Hi, everyone. I'm very excited to present to you on ICU burnout, a refractory problem by discussing assessment of ICU workload and outcomes. I'm Ankita Agarwal, an assistant professor at Emory University in pulmonary and critical care. I have a research interest in ICU workload and administration with the goal to understand how to optimize care delivery in the intensive care unit. At this point, almost three years out from the start of the COVID-19 pandemic, to many and often to me, the words burnout, mandatory wellness, and resilience can spark a flicker of annoyance similar to when the dog gets into the garbage or a patient's arterial line falls out yet again. It's unavoidable, maybe a waste of time, and can feel like nothing is going to change. And yet, in healthcare, for most or all of us, where we are right now is morally uninhabitable, and we need to find a way out. But tackling burnout can seem like a Sisyphean task, with the rock growing ever heavier, the mountain taller, and no end in sight. So can we even get from there, endlessly rolling the boulder up the mountain, to here, where we stand on top? Our session as a whole focuses on strategies that might get us there, or here, or at least a little closer, and we discuss a few different strategies targeting different drivers of healthcare burnout in the ICU. I'd like to spend the next several minutes or so discussing the roles of institutions and systems processes in mitigating healthcare worker burnout, particularly considering how workload may impact burnout. Much of the language focuses not only on burnout and moral distress, which are two distinct but often mentioned together concepts, but also on resilience and professional fulfillment. Shown here is the Stanford model of professional fulfillment, which describes the framework where well-being and professional fulfillment is driven in part by individuals, i.e. personal resilience, as well as the organization and institution fostering a culture of wellness and efficiency of practice. Culture of wellness is mostly what it sounds like. It reflects the overall work environment, values, and compassion, and support for wellness. Efficiency of practice is what I'm going to be focusing on. This represents the workplace systems, processes, and practices that promote safe, quality, and effective care. In 2019, the National Academies of Sciences, Engineering, and Medicine published a consensus report as a critical follow-up to two landmark reports by the Institute of Medicine, calling for attention to the issue of patient safety and quality of care. This consensus report served to call attention to the safety, health, and well-being of healthcare providers, focusing on themes of healthcare burnout as a major problem, and more importantly, a multifactorial problem with many interacting factors producing imbalances in job demands and resources. The report also highlighted the role of stakeholders in the external environment, including hospital leadership, medical societies, and healthcare delivery organizations, in the realignment of job demands and resources, i.e. work-life balance. The report puts forth a framework to describe the systems aspect of clinician burnout and professional well-being, highlighting that the systems factors and individual mediating factors lead to outcomes of burnout and well-being, which subsequently impact patients, clinicians, workers, healthcare organizations, and ultimately society. The model also recognizes that we have the ability to learn and improve from these consequences. There are three levels the system describes. First, or there are three levels of the system described by the model. First, the inner tier, which is the frontline of care delivery. This includes the care team members, as well as the local physical work environment, technologies, and organizational conditions. The second tier is leadership and management at the organizational level and associated governance and policies. And the third tier describes the overall healthcare industry, the laws and regulations of the industry, and even societal values. These together create work system factors. Other factors, which are unique to each individual, such as personality, coping strategy, resilience, social support, mediate the effect of the work system factors on burnout and professional well-being. The report then goes on to describe potential areas and recommendations to target burnout at each level. Many of the recommendations, especially those related to the first and second tier, are similar to themes and interventions for greater efficiency of practice discussed in the Stanford model. For example, organizational and structural interventions to reduce burnout can include duty hour requirement changes, delivery changes, strategies to reduce workload intensity by changing schedules or staffing, team-based interventions so that everyone is practicing in the role they trained for, and technology-related factors, such as reducing work required by the electronic medical record, use of medical scribes, or simply decreasing email burden. I think it's important to highlight that both the National Academy's consensus report and the Stanford model were published well before the COVID-19 pandemic, and it's probably not news for most attending this session that burnout is not a COVID problem, but certainly it's been exacerbated and perhaps reached a fever pitch during the pandemic. Similar to these reports, recommendation on pandemic, similar to these reports, recommendations on pandemic preparedness and critical care have included a call for expanding the workforce, protecting intensivists, and promoting resilience, all while ensuring adequate care for patients. Sounds simple. Yeah, but I would argue that we were not successfully doing these things even before the pandemic, and now with the worsening resource shortages, increase in burnout, implementing these recommendations after 2020 is even more difficult, especially when we cannot simply add more people to the workforce and or resources. The solution then, as these reports sort of suggest, is to optimize and protect the workforce we do have, and to consider optimizing their workload. One aspect of ICU structure I think ripe for intervention to potentially impact burnout and well-being is workload. Prior work suggests that there is likely a certain amount of work needed for best patient outcomes. In other words, like Goldilocks, not overworked nor underworked. Currently there is so little consensus on optimal ICU workload and how to define it, and there's a lot of variation in ICU structure and organization nationally and globally. Examples of these differences across the nation in ICUs include the number of patients physicians take care of, the size of the ICU or number of available beds, use of advanced practice providers, nurse practitioners, or PAs, 24-hour intensivist coverage, mandatory staffing of ICU patients, biointensivists, and telemedicine or EICU. And most of the prior studies have not shown a consistent benefit of these approaches on staffing models or on ICU patient outcomes. And society guidelines offer little guidance in defining an optimal staffing and structural models. So while we can think about and consider interventions to improve the workplace and ICU staffing, how exactly do we do that? And what proof do we need to know that what we are doing is actually working? It's important to know one, is what we are doing right now actually linked to outcomes of burnout and well-being as well as patient outcomes? And two, if proposed interventions will actually improve these outcomes. As a field that recognizes and the importance of evidence-based practices, it's surprising that the same rigor is not applied to understanding delivery of one of the most important treatments the ICU offers, the providers of ICU care. So while there's increasing attention being paid to ICU structure and staffing models, we need to have a good understanding of how to define and measure ICU work so that we can assess if changing models of ICU work actually provides meaningful changes. It was in this area of limited evidence that we embarked on a small observational study, SWET-ICU, which stands for the Study of Work and the Association of Outcomes in the ICU. We sought to understand if ICU workload, defined as the number of patients a physician was taking care of, was linked to physician burnout. We hypothesized that as the number of patients increased, presence of burnout symptoms would also increase, and we would identify a threshold workload or number of patients above which we would identify a greater prevalence of burnout syndrome. We designed SWET as an observational study and defined workload, as I said, as the number of patients a physician took care of on a single day. We dichotomized workload to high workload, which was greater than 14 patients, and low, which was 14 or less. And we measured burnout using the Mayo Clinic's Wellbeing Index, which directly measures distress, but it is well validated to predict burnout, and it's convenient due to its easy access and short time to complete. We a priori identified other factors that might confound the relationship between workload and burnout using directed acyclic graph theory, and in our final model adjusted for patient and ICU related variables as confounders. We also examined if there was any interaction of physician-related variables on the relationship between workload and burnout. Our primary analysis did not find any difference in the odds of burnout in those with high workload compared to low. Even when adjusted for patient illness severity, number of new patients, number of providers, number of providers on the ICU team, or presence of ICU strain or conflict with patient and families. Our study did identify a lower than expected prevalence of burnout at about 26% in the entire sample. We suspect this is related to selection bias, perhaps analogous to the healthy worker effect, where people with less burnout were more likely to complete the study due to time or interest, as well as sampling physicians likely at times of COVID-19 lulls in their individual ICUs or hospitals. We did identify some significant associations with burnout, including female sex, presence of ICU strain, and a higher median SOFA score predicting a higher prevalence of burnout. While sweat ICU did not show the relationship between the number of patients and burnout that we had hypothesized, our work has led us to ask more questions about understanding workload in the ICU, particularly how to incorporate the entire ICU team, such as nurses, NPs, PAs, trainees, respiratory therapists, when we consider the ICU work, and other aspects of the environment and ICU work, such as the electronic medical record and individual characteristics. We're currently in the process of starting an observational study to examine whether cognitive load, a theory that describes the amount of working memory used for any given task, can serve as a useful framework to understand and define ICU work. The landscape of intensive care medicine has changed significantly in the latter half and first half of the 20th and 21st century, respectively. We have seen dramatic advancement in the management of critically ill conditions, such as sepsis and ARDS, and a significant portion of improvement in patient outcomes can be attributed to improved care bundles and adherence to standard of care practices. It is now time to focus our energies on understanding how to improve outcomes of the ICU team members, and to do so requires individuals, leaders, health care organizations, and society to name this as a priority, and to devote time and resources to the research so that we may understand, implement, and evaluate workplace-based interventions to reduce burnout in the ICU. I'd like to end by highlighting a task force I've had the pleasure on serving this year as a partnership between SCCM and the CDC. The SCCM Wellbeing Task Force has created an ICU toolkit for staff and leaders to be able to identify specific areas that might be driving their own well-being or burnout or distress, and specific areas for potential improvement on an individual, team, and organizational level. And what's best about the toolkit is that not only does it offer an assessment, but also potential solutions in those areas. To me, this is what's needed for ICU burnout. Our leadership and organization to not only say this is important, but back it up with action. Thank you all for listening, and I hope you enjoy the rest of Congress.

ICU burnout

healthcare

Stanford model

workload

interventions