Well, thank you so much for that introduction, and thanks again to all of you for sticking around for this last session. It's been so wonderful so far, and I'm excited to be able to be contributing. So I have no disclosures. So symptoms of post-traumatic stress are a core outcome measure following ICU admission, both for patients and their family caregivers. Post-traumatic stress is common. We know that up to one in three family caregivers will experience clinically significant post-traumatic stress symptoms at some point. And while reducing caregiver stress is a stated priority of the SCCM and other leaders in critical care research, interventions to date trying to address post-traumatic stress, which have mostly focused on improved support with communication and decision-making in the ICU, have shown limited success. So my goals with my WILD grant were to try to fill some key knowledge gaps, building upon the foundation of these other completed studies to help develop effective support interventions for these patients and their families after ICU discharge. And so under this award, I completed two connected research studies I'm going to present the results of here today. The first was a prospective cohort study to measure post-traumatic stress symptom trajectories over the six months after a loved one's ICU admission for family caregivers. And then in part two, we did a qualitative analysis to learn more about how family caregivers define their own outcomes and to try to get a sense of how much caregiver-defined distress and how they perceive their own distress and their own words aligns with the quantitative measures that we so often use to measure and define caregiver outcomes, quantitative measures of things like post-traumatic stress and other psychological symptoms. So for part one, we enrolled patients experiencing acute cardiorespiratory failure as defined pretty broadly by needing only one or more of the following four things, pressures for shock, non-invasive mechanical ventilation for an acute indication, high-flow nasal cannula, or invasive mechanical ventilation. We enrolled those patients along with their primary family caregiver within the first 48 hours of ICU admission, and we performed four assessments over time, shown here on this timeline. The first one was at the time of enrollment, again, early in the admission, hospital discharge, and then three and six months later. And at each of these assessment points, we measured caregiver post-traumatic stress symptoms using the Impact Event Scale Revised, or the IESR, and then also measured key patient and caregiver characteristics and then three and six-month outcomes. After data collection was complete, we performed latent class analysis to identify caregiver post-traumatic stress trajectories over that six months, and then went on to look for trajectory predictors and also the association between caregiver trajectories and three- and six-month patient and caregiver outcomes. So a little bit about our cohort. We enrolled 95 patient caregiver dyads. For the caregivers, mean age is about 54 years, three-quarters were women, 38% of the caregivers were the spouse or the partner, while 35% were the adult child, and then a small collection of parents and other relationships. About a quarter were black and three-quarters were white. For the patients, mean age is about 62 years, 43% were women. I didn't copy the racial characteristics here because the breakdown was almost identical to the caregiver sample, but I'll note that one patient identified as Hispanic ethnicity. 40% of patients were admitted with COVID-19, and then 58% were required in basic mechanical ventilation at the time of enrollment. So this slide here with this figure shows the main set of findings from this portion of the study. And so along the X-axis are the four time points of measurement that I mentioned earlier. The Y-axis shows the mean IESR score or the PTSS symptom severity. That score runs from 0 to 88. Higher scores indicate worse PTSS. So as you can see, starting at the bottom with the red line, about 51% of our cohort of the caregivers belong to what we call the resilient trajectory. So they started with pretty much absent post-traumatic stress symptoms at enrollment and maintained very low symptom levels over the entire six months. Shown in the green line in the middle, about 33% of caregivers belong to the resolving trajectory. They started with pretty high symptom levels. So on the IESR, scores of 33 indicate clinically significant symptoms. So this mean in this middle group was about 30, indicating, again, pretty high symptoms that then resolved over time without intervention, so that by six months, the symptoms were basically absent. But then a small but important part of our cohort, 16% of these caregivers belong to what we call the chronic post-traumatic stress trajectory. So they started with very high symptom levels within the first 48 hours of the patient's ICU admission, and they stayed high the entire time. We looked at trajectory predictors, and so we found that a set of four characteristics, two belonging to the patient, two belonging to the caregiver, best identified caregivers who were belonging to the chronic trajectory. So we found that if the patient had a high severity of illness at the time of enrollment and a good premorbid functional status, combined with the caregiver having a prior history of trauma for themselves and also low resilience, those four things in combination best predicted caregivers belonging to the chronic post-traumatic stress trajectory. And then looking at outcomes, we did not find any statistically significant associations between patient outcomes and caregiver trajectory, but we did find that caregivers in the chronic trajectory suffered from increased challenges at work, so reduced self-perceived efficacy at work and also having to change jobs more often after the loved one's hospital stay. And then also these caregivers in the chronic trajectory had a reduced health-related quality of life, as indicated by lower SF-36 scores. And so while we felt that we generated some results that shed some really important light on the risks and contributors to caregiver post-traumatic stress, we kept coming into this sort of discrepancy over and over that was best summarized by this quote here. So one of our family caregivers in her REDCap survey assessing her post-traumatic stress symptoms said that she typed this, these questionnaires are worded so poorly that I don't think you're going to get valid and reliable results. And so while this was but one quote, this concept came up over and over where these caregivers would be subjectively endorsing tremendous distress, but having really low PTSS scores or vice versa, said they were doing well, but high symptom levels. And so we wanted to try to kind of let caregivers tell us things in their own words and try to get at these discrepancies. And so that brings us to part two, this qualitative analysis. So all caregivers who were in the prior cohort that I just presented, everybody who hung with us for the whole six months and completed all their surveys, were invited to participate in a semi-structured interview. We had 21 caregivers who took us up on the offer and completed an interview. And this was just one-on-one, our study team and the family caregivers. Interviews took about 45 minutes and covered global perceptions of the caregiver's well-being or lack thereof, and then factors that increased or decreased stress all throughout the hospital stay and after. All interviews were coded independently by two coders and all disagreements were resolved. And then we analyzed the data using the framework analysis using a conceptual framework called the Chronic Traumatic Stress Framework. So our kind of main findings from this was that distress was both common and also multidimensional. And so the 21 caregivers interviewed, about half reported that at six months they were feeling quite distressed, whereas the other half approximately reported they were doing well. And then there were two caregivers who really had sort of this ambivalent mixed wellness and distress outcome that couldn't be neatly categorized. We also found that distress was multidimensional, and we found four key dimensions of distress. So I do want to note that all nine of the distressed caregivers did endorse some element of psychological or physical distress symptoms. So there was some post-traumatic stress, anxiety, depression, chronic pain, insomnia, something along those lines. But there were also three other elements of distress these caregivers identified as being really important, and those are shown here. Caregivers felt they had a low capacity for their own self-care. Many of them felt like they were struggling in the caregiving role, struggling to care for their loved one. And then an ongoing sense of life disruption, which was often endorsed as sort of the sense of not getting back to normal. I'm going to show a few key quotes here on the following slides to summarize these outcomes. So in a few cases, caregivers did describe symptoms that sounded a lot like post-traumatic stress. And so this was the mother of a woman who's, excuse me, a mother whose daughter died in the ICU on the ventilator of liver failure and septic shock. And she said, I think that it's kind of like when you get hit really hard and you don't feel it at first, and later the feeling starts to come back. We've had some enlightening, brutal memories of that time. We're getting there, but I still haven't removed her slippers from my living room, and I haven't wrestled her bedroom yet. And so as I pointed out here with the arrows, these statements really do align quite nicely with very typical post-traumatic stress symptoms, things like emotional numbing, flashbacks and avoidance. More often, this distress was sort of more bundled and went well beyond symptoms. And so this is a woman whose husband survived an ICU stay for severe COVID-19, but he came home with lots of physical, cognitive and emotional issues. And she said, I've been doing a lot of emotional eating in step two, so I've gotten fat. That makes me usually feel worse. Just a lot of domino effect things. It affects my eating, my sleeping, my time management relationships. It affects everything right now. Sometimes I get with him and other people more short, pissed off answers and comments. I sleep a lot, but I think that's a sign of the just tired and stressed and exhausted, so I don't take care of myself. And so again, she's endorsing some psychological and physical distress and also just lots more to what she needs help with and what she's going through. And then the other key thing that we picked up on in these interviews was that while a lot of the caregivers did endorse high quality emotional support in the ICU, this support was not enough to override what they were going through after the ICU stay was over. And so one man whose wife died in the ICU said, the staff did what they could, I think, make me feel better. They let me know what to expect. They told me what they were going to do to make my wife comfortable so that she didn't suffer. And that was wonderful. That was great. But it doesn't keep you from suffering. And so to wrap this up and bring these two studies together, we found that one in six family caregivers, so a small but important subgroup of caregivers, experience serious persistent chronic post-traumatic stress at six months. These caregivers can be screened to identify high symptom levels or being at risk for high symptoms in the first 48 hours of ICU admission. And the interventions for these caregivers to reduce post-traumatic stress should incorporate elements of trauma-informed care and resilience building based on our findings that these caregivers have a history of prior trauma and low resilience. But beyond this, we found that about half of caregivers describe an overall sensation of distress at six months that includes, in many cases, but often extends beyond post-traumatic stress. We found that in-ICU support is important but may need to be paired with post-discharge support to improve caregiver outcomes, and that in addition to psychological symptoms like post-traumatic stress, intervention should focus on addressing adaptation to help caregivers adapt to their new normal after an ICU stay and also skill building for patient and self-care. So I will stop there. I have a wonderful group of mentors and collaborators both at UNC and elsewhere, and I thank you all for your time. Thank you.

post-traumatic stress

ICU caregivers

trauma-informed care

resilience-building

quality of life