OBJECTIVE:To examine characteristics associated with formal ethics consultation (EC) referral in pediatric extracorporeal membrane oxygenation (ECMO) cases, and document ethical issues presented. DESIGN:Retrospective cohort study using mixed methods. SETTING:Single-center quaternary pediatric hospital. PATIENTS:Patients supported on ECMO (January 2012 to December 2021). INTERVENTIONS:We compared clinical variables among ECMO patients according to the presence of EC. We defined optimal cutoffs for EC based on run duration, ICU length of stay (LOS), and sum of procedures or complications. To identify independent explanatory variables for EC, we used a forward stepwise selection multivariable logistic regression model. EC records were thematically characterized into ethical issues. MEASUREMENTS AND MAIN RESULTS:Of 601 ECMO patients and 225 patients with EC in 10 years, 27 ECMO patients received EC (4.5% of ECMO patients, 12% of all ECs). On univariate analysis, use of EC vs. not was associated with multiple ECMO runs, more complications/procedures, longer ICU LOS and ECMO duration, cardiac admissions, decannulation outcome, and higher mortality. Cutoffs for EC were ICU LOS >52 days, run duration >160 hours, and >6 complications/procedures. Independent associations with EC included these three cutoffs and older age. The model showed good discrimination (area under the curve 0.88 [0.83, 0.93]) and fit. The most common primary ethical issues were related to end-of-life, ECMO discontinuation, and treatment decision-making. Moral distress was cited in 22 of 27 cases (82%). CONCLUSION:EC was used in 4.5% of our pediatric ECMO cases, with most ethical issues related to end-of-life care or ECMO discontinuation. Older age, longer ICU LOS, prolonged runs, and multiple procedures/complications were associated with greater odds for EC requests. These data highlight our single-center experience of ECMO-associated ethical dilemmas. Historical referral patterns may guide a supported decision-making framework. Future work will need to include quality improvement projects for timely EC, with evaluation of impacts on relevant endpoints.

Managing mechanical ventilation is a core component of an intensivist's daily work, and optimizing ventilator management continues to be a major focus of pediatric critical care research (1). Physiologic models of human respiration help illustrate core principles and can guide clinicians in titrating ventilator parameters (2). Such physiologic models, when implemented in software with graphical user interfaces, can give rise to virtual patient simulators, allowing clinicians to test the impact of potential ventilator changes, to suggest ventilator changes, or even to make ventilator changes autonomously (3). However, accurate implementation of such models, particularly for children, has proven challenging (2). Are the models reliable? Do they respond to inputs the way real patients do? In this issue of Pediatric Critical Care Medicine, Pelletier et al (4) highlight these challenges by testing how well one classic physiologic equation, described by Wexler and Lok (5) and originally evaluated in 50 adult patients, performed in a large pediatric cohort. The equation demonstrates that, assuming constant carbon dioxide production (V̇co2), unchanged ratio of dead space to tidal volume (Vd/Vt), and no spontaneous breathing, Paco2 varies inversely with minute ventilation. Pelletier et al (4) examined 484 patients who were mechanically ventilated while under neuromuscular blockade. Among 15,121 arterial blood gas (ABG) values, there was a median prediction error of 0.00 mm Hg (interquartile range, –3.07 to 3.00 mm Hg). However, practically speaking, the authors rightly note that only 68% of measured ABG values were within ± 5 mm Hg of the predicted values. Pelletier et al (4) made a pragmatic choice to include ABG values up to 6 hours after ventilator changes. In their particularly ill patient population—with frequent comorbidities (87% of patients), prolonged neuromuscular blockade (median 4 d), and high mortality (12%)—other changes in patient physiology, including changes in V̇co2, bicarbonate clearance, or alveolar dead space, likely also impact ABG values, independent of the product of respiratory rate and tidal volume (6). Pelletier et al (4) also examined clinicians' ventilatory rate changes in a subset of ABGs in which adjustment of minute ventilation to maintain an acceptable pH was indicated. Of note, the authors constrained the problem space by excluding scenarios where ventilator parameters other than the set rate (e.g., tidal volume) were adjusted. They compared the clinician's change in respiratory rate vs. the change recommended by the equation and used the subsequent blood gas to determine whether the clinician or the calculator made the "better" choice. In most scenarios (75% vs. 23%), the authors found that the calculator's recommendation outperformed the clinician's judgment. Furthermore, there was a marked difference when comparing ABGs that required escalation vs. weaning of minute ventilation. When the ABG indicated respiratory acidosis requiring increased minute ventilation, clinicians' rate changes outperformed calculator recommendations for 53% of ABGs. In contrast, when the ABG suggested a wean was indicated, the clinician's choice was better in only 21% of scenarios. This finding is concordant with prior literature that computerized clinical decision support (CDS) systems for mechanical ventilation management are particularly effective for the weaning phase (7). Prior work suggested that these benefits may be achieved because closed-loop systems, such as those reviewed in (7), can effect ventilator changes more frequently and more consistently than humans. The study by Pelletier et al (4) suggests that an additional factor may be that PICU clinicians tend to wean the ventilator in smaller increments than might be physiologically warranted. A key element of well-implemented CDS systems is users' understanding of and faith in the algorithm's recommendations (8). The findings by Pelletier et al (4) could be incorporated into a future CDS system that nudges clinicians toward making larger ventilator weans by showing the mathematical model behind the recommendation and reminding clinicians that in most cases in the study by Pelletier et al (4) the larger wean resulted in a "better" subsequent ABG than the clinicians' intuition. By limiting their analysis to patients receiving neuromuscular blockade, Pelletier et al (4) removed the effects of varying breath-to-breath minute ventilation in spontaneously breathing patients. This constraint is necessary to apply the Wexler and Lok (5) equation but reduces applicability at the bedside. Additional limitations of the current work, also appropriately noted by the authors, include the lack of clinical detail about the patient population and, in particular, whether some patients may have had clinical indications for the clinicians' more conservative weaning choices. For instance, in patients with pulmonary hypertension, where respiratory acidosis may be more poorly tolerated, decreasing the ventilatory rate in smaller increments may be logical. It is important to note that serial ABGs in patients are not statistically independent. Indeed, a future computerized CDS system could use modern machine learning methods like reinforcement learning (9) to adjust the system's recommendation for subsequent ventilator changes based on whether a Paco2 after an index ventilator change resulted in an ABG that was higher or lower than expected, much as clinicians do at the bedside. Perhaps the most intriguing application of the work by Pelletier et al (4) may be for informing physiologically based computerized teaching platforms for critical care. Such medical simulators are prevalent (10), and some studies show that they can be effective teaching aids (11). Still, less emphasis has been placed on validating the physiologic accuracy of the models underlying these teaching tools (12). The authors provide a link to a prototype teaching platform that they designed and are testing. A formal evaluation of the educational effectiveness of the platform will be a valuable contribution. Insightfully, on their website, the authors explain the effects of changes to V̇co2 and Vd/Vt on their estimates of Paco2. Thus, the authors turn limitations into teachable moments. Some may wonder why CDS systems and "serious games" continue to rely on simplified physiologic models when our electronic health records provide a wealth of data on ventilator changes and resulting blood gas values from actual patients. These records, however, provide information only on what clinicians actually did. If clinicians never reduced the respiratory rate by 10, the model will struggle to extrapolate the impact of such a reduction in a future scenario. Physiologic equations like that of Wexler and Lok (5) certainly have their limitations, but the work of Pelletier et al (4) suggests that validating and incorporating such models into our training and CDS systems could start nudging us toward bigger ventilator weans. These bigger weans will generate more data on the effects of such weans for future models and could ultimately shorten the time to liberation from mechanical ventilation.

OBJECTIVE: Online learning activities are used in medical school clinical clerkships, but studies report variable learner utilization. This study investigated the effect on lesson completion and knowledge gains when providing protected time and when making a video-based curriculum mandatory during the pediatric clerkship. METHODS: From March 2019 to March 2020, a multicenter, prospective, randomized trial was conducted at 7 medical schools. Students were randomized by clerkship block to receive or not receive protected time and to mandatory versus optional assignment of a 6-video curriculum. Lesson completion, difference between pre- and post-clerkship knowledge tests, and student experience were assessed. RESULTS: One-hundred and sixty students completed the study. Students given protected time completed more lessons (mean = 4.89 [standard deviation = 2.15] vs 2.7 [2.87]; P < .001) and were more likely to complete all 6 lessons as compared to students without protected time (79.2% vs 39.8%; P < .001), with no difference in lesson completion observed between students in mandatory completion versus optional arms (P = .250). There was no difference in knowledge gains across arms (P = .957), but students who completed all 6 lessons had higher knowledge gains as compared to those who viewed fewer or none (P = .002). Students appreciated protected time, although most did not complete lessons during protected time. Critics of protected time encouraged prioritization of patient-related clinical time and desired better integration into the clerkship. CONCLUSIONS: Protected time may improve utilization of supplemental learning activities but should be integrated to avoid competition with patient care. Optimal provision of protected time warrants further study.

Background Online education has experienced explosive growth, particularly in the wake of the COVID-19 pandemic. We explored the current state of the evidence base for online education targeted towards healthcare professionals working in pediatric intensive care units (PICUs), to report how we are using online education in our field. Materials and Methods We performed a literature review by systematically generating a list of publications indexed in PubMed describing online educational interventions in the PICU, using Medical Subject Header (MeSH)-based search terms and the following inclusion criteria: studies published after 2005 that describe online educational interventions aimed at healthcare professional working in the PICU. We reviewed the full text of all included articles, and summarized the study aims, design, and results. Results Our initial search yielded 1,071 unique articles. After screening abstracts and titles, then full texts, eight articles were included in the review. Many online learning modalities are represented, including websites, self-study modules, videos, videoconferencing, online self-assessment with feedback, virtual patient cases, screen-based simulation, and podcasts. Three studies focused on residents, two studies on nurses, two studies on a multidisciplinary team, and one study on transport nurses and paramedics. Most studies utilized participant surveys to assess satisfaction, and half included pre- and post-intervention multiple-choice question tests. Only one study included a patient-related outcome measure. Conclusions Despite growth in online medical educational intervention research, there are relatively few published studies in pediatric critical care, and only one study evaluated the impact of online learning on patient outcomes. There remain significant opportunities for PICU educators to assess the impact of online educational interventions, especially related to clinician behaviors and patient outcomes.

The COVID-19 global pandemic disrupted healthcare, society, and medical education. Use of online video educational content increased at the onset of the COVID-19 pandemic, across two platforms. This demonstrates the potential of online videos to provide timely information in a scalable fashion, quickly meeting clinical information needs.

Daniel, Dennis1; Zalieckas, Jill1; Vitali, Sally1; Brediger, Steven1; Lehmann, Sonja2; Manning, Mary-Jeanne1; Alexander, Peta3; Dalton, Heidi4; Li, Liza1; Priest, John1; Wolbrink, Traci1 Author Information

Aims & Objectives: Pediatric critical care medicine (PCCM) fellows must provide acute, complex clinical care while addressing numerous knowledge gaps, and so may particularly benefit from online educational support. Limited data illustrate the e-learning behaviors or interests of PCCM fellows specifically. Such information may help guide design and targeting of educational interventions. To explore this, we characterized PCCM fellow usage of OPENPediatrics, a worldwide open-access e-learning resource, from June 2018 through June 2019. Methods: IRB-approved retrospective analysis of automatically collected activity log data from users who self-identified as PCCM fellows. The most popular resources (based on percent of fellows who accessed the resource) are reported. Timestamps were used to calculate the duration of time spent by individual users on the website. Results: 299 fellows from 39 countries were included. The distribution of time spent on the site was bimodal, with 105/299 (35%) of users logging <=60 minutes on the site in the analyzed interval, and 100/299 (33%) logging >=240 minutes (Figure 1). These users accessed an average of 9 resources on the site (range 1-89). Table 1 reports the most frequently accessed resources by this cohort of users.Conclusions: The majority of PCCM fellows spent at least 30 minutes on the site. Over 33% of included fellows spent >=4 hours using the site in one year. Courses and simulators were especially popular. Specific popular items included PCCM fellowship-specific curricula, materials for cardiology and congenital heart disease, and interactive online simulators teaching mechanical ventilation and hemodialysis. Our findings can help shape educational interventions and strategies for PCCM fellows.

Aims & Objectives: Extracorporeal life support (ECLS) is used to keep critically ill patients alive when conventional treatments have failed. Safe and effective delivery of ECLS requires multidisciplinary collaboration, and significant training and education, in order to optimize patient outcomes. Previous reports have described that high-quality, evidence-based and peer-reviewed online educational materials for ECLS are lacking. Leveraging our prior experience in developing online and interactive medical education, we have developed a multidisciplinary solution for education on basic concepts of ECLS. Methods: A multidisciplinary group consisting of intensivists, nurses, surgeons, cardiologists, ECLS-trained respiratory therapists, and pharmacists identified essential domains of ECLS knowledge for each healthcare professional group represented. ECLS physiology modeling was developed with input from clinical experts, published literature, and manufacturer-provided specifications. In collaboration with an application developer, best practices in software development and online learning were applied to user-experience design and software development. Results: We developed a screen-based interactive ECLS simulator (figure 1), centered on high-quality, peer-reviewed visual content (figure 2) supporting text-based, interactive exercises that train clinicians on the basics of ECLS (figure 3). The simulator models patient-circuit interactions in response to user inputs. Case-based interactive exercises allow for experiential learning in management of both common and rare ECLS situations, with individualized feedback provided at the end of each scenario.Conclusions: We have successfully created an online interactive simulator for educating clinicians on the essentials of ECLS. Next steps will include usability testing, validation of the content, investigation of educational efficacy, and deployment on the OPENPediatrics website.

Objectives: Residents are often assigned online learning materials as part of blended learning models, superimposed on other patient care and learning demands. Data that describe the time patterns of when residents interact with online learning materials during the ICU rotation are lacking. We describe resident engagement with assigned online curricula related to time of day and ICU clinical schedules, using website activity data. Design: Prospective cohort study examining curriculum completion data and cross-referencing timestamps for pre- and posttest attempts with resident schedules to determine the hours that they accessed the curriculum and whether or not they were scheduled for clinical duty. Residents at each site were cohorted based on two differing clinical schedules-extended duration (>24 hr) versus shorter (maximum 16 hr) shifts. Setting: Two large academic children's hospitals. Subjects: Pediatric residents rotating in the PICU from July 2013 to June 2017. Interventions: None. Measurements and Main Results: One-hundred and fifty-seven pediatric residents participated in the study. The majority of residents (106/157; 68%) completed the curriculum, with no statistically significant association between overall curriculum completion and schedule cohort at either site. Residents made more test attempts at nighttime between 6 pm and 6 am (1,824/2,828; 64%) regardless of whether they were scheduled for clinical duty. Approximately two thirds of test attempts (1,785/2,828; 63%) occurred when residents were not scheduled to work, regardless of time of day. Forty-two percent of all test attempts (1,199/2,828) occurred between 6 pm and 6 am while off-duty, with 12% (342/2,828) occurring between midnight and 6 am. Conclusions: Residents rotating in the ICU completed online learning materials mainly during nighttime and off-duty hours, including usage between midnight and 6 am while off-duty. Increasing nighttime and off-duty workload may have implications for educational design and trainee wellness, particularly during busy, acute clinical rotations, and warrants further examination.

IMPORTANCE:Online learning is increasingly prevalent throughout all stages of medical education. There is little published literature exploring what motivates healthcare professionals to engage with different types of e-learning content. Learner motivations must be understood in order to design effective educational solutions and to optimize the overall online learning experience.OBJECTIVE:Examine engagement, satisfaction, and motivations of healthcare professionals using OPENPediatrics, an open-access medical e-learning platform.METHODS:Retrospective analysis of online survey data. Users were asked to report engagement and satisfaction with the platform, as well as to select motivations for using different types of content on the site: Courses, Simulators, and World Shared Practice Forum videos.RESULTS:Majority of respondents were physicians and nurses in North America and Europe. Overall satisfaction with the platform was high. Most frequently cited motivations for using Courses and Simulators were: learn basic and in-depth information around topics, and learn how to deliver safer or more effective patient care. For World Shared Practice Forum videos, most commonly cited motivations were: learn in-depth information about a topic, learn the latest advances or developments in an area, and learn how to deliver safer or more effective patient care.INTERPRETATION:We appreciated both commonalities and differences in learning motivations among clinicians accessing different kinds of medical e-learning content. Respondents were consistently motivated to learn in order to deliver safer or more effective patient care, but they reported using different types of educational content depending on whether they were learning basic information versus updating or changing their knowledge.

Background and objectives Despite the increasing prevalence of childhood kidney disease worldwide, there is a shortage of clinicians trained to provide peritoneal dialysis (PD). E-learning technologies may provide a solution to improve knowledge in PD. We describe the development of a virtual PD simulator and report the first 22 months of online usage. Design, setting, participants, & measurements The PD simulator was developed and released on OPENPediatrics in January of 2016. A prospective study of international, multidisciplinary healthcare providers was conducted from January of 2016 through October of 2017. User action data were analyzed with descriptive statistics and linear regression. Paired t tests compared user pre- and post-test scores. User satisfaction was assessed by survey. Results The simulator was accessed by 1066 users in 70 countries. Users spent a median of 35 minutes (interquartile range [IQR] 14-84) in the simulator. Users who completed the structured learning curriculum (n=300) spent a median of 85 minutes (IQR 46-95), and those who completed the entire simulator (n=63) spent a median of 122 minutes (IQR 69-195). Users who completed the simulator were more likely to scroll through text and access the simulator in multiple sessions. The 300 users that completed testing showed statistically significant increases in the post- versus pretest scores, with a mean increase of 36.4 of 100 points, SD 19.9 (95% confidence interval, 34.1 to 38.6, P<0.001). Eighty-seven percent (20 of 23) of survey respondents felt the simulator was relevant to their clinical practice, and 78% (18 of 23) would recommend it to others. Conclusions This is the first reported virtual PD simulator. Increased test scores were observed between pre- and post-tests by clinicians who completed testing, across disciplines, training levels, and resource settings.

Background Usage of online resources by clinicians in training and practice can provide insight into knowledge gaps and inform development of decision support tools. Although online information seeking is often driven by encountered patient problems, the relationship between disease prevalence and search rate has not been previously characterized. Objective This article aimed to (1) identify topics frequently searched by pediatric clinicians using UpToDate (http://www.uptodate.com) and (2) explore the association between disease prevalence rate and search rate using data from the Pediatric Health Information System. Methods We identified the most common search queries and resources most frequently accessed on UpToDate for a cohort of 18 children's hospitals during calendar year 2012. We selected 64 of the most frequently searched diseases and matched ICD-9 data from the PHIS database during the same time period. Using linear regression, we explored the relationship between clinician query rate and disease prevalence rate. Results The hospital cohort submitted 1,228,138 search queries across 592,454 sessions. The majority of search sessions focused on a single search topic. We identified no consistent overall association between disease prevalence and search rates. Diseases where search rate was substantially higher than prevalence rate were often infectious or immune/rheumatologic conditions, involved potentially complex diagnosis or management, and carried risk of significant morbidity or mortality. None of the examined diseases showed a decrease in search rate associated with increased disease prevalence rates. Conclusion This is one of the first medical learning needs assessments to use large-scale, multisite data to identify topics of interest to pediatric clinicians, and to examine the relationship between disease prevalence and search rate for a set of pediatric diseases. Overall, disease search rate did not appear to be associated with hospital disease prevalence rates based on ICD-9 codes. However, some diseases were consistently searched at a higher rate than their prevalence rate; many of these diseases shared common features.

Copyright © 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.

Copyright © 2016 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.