Podcast Episode Transcript: Frailty Index: Predicting length of stay and complications

With L. Gordon Moore, MD, Victoria Sharp, MD, MBA

Dr. Gordon Moore: Hello and welcome to 3M’s Inside Angle Podcast. This is your host, Gordon Moore, and with me today is Dr. Victoria Sharp. Dr. Sharp is a clinical transformation physician consultant with 3M Health Information Systems where she develops patient-centered transformation roadmaps for a variety of health plans, payers, and clinical system partners. In her previous role, she’s been the chief medical officer of an integrated Medicaid managed care organization, clinical professor of urology and family medicine, and chief of staff at the University of Iowa Hospitals and Clinics. Welcome, Dr. Sharp.

Dr. Victoria Sharp: Thank you very much, Dr. Moore.

Dr. Moore: So I want to talk with you about the frailty index, but before we go there, I want to set up why that’s interesting to me. I think a lot about how it is we know something about the people who are coming to us for care and how the different layers of knowledge and information provide context and nuance that allows me to better focus interventions to think about risk factors in where we ought to apply resources to the best possible ability to help people get good outcomes.

For instance, we know that people with chronic conditions, especially when they have multiple chronic conditions, are much more likely to have a tough time in terms of the probability of hospitalization or ending up in the emergency department. That may be due, in some parts of the degree, that people are overwhelmed by their conditions. And the clinical conditions, the diagnoses and their severity, are interesting and useful bits of information, but we layer, on top of that, things like a person’s age. Then we start thinking even more broadly about other predisposing factors that can impact a person’s health trajectory.

And you and I were having a conversation where you talked about the frailty index and how important and useful that is and how it’s intriguing you in your work. So Dr. Sharp, I’d like to hear what is the frailty index and how does it work?

Dr. Sharp: So I first became interested in frailty and the frailty index, how it came to my attention was I was city VA. I’ve been involved in that for many years, but they’re implementing a pre-operative frailty screening and a pre-habilitation program. That’s where I first thought or heard about the frailty score. They use the RAI, which is a risk analysis index. But I also noticed in journals and headlines recently that frailty is becoming something that is becoming more noticed and seems to be applicable in many different areas. I also, on a personal note, I think that I’ve often thought about people or patients that I’ve known that some in their 30s seem very old or frail, and then some in their 90s and very young and agile. And so there’s something more than just their medical condition.

Also, I have a conflict of interest in this in that I have a 93-year-old mother who is very active and she doesn’t seem very frail. So anyway, there are many, many tools. There was a review in 2017 where it identified 26 different structured questionnaires and brief assessments. And so there’s many, many tools that can be used. I decided to look and see, at least for this discussion, maybe two specific tools. So one would be the clinical frailty scale, which it evolved from a Canadian study of health and aging initially in 2005, and it is a scale of one, very fit to seven, severely frail. And then they updated it in 2007 to include severely frail and terminally ill.

This scale is based on a clinician’s clinical judgment, and so it’s used to assess frailty and fitness, and then there’s a visual chart that can be used to assist with this classification. On this scale, they consider a person with a score of five or greater to be frail. Okay. So the second scale that I have interest in is the risk analysis index. It’s a 14-item instrument to measure surgical frailty. On the clinical frailty scale that I previously talked about, it requires a clinician to watch the patient mobilize, ask about physical activity, can they bathe themselves, dress, things like that. Can they go upstairs? Whereas, this risk analysis index, it doesn’t require any physical performance measures, and it takes about one to two minutes to complete.

It can be administered where you give them questionnaire to a patient, or it can be where a health care provider would go through the checklist. So it has 14 items, and in there, a few example would be unintentional weight loss of 10 pounds, history of renal failure, congestive heart failure, poor appetite, memory loss, functional deficits in the past three—and then if they live in some type of setting other than home, such as assisted living ,a skilled nursing facility, and then if they have cancer, things like that, and then just more on their, do they need assistance with mobility eating.

So this can be calculated prospectively using a clinical questionnaire or retrospectively, and it’s been proven. It was initially developed in the VA system, but then it’s been more generalized. So now, it can be used multiple, different, surgical specialties and it can be used in facilities, health care systems outside of the VA system. They use the Surgeons National Surgical Quality Improvement Projects and it is equipped database to help with that.

Dr. Moore: You’re an interesting clinician because you’re both family medicine and a surgeon. And here it is, you’re talking about using this in the surgical context, and why is that important and what do you know from literature about the importance of using a tool like this?

Dr. Sharp: So this risk analysis index was originally developed in VA. What they found was, this was in Omaha, and they found that they were having high mortality rates post-operatively. It was developed there to look at surgical post-op outcomes at 30, 180, and 365 days. Anyway, they’ve done studies and they’ve proven that it makes a difference, and a patient would present to the outpatient clinic and then they do this risk stratification. And they have a specific cutoff of where they would say that the patient is frail and you need to pause. So they have a certain cutoff and then they do a surgical pause.

At that point, they make no promises to the patient, but they do a further risk assessment, and then they have a panel or a board where it may be that the individual surgeon would just need a pause and then they would say, yes, this is in the best interest of the patient. So then they need to fill out a form saying, yes, I have paused, and this is in the best interest of the patient. If for some reason they’re not sure and they need additional help, then there’s a panel which is set up of surgeons, anesthesia, palliative care, a geriatrician and critical care, and then they can present the patient to this group and then they can discuss it and then decide does this make sense. But it always goes back to having the discussion with the patient as well.

So this can be an adjunct, it doesn’t take the place of a surgeon’s clinical judgment, but it can just be a tool that can help in these. And also sometimes, it can be that knowing that the patient is frail, that you may want to do some things to tune them up a little bit or make them stronger before they have surgery so that they will have better outcomes post-operatively.

Dr. Moore: For instance?

Dr. Sharp: Like doing some exercises. Just some strengthening exercises, things like that. And it may be that it’s important to have them see palliative care, because one thing that’s important is what is the patient’s goals with the surgery? What are the patients’ and the surgeons’ goals with performing a surgery? It may be that it is to cure cancer, but it may be a palliative type surgery, but it’s really to get everybody on the same page. In some cases, it would be that the best decision may be no surgery.

Dr. Moore: That’s interesting. So I think about this like a pre-flight checklist where pilots are highly skilled and they know what they’re doing, but still they break out the checklist, every single flight and they go through it just to make sure we have everything in the right order, and this sounds like a pre-op checklist in a sense.

I know that there are other checklists that are used in the OR and the like, but this sounds like this is the frailty is another layer on top of what we know about a person’s medical conditions. Although it sounds like this index also folds those conditions in as well. And so is this idea that a surgeon would do this evaluation or the team would do the evaluation, then a surgeon would pause, the pause, is that a common thing? Is this rare? Is this only in the VA in Omaha? What’s your sense of that?

Dr. Sharp: No. I think that that’s where it started, but I think that it’s really spread. I think it’s spreading throughout the VA system. But there have been many studies to make sure that it’s applicable outside the VA system and also across multiple surgical specialties and that it could be implemented in large health care systems, so there’s a lot of research on it.

Dr. Moore: Yeah. That was also a question I was going to go to, which is, is this specific to certain types of surgeries? I mean, would you do this with a hernia repair or bypass surgery or everything in between?

Dr. Sharp: So I think it’s applicable to multiple specialties within surgery, so different types of operation. I was looking at the literature and it can be used for vascular surgical patients, and descending about repair of abdominal aortic aneurysm surgery, carotid endarterectomy. One thing about the surgical specialties, in some systems, they use the same assessment tool, but in the literature, it seems like that some specialties have been finding that it doesn’t exactly fit their patients. And so vascular surgery, they’ve added or adjusted the screening assessment tool to be more appropriate for their patient population.

So there’s a lot going on, and in the surgical world, each month, I looked at the vascular surgery journals over the last several month, and every month there is an article or two related to frailty in association with surgical outcomes. It’s a very hot topic and really being looked into to see how it can really help with deciding on performing surgeries or how to optimize patients to have the best outcomes.

Dr. Moore: Do you have the sense that there’s a greater probability overtime now that some people would then therefore not opt to have surgery because they’ve gone through this frailty assessment and in discussion with their surgeon, they say, as you were describing before, it’s like, well, the risk is high for me because of how frail I am, or maybe a family member’s weighing in and saying, I think my mom’s in a rocky place and maybe this is not the best idea. Do you have the sense that that pulling back and opting out is happening more?

Dr. Sharp: Yes. I think that taking that pause can have outcomes. I mean, one thing is, should the patient get this operation because they are not going to have the best outcomes in 180 or 365 days, that would be mortality, morbidity. There’s also costs associated with all these type of having surgery or one thing is the preoperative assessment, just through the whole having the surgery, and then if they are frail, then the higher chance that they would need to be hospitalized, maybe be in the intensive care unit, and it definitely affects length of stay, readmissions, and other performance metrics.

Dr. Moore: It’s interesting. We talk about certain people, because of their frailty, are more likely to use intensive care unit beds. And I’m thinking right now, we’re in the midst of a really dramatic surge in COVID, and ICU beds can be scarce or not existed in some institutions at this point. Do you think that this is part of the decision tree as hospitals are thinking about how they screen people and maybe say not now?

Dr. Sharp: There is literature, and in the news, I just saw an article the other day. I mean, I guess this risk analysis index could be used in the pandemic, but what I’ve seen in the literature recently is the clinical frailty scale, which seems to be more related to medicine-type patients compared to surgical. But in the UK, they are using it, so it’s the National Institute for Health and Care Excellence in UK, they have in their COVID-19 rapid guidelines for critical care in adults. So they recommend the use of the clinical frailty scale in patients 65 or older. And it states that the decisions about admission to ICU should be made in the basis of potential for medical benefit. And so it is used there.

There are editorials that—several of them that I read that were saying, oh, my gosh, don’t use that just as the whole source of triaging patients that you need to take a look at some other factors as well. But I think that it is being used as a component of the triage when there’s limited beds, limited treatment options available.

Dr. Moore: Wow. That sounds like that could be certainly a very challenging conversation and maybe scary, as people think, are using this in a way that might deny me care.

Dr. Sharp: Right. And one thing about the clinical frailty scale, there could be inter-observer differences because it is having the patient walk, you’re observing, you’re doing things like that when you are scoring the patient. Compared to the risk—the other one for surgery is more concrete to some extent. Although there is the part about, can they get around and eat and toilet and things like that. But yeah, so there could be inter-observer variability that could affect their ultimate frailty score.

Dr. Moore: Yeah. It’s awful that we would even have to have this conversation and think about that. I hear that it may be used in some way like this in the UK, but that we would have such a scarcity of ICU beds or capacity to treat very sick people that we would have to decide those who are so unlikely to survive, maybe not consuming resources that could be applied to somebody who could benefit long-term from that. So that’s a scary topic. If you don’t mind, I want to pivot to one that’s maybe a little bit less scary.

You’ve been describing this as something that’s really, really useful at a person level, as we expose a person’s frailty and think, how can I deploy resources to help this person get better, I could see how, in the consideration of maybe going to the OR, we could do some things that mitigate some of the factors that improves a person’s score so that they could then proceed forward, or we could bring special resources to bear to assist that person in the perioperative time to help them get better outcomes.

But I’m wondering about, now, aggregating the data and understanding something. We do a lot of work, you and I, and others, HAS, and thinking about populations, segments with special needs and where we can think about deploying resources, and adding a frailty index on top of that, I would think would provide a lot more nuance. Is there anything that you’ve read about the frailty index being used at a population level?

Dr. Sharp: No, I haven’t. But there’s so much in the literature and it’s so fascinating that, right, I haven’t really looked for that, but I have thought about it. And I think that it could be a very interesting component, because I think in a lot of the methodologies and different things that are used currently, that that frailty component is absent, and I think that it could really enhance what has been done on a population health level.

Dr. Moore: Yeah. This reminds me, I know that there are people who have the activities of daily living or independent activities, daily living scale performs because that is used as a threshold for access to certain types of devices or services, do we justify paying for help in the home because of ADL or IADL in capacity. And so it makes me think that, I know that when we have that information on some people, we can layer that on a person’s burden of illness and know much more because as you would expect, a person who has incapacity and IADL or ADLs is much more likely to be frail in this context, and that frailty represents as probability of hospitalization, ED utilization and the like, in terms of outcomes.

But that’s applied because of specific reimbursement for certain types of services. I’m wondering then about the future of a frailty index, do you have a sense that it’s going to spread to all surgery in all instances, and do you think that would even be appropriate, and where do you see the future of this going?

Dr. Sharp: I think that it is spreading, and I think that systems are looking at it because they have shown that it really affects the postoperative outcomes. And so I think that some systems have felt that it’s in their best interest to have this system-wide. So they’re really doing a lot to try to make sure that there’s a hundred percent compliance, and all patients that are screened for surgery, potentially going to have surgery, so across all service lines. So I think that they are looking at that.

The other thing that just came to mind was, one thing about these screening tools there are components of activities, of daily living, that type of thing, but one other thing take to keep in mind is that you’re assessing frailty, but the other component is disability. And so they overlapped to some extent, but also trying to keep them separate as well, so that’s one of the challenges. And I think that there’s still not the perfect assessment tool, and so I think that those continue being researched to try to determine the optimal components for the pre-op frailty assessments and also for—and other type of assessment tools for patients hospitalized with medical conditions as well.

Dr. Moore: Yeah. It occurred to me, as you were describing that, to think about are there problems with these tools in any way that there are limitations and barriers in neither the way the tools work or the way they’re being used that you’ve seen or read about?

Dr. Sharp: Sure. Well, one thing is you have to do the assessment. Depending on how you do that, one thing is who does the assessment? So should it be the nurse, the scheduler, if you have a teaching institution, should it be the resident or should it be the attending physician? So trying to figure out who best to do it. Once you have completed the assessment, you need to make sure that it is accurate, and then you can interpret it or decide whether you need the pause, so just trying to figure out the time. In general, the tools don’t take very long, usually one to two minutes, but still the manpower, you just adding one more thing on.

The other thing is, where does information go, and do you have a system set up that it can go into the electronic medical record? I know that in Epic, that some institutions have within their Epic Systems they have it, so that it can go into their electronic medical record within the VA system, they have that set up, where’s the information accessible and how can you use it are some of the potential limitations. But once you figure those things out, then it can move along and you can use it.

Dr. Moore: Just on the documentation part, have you either experienced or read about it becoming a structured data element, or is it going in as text?

Dr. Sharp: It is structured, so you just check the box. So they would have discrete data elements in there.

Dr. Moore: Being a bit of a data geek and thinking about how that is useful information, that then attracts me to the idea that we could begin to aggregate data over lots of people, both to assessing impacts that I’m guessing that’s a lot of the literature that you’ve been reading about with the frailty index, but then it also lends itself. Because as we know and understand things about people with these extra layers of information, we can then use that information to identify people at risk and then create those resources and target those resources to those individuals as you’ve described at the ad hominem level.

But then, as we see patterns, we can say, for people like this, we should always have that stop, that pause. We should think about how we would deploy resources in the hospital or in the home to help this person who’s extra frail. So that’s really useful information. Do you know, does this information work its way into the code set? Is it coded for any reason?

Dr. Sharp: No, not that I know of. So one thing that I thought was really interesting: There was this preoperative frailty surgical outcome presentation that was presented in 2008, at the Texas American College of Surgeons meeting. One of the things that they said was frailty affects morbidity and mortality, health care resource use, and they found that it was both inpatient and outpatient operations. Additionally, there was inpatient costs associated with frailty, which makes sense.

And they thought that improving the post-op outcomes was important, but it also reduces cost, and they thought that that was a cornerstone to successful strategy to provide value. So they talked about how it could be useful in value-based care. I think that that would be an interesting place to see as it goes along because we are talking about that all the time of trying to have value-based care. And so I think that this could be a component that could be used.

Dr. Moore: So we identify an individual who’s frail and we want to deploy some resources to help them get better outcomes. And sometimes to say, maybe the surgery is a bad idea because of the high risk from your frailty, but the deploying of resources comes with an added cost, which is good, we would think of that as expenditures well applied. And what I’m hoping is that we will begin to see in the literature, some nice studies that look at unintended costs or the consequences of not attending to this information set, not knowing about it and the problems that can arise from that, which would be extended length of stay, unnecessary ICU bed days, applying all sorts of testing and procedures to help a person because of unrecognized or managed frailty as well, and sadly, up to including mortality.

Dr. Moore: Well, Dr. Sharp, this has been a fascinating journey and exploration of frailty index. I’m glad you brought it to my attention, and I am pleased to learn that it sounds like a incredibly valuable thing that’s spreading.

Dr. Sharp: Yes. Thank you very much. And I think we’ll see a lot more about it as days go by in all aspects of medicine and health care.

View Session Spotlight (PDF)