.

Linda: So what I did was to then take a deep drive trying to figure out the science that could explain what we were seeing clinically, and in fact ended up with evidence that we’re looking at a phenotypic group of a clinical syndrome. We’re used to thinking about angina as a clinical syndrome. That’s a no-brainer.

Linda: What’s the medical definition of a clinical syndrome? A constellation of symptoms and signs that co-occur in a cluster and, when there’s a cluster present, gives you an indication of a pathophysiologic specificity of something going wrong in the body that results in that cluster. I wasn’t reading that, so I was improvising [laughter]

Linda: That’s basically the medical dictionary definition of a clinical syndrome. In fact, what we’re looking at here is seemingly a cluster of symptoms and signs that, when they co-occur in a group, help mark something going on physiologically in that person which produces that cluster of symptoms and signs.

Linda: What we’ve gone on to do is to show actually that that follows the characteristics of a clinical syndrome, that in fact there’s an underlying physiologic explanation for what’s happening and biologic explanation for what’s happening, which is what you would need to know if you really believe that this phenotype mattered.

Eric: I guess, again, taking another step back, thinking about this definition, I feel like when I also open up that JAGS article, I often see sometimes a collection of different things that people call frailty that doesn’t look like this. It may be like an index of a lot of different lab values and other markers which often looks like, if I go to ePrognosis, a prognostic index, like as all these things that are associated with bad things, and they’re calling it frailty. Is it the accumulation of deficits model, or is there another word that people use for that?

Linda: So I think your description is a fair one. What I’ve been trying to understand, and my collaborators, is what we think is a distinct physiologic process that underpins this phenotype. What you’re talking about is something that is also carrying the name of frailty. Whether that matches what it’s indicating or not I guess is for the viewer to decide, but really derives from what we as geriatricians do in creating a problem list.

Linda: So you have a clinical problem list and they’re counting up the numbers of problems on that list and demonstrating that if people have a lot of problems, they’re probably going to have some poor outcomes. Now the poor outcome that that problem list has been tied to is mortality.

Linda: As a clinician, that doesn’t surprise me that if you have a lot of bad problems, your mortality risk goes up. Just knowing that someone has congestive heart failure, you know they have a five-year mortality rate associated with that. So that’s not so good in itself, and you add in a few other problems, and probably they’re not going to help out the mortalities.

Linda: Is that frailty? I would, as a scientist, call that multimorbidity. As a clinician, I would call that a marker multimorbidity. But there’s a wide variation in what people put into the problem list. Some people put their clinical diagnosis, some clinicians put in the self-reported concerns, some people add in their lab values, some people put in their health behaviors, yes/no, we’re good.

Linda: What you’ll find is that there’s high variation from one person’s problem list to another, and that actually comes out, just as you were suggesting, Eric, in a lot of reports and publications. There are many things that predict the same bad outcomes. You also just made that point.

Linda: Lots of different things predict whether somebody’s at risk of mortality, of dying. That doesn’t mean that they are the same clinically. If you’re at high risk of cancer mortality in five years, at high risk of congestive heart failure mortality in five years, I would venture to guess that the treatment would be different.

Eric: Yeah. I think from … I mean correct me if I’m wrong. So your definition of frailty, the phenotypic definition associated with a lot of bad outcomes, so falls, fractures, hospitalizations, mortality, disability … And I want to go back to a point that you said, is frailty is not disability. They are two different things.

Eric: We actually just had a podcast about CAPABLE study. We talked a lot about what disability is and there’s this mismatch between what somebody can do and what their environment is. Do you want to talk a little bit more how you think about frailty and disability?

Linda: Sure. I’d be pleased to. So there are a lot of ways to talk about it. First of all, I think that we have shown over the years that the likelihood of disability increases with age, but its causes clinically are a lot of chronic diseases which have different fingerprints, if you will, on the kinds of disabilities they cause.

Linda: To be obvious, bad osteoarthritis of the knees is going to cause one kind of disability, immobility, and a stroke may cause a very different kind of disability. If you have both disability from each, the compromise function that might result from the interplay of both kinds of disability could escalate the severity of the outcomes. So chronic diseases have very distinct fingerprints on function, and they are major risk factors for disability.

Linda: We also have shown over the years that frailty distinct from chronic diseases causes disability. Here I’m talking about understanding what I think we’ve learned at least this definition of frailty is, which is that when people have three, four, or five of the five phenotypic criteria, they have deeply diminished physiological function in a way that compromises their reserves, their resilience, and their ability to maintain homeostasis, and, because of those, their ability to bounce back from different kinds of insults.

Linda: Those decreased reserves also contribute to function and disability. In fact, what I and my colleagues have shown, and many others have shown over the years, is that someone who’s frail who goes into the hospital is at very high risk of coming out more functionally compromised, more disabled than they came in compared to someone who goes into the hospital and isn’t frail. Similarly … Anyway, so I’ll stop there. But they are closely related, but not the same.

Eric: So in a way, one way to think about it is you can have somebody who’s frail, but if they don’t have any stressors, they’re not going to have any disability from that. But if you start putting stressors on them, for example, like a hospitalization, then things start falling apart.

Alex: I wanted to build on this precise point here and ask about resilience. You mentioned this word just now. George Kuchel has this wonderful image, which Eric and I are quite partial to because he explains phenotypic frailty, what he calls stochastic frailty or the deficit accumulation model, and resilience as related but distinct concepts, and that conceptual model he uses, this is why Eric and I love this, is the Golden Gate Bridge.

Alex: He talks about phenotypic frailty being loss of the major supports, the pillars, or the major cable that runs along and stochastic frailty as loss of those little support cables that run up and down between the main cable and support the bridge, those individual little pieces. Then resilience as a measure of stress. Oh, Eric’s pulled it up for those of you who are watching on YouTube. We’ll have the link to it in the show notes associated with this podcast.

Alex: Stress is, in the case of the Golden Gate Bridge, wind, traffic, water. In the case of an older adult, as we just discussed, hospitalization or it could be chemotherapy, and that these are each related and distinct conceptually. They also have different uses clinically. Interested in your thoughts and reflections. This is what I use when I’m teaching about frailty and resilience and the relationship between these different models. Interested in your reflections on this.

Linda: So, conceptually, I think this is very useful. I guess I would add some additional thoughts. I actually think that phenotypic frailty is what happens when resilience diminishes past a threshold level, and the person’s physiologic function sinks to a different level. It is a product of the unraveling of resilience. When that reaches a threshold, a person emerges as frail.

Alex: So the relationship between resilience and frailty is that once a person passes some threshold, they’re no longer able to withstand the stressors and they become frail.

Linda: Right. That’s what our data would suggest. You could think about it as the unraveling of the body’s ability to maintain homeostasis. That kind of resilience is not tied to a single organ system. It’s not tied to a single disease. It’s our core vitality as an organism that is built on robustness and resilience, which actually has a lot to do with the structure of that bridge and the fact that it doesn’t collapse. I have come to think about frailty has won over resilience. I hope that communicates.

Alex: Yeah. I’m interested also in your thoughts about … So, as Eric mentioned, there’s been an explosion in frailty research, in large part due to your incredible research and groundbreaking work in this area. I took over … I started as executive editor. I shouldn’t say took over. It sounds like a hostile takeover.

Alex: I started as executive editor at JAGS in January, and I would agree that the most common submission topic is frailty. I wonder what you think about this explosion in frailty research and thinking about some contemporaries who are building some successful careers. I’m thinking of Dae Kim, who’s at Harvard, who’s developed a frailty model that can be derived from Medicare claims, and other people who are assessing frailty clinically as predictors of COVID mortality among older adults. What is your take on the current landscape and multiple applications of frailty for older adults?

Linda: So I think it’s fabulous. I think that it is consistent with what I was taught when I was training to be a geriatrician, which is that frailty is the raison d’etre of geriatric medicine and the heart of the problem that we’re trying to solve for.

Linda: I think the challenge for the field is how to be meticulous about the concept that we’re interrogating, and communicating so that we each understand … You can put the same name on a million things, and that generally doesn’t, past a certain point in the field, help us advance. Mary Tinetti struggled with that in doing falls research, that how do you even define a fall? That was 20 years ago. But it required coming to some agreement at a point about what we all mean when we’re saying this.

Linda: Conceptually, what I have learned is that frailty and disease are two levels of health. Frailty is about our intrinsic vitality and homeostatic capabilities and reserves and resilience. What happens when, with aging, too much of diminishes and we lose our bounce-back ability basically?

Linda: Disease is another level of health or ill-health that is superimposed on that background vitality, and they affect each other. There’s no question about it. But to me they are not the same. If we can distinguish those, then I think … If we agree that that distinction was valid, that’s step one. There may be good reason for people to disagree. But if we find the right distinctions in terms of the processes that affect aging itself and that affect health, then we can at least know which ones we’re talking about with the same label. Now there may be a point when we give them different labels.

Alex: I wanted to ask also … As we mentioned, we had a podcast with one of your former mentees, Sarah Szanton. She was optimistic, that the times they are a-changing, and that in fact we will have CMS-mandated assessment of function payment for services to help people age in place in their homes. I wonder where frailty fits into this. Where should it be assessed clinically, thinking of our audience of clinicians here, and how can they intervene on it?

Alex: I think of frailty, given your conceptual framework, as more of like hypertension. People don’t get symptoms from hypertension generally, unless it’s quite high, but it certainly leads to bad outcomes, and we should treat it. Similarly, where should we be assessing frailty and how should clinicians be treating it? That’s my last question.

Linda: So there was an international consensus conference which published a report in 2013 that said, absolutely, that clinicians should be screening for frailty for everyone over 70. I agree.

Eric: Did they say how?

Linda: Well, they ducked that a little bit. [laughter]

Eric: Because I’ve got to say, probably most people, even most geriatricians, they’re not assessing grip strength. They don’t have one of those hand dynamometers. It would be great if they’re assessing walking speed, but I’m not sure how often that’s happening.

Linda: So there are places that do it routinely. Certainly, my old division at Hopkins does.

Eric: Yeah, I can imagine.

Linda: But it is possible to set that up well and easily so that non-clinician providers do the initial screening.

Eric: How long does a frailty assessment actually … If you’re going to do the frailty assessment, including physical activity, slow walking speed, what’s the time commitment?

Linda: It’s a five-minute time commitment for screening, plus a self-administered questionnaire.

Eric: Yeah. So feasible.

Linda: From my point of view, it’s feasible. I think the next question is what to do with it. I think there are several approaches to that. One is that prevention really matters. At the least, I think prescriptions, so that people who are at risk of frailty or are progressing in terms of showing early signs are engaged in physical activity programs to maintain strength and muscle mass and function. Really important.

Linda: Physical activity is like the magic pill as a model for what happens in the diminution of resilience, because it improves … It’s like taking your car in for a tune up and every single thing gets fixed. Physical activity improves every dimension of what goes wrong in frailty, from upregulating mitochondrial function to lowering inflammation, to improving strength. So that’s a winner. Diet, important with physical activity, not so successful by itself.

Linda: Then we need to talk about treatment. Certainly ruling out diseases that could contribute to frailty and treating them well effectively is important. Distinguishing what’s a disease process from what’s a true frailty matters. But then I think frailty is really critical to understand in the context of moments of risk.

Linda: So if you’re putting somebody in the hospital for elective surgery or for non-elective reasons, how we go about protecting them from functional decline, how we assure that they are not immobilized unnecessarily, how we try and maintain strength and diet and nutrition, and then what our expectations are for physical activity and rehabilitation all, I think, are affected by whether somebody is frail or at risk of being frail.

Linda: It’s going to take longer to bounce back where a person is frail, but unless they have four or five of the frailty criteria, they should be able to bounce back.

Eric: Yeah. I’ve seen a lot of literature, especially around prehab for surgery, and the surgeons seem to really love frailty right now, highlighting that that’s a great place to potentially target assessing for frailty and then deciding what do we want to do about it to either improve the surgical outcomes, to think about does this elective surgery make sense, or even to think about, okay, what if things go wrong? Let’s do some advanced care planning.

Linda: I think we need new medical care support systems in-hospital and after-hospital and better rehab approaches for people who are frail.

Eric: Yeah. I guess my last question for you too is it sounds like it’s not just a zero or a one. It’s not off or on about frailty, that there is a continuum. I read like you have your pre-frail components, you have your frail, so those people who are scoring more than three, and then you mentioned four to five. Tell me more about that four to five on that marker or those five phenotypic traits.

Linda: So this week, Qian-Li Xue published an article that actually explicates exactly what’s going on. But people who have four, and particularly five, frailty criteria are really at very high risk of mortality in short term, the next six months. Particularly those with five criteria are at a falling off point and really it would appear in a pre-death phase. Their mortality risks are quite different than those who have three frailty criteria.

Linda: People with three frailty criteria are less likely, and Tom Gill showed this a number of years ago, to revert back to fewer, less likely than people who have two. But they’re different in terms of their … Their mortality risk is not as imminent as those four and particularly five.

Eric: Yeah. That was like end-stage frailty what we’re dealing with there.

Linda: Right. Look at Qian-Li’s elegant article. You’ll see it’s a breathtaking difference.

Eric: Well, we’ll have a link to that. We want to end … I know we’re running out of time. But, Linda, if you had a magic wand, if you can fix one thing in the healthcare system around frailty, what thing would you do right now?

Linda: I would make it so that frail older adults could have rehab for three times longer than Medicare pays for.

Eric: I love the specifics on that. Well, Linda, times are a-changing. Before we end, maybe we can hear a little bit more, so hopefully that little magic wand will work. Alex?

Alex: (singing)

Linda: Thank you, guys.

Eric: Thank you, Linda. That was wonderful. I learned a lot. I think I got a better hold of my head as far as the concept of frailty. So thank you and thank you for everything that you’ve done for geriatrics and the field.

Alex: Thank you so much.

Linda: Thank you so much. This was a joy.

Eric: And a big thank you to Archstone Foundation, for the continued support and to all of our listeners. Have a wonderful night.

Back To Top