Care Fellows Research Articles

Postgraduate education for advanced practice team members has emerged as a viable way of gaining critical experience needed for specialty practice. Residency and fellowship training programs have been part of the landscape for medical education for decades but have only recently become available to advanced practice team members. Most programs provide a year of structured learning, but there are few published reports of their value. We provide a description and outcome data of our specialty practice fellowship at an academic medical center over a 5-year period. A fellowship advisory board with representation from numerous specialties defined entry criteria, curriculum, and clinical site placement. Participant and preceptor experience was measured monthly to optimize rapid acquisition of skills. The curriculum was modified through rapid cycle quality improvement in response to scores lower than 70%. In total, 30 fellows have participated in the advanced practice nurse practitioner and physician assistant fellowship program between 2019 and 2023, among 8 specialty tracks. Of the participants from 2019 to 2023, 28 fellows graduated with successful program completion. Ninety-six percent were employed in specialty care settings at the time of graduation or within six months post program completion. The overall postfellowship retention rate at the training site was 79%. This outcome data supported the return on investment. This article provides data and insight into postgraduate fellowship programs with 5 years of cohort data, long-term outcomes, retention rates, and return on investment. In addition, it informs educators and employers with insight into programmatic design, evaluation, and outcomes in transition to practice programs.

Gastroesophageal Balloon Tamponade Simulation-Based Mastery Learning Curriculum for Critical Care Fellows.

BackgroundEscape Rooms in medical education have been shown to improve learner satisfaction and reinforce knowledge in pulmonary and critical care medicine, yet virtual escape rooms (VER) as instructional strategies have remained understudied.ObjectiveWe aimed to evaluate participants’ perception and knowledge acquisition after completing a VER as part of a national fellows’ course.MethodsWe designed a prospective study conducted during the 2022 CHEST Pulmonary and Critical Care Medicine Fellow's Virtual course. This was a 2-day back-to-back national course targeting pulmonary and critical care fellows. The course had 5 virtual and 1 VER session, each lasting 50 min. A pre- and post-course knowledge assessment included 9 clinical questions (1 for each of the 5 virtual sessions and 4 for the VER session). A summative score and an individual session score were obtained. Participant's VER perceptions were analyzed using a 5-point Likert scale (1 = strongly disagree and 5 = strongly agree).ResultsFifty-nine (34.5%) of the 165-course participants completed the pre-test, and 29 (17.8%) completed the post-course test. Higher clinical knowledge assessment total score and VER scores were seen in the post-test compared to the pre-test (77.8 ± 14.9 vs 59.6 ± 23.1 P < .001) and (77.3 ± 26.6 vs 62.5 ± 32.5, P < .05) respectively. VER perceptions in the course were positive across participants and sustained after deployment.ConclusionsImplementing an educational VER in a national pulmonary and critical care fellows’ virtual course may promote knowledge acquisition and could potentially lead to a positive perception among participants. Longitudinal studies in various contexts are required to further elucidate direct causality between gamification and learner centric outcomes.

IntroductionThere is a lack of a standardized curriculum for the appropriate use of noninvasive ventilation (NIV), which is readily accessible. Management of NIV is a core competency for physicians training in pulmonary and critical care medicine (PCCM). We present a blended model of instruction that was highly successful in our pilot program.MethodsThe curriculum targeted eight first-year PCCM fellows to assess knowledge and confidence in key competencies of NIV management. After a baseline assessment, fellows engaged in both hands-on instruction and traditional didactics in NIV. Following, fellows were encouraged to use the e-learning modules for enhanced instruction. The modules were designed to cover all major aspects of NIV management and with unique interactive patient cases for both inpatient and outpatient uses of NIV.ResultsEight first-year PCCM fellows completed the training and responded to the posttest assessment 4 weeks later. The average multiple-choice questions (MCQs) score increased from 13.5 ± 3.2 (54.0%) to 18.4 ± 1.6 (73.6%) and was significant (p = 0.004). A Likert assessment of learner confidence also showed significant improvement across several key competency domains.ConclusionThis curriculum represents a successful and novel approach to NIV education, a critical but challenging core competency in pulmonary medicine for physicians training in PCCM.

Neurologists, especially neurointensivists, may be expected to lead cardiac arrest resuscitations. However, neurocritical care (NCC) fellows may face barriers to acquiring the necessary skills and knowledge needed for successful leadership in these scenarios. Whether a simulation course created for one group of learners and applied in a new context (a "shared simulation") could facilitate acquisition of desired outcomes among neurology learners is unclear. In this prospective, pre-post educational intervention study, NCC fellows at 2 centers completed precourse knowledge and confidence assessments and reported barriers to resuscitation leadership. Fellows then led 2 simulated cases of cardiac arrest, initially developed for internal medicine residents with vignettes adapted to better reflect an NCC patient population. Postcourse knowledge and confidence assessments were administered immediately after the intervention and again one to 4 months later. Pre-, immediate post-, and delayed post-confidence and knowledge assessments were compared. Thirteen NCC fellows participated in the study. Limited experience leading a resuscitation and not being preassigned the resuscitation leader role were the most highly cited barriers to leading resuscitations (n = 8/13, 61.5%). Lack of confidence and lack of knowledge were barriers for 38.5% (n = 5/13) and 22.2% (n = 2/9) of participants, respectively. Both confidence and knowledge scores improved on the immediate postassessments (5-point Likert median [IQR] 3.5 [3.1-3.9] vs 4.1 [4.0-4.7], p = 0.005; mean [SD] 69.8% [8.8%] vs 88.5% [7.6%], p = 0.004). No confidence or knowledge decay was observed between the immediate and delayed postcourse assessments (4.1 [4.0-4.7] vs 4.1 [3.2-4.4], p = 0.50; 91.7% [6.8%] vs 81.3% [8.0%], p = 0.08). Shared simulation training improved learner confidence and knowledge in cardiac arrest resuscitation leadership and may yield similar benefits in other simulated scenarios. Low confidence, identified as a barrier to resuscitation leadership for over one-third of NCC fellows, was improved by brief simulation training.

Background: Traditional mannequin-based simulation has been previously demonstrated to be a versatile and effective modality for the education of trainees in the pediatric cardiac intensive care unit (CICU). Unfortunately, creating and maintaining high-fidelity mannequin simulations is resource-intensive. Virtual reality (VR) is an emerging alternative to traditional high-fidelity mannequin simulation for medical education. A previous pilot study by these authors delineated the feasibility of VR in the CICU for the diagnosis and management of patients with hemodynamic compromise. This study sought to compare VR to classic mannequin simulations.Methods: This prospective, single-center study was approved by the Institutional Review Board (IRB) of a quaternary pediatric center and conducted from September to December of 2021. Four common CICU patient scenarios were developed for both VR and mannequins, including supraventricular tachycardia (SVT), postoperative hypotension after a Norwood procedure, pulmonary hypertensive crisis, and apnea with bradycardia. The VR logic was created by the authors, and the programming was completed by SimX (Mountain View, California, USA). Pediatric cardiology and advanced cardiac critical care fellows completed the first two simulations using either VR or mannequins and then crossed over to complete the last two simulations on the other modality. Fellows were assessed on their completion of a critical action checklist for each scenario as well as by an eight-question post-simulation knowledge test. The average number of checklist items completed was calculated for both mannequin and VR-based scenarios, stratified by type of simulation and year of fellowship.Results: A total of 14 fellows completed the study. Overall, the average number of checklist items completed and post-test score increased with each postgraduate year. When comparing the overall number of checklist items completed for all of the simulations between mannequin and VR, there was no significant difference in the means (p = 0.463). On average, fellows completed two to three out of five critical actions, and the average post-test score was 87%. Fellow questionnaires reflected an interest in VR, a lack of previous VR experience, and a general feeling that VR was more immersive than mannequin simulation. The most common complaint by participants was mild nausea.Conclusion: Using VR, multiple common CICU scenarios were designed to accurately reflect complex physiologic changes in real-time, creating an immersive and highly realistic simulation environment. Trainees performed no differently with VR simulations as compared to high-fidelity mannequin simulations in the pediatric CICU. Further investigation is required to demonstrate how VR compares to traditional simulation modalities for long-term knowledge retention.

Health disparities persist among marginalized racial, ethnic, and lower socioeconomic status groups, extending to pediatric intensive care units. Complex interactions between structural forces and provider biases influence patients' risk for critical illness, access, timing, and quality of care. An understanding of how these social influencers of health directly impact health outcomes should be necessary components of pediatric critical care medicine (PCCM) fellowship. Currently, no uniform guidelines exist that provide training in diversity, equity, and inclusion (DEI) in PCCM. We therefore sought to describe the current state of DEI education in PCCM fellowship programs from the program director (PD) perspective. A national survey was conducted among Accreditation Council for Graduate Medical Education-accredited PCCM fellowship PDs to evaluate the state of DEI education. The survey, developed collaboratively and iteratively, encompassed program details, DEI teaching modalities, barriers, and PD perspectives. Statistical analysis was performed using descriptive statistics. Of the 76 PDs surveyed, 47 responded (62% response rate). Although 57% of PDs recognized DEI education as a divisional priority, only 34% had a formal DEI curriculum. Barriers to DEI education included lack of trained faculty, attending time, and engagement. Notably, PDs expressed concerns about faculty competence in delivering DEI education and the scarcity of underrepresented in medicine physicians across trainees and faculty. The findings reveal significant gaps in DEI education within PCCM fellowship programs. Despite recognition of DEI importance, formalized curricula are lacking. The study underscores the necessity for tailored educational interventions and strategies to optimize care.

This study examines trends in application and match rates for the United States Hospice and Palliative Care Fellowship Match, focusing on demographic variations, competitiveness, and match success across applicant types. This retrospective study analyzed data from the National Resident Matching Program for Hospice and Palliative Care fellowships from 2016 to 2024, including applicant demographics and match outcomes. Chi-square tests were used to assess differences and temporal trends were evaluated with Mann–Kendall tests, and statistical significance was set at p < 0.05. Our results were that the number of fellowship programs increased from 122 in 2016 to 185 in 2024 (51.6% increase), with positions rising from 280 to 451 (61.1% increase). Applicants grew by 75.2%, from 259 to 454. US MDs had significantly higher match rates than non-US MDs (86.1% vs. 79.8%, p < 0.001). Female applicants were significantly overrepresented in Hospice and Palliative Care fellowships compared to all fellowships (64.9% vs. 47.0%, p < 0.001), while males were underrepresented (33.4% vs. 51.4%, p < 0.001). A higher proportion of matched applicants were from rural areas compared to all fellowships (16.0% vs. 13.1%, p = 0.040). In conclusion, the number of Hospice and Palliative Care fellowship programs, positions, and applicants has increased. Female applicants, compared to males, and White applicants, relative to their representation in all fellowships, accounted for a larger proportion of applications and had higher match rates. Heterosexual applicants were less likely to pursue Hospice and Palliative Care fellowships compared to all specialties.

Point-of-care ultrasound (POCUS) has become a central component in the assessment and management of critically ill patients. Despite its widespread application, there is no standardized curriculum across critical care fellowships. Previous studies have examined the efficacy of ultrasound curricula in enhancing provider comfort and expertise with POCUS. However, these educational interventions are typically limited to specific critical care subspecialties and do not evaluate the longitudinal impact of POCUS training.In this study, we assessed the impact of a longitudinal pulmonary ultrasound curriculum on a multidisciplinary group of critical care fellows. Participants trained in internal medicine, emergency medicine, anesthesia, and neurology first completed a pre-training knowledge assessment. They then attended two one-hour didactic lectures on lung ultrasound (LUS). Hands-on training sessions were provided, with each fellow receiving two-hour scanning sessions under the guidance of critical care ultrasound faculty.Confidence among critical care fellows in acquiring and interpreting images increased from 57% (n = 21) to 79% (n = 19). Similar increases in fellow confidence were noted in using LUS to identify the etiology of respiratory distress and using ultrasound for the diagnostic and therapeutic management of pleural effusions. All study participants were able to accurately complete a comprehensive LUS exam within 10 minutes after training. Quality improvement initiatives and scan reviews provided ongoing feedback over the next year. Upon follow-up with graduates within a year of completing their fellowship, pulmonary ultrasound was routinely used in their practice, with an average frequency of two to three times per week. All respondents reported that pulmonary ultrasound training during fellowship had meaningfully changed how they integrate the tool into their clinical practice.

How intensive care fellows experience end-of-life communication, and how they (wish to) learn it

Background: Pediatric critical care medicine (PCCM) fellowship structures vary widely among programs and have not been examined since 2006. The development of scholarly knowledge and skills is an important part of fellowship training. However, the optimal PCCM fellowship structure to enable fellows' scholarly productivity is unknown. Objective: We examined PCCM fellowship time structure and resources to better understand their association with scholarly productivity. Methods: This is a secondary analysis of cross-sectional survey data from PCCM fellowship program directors. We defined highly productive programs as those in which greater than 75% of fellows published papers from their scholarly projects in the previous 5 years. Analyses investigated the association of scholarly productivity with dedicated scholarship time and program resources and barriers. Results: Forty-nine of 65 PCCM fellowships (75%) completed the survey. Only 20% of fellowships reported that greater than 75% of fellows published papers from their scholarly projects. Median total scholarly activity time was 16 months (interquartile range, 10-18 mo). The total amount of time devoted to scholarly activity was not associated with highly productive programs. Among resources and barriers, only T32 training grants were associated with highly productive programs. Conclusion: We found no relationship between time allocated to scholarly activity and high scholarly productivity in PCCM fellowships. Hence, programs that wish to increase fellows' scholarly productivity should not rely solely on providing more time for scholarly activity. Aside from T32 training grants, no specific resource or barrier that we evaluated is related to productivity.

Critical Care Fellows’ Training Experiences with Obstetric Critical Care: A Cross-Sectional Survey

Continuous Renal Replacement Therapy (CRRT) is emerging as an essential component of organ support in critically ill children. In low- and middle- income countries (LMIC), limited resources, lack of technical support, cost, and administrative issues are major barriers in initiating and sustaining a CRRT program. A core team, comprising a consultant and two pediatric intensive care fellows, was assigned additional responsibility of initiating and sustaining a CRRT program in the PICU of a tertiary care teaching and referral hospital. We retrospectively reviewed the data from initiation in February 2019 till May 2023 to understand the indications, prescription details, challenges, and their relation to outcomes. During this period, 52 children with mean (SD) age of 7.6 (3.2) years and median (IQR) weight of 20 (17, 30) kg underwent CRRT in 71 sessions. The mean PRISM-III score was 18 (5.2), with 98.1% having multiorgan dysfunction at CRRT initiation. Acute kidney injury (53.8%), hyperammonemia (21.2%), and a combination of both (17.3%) were common indications. Continuous venovenous hemodiafiltration was the most used modality (61.5%). Median CRRT duration was 36 (20.3, 58) hours. Filter usage averaged 1.4 per patient with a median life of 35 (17, 48) hours, improving from 24 to 36h over time. Filter clotting (33.8%), access flow issues (7%), and hemodynamic instability (4.2%) were complications encountered. Survival to discharge was 25%, with serum lactate [Formula: see text]3 mmol/L at CRRT initiation being an independent predictor of mortality (adjusted OR 6.1, 95% CI: 1.1-34.9; P = 0.04). Major challenges faced in our program included the SARS-CoV-2 pandemic, non-availability of technical support, and out-of-pocket expenses. These were circumvented by involvement of fellows and nurses, training them with internal and external experts, and mobilizing resources from governmental and non-governmental organizations. Initiating a CRRT program in LMICs is feasible despite challenges. Creating a team with members willing to shoulder additional responsibility and training them gave impetus to our program. Tapping governmental and non-governmental support helped us circumvent financial challenges. However, in a resource limited setting, sustainability requires in-house technical and financial support. Survival to discharge was 25%, with hyperlactatemia at CRRT initiation predicting mortality.

Simulation-based training can be valuable for teaching bronchoscopy to intensivists, providing a risk-free training environment. We developed, implemented and evaluated a simulation-based flexible bronchoscopy training program for intensive care Fellows and intensivists. This paper presents the development of its design and lessons learned. We used the Analyse, Design, Develop, Implement and Evaluate model for developing and evaluating the training program (Analysis and Design - phase 1, Development - phase 2, Implementation - phase 3, Evaluation - phase 4). In phase 1, two intensivists formulated learning objectives for bronchoscopy in an intensive care setting, which guided the identification and development of training materials and the preliminary training program (phase 2). In phase 3, we tested this program and gathered feedback from participants to guide program modifications. After implementing the adjusted training, we measured participants' satisfaction using a survey based on closed- and open-ended questions (phase 4). Fifty-seven participants attended the training, with 18 (32%) responding to the questionnaire. Respondents highly appreciated the training program, with median satisfaction scores of 4 or higher on a five-point scale for all closed-ended questions. Respondents appreciated the supervision and feedback and found the simulator equipment relevant for learning bronchoscopy. This description of the program's development and its evaluation results can serve as a valuable resource for those wishing to establish similar training programs. We recognise that further implementation of evidence-based instructional design principles might enhance the training program's scientific foundation and effectiveness. We therefore recommend a more evidence-based approach for the design of future bronchoscopy simulation training programs.

ABSTRACT Simulation training aims to increase exposure to high-stakes low-frequency events like cardiac arrest. However, within our laboratory-based simulation program, we have observed limited buy-in from internal medicine (IM) residents due to competing patient care obligations and a limited fidelity environment. Mirroring patient data within in situ simulation may provide relevance to ongoing resident patient care obligations, increasing buy-in and confidence in management. Clinical data from presently admitted patients in our institution’s medical intensive care unit (MICU) was ‘mirrored’ to create cardiac arrest simulations. Simulations took place in a vacant MICU patient room with resuscitation equipment, including a code cart, saline-substituted medications, and a mannequin capable of endotracheal intubation. The trainee team consisted of one post-graduate year (PGY) 3 IM resident, two PGY-1 residents, and a critical care fellow. A pre- and post-survey was administered to the PGY-3 IM resident to assess confidence in performing technical and non-technical skills. An advanced cardiac life support (ACLS) instructor evaluated PGY-3 IM resident performance using a skills checklist. Eighty-three percent of PGY 3 residents endorsed changes to their practice following the simulation. Confidence increased in skills related to flexible decision-making skills but not for fixed skills such as following ACLS protocol. Qualitative feedback highlighted realism, spontaneity, and debriefing sessions as the most valuable aspects of the program. In situ mirror simulation may be a useful adjunct for IM residency programs suffering similar concerns with learner motivation during laboratory simulation.

The Importance of a Transport Medicine Rotation for a Pediatric Critical Care Fellowship.

Improving Critical Care Communication with the Serious Illness Conversation Guide at an Academic Medical Center.

A National Assessment of Pediatric Surgical Critical Care Fellowship Structure: Defining the Training Experience.

Extracorporeal cardiopulmonary resuscitation (eCPR) has demonstrated patient outcome-driven benefits for those with out-of-hospital cardiac arrest in refractory ventricular fibrillation/pulseless ventricular tachycardia but remains an infrequent procedure requiring hands-on training. We created a high-fidelity simulation utilizing a cannulation manikin to simulate cardiac arrest in a 57-year-old patient in ventricular fibrillation refractory to standard resuscitation. Participants (consisting of emergency medicine and critical care resident and attending physicians, critical care fellows, advanced practice providers, nurses, pharmacists, and respiratory therapists) were instructed to respond to the simulation by recognizing the indication for eCPR and performing ultrasound-guided percutaneous extracorporeal membrane oxygenation (ECMO) cannulation to facilitate patient transfer to the cardiac catheterization lab. Participants rated their comfort level with various aspects of eCPR on a 5-point Likert scale, both presimulation (N = 27) and postsimulation (n = 17). A total of 27 participants with varied levels of training completed the simulation, with positive feedback from all respondents on the postsimulation survey. A statistically significant increase in comfort scores from pre- to postsimulation was observed across all domains, including knowledge of eCPR candidacy (p < .001), cannulation procedures (p < .001), and overall process (p = .001). Simulation is a valuable tool for ensuring procedural competency, especially for rarely performed and high-risk procedures such as ECMO cannulation. As eCPR becomes more prevalent, it is vital that simulation models be available and practiced on a multidisciplinary level to ensure general knowledge of the indications, procedures, and overall process of eCPR.

Abstract Deep venous thrombosis (DVT) is an important cause of morbidity and mortality in the intensive care unit (ICU) and it is often missed in the neurocritical care unit (NCCU) as patients might not manifest with obvious symptomatology. Screening of these patients may provide a preventive benefit. The routine use of formal lower extremity venous Doppler (LEVD) scan, which is done by a vascular laboratory technician and comprises the imaging of the entire lower extremity venous system plus Doppler imaging, can be resource intensive. Alternatively, the two-point compression ultrasound (2-CUS), which consists of the imaging and compression at the common femoral vein and popliteal fossa, is quick and can be done by the clinician at bedside with minimal training. The aim of this study is to evaluate the use of 2-CUS as a screening tool for all patients in the NCCU.We performed a retrospective analysis of prospectively collected data for consecutive patients admitted to the NCCU at Baylor College of Medicine between August 2020 and February 2021. A 2-CUS scan was performed by neurocritical care fellows, evaluating the bilateral common femoral veins and popliteal veins without color Doppler, on admission and every third day. An LEVD scan was performed in all NCCU patients every 7 days and sooner if 2-CUS scan was positive. The inclusion criteria were patients admitted to the neurocritical care service and age ≥18 years. The exclusion criteria were expected length of stay (LOS) of less than 48 hours, known recent DVT, or positive test for COVID-19.A total of 298 patients were admitted to the NCCU. 2-CUS was done in 256 patients, while 64 patients underwent both 2-CUS and an LEVD scan. The overall incidence rate of lower extremity DVTs was 10.1%. Nine patients undergoing 2-CUS were positive, of whom six were confirmed by LEVD, and three were false positives (positive predictive value = 66.7%). Two patients had false-negative 2-CUS (negative predictive value = 96.6%). When compared with LEVD, the sensitivity of 2-CUS was 75% (6/8) and the specificity was 95% (57/60). The sensitivity for proximal DVTs was 100%. The two DVTs missed were in the posterior tibial veins not scanned with 2-CUS.This is the first study to evaluate the use of 2-CUS as a screening tool for lower extremity DVTs. An incidence of 10.6% was found on patients with a LOS ≥7 days. With a sensitivity of 100% for proximal DVT, a specificity of 95%, and a diagnostic accuracy of 95%, the authors conclude that 2-CUS is a feasible, easily learnt, quick and cost-effective tool for screening DVT in NCCU patients.