A Least Absolute Shrinkage and Selection Operator-Derived Predictive Model for Postoperative Respiratory Failure in a Heterogeneous Adult Elective Surgery Patient Population.

BACKGROUND: Postoperative respiratory failure (PRF) is associated with increased hospital charges and worse patient outcomes. Reliable prediction models can help to guide postoperative planning to optimize care, to guide resource allocation, and to foster shared decision-making with patients. RESEARCH QUESTION: Can a predictive model be developed to accurately identify patients at high risk of PRF? STUDY DESIGN AND METHODS: In this single-site proof-of-concept study, we used structured query language to extract, transform, and load electronic health record data from 23,999 consecutive adult patients admitted for elective surgery (2014-2021). Our primary outcome was PRF, defined as mechanical ventilation after surgery of > 48 h. Predictors of interest included demographics, comorbidities, and intraoperative factors. We used logistic regression to build a predictive model and the least absolute shrinkage and selection operator procedure to select variables and to estimate model coefficients. We evaluated model performance using optimism-corrected area under the receiver operating curve and area under the precision-recall curve and calculated sensitivity, specificity, positive and negative predictive values, and Brier scores. RESULTS: Two hundred twenty-five patients (0.94%) demonstrated PRF. The 18-variable predictive model included: operations on the cardiovascular, nervous, digestive, urinary, or musculoskeletal system; surgical specialty orthopedic (nonspine); Medicare or Medicaid (as the primary payer); race unknown; American Society of Anesthesiologists class ≥ III; BMI of 30 to 34.9 kg/m2; anesthesia duration (per hour); net fluid at end of the operation (per liter); median intraoperative FIO2, end title CO2, heart rate, and tidal volume; and intraoperative vasopressor medications. The optimism-corrected area under the receiver operating curve was 0.835 (95% CI,0.808-0.862) and the area under the precision-recall curve was 0.156 (95% CI, 0.105-0.203). INTERPRETATION: This single-center proof-of-concept study demonstrated that a structured query language extract, transform, and load process, based on readily available patient and intraoperative variables, can be used to develop a prediction model for PRF. This PRF prediction model is scalable for multicenter research. Clinical applications include decision support to guide postoperative level of care admission and treatment decisions.

Quality of Life Improvement: A Novel Framework and Approach to Well-Being.

Physician burnout is a complex problem that requires creative solutions. Despite increasing awareness of the importance of systems approaches to address physician well-being, few tools exist for organizations to bridge the gap between well-being theory and practice. As demonstrated during the COVID-19 pandemic, new threats to well-being can arise at any time, necessitating an iterative approach. Here we outline the quality of life improvement (QOLI) approach, a novel framework and approach that incorporates principles of human-centered design (HCD), quality improvement (QI), and implementation science (IS) to address clinician well-being. Additionally, we share our experience using this approach in a large academic anesthesiology department.

Outcomes and risk factors for delayed-onset postoperative respiratory failure: a multi-center case-control study by the University of California Critical Care Research Collaborative (UC3RC).

BACKGROUND: Few interventions are known to reduce the incidence of respiratory failure that occurs following elective surgery (postoperative respiratory failure; PRF). We previously reported risk factors associated with PRF that occurs within the first 5 days after elective surgery (early PRF; E-PRF); however, PRF that occurs six or more days after elective surgery (late PRF; L-PRF) likely represents a different entity. We hypothesized that L-PRF would be associated with worse outcomes and different risk factors than E-PRF. METHODS: This was a retrospective matched case-control study of 59,073 consecutive adult patients admitted for elective non-cardiac and non-pulmonary surgical procedures at one of five University of California academic medical centers between October 2012 and September 2015. We identified patients with L-PRF, confirmed by surgeon and intensivist subject matter expert review, and matched them 1:1 to patients who did not develop PRF (No-PRF) based on hospital, age, and surgical procedure. We then analyzed risk factors and outcomes associated with L-PRF compared to E-PRF and No-PRF. RESULTS: Among 95 patients with L-PRF, 50.5% were female, 71.6% white, 27.4% Hispanic, and 53.7% Medicare recipients; the median age was 63 years (IQR 56, 70). Compared to 95 matched patients with No-PRF and 319 patients who developed E-PRF, L-PRF was associated with higher morbidity and mortality, longer hospital and intensive care unit length of stay, and increased costs. Compared to No-PRF, factors associated with L-PRF included: preexisiting neurologic disease (OR 4.36, 95% CI 1.81-10.46), anesthesia duration per hour (OR 1.22, 95% CI 1.04-1.44), and maximum intraoperative peak inspiratory pressure per cm H20 (OR 1.14, 95% CI 1.06-1.22). CONCLUSIONS: We identified that pre-existing neurologic disease, longer duration of anesthesia, and greater maximum intraoperative peak inspiratory pressures were associated with respiratory failure that developed six or more days after elective surgery in adult patients (L-PRF). Interventions targeting these factors may be worthy of future evaluation.

A Tale of Two Supercities: Early COVID-19 Outcomes Reflect Social Vulnerability in New York City and San Francisco.

In this narrative review, anesthesiologists at 2 large hospital systems in New York City and San Francisco compare early coronavirus disease 2019 (COVID-19) public health outcomes while considering the role played by social vulnerability and relevant approaches in their 2 cities. An iterative search process allowed for a broad review of medical and public policy research, as well as newspaper reports, expert opinion, and multimedia sources, with the goal of exploring the importance of crowding, the labor force, and social identity in pandemic experiences. Related struggles, pitfalls, and successful interventions in both locales are summarized. Although technology in the form of vaccination will likely play an outsize role in the next phase of the pandemic, our review concludes that we must carefully consider how social vulnerabilities have and will continue to inform equitable and effective access to life-saving resources.

Dexmedetomidine or Propofol for Sedation in Mechanically Ventilated Adults with Sepsis.

BACKGROUND: Guidelines currently recommend targeting light sedation with dexmedetomidine or propofol for adults receiving mechanical ventilation. Differences exist between these sedatives in arousability, immunity, and inflammation. Whether they affect outcomes differentially in mechanically ventilated adults with sepsis undergoing light sedation is unknown. METHODS: In a multicenter, double-blind trial, we randomly assigned mechanically ventilated adults with sepsis to receive dexmedetomidine (0.2 to 1.5 µg per kilogram of body weight per hour) or propofol (5 to 50 µg per kilogram per minute), with doses adjusted by bedside nurses to achieve target sedation goals set by clinicians according to the Richmond Agitation-Sedation Scale (RASS, on which scores range from -5 [unresponsive] to +4 [combative]). The primary end point was days alive without delirium or coma during the 14-day intervention period. Secondary end points were ventilator-free days at 28 days, death at 90 days, and age-adjusted total score on the Telephone Interview for Cognitive Status questionnaire (TICS-T; scores range from 0 to 100, with a mean of 50±10 and lower scores indicating worse cognition) at 6 months. RESULTS: Of 432 patients who underwent randomization, 422 were assigned to receive a trial drug and were included in the analyses - 214 patients received dexmedetomidine at a median dose of 0.27 µg per kilogram per hour, and 208 received propofol at a median dose of 10.21 µg per kilogram per minute. The median duration of receipt of the trial drugs was 3.0 days (interquartile range, 2.0 to 6.0), and the median RASS score was -2.0 (interquartile range, -3.0 to -1.0). We found no difference between dexmedetomidine and propofol in the number of days alive without delirium or coma (adjusted median, 10.7 vs. 10.8 days; odds ratio, 0.96; 95% confidence interval [CI], 0.74 to 1.26), ventilator-free days (adjusted median, 23.7 vs. 24.0 days; odds ratio, 0.98; 95% CI, 0.63 to 1.51), death at 90 days (38% vs. 39%; hazard ratio, 1.06; 95% CI, 0.74 to 1.52), or TICS-T score at 6 months (adjusted median score, 40.9 vs. 41.4; odds ratio, 0.94; 95% CI, 0.66 to 1.33). Safety end points were similar in the two groups. CONCLUSIONS: Among mechanically ventilated adults with sepsis who were being treated with recommended light-sedation approaches, outcomes in patients who received dexmedetomidine did not differ from outcomes in those who received propofol. (Funded by the National Institutes of Health; ClinicalTrials.gov number, NCT01739933.).

Risk Factors Associated With Early Postoperative Respiratory Failure: A Matched Case-Control Study.

BACKGROUND: Postoperative respiratory failure is the most common serious postoperative pulmonary complication, yet little is known about factors that can reduce its incidence. We sought to elucidate modifiable factors associated with respiratory failure that developed within the first 5 d after an elective operation. MATERIALS AND METHODS: Matched case-control study of adults who had an operation at five academic medical centers between October 1, 2012 and September 30, 2015. Cases were identified using administrative data and confirmed via chart review by critical care clinicians. Controls were matched 1:1 to cases based on hospital, age, and surgical procedure. RESULTS: Our total sample (n = 638) was 56.4% female, 71.3% white, and had a median age of 62 y (interquartile range 51, 70). Factors associated with early postoperative respiratory failure included male gender (odds ratio [OR] 1.72, 95% confidence interval [CI] 1.12-2.63), American Society of Anesthesiologists class III or greater (OR 2.85, 95% CI 1.74-4.66), greater number of preexisting comorbidities (OR 1.14, 95% CI 1.004-1.30), increased operative duration (OR 1.14, 95% CI 1.06-1.22), increased intraoperative positive end-expiratory pressure (OR 1.23, 95% CI 1.13-1.35) and tidal volume (OR 1.13, 95% CI 1.004-1.27), and greater net fluid balance at 24 h (OR 1.17, 95% CI 1.07-1.28). CONCLUSIONS: We found greater intraoperative ventilator volume and pressure and 24-h fluid balance to be potentially modifiable factors associated with developing early postoperative respiratory failure. Further studies are warranted to independently verify these risk factors, explore their role in development of early postoperative respiratory failure, and potentially evaluate targeted interventions.

Consensus-based perioperative protocols during the COVID-19 pandemic.

OBJECTIVE: During the COVID-19 pandemic, quaternary-care facilities continue to provide care for patients in need of urgent and emergent invasive procedures. Perioperative protocols are needed to streamline care for these patients notwithstanding capacity and resource constraints. METHODS: A multidisciplinary panel was assembled at the University of California, San Francisco, with 26 leaders across 10 academic departments, including 7 department chairpersons, the chief medical officer, the chief operating officer, infection control officers, nursing leaders, and resident house staff champions. An epidemiologist, an ethicist, and a statistician were also consulted. A modified two-round, blinded Delphi method based on 18 agree/disagree statements was used to build consensus. Significant disagreement for each statement was tested using a one-sided exact binomial test against an expected outcome of 95% consensus using a significance threshold of p < 0.05. Final triage protocols were developed with unblinded group-level discussion. RESULTS: Overall, 15 of 18 statements achieved consensus in the first round of the Delphi method; the 3 statements with significant disagreement (p < 0.01) were modified and iteratively resubmitted to the expert panel to achieve consensus. Consensus-based protocols were developed using unblinded multidisciplinary panel discussions. The final algorithms 1) quantified outbreak level, 2) triaged patients based on acuity, 3) provided a checklist for urgent/emergent invasive procedures, and 4) created a novel scoring system for the allocation of personal protective equipment. In particular, the authors modified the American College of Surgeons three-tiered triage system to incorporate more urgent cases, as are often encountered in neurosurgery and spine surgery. CONCLUSIONS: Urgent and emergent invasive procedures need to be performed during the COVID-19 pandemic. The consensus-based protocols in this study may assist healthcare providers to optimize perioperative care during the pandemic.

Postoperative respiratory failure: An update on the validity of the Agency for Healthcare Research and Quality Patient Safety Indicator 11 in an era of clinical documentation improvement programs.

BACKGROUND: Administrative data can be used to identify cases of postoperative respiratory failure (PRF). We aimed to determine if recent changes to the Agency for Healthcare Research and Quality Patient Safety Indicator 11 (PSI 11) and adoption of clinical documentation improvement programs have improved the validity of PSI 11. We also analyzed reasons why PSI 11 was falsely triggered. STUDY DESIGN: Cross-sectional study of all eligible discharges using health record data from five academic medical centers between October 1, 2012 and September 30, 2015. RESULTS: Of 437 flagged records, 434 (99.3%) were accurately coded and 414 (94.7%) represented true clinical PRF. None of the false positive records involved respiratory failure present on admission. Most (78.3%) false positive records required airway protection but did not have respiratory failure. CONCLUSION: The validity of PSI 11 has improved with recent changes to the code criterion and adoption of clinical documentation improvement programs.

Interprofessional Care and Teamwork in the ICU.

OBJECTIVES: We describe the importance of interprofessional care in modern critical care medicine. This review highlights the essential roles played by specific members of the interprofessional care team, including patients and family members, and discusses quality improvement initiatives that require interprofessional collaboration for success. DATA SOURCES: Studies were identified through MEDLINE search using a variety of search phrases related to interprofessional care, critical care provider types, and quality improvement initiatives. Additional articles were identified through a review of the reference lists of identified articles. STUDY SELECTION: Original articles, review articles, and systematic reviews were considered. DATA EXTRACTION: Manuscripts were selected for inclusion based on expert opinion of well-designed or key studies and review articles. DATA SYNTHESIS: "Interprofessional care" refers to care provided by a team of healthcare professionals with overlapping expertise and an appreciation for the unique contribution of other team members as partners in achieving a common goal. A robust body of data supports improvement in patient-level outcomes when care is provided by an interprofessional team. Critical care nurses, advanced practice providers, pharmacists, respiratory care practitioners, rehabilitation specialists, dieticians, social workers, case managers, spiritual care providers, intensivists, and nonintensivist physicians each provide unique expertise and perspectives to patient care, and therefore play an important role in a team that must address the diverse needs of patients and families in the ICU. Engaging patients and families as partners in their healthcare is also critical. Many important ICU quality improvement initiatives require an interprofessional approach, including Awakening and Breathing Coordination, Delirium, Early Exercise/Mobility, and Family Empowerment bundle implementation, interprofessional rounding practices, unit-based quality improvement initiatives, Patient and Family Advisory Councils, end-of-life care, coordinated sedation awakening and spontaneous breathing trials, intrahospital transport, and transitions of care. CONCLUSIONS: A robust body of evidence supports an interprofessional approach as a key component in the provision of high-quality critical care to patients of increasing complexity and with increasingly diverse needs.

Developing a Comprehensive Model of Intensive Care Unit Processes: Concept of Operations.

OBJECTIVES: This study aimed to use a systems engineering approach to improve performance and stakeholder engagement in the intensive care unit to reduce several different patient harms. METHODS: We developed a conceptual framework or concept of operations (ConOps) to analyze different types of harm that included 4 steps as follows: risk assessment, appropriate therapies, monitoring and feedback, as well as patient and family communications. This framework used a transdisciplinary approach to inventory the tasks and work flows required to eliminate 7 common types of harm experienced by patients in the intensive care unit. The inventory gathered both implicit and explicit information about how the system works or should work and converted the information into a detailed specification that clinicians could understand and use. PROTOTYPE CONOPS TO ELIMINATE HARM: Using the ConOps document, we created highly detailed work flow models to reduce harm and offer an example of its application to deep venous thrombosis. In the deep venous thrombosis model, we identified tasks that were synergistic across different types of harm. We will use a system of systems approach to integrate the variety of subsystems and coordinate processes across multiple types of harm to reduce the duplication of tasks. Through this process, we expect to improve efficiency and demonstrate synergistic interactions that ultimately can be applied across the spectrum of potential patient harms and patient locations. CONCLUSIONS: Engineering health care to be highly reliable will first require an understanding of the processes and work flows that comprise patient care. The ConOps strategy provided a framework for building complex systems to reduce patient harm.

Preventing Harm in the ICU-Building a Culture of Safety and Engaging Patients and Families.

OBJECTIVE: Preventing harm remains a persistent challenge in the ICU despite evidence-based practices known to reduce the prevalence of adverse events. This review seeks to describe the critical role of safety culture and patient and family engagement in successful quality improvement initiatives in the ICU. We review the evidence supporting the impact of safety culture and provide practical guidance for those wishing to implement initiatives aimed at improving safety culture and more effectively integrate patients and families in such efforts. DATA SOURCES: Literature review using PubMed including evaluation of key studies assessing large-scale quality improvement efforts in the ICU, impact of safety culture on patient outcomes, methodologies for quality improvement commonly used in healthcare, and patient and family engagement. Print and web-based resources from leading patient safety organizations were also searched. STUDY SELECTION: Our group completed a review of original studies, review articles, book chapters, and recommendations from leading patient safety organizations. DATA EXTRACTION: Our group determined by consensus which resources would best inform this review. DATA SYNTHESIS: A strong safety culture is associated with reduced adverse events, lower mortality rates, and lower costs. Quality improvement efforts have been shown to be more effective and sustainable when paired with a strong safety culture. Different methodologies exist for quality improvement in the ICU; a thoughtful approach to implementation that engages frontline providers and administrative leadership is essential for success. Efforts to substantively include patients and families in the processes of quality improvement work in the ICU should be expanded. CONCLUSIONS: Efforts to establish a culture of safety and meaningfully engage patients and families should form the foundation for all safety interventions in the ICU. This review describes an approach that integrates components of several proven quality improvement methodologies to enhance safety culture in the ICU and highlights opportunities to include patients and families.

An Ethnographic Study of Health Information Technology Use in Three Intensive Care Units.

OBJECTIVES: To identify the impact of a full suite of health information technology (HIT) on the relationships that support safety and quality among intensive care unit (ICU) clinicians. DATA SOURCES: A year-long comparative ethnographic study of three academic ICUs was carried out. A total of 446 hours of observational data was collected in the form of field notes. A subset of these observations-134 hours-was devoted to job-shadowing individual clinicians and conducting a time study of their HIT usage. PRINCIPAL FINDINGS: Significant variation in HIT implementation rates and usage was noted. Average HIT use on the two "high-use" ICUs was 49 percent. On the "low-use" ICU, it was 10 percent. Clinicians on the high-use ICUs experienced "silo" effects with potential safety and quality implications. HIT work was associated with spatial, data, and social silos that separated ICU clinicians from one another and their patients. Situational awareness, communication, and patient satisfaction were negatively affected by this siloing. CONCLUSIONS: HIT has the potential to accentuate social and professional divisions as clinical communications shift from being in-person to electronically mediated. Socio-technically informed usability testing is recommended for those hospitals that have yet to implement HIT. For those hospitals already implementing HIT, we suggest rapid, locally driven qualitative assessments focused on developing solutions to identified gaps between HIT usage patterns and organizational quality goals.

A Typology of ICU Patients and Families from the Clinician Perspective: Toward Improving Communication.

This paper presents an exploratory case study of clinician-patient communications in a specific clinical environment. It describes how intensive care unit (ICU) clinicians' technical and social categorizations of patients and families shape the flow of communication in these acute care settings. Drawing on evidence from a year-long ethnographic study of four ICUs, we develop a typology of patients and families as viewed by the clinicians who care for them. Each type, or category, of patient is associated with differing communication strategies, with compliant patients and families engaged in greater depth. In an era that prioritizes patient engagement through communication for all patients, our findings suggest that ICU teams need to develop new strategies for engaging and communicating with not just compliant patients and families, but those who are difficult as well. We discuss innovative methods for developing such strategies.

Predicting mortality in the intensive care unit: a comparison of the University Health Consortium expected probability of mortality and the Mortality Prediction Model III.

BACKGROUND: Quality benchmarks are increasingly being used to compare the delivery of healthcare, and may affect reimbursement in the future. The University Health Consortium (UHC) expected probability of mortality (EPM) is one such quality benchmark. Although the UHC EPM is used to compare quality across UHC members, it has not been prospectively validated in the critically ill. We aimed to define the performance characteristics of the UHC EPM in the critically ill and compare its ability to predict mortality with the Mortality Prediction Model III (MPM-III). METHODS: The first 100 consecutive adult patients discharged from the hospital (including deaths) each quarter from January 1, 2009 until September 30, 2011 that had an intensive care unit (ICU) stay were included. We assessed model discrimination, calibration, and overall performance, and compared the two models using Bland-Altman plots. RESULTS: Eight hundred ninety-one patients were included. Both the UHC EPM and the MPM-III had excellent performance (Brier score 0.05 and 0.06, respectively). The area under the curve was good for both models (UHC 0.90, MPM-III 0.87, p = 0.28). Goodness of fit was statistically significant for both models (UHC p = 0.002, MPM-III p = 0.0003), but improved with logit transformation (UHC p = 0.41; MPM-III p = 0.07). The Bland-Altman plot showed good agreement at extremes of mortality, but agreement diverged as mortality approached 50 %. CONCLUSIONS: The UHC EPM exhibited excellent overall performance, calibration, and discrimination, and performed similarly to the MPM-III. Correlation between the two models was poor due to divergence when mortality was maximally uncertain.

The Feasibility and Utility of Continuous Sleep Monitoring in Critically Ill Patients Using a Portable Electroencephalography Monitor.

BACKGROUND: Sleep disruption in critically ill adults can result in acute decrements in cognitive function, including delirium, but it is underdiagnosed in the setting of the intensive care unit (ICU). Although sleep stages can be assessed by polysomnography (PSG), acquisition and interpretation of PSG is costly, is labor intensive, is difficult to do over an extended period of time with critically ill patients (multiple days of continuous recording), and may interfere with patient care. In this pilot study, we investigated the feasibility and utility of monitoring sleep in the ICU setting using a portable electroencephalography (EEG) monitor, the SedLine brain monitor. METHODS: We first performed a baseline comparison study of the SedLine brain monitor by comparing its recordings to PSG recorded in a sleep laboratory (n = 3). In a separate patient cohort, we enrolled patients in the ICU who were monitored continuously with the SedLine monitor for sleep disruption (n = 23). In all enrolled patients, we continuously monitored their EEG. The raw EEG was retrieved and sleep stages and arousals were analyzed by a board-certified technologist. Delirium was measured by a trained research nurse using the Confusion Assessment Method developed for the ICU. RESULTS: For all enrolled patients, we continuously monitored their EEGs and were able to retrieve the raw EEGs for analysis of sleep stages. Overall, the SedLine brain monitor was able to differentiate sleep stages, as well as capture arousals and transitions between sleep stages compared with the PSG performed in the sleep laboratory. The percentage agreement was 67% for the wake stage, 77% for the non-rapid eye movement (REM) stage (N1 = 29%, N2 = 88%, and N3 = 6%), and 89% for the REM stage. The overall agreement was measured with the use of weighted kappa, which was 0.61, 95% confidence interval, 0.58 to 0.64. In the ICU study, the mean recording time for the 23 enrolled patients was 19.10 hours. There were several signs indicative of poor-quality sleep, where sleep was distributed throughout the day, with reduced time spent in REM (1.38% ± 2.74% of total sleep time), and stage N3 (2.17% ± 5.53% of total sleep time) coupled with a high arousal index (34.63 ± 19.04 arousals per hour). The occurrence of ICU delirium was not significantly different between patients with and without sleep disruption. CONCLUSIONS: Our results suggest the utility of a portable EEG monitor to measure different sleep stages, transitions, and arousals; however, the accuracy in measuring different sleep stages by the SedLine monitor varies compared with PSG. Our results also support previous findings that sleep is fragmented in critically ill patients. Further research is necessary to develop portable EEG monitors that have higher agreement with PSG.

Interprofessional rhetoric and operational realities: an ethnographic study of rounds in four intensive care units.

Morning interprofessional rounds (MIRs) are used in critical care medicine to improve team-based care and patient outcomes. Given existing evidence of conflict between and dissatisfaction among rounds participants, this study sought to better understand how the operational realities of care delivery in the intensive care unit (ICU) impact the success of MIRs. We conducted a year-long comparative ethnographic study of interprofessional collaboration and patient and family involvement in four ICUs in tertiary academic hospitals in two American cities. The study included 576 h of observation of team interactions, 47 shadowing sessions and 40 clinician interviews. In line with best practices in ethnographic research, data collection and analysis were done iteratively using the constant comparative method. Member check was conducted regularly throughout the project. MIRs were implemented on all units with the explicit goals of improving team-based and patient-centered care. Operational conditions on the units, despite interprofessional commitment and engagement, appeared to thwart ICU teams from achieving these goals. Specifically, time constraints, struggles over space, and conflicts between MIRs' educational and care-plan-development functions all prevented teams from achieving collaboration and patient-involvement. Moreover, physicians' de facto control of rounds often meant that they resembled medical rounds (their historical predecessors), and sidelined other providers' contributions. This study suggests that the MIRs model, as presently practiced, might not be well suited to the provision of team-based, patient-centered care. In the interest of interprofessional collaboration, of the optimization of clinicians' time, of high-quality medical education and of patient-centered care, further research on interprofessional rounds models is needed.