Good morning, everyone. So nice to be here with all of you. And what an amazing talk by Jeremy. I think that was a great segue into starting to think about some of the long-term outcomes among older ICU survivors. And actually, as you could probably see from the description in the program, I am going to focus on actually older adults who've survived a COVID hospitalization. So thinking about what we've all been through in the last few years. I have no conflicts of interest to disclose. I'm funded by the National Institute on Aging through these sources. And the objectives for today's talk are threefold. So first, just to explain briefly the rationale behind studying the outcomes that I'll talk about in this talk, those that matter most to older adults after a COVID hospitalization. I'm going to present some results from a new longitudinal cohort study of older adults who've survived a COVID hospitalization and underwent assessments of patient-centered outcomes. So this will actually be the first time I'm presenting these results. And then I'll discuss these findings in the context of some other studies focused on outcomes among older adults who've been hospitalized with COVID. And so I want us all to just think back to April of 2020. I'm sure we all remember that time vividly. What we were seeing at the time were these alarming reports about the increased risk of death as it was due to age. And this was actually published in the Lancet Infectious Diseases on March 30, 2020. And it's pretty easy to see how the risk of death just increases exponentially with each decade. Yet when we looked at the CDC COVID data tracker at the time, what really stood out to us was that when you looked at the numbers, most older adults will actually survive COVID even at the time back in the first wave of 2020. And at the time, nobody was actually talking about outcomes other than mortality, especially for the age group, which was being disproportionately affected by COVID. And when we look back at the literature, this is a study I often like to mention. This was seminal work by Terry Freed and others showing that of all outcomes, actually it's not mortality that matters most to older adults. It's maintaining functional independence. Yet again, at the time, we weren't thinking at all about these outcomes that mattered most to older adults. And so my colleagues and I spoke, and we decided that we wanted to take action and to study these patient-centered outcomes. And what also struck us was, as we just heard from Jeremy, and as was discussed during the Q&A, we know that even before COVID, that rates of frailty, disability, and multimorbidity, which are known geriatric vulnerability factors for older adults, that these rates are actually increasing and have been increasing over time. This is a study that I was a part of that was published in CHESS last year, where we looked at over 6,000 patients between 1998 and 2015, and showed that the rates of those three geriatric vulnerability factors have been increasing among ICU patients during that timeframe. And we know that those geriatric vulnerability factors from ours and others' work increases the likelihood of poor functional outcomes and other patient-centered outcomes after critical illness. Yet, at the time, we weren't really talking about this. And we all know that, as we talked about, older adults plus COVID units are just a perfect storm. So there is isolation, and especially in the first wave, a lot of isolation, people weren't going in the rooms as much, immobility, many rehab programs were suspended in the first wave, and medication shortages, like we talked about, resulting in an increased use of benzos in the ICU, and we just heard what that can result in. And so we felt there was reason to be concerned about long-term outcomes among older survivors of COVID, not just in the ICU, but throughout the hospital, because we were seeing this throughout the hospital. And so this led us to proposing a study to the National Institute on Aging, which was funded by a COVID supplement in the first wave, called VALIANT, which stands for COVID-19 in Older Adults, a Longitudinal Assessment. There were three multi-PIs, Dr. Cohen, who's a geriatrician at Yale, Dr. Hajduk, who's an epidemiologist, and me, and I did not want to put my picture there, so you can look at the smiley face instead. And we had a number of co-investigators across disciplines, so that everybody could really focus on studies that matter to older adults in their own areas. And our objective was to evaluate the patient-centered outcomes that matter most to older adults, so function, cognition, and freedom from burdens and symptoms primarily, but also a number of other outcomes. And so the cohort profile was just published in JAGS this fall, actually, I think just last month. And just to talk briefly, we included adults age 60 and older. We chose that cutoff rather than 65, because at the time, we weren't sure if we would have enough patients, which we all know was not an issue, but when we were designing this study, we weren't sure. And for older adults, different cutoffs have been used, 60, 65, 70. It had to be the person's index admission with COVID-19, and we enrolled from five different Yale New Haven hospitals. The Yale New Haven Health System Network extends from Connecticut into Rhode Island. And the participant or their proxy had to be able to provide informed consent. We excluded participants who were transitioning to hospice or comfort measures, or who were long-term residents in a nursing home, or who had advanced dementia, and we did enroll English and Spanish-speaking patients. All participants underwent an in-hospital interview that was a very thorough and essentially a comprehensive geriatric assessment. Our research nurses would remote into the patient's room and talk to them either via phone or through video chat, and we gathered data across a broad range of domains and linked these to EMR data through a cleaned data registry of variables relevant to COVID-19, so not just your traditional hospital variables, but also biomarkers and COVID treatments. We did have to perform some manual EMR abstractions, such as the ChartDEL and ChartDEL-ICU for delirium. And we were very careful to enroll a sample diverse in race and ethnicity, and we also gathered data on neighborhood socioeconomic disadvantage as measured by the Area Deprivation Index and individual-level socioeconomic disadvantage. And then all participants underwent repeated assessments in these outcomes at one, three, and six months. These are the characteristics of the Valiant cohort, and just to highlight a few things, you can see that the mean age was 71 years, which I'll talk in a little bit about how the findings I'm going to present compare to the existing COVID literature. But most COVID studies have actually been in the entire population, and only a small subset have focused on older adults, so the mean age here is older than most COVID studies. About half were women. About 37% were of black race or Hispanic ethnicity, and 31% were on Medicaid. This was pretty representative of older adults in general, and I do want to highlight that this is a hospital cohort, of which about 20% were in the ICU, but you could see that nearly half were disabled in one of 15 functional activities, and the median comorbidity count was 3. Because we enrolled from June 2020 through June 2021, most patients did qualify and receive medications such as dexamethasone, which at the time had already been shown to be beneficial in acute COVID. And you'll see there also that about 18% suffered delirium during the hospitalization, as measured by the chart dial. So this is a first look at descriptive outcomes in function, cognition, and symptom burden. This was published in JAGS, and I just want to highlight a few things. So first, you know, when we look here, we actually just wanted to present the number of participants who experienced increased disability after being discharged from COVID, a COVID hospitalization and surviving. And you can see here that the rates are still quite high at one, three, and six months after discharge. I've listed the 15 functional activities below in footnote A. Second, we actually did the MOCA in the hospital, which we knew would be confounded by acute illness, but there's no way to retroactively do a MOCA. I do want to mention, by the way, for the first box for disability, we did have functional status from one month before the admission. So that is a comparison to their true baseline. So here for the MOCA, although you see that of course there is some recovery after the admission, the overall prevalence of cognitive impairment is actually still higher than you would expect for a population of this mean age. The prevalence of cognitive impairment among older adults should not be close to 20% when the mean age is 71. So that was concerning. And then here, although we have 14 symptoms, which I'll show you briefly in a moment, these were the three symptoms with the greatest prevalence, and we were really struck by just how high some of these were, with more than half of participants still experiencing fatigue at six months after discharge, over a third still experiencing dyspnea, and about 15% still experiencing dizziness. We are doing some work. So as I said, each of us are leading different studies, and I'm leading a study where we're going to look at functional trajectories, which we had done a while ago in ICU survivors, but we really wanted to look at it among older adults who survived COVID. And for this, we used the disability measures at one, three, and six months after discharge, and our analytic sample included those who survived and had at least one follow-up interview. And then we wanted to understand which factors in various domains were associated with membership in a given trajectory. This will make more sense in a moment. But we chose this based on the COVID literature and the non-COVID literature. So for example, age and comorbidities were shown early on to be really important in COVID and poor COVID outcomes. And as we talked about, frailty is incredibly important, both in non-COVID, but also especially in COVID. There were lots of studies like COPE and others showing the importance of frailty with outcomes after COVID. Delirium also was shown to be highly prevalent during COVID. And then of course, we wanted to look at severity of illness and pre-admission functional status, just because we know that's obviously tied to post-ICU or post-COVID functional outcomes. So this is what trajectories are looking like for older adults who've survived a COVID hospitalization. We found that actually there were four different functional trajectories, and there were some really striking things to us when we looked at this. So first of all, what really struck me was that 43% of older adults who have been in the hospital for COVID actually did not have any disability before or afterwards. Again, this is a hospital cohort, not an ICU cohort, so just keep that in mind. There was a second trajectory group here, about 16%, that did experience some worsening, but then recovered over the six months after discharge. But then there were two more trajectories that worsened. So this is the pre-admission disability count here, which is just here as a reference. And you can see that these top two groups started out lower and ended up with worse disability one month after their COVID hospitalization, and did not recover over the six months afterwards. This group experienced essentially no recovery at all, and this one kind of bounced around a little bit. But in our model, you could see that it's essentially flat. So this was concerning. And when we looked at the six factors that I mentioned, all six factors, including severity of illness, were associated with the worst functional trajectory. In the moderate trajectory, delirium, frailty, and comorbidities, and pre-ICU functional status were all associated with this. Yet in the most mild group, just pre-admission functional status was. So this was really striking to us because we actually expected age to be associated with all of the trajectories, and it was not. That is consistent with what has been found in the prior non-COVID literature. And delirium really stood out as being an important factor in the worst trajectories. One of my colleagues, Dr. Brianne Minor, who's a geriatrician, is diving into symptom burden. And so this is just some of her preliminary data that she wanted me to share at this meeting, where we're finding that actually symptom trajectories and burden over the six months after the COVID hospitalization remain quite high. And there's more to come in this area. And I do want to just note that our symptoms, we use the Edmonton Symptom Assessment Scale, but modified it to include COVID symptoms, such as loss of smell or loss of taste. And then a mentee of mine, who will actually be presenting at this meeting tomorrow on a different study, she's actually diving into looking at and understanding the association of delirium with functional and cognitive outcomes using these data. This is the first look and the first time, because this is actually a work in progress right now. But if you look just at this figure that she made, you can see here that delirium during the COVID hospitalization really seems to be determining, or at least contributing to the post-COVID disability over the six months after discharge, with a very clear separation in the figure here. These are unadjusted results in this figure, and we'll be working on the multivariable modeling for this. It will probably be a late breaker at ATS. We have finished the cognitive model. And again here, the exposure was the chart DEL. The outcome was cognitive impairment defined as a MOCA less than 22, measured at 1, 3, and 6 months after discharge. And we adjusted for age, sex, low education. The MOCA score measured during hospitalization, hospital length of stay, and follow-up time. And as you can see here, there was a strong association between delirium during the COVID hospitalization and cognitive impairment over the six months after discharge. So more to come here. So how did these findings compare to other studies of COVID in hospitalized older adults? So there was the Gero COVID study, which was published from eight Italian hospitals in 2020. This study measured six ADLs. And when we look at disability, there is a hierarchy of how you lose the ability to function in an area. So most older adults will lose mobility and higher order activities first, and then lose basic ADLs. So since basic ADLs were measured in this study, we felt this was pretty consistent with our findings, although we still have to dive into ADLs versus the higher order activities. There is also, in terms of delirium, the CMAJ study in eight Toronto hospitals from March 2020 to April 2021. The median age was slightly older, so 79 as opposed to 71. But the delirium prevalence was strikingly higher. So over half of the participants in this study were delirious. We thought a little bit about why this might be, and maybe part of that is being driven by their enrollment starting three months earlier. But I'm not sure that can explain the whole thing. And in any case, our delirium prevalence was lower, but we know that delirium is a big problem in COVID. In terms of ICU patients, the COVID study, COVIP study, which was done in Europe through the VIP network, included only ICU patients. So that's a little different from ours. But they also found that many patients did not have any disability prior to ICU admission. And I think that's really what we're seeing in COVID. Like a lot of times in the ICU, we might see patients who are in and out of the hospital or more chronically ill, but of course in COVID, it didn't discriminate and just, you know, you could have patients who weren't disabled at all being hospitalized. So they also had a number of patients who didn't have any disability prior to ICU admission. And when they looked at frailty and pre-ICU disability, it was associated with mortality. I don't think they've published yet on other outcomes. And one other thing I just wanted to highlight since frailty came up is that we really are seeing a variable prevalence of frailty across all of these cohorts. So Gero COVID had a prevalence of about 40% using the 30-item frailty index. The CMAJ study prevalence used the clinical frailty scale was about 62%. That was a different study. In our cohort, about 19% were frail and nearly half were pre-frail that used the free frailty index. And in the COVIP study, which again was all ICU patients, unlike the other studies presented here, the prevalence was about 21% using the CFS. So really all over the place, but we know that frailty, if it is present, is strongly associated with outcomes. If you look at all these data. And then in terms of cognitive impairment, there's definitely less in the literature in terms of long-term outcomes after COVID. But in this Wuhan cohort published, which was enrolled during the spring of 2020, over 3,000 participants and their age matched, well, not age matched, but they matched them to their uninfected spouses and then looked at 12 months. There was a 12.5% incidence of cognitive impairment at 12 months. So our prevalence was a little bit higher at 6 months, but pretty close to what they found. So take-home points. So first, new or increased disability is present in over a third of older survivors of a COVID hospitalization. Frailty, delirium, and comorbidities are associated with both trajectories of worsening disability over the 6 months after discharge. But age and severity of illness were only associated with the worst trajectory. And then again, interestingly, 43% of older COVID survivors actually didn't have any disability. We have found that the prevalence of cognitive impairment, and actually the Wuhan group too, after discharge is higher than expected for age, and burdens and symptoms are prevalent. And delirium in hospital is associated with cognitive impairment over the 6 months after a COVID hospitalization among older adults. So all really cause for concern and things we need to explore more. And then in terms of bigger picture take-home points. So I just want to say it's important that we evaluate patient-centered outcomes among older adults who survive a COVID hospitalization. We can't just focus on mortality. And we already know this from all of our years of studying the post-ICU syndrome, but I thought it was worth mentioning since we have focused on mortality so much over the last 3 years. And then our findings are consistent with the few other studies that have described non-mortality outcomes among older adults after a COVID, though again, most studies have focused on mortality around younger populations. And so with that, I just want to acknowledge our team. All of the people who are underlined here really did a lot of heavy lifting with regard to this study. So I just want to thank them for all their hard work and to thank the whole team here and to thank all of you for your attention. I'll take any questions.

COVID hospitalization

functional independence

disability

cognitive impairment

delirium

patient-centered outcomes