Discovery Grant Recipients

Catherine Chen, MD

2024 Discovery Research Grant Recipient

University of Texas Southwestern Medical Center

Topic: Effect of Texas Senate Bill 8 on Critical Illness Among Pregnant Women

Abstract: In September 2021, Texas became the first state to restrict abortion with the passage of Texas Senate Bill 8 (SB8), which criminalized abortion after detection of fetal cardiac activity. Following the Supreme Court’s Dobbs v Jackson ruling in June 2022, 14 additional states have implemented similar abortion restrictions. Abortion restrictions are expected to result in forced continuation of high-risk pregnancies, unsafe self-managed abortions, and delayed medically necessary abortions, which may culminate in higher incidence of maternal critical illness and mortality. To date, there are no studies that estimate the impact of recent abortion restrictions on maternal critical illness and mortality. Texas’ early adoption of abortion restriction may offer insight into the early effects of the Dobbs ruling.

Our proposed study will quantify the impact of Texas SB8 on the incidence of critical illness among pregnant women in Texas. Using the 2021-2022 Texas Inpatient Public Data Use Files, which contain information about all discharges from Texas hospitals, we propose conducting a difference-in-difference analysis to estimate the causal effect of Texas SB8 on clinical outcomes of hospitalized pregnant women. Control populations will include non-pregnant women and, separately, men, which will be matched to pregnant women on key characteristics that influence illness acuity. The primary outcome will be incidence of critical illness, defined as admission to an intensive care unit with a diagnosis of sepsis or receipt of mechanical ventilation. Secondary outcomes will include in-hospital mortality, hospital length of stay, and hospital charges. The study results will provide an early glimpse at the national effects on maternal health following the implementation of abortion restrictions.

Ankita Agarwal, MD

2023 Discovery Research Grant Recipient

           
Topic: Study of Physician Cognitive Load and Work in the ICU

Abstract: Intensive care units (ICUs) are robustly staffed with skilled professionals who care for patients with life-threatening illnesses and injuries. Currently, there is tremendous variation in ICU staffing models across the United States and the world. Prior work examining ICU staffing has shown mixed results with nighttime staffing, telemedicine use, and 24-hour intensivist presence. Given the expense and limitations of clinical personnel, it follows that providing the optimal staffing model—neither overextended nor overstaffed—should be an important goal for hospital systems, allowing them to develop models to deliver effective and efficient care. Guidelines for ICU staffing typically focus on patient-to-physician ratios and are based on low-quality evidence (Ward 2013). Patient-to-physician ratios likely miss important aspects of physician workload in the ICU and may not provide an accurate measurement of ICU work. In our study, we will assess whether physician cognitive load in the ICU, as measured by the NASA task load index, is associated with physician and ICU characteristics as well as other measures of workload (patient-to-physician ratio). We will qualitatively explore factors that impact physician workload through dedicated, semistructured focus groups. We will explore the association between cognitive load and physician and ICU factors and develop a predictive model to explain variation in cognitive load in similar patient-to-intensivist ratios. The study results will help generate testable hypotheses of areas for intervention to optimize workload and future studies examining the association of cognitive load and patient and physician outcomes.    

Whitney Gannon, APRN, MSN, RN

2022 Discovery Grant Award Recipient

          
Topic: ECMO-Free Trial: A Multicenter Pilot Feasibility Study

Abstract: A dramatic increase in the use of venovenous extracorporeal membrane oxygenation (VV ECMO) for adults with acute respiratory failure has occurred over the past decade, punctuated by the COVID-19 pandemic. Advances in technology, increasing clinical experience, and data from randomized trials have improved the process by which patients are selected for VV ECMO and the manner in which care is delivered. In contrast, the process by which patients are liberated from VV ECMO represents an urgent and important knowledge gap.

While VV ECMO is a potentially lifesaving therapy, it exposes patients to risks, and the provision of ECMO at many centers is hampered by resource limitations. Decannulation from VV ECMO at the earliest and safest possible time would be expected to improve outcomes, optimize resource allocation, and reduce cost. However, approaches to weaning and liberating patients from ECMO are inconsistent within and across centers.

Our research team designed a simple bedside protocol for the evaluation of patients’ readiness for liberation from VV ECMO that was shown to be safe and feasible in a prospective, single-arm study including 385 daily assessments (Gannon WD et al. Chest. 2021;160:1693-1703). Additional data from a large, multicenter, randomized controlled trial are needed to evaluate the effects of a protocolized daily assessment of readiness for liberation from VV ECMO on duration of ECMO and other clinical outcomes, compared to usual care. Before such a large multicenter trial can be conducted, however, a multicenter pilot trial is needed.

This proposed ECMO-free trial is a multicenter, parallel-group, randomized pilot trial comparing a protocolized daily assessment of readiness for liberation from VV ECMO to usual care. All patients who receive VV ECMO at Vanderbilt University Medical Center, Toronto General Hospital, and Hennepin County Medical Center who meet all inclusion criteria and no exclusion criteria will be eligible for participation. Participants will be randomly assigned to receive the ECMO-free protocol daily or usual care. Feasibility outcomes and primary and secondary outcomes specified for the larger trial will be examined and described between groups. Physiologic parameters of respiratory mechanics, surrogates of respiratory effort, and parameters of hemodynamic loading will be collected at distinct time points and evaluated for their ability to predict which patients will be candidates for decannulation.    

Paul E. Wischmeyer, MD, EDIC

2021 Discovery Grant Award Recipient

          
Topic: REmotely Monitored, Mobile-Health High-Intensity Interval Training After Critical Care (REMM-HIIT)

Abstract: The number of adults surviving critical illness is increasing, with over 4 million adults surviving in the United States alone. More than half of patients will experience persistent neuromuscular weakness, exercise intolerance, and cognitive dysfunction. Currently there are few effective treatments available for this constellation of symptoms, known as post-intensive care syndrome (PICS), once patients leave the hospital. Post-discharge exercise rehabilitation has been studied as a potential treatment, yet few studies have demonstrated benefit. Previous studies have included home-based exercise training, but these are limited by the lack of individualized exercise prescription and monitoring. Because of the current global COVID-19 pandemic, delivery of rehabilitation is more difficult than ever, creating a significant public health burden with limited treatment options available for reducing disability and impairment after an intensive care unit (ICU) stay. 

Therefore, our proposed study will evaluate a novel personalized, remotely monitored, and coached aerobic interval exercise intervention in ICU survivors using wearable mobile devices and guided via mobile device direct patient interaction by physical therapists trained to facilitate ICU recovery. The specific aims of this proposal are to 1) evaluate adherence to and feasibility of the exercise intervention, 2) analyze the association of the intervention with exercise capacity and physical function, and 3) analyze the association of the exercise intervention with cognitive function. This multidisciplinary proposal will provide a unique opportunity to build a new collaboration integrating the combined expertise of two institutions (and a team of both experienced and mentored ICU investigators with significant experience in conducting critical care trials in survivors with PICS, using an innovative yet pragmatic approach to rehabilitation. This proposal will set the stage for larger, multicenter National Institutes of Health trial proposals and advance the scientific knowledge of rehabilitation following critical illness.

Brian M. Varisco, MD

2020 Discovery Grant Recipient

Topic: Multi-Omic Approach to Identifying PARDS Endotypes

Abstract: Acute respiratory distress syndrome (ARDS) and pediatric ARDS (PARDS) are severe, rapid- onset lung injury conditions that can arise following surgery, infection, or severe injury. ARDS and PARDS typically require prolonged admission to the intensive care unit and are associated with substantial morbidity, mortality, and medical costs. Very few therapies have been shown to be effective, and researchers have had modest success determining subtypes of ARDS and PARDS that might respond better to different therapies. To better identify the different biological categories of PARDS (termed endotypes), we have undertaken a study using nasal epithelial cell gene expression to categorize patients with PARDS. In diseases such as asthma, lung cancer, and chronic obstructive pulmonary disease, nasal epithelial cell gene expression has been predictive of lung pathology. Preliminary results show that this technique accurately identifies patients who go on to experience an uncomplicated PARDS course and identified two patients with complicated course and different gene expression profiles. We are currently funded to perform this study at our institution. This proposal: 1) expands our study to the Children’s Hospital of Philadelphia (CHOP), 2) integrates findings from both our institution and CHOP with findings of a PARDS peripheral blood transcriptomic endotyping study, 3) generates data necessary for planning future, multisite trials, and 4) investigates how two additional techniques (assay for transposase-accessible chromatin sequencing (ATAC-seq) of nasal brushings and proteomic analysis of expiratory limb filters might complement the mRNA-seq approach. The findings from this study will help us plan a definitive endotyping study that will include adult patients with ARDS and ultimately help clinicians better execute clinical trials in ARDS and PARDS and better care for critically ill patients with ARDS and PARDS.

Amelia Barwise, MB, BCh, BAO

2019 Discovery Grant Recipient

Topic: Nested Observational Cohort to Evaluate Diagnostic Error and Delay in the Critically Ill

Abstract: Despite the recognition that diagnostic errors and delays are major contributors to preventable deaths and avoidable illness in the United States, little progress has been made in understanding the risk factors, prevalence, and impact on mortality and morbidity outcomes. Although there is a considerable body of literature outlining contributing causes, an effective strategy leading to a meaningful reduction in diagnostic error and delay rates has not made its way into practice. The process of reporting errors and near misses remains underdeveloped and lacks standardized measurement tools. Until the identification of diagnostic error improves so that timely inquiry can be performed, reductions in harm from diagnostic error and delay will be difficult to achieve. This proposal combines mixed methods research with comprehensive and systematic analysis of electronic health records (EHRs) to improve the identification of diagnostic error and delay in patients undergoing rapid response team (RRT) evaluations in four structurally and geographically diverse hospitals. We are seeking funds to conduct a nested observational multicenter cohort study to better understand diagnostic error and delay, a significant and understudied problem in healthcare. Using a previously retrospectively validated taxonomy-based EHR approach that successfully identifies diagnostic error and delay, we will prospectively screen hospitalized patients, triggering RRT activation. This enriched population has a high rate of diagnostic error and delay (5%-20%), according to our preliminary data and previously published studies. In parallel with external validation of the EHR approach, we will also engage clinicians and key stakeholders in mixed methods research to establish the factors that contribute to diagnostic error and delay and the potential solutions to reduce it. The information gained from this research will inform a future extramural funding application to test potential preventive strategies.    

Lauren Van Scoy, MD

2019 Discovery Grant Recipient

Topic: Preparing Families for the Process of Shared Decision-Making During ICU Family Meetings

Abstract: Many studies have found that families find communication with clinicians in the intensive care unit (ICU) to be deeply flawed. This poor communication has significant health consequences for families, including long-term anxiety, depression, and posttraumatic distress disorder. The conceptual framework of shared decision-making (SDM) has been endorsed by many professional societies as an evidence-based communication strategy, suggesting that clinicians partner with family members during the decision-making process. To date, interventions intended to improve communication during the SDM process have focused on preparing clinicians for their side of the communication. Few interventions, however, prepare families for the stresses associated with decision-making. This is important to address because, only when families are emotionally and psychologically ready to engage with clinicians, can they be effective partners in SDM. This project will evaluate a novel tool for family members of ICU patients to use in advance of their first family meeting. Designed with emotionally overwhelmed family members in mind, the tool is an intentionally simple, open-ended workbook to allow families to acknowledge emotions and reflect before their family meeting so they can be effective partners during SDM. The workbook directly aligns with the SDM conceptual framework and the multiple goals theory of communication. This pilot project uses a mixed methods, randomized controlled design at two academic medical ICUs. Phase 1 involves a qualitative exploration of the tool’s acceptability and usability with family members of ICU patients to inform further refinement. Phase 2 involves a small, randomized controlled trial to evaluate feasibility and effectiveness outcomes, short-term efficacy, and contamination rates in the usual care group. The primary outcome is based on the Impact of Events Scale to assess stress and risk of posttraumatic stress disorder. Secondary outcomes are based on the family satisfaction in the ICU tool and Hospital Anxiety and Depression Scale. We will also begin to explore the relationship of attention to communication goals with the study outcomes. Communication assessments will include the Communication Quality Analysis tool and a determination of the extent to which SDM occurred. When the study aims are met, we will be prepared to apply for a National Institutes of Health R01 research grant to conduct a definitive randomized controlled trial aimed at better preparing family members for the stress of ICU decision-making and improving their long-term psychosocial health.    

Tài Pham, MD, PhD

2018 Discovery Grant Recipient

Topic: Breathing Efforts in ARDS (BEARDS)

Abstract: Acute hypoxemic respiratory failure (AHRF) is one of the most frequent reasons to use invasive mechanical ventilation (MV) in the intensive care unit. Among potentially modifiable MV factors, the interaction between the patient and the ventilator (including the sedation delivered to control it) has raised major interest. MV is indispensable but can induce or worsen lung injury (ventilator-induced lung injury); the interaction between the delivered breaths and the patient’s spontaneous breathing activity can also create what has been recently referred to as patient self-inflicted lung injury. Asynchrony is also strongly associated with mortality. Additionally, respiratory muscle dysfunction potentially induced by MV can impact the weaning process. This multicenter study will focus on assessing patient-ventilator interaction and detecting specific asynchronies such as breath stacking, flow starvation, and reverse triggering, by continuously recording ventilator waveforms, indexes of respiratory drive, and patients’ clinical characteristics and outcomes. Patients with AHRF will be monitored with an esophageal catheter or by using diaphragm electrical activity with a neurally adjusted ventilatory assist catheter. Recordings will be made four times a day (for 30-60 minutes) for at least three days during patients’ first week of AHRF. The same software as is dedicated to dyssynchrony detection will be used to centralize and analyze patients’ interactions with the ventilator settings and patients’ spontaneous activity and measure patients' efforts. Current and/or newly developed algorithms will be validated against a visual analysis of the waveforms. The study’s primary objective is to determine the incidence of each type of asynchrony in intubated hypoxemic patients, associated risk factors, and their relationship to outcomes. We will evaluate the delta between delivered ventilation and lung-protective ventilation across different ranges of severity as a result of these interactions and assess the impact on outcome. This will be the first large study of dyssynchrony epidemiology, informing the need to better control this factor, such as with neuromuscular blockers, implementing new monitoring techniques, titrating sedation on patients with high respiratory drive or using different MV settings. A better understanding of the impact of each type of dyssynchrony will eventually allow clinicians to personalize MV for each patient.