OBJECTIVES/GOALS: Harnessing the EHR to support clinical research is critical for the study of rare diseases such as interstitial lung disease (ILD). However no studies have compared EHR and research-quality data in the ILD population. Our objectives were to (1) identify ILD patients and extract clinical data from an EHR system and (2) assess the performance of ILD data capture. METHODS/STUDY POPULATION: Case validated algorithms were implemented to identify patients from the University of California San Francisco EHR and extract key ILD clinical information including, demographic variables, process measures and patient outcomes. Key clinical information were defined based on consensus statements and ILD clinical trials. A subset of ILD patients, had variables recorded in both the EHR and a separate ILD longitudinal research database. The completeness of EHR data capture and level of agreement were compared between three data collection methods: (1) data manually and systematically collected for an ILD research database (gold standard), (2) data automatically extracted from structured fields in the EHR, and (3) data extracted from unstructured data sources. RESULTS/ANTICIPATED RESULTS: We identified 5857 ILD patients in the EHR, of which 2100 patients had data available in the both the EHR and research database. Baseline demographic variables, co-morbidities, use of diagnostic testing, pharmacotherapy were accurately extracted from structured fields. Outcome measures, including lung physiology, radiographic patterns, pathology results, and health related quality of life (HRQoL) were unevenly extracted from structured fields alone. With the exception of HRQoL, these measures were accurately captured in unstructured EHR sources. Notably, certain metrics were better defined in the EHR, including health care resource utilization metrics, acute exacerbations, medication side effects, supplemental oxygen use and specialty care referrals (rheumatology, lung transplant, palliative care, etc). DISCUSSION/SIGNIFICANCE OF IMPACT: A large real-world ILD cohort can be algorithmically extracted from the EHR along with key clinical variables with accuracy comparable to protocol-driven research databases. Rigorous assessment of the types of disease-specific variables that are present in EHR-derived data will inform future interventions to improve the fidelity, accessibility and use of the EHR in clinical research.

OBJECTIVES/SPECIFIC AIMS: To evaluate the ability of various techniques to track changes in body fluid volumes before and after a rapid infusion of saline. METHODS/STUDY POPULATION: Eight healthy participants (5M; 3F) completed baseline measurements of 1) total body water using ethanol dilution and bioelectrical impedance analysis (BIA) and 2) blood volume, plasma volume and red blood cell (RBC) volume using carbon monoxide rebreathe technique and I-131 albumin dilution. Subsequently, 30mL saline/kg body weight was administered intravenously over 20 minutes after which BIA and ethanol dilution were repeated. RESULTS/ANTICIPATED RESULTS: On average, 2.29±0.35 L saline was infused with an average increase in net fluid input-output (I/O) of 1.56±0.29 L. BIA underestimated measured I/O by −3.4±7.9%, while ethanol dilution did not demonstrate a measurable change in total body water. Carbon monoxide rebreathe differed from I-131 albumin dilution measurements of blood, plasma and RBC volumes by +0.6±2.8%, −5.4±3.6%, and +11.0±4.7%, respectively. DISCUSSION/SIGNIFICANCE OF IMPACT: BIA is capable of tracking modest changes in total body water. Carbon monoxide rebreathe appears to be a viable alternative for the I-131 albumin dilution technique to determine blood volume. Together, these two techniques may be useful in monitoring fluid status in patients with impaired fluid regulation.

ABSTRACT IMPACT: This review will encourage further development of novel uses of REDCap for the benefit of the research community. OBJECTIVES/GOALS: REDCap is a clinical research data collection platform that is primarily used as intended. However, little is known about its more novel uses, specifically in clinical decision support in patient care and in clinical research management. Thus, the purpose of this review is to examine peer reviewed literature identifying and describing such novel uses. METHODS/STUDY POPULATION: A systematic search was conducted in both PubMed and Google Scholar using the equation ((REDCap) OR ('Research Electronic Data Capture')) AND ((Clinical Trial Management) OR (Clinical Research)).’ Articles were screened by title, then abstract, and then were reviewed in full if they met inclusion criteria. Articles were included if they had potential relevance to the topic of REDCap or if they mentioned activities related to fields of clinical and translational science including operational support in areas such as clinical research management. Articles were excluded if they focused on common clinical research activities relating to data collection software such as survey administration, database building or data collection for clinical trials, registries, and cohort studies. RESULTS/ANTICIPATED RESULTS: The initial search yielded 390 results, of which 40 underwent an abstract review; only 8 of these underwent full text review. Of these, 5 discussed uses of REDCap in the context of operational support in clinical research management; 3 were related to clinical decision support in patient care. For the 5 articles focused on operational support in clinical research management, topics include e-consenting procedures, collection and storage of protected health information (PHI), patient recruitment and tracking stakeholder engagement. The 3 articles about clinical decision support discuss REDCap tools for generating risk predictions for post-surgical clinical outcomes, generating recommendations and STI test orders, and increasing efficiency in hand-offs to enhance care of surgical oncology patients. DISCUSSION/SIGNIFICANCE OF FINDINGS: Considering that only a small percentage of peer reviewed research reports out on novel uses of REDCap, there is a need for the REDCap consortium to do further work to fulfill its mission to adopt, innovate, and suggest novel uses of REDCap, thus expanding the understanding of its functionalities and therefore its utility in the research community.

OBJECTIVES/GOALS: Identify the interactions between nasopharyngeal bacterial pathogens, commensals, and patient clinical characteristics in relation to the development of viral upper respiratory tract infections (URI) and acute otitis media (AOM) in infants. METHODS/STUDY POPULATION: The subjects were part of a prospective, longitudinal study (2008–2014) of infants to evaluate the prevalence and risks for the development of URI and AOM. Healthy infants (n = 362) were enrolled from near birth and followed to the first episode of AOM up to 12 months of age. Nasopharyngeal specimens and clinical traits were collected at monthly intervals between 1-6 months, month 9, and during viral URI episodes. Subjects were closely followed for AOM development. 16S rRNA sequencing was performed on the nasopharyngeal swabs to identify their bacterial composition. Multidimensional (2, 3, and 4 dimensional) co-presence, co-exclusion, and one-way relation patterns were identified between the microbiome compositions, health status, and other collected clinical traits. RESULTS/ANTICIPATED RESULTS: We analyzed 971 specimens collected monthly and during URI and AOM episodes from 139 infants. Of the 139 enrolled subjects, 96% had 2 or more healthy samples, 77% contracted URI/AOM during the study period, and 60% had at least 1 healthy sample before URI/AOM onset. Otopathogens (Moraxella, Haemophilus, and Streptococcus), Staphylococcus, and Pseudomonas were the most common pathogenic genera. Corynebacterium, Dolosigranulum, and Acinetobacter were 3 most abundant commensal bacterial genera. Samples from infants with AOM in the first year had a significantly higher relative abundance of Haemophilus, Enterobacter, and Yersinia, and lower relative abundance of Corynebacterium, and Pseudomonas compared to samples from infants who did not develop AOM. DISCUSSION/SIGNIFICANCE OF IMPACT: Identification of complex multidimensional interaction patterns within microbial communities and environmental factors is vital to understanding disease onset risk and prevention. Prophylactic microbiome and environmental factor modulation between enterotypes could be used to reduce URI/AOM onset in infants.

OBJECTIVES/SPECIFIC AIMS: To demonstrate the prevalence of ocular complications in patients suffering with Mycoplasma Induced Rash and Mucositis (MIRM). METHODS/STUDY POPULATION: In this retrospective observational study, we identified all patients in our hospital database who were diagnosed with MIRM. Diagnosis was confirmed by clinical information and positive Mycoplasma pneumoniae serology. Only patients with available records with formal ophthalmology consults were included. Clinical and laboratory data were collected from our electronic medical record system to capture key components of their clinical course. RESULTS/ANTICIPATED RESULTS: A total of 12 patients satisfied all inclusion and exclusion criteria and were included in our study. The average age of our included patients was 21.2 ± 14.7 years, and the majority were male vs. female (66.7% vs. 33.3%). In all 24 eyes, the only acute ocular findings included conjunctival hyperemia (n=20, 83.3%), meibomitis (n=4, 16.7%), and conjunctival epithelial defects (n=1, 4.2%). None of the patients required or were recommended to receive an amniotic membrane transplantation in the acute phase. Only 2 patients were followed in the chronic phase, one of whom showed evidence of meibomitis in both eyes. Otherwise, no other chronic complications were seen in either patient with chronic follow-up. DISCUSSION/SIGNIFICANCE OF IMPACT: Ocular complications from MIRM may be much milder in comparison to ocular complications found in other bullous and inflammatory conditions such as Stevens-Johnson Syndrome or Toxic Epidermal Necrolysis. Understanding MIRM’s specific sequelae is important in understanding disease manifestation and prognosis in order to better inform acute and chronic management.

OBJECTIVES/GOALS: Well-appearing febrile infants are a model for exploring communication, bias, and health disparities in the pediatric emergency department (ED). Using mixed methods, we will perform an in-depth analysis of disparities and shared decision making, a potentially modifiable driver of inequities. METHODS/STUDY POPULATION: We will conduct a multicenter cross-sectional chart review study of well-appearing febrile infants 29-60 days old treated in the ED and apply multivariable logistic regression to assess the association between 1) race/ethnicity and 2) limited English proficiency with the primary outcome, discharge to home without lumbar puncture and without antibiotics (standard of care). We will concurrently perform an interpretive study using purposive sampling to conduct interviews with: 1) minority parents of febrile infants and 2) ED physicians. By capturing dyadic data, we will triangulate perspectives to elucidate disparities and bias that can emerge in the shared decision making process. RESULTS/ANTICIPATED RESULTS: Forty member institutions of the Pediatric Emergency Medicine Collaborative Research Committee are participating, providing a projected cohort of 3000 infants. In the 6 months since site recruitment, 235 eligible infants have been entered into the dataset (43% minority race/ethnicity, 6% language other than English), 61% of whom received the primary outcome. Chart review has the benefits of 1) ensuring exclusion of ill infants, 2) providing data on interpreter use that is unavailable in administrative datasets, and 3) allowing an analysis of shared decision making. These findings will inform an interpretive study of parent and provider experiences of bias in shared decision making. DISCUSSION/SIGNIFICANCE: We demonstrate the feasibility of a large-scale manual chart review to analyze disparities within a shared decision making context. Partnered with qualitative scholarship, this research will support the development of communication interventions to mitigate implicit bias in the clinical encounter.

OBJECTIVES/SPECIFIC AIMS: Autism spectrum disorder (ASD) is a neurodevelopmental disorder that affects one in 68 children. Children with ASD have 2 core areas of difficulty: social communication skills and restricted and repetitive interests and patterns of behavior. Children with social skills deficits are at higher risk of developing mental health problems, and underdeveloped social skills predict poorer quality of life in adulthood. Therapies have been developed to help people with ASD improve social abilities in childhood, often involving a clinician directly teaching social skills lessons, either one-on-one or in a group setting. However, children with ASD can become anxious when interacting with other people and have an intrinsic motivation to interact with technology. To capitalize on this interest, this research team developed a robot, the socially animated machine (SAM) to teach social skills to children with ASD. Previous research found that this intervention was feasible and enjoyable for children with ASD and average cognitive ability, and participants improved in complex emotion recognition following intervention. The purpose of this study was to determine whether participants of all IQ levels were motivated by the SAM intervention, and whether they improved on emotion identification, facial recognition, social skills, and adaptive behavior. METHODS/STUDY POPULATION: This study recruited 20 children with ASD ages 5–14. Children completed tasks measuring ASD symptoms, IQ, receptive language, facial recognition, and emotion identification and were assigned to the control group (nonemotion dance games with SAM robot) or the intervention group (emotion games with SAM robot). Parents and teachers completed questionnaires about the child’s social skills. Following the robot intervention, facial recognition, emotion identification, and social skills were measured again, and parents and children rated participant enjoyment during the robot interaction. RESULTS/ANTICIPATED RESULTS: Overall, parents and children in both groups rated the robot interaction as highly enjoyable and motivating (parent ratings: M=26.4 out of 30, child ratings: M=17.5 out of 20). There were no differences between groups on post-test measures when controlling for pre-test scores (all p>0.05). Both groups improved over time on emotion identification accuracy (intervention: M=13.0% improvement, t=2.57, p<0.05; control: M=10.2% improvement, t=2.38, p<0.05) and parent-rated social skills (intervention: pre-test M=113.8, post-test M=100.6, t=−3.37, p=0.01; Control: pre-test M=107.9, post-test M=89.0, t=−2.83, p<0.05; decrease in scores indicates improvement). Teachers saw a decrease in problem behaviors for the intervention group (pre-test M=127.4, post-test M=119.6, t=−3.79, p<0.01, decrease in scores indicates improvement). DISCUSSION/SIGNIFICANCE OF IMPACT: This study shows that children with ASD and all levels of cognitive ability enjoyed and were motivated by the SAM robot intervention. This is particularly important for children with ASD who often have difficulty with attention and motivation. Children who are intrinsically motivated by the learning process will be more likely to benefit from it; therefore, continuing to pursue the methodology of robot-based interventions with this population is a worthwhile endeavor.

OBJECTIVES/GOALS: Tobacco use remains a significant public health problem in rural America. Community health centers (CHCs) can help reduce the burden of tobacco use in rural areas, but we know little about their knowledge and implementation of best practices for tobacco control. This study assessed the knowledge and existence of tobacco control practices in rural CHCs. METHODS/STUDY POPULATION: Using a cross-sectional study design, we electronically surveyed health administrators and providers (n=64) in six rural CHCs in Louisiana between March 2021 and June 2023. The assessment measured 1) knowledge of the U.S. Public Health Service Clinical Practice Guideline for Treating Tobacco Use, 2) the priority given to smoking cessation programming, 3) the presence of best practices for tobacco control programming, such as having a tobacco control champion and team, 4) establishment of treatment and smoke-free campus policies, and 5) existence of referral procedure to external cessation services. We used descriptive statistics to characterize survey participants and responses. RESULTS/ANTICIPATED RESULTS: Most of the respondents were female (80%), White (61.8%), between 35 and 49 years of age (48.1%), and non-smokers (72.7%). Only half (51.6%) reported knowledge of the guideline for treating tobacco use among all respondents. Only a third (32.8%) said their health center gave smoking cessation high priority relative to other health priorities. Only a third (35.9%) reported having a tobacco champion; less than a fifth (18.8%) said they had a tobacco control team at their health center. Although all health centers had a smoke-free campus policy, about a quarter (23.4%) were unaware of the policy. Less than a quarter (23.4%) reported having a written policy for smoking cessation treatment at their health center, and less than half (46.7%) knew about cessation services to which they could refer patients. DISCUSSION/SIGNIFICANCE: Rural CHCs had limited knowledge of the guideline for tobacco use treatment. Smoking cessation lacked priority, and the prevalence of tobacco control best practices implementation was low. Rural CHCs must improve their implementation of guideline-recommend policies and clinical strategies to promote treatment and reduce the burden of tobacco use.

OBJECTIVES/GOALS: At least 15 stroke thrombectomies per proceduralist per year are required for Thrombectomy-Capable and Comprehensive Stroke Centers. We sought to estimate the potential impact of these minimum volume thresholds on geographic access to stroke thrombectomy (ST). METHODS/STUDY POPULATION: Using the Florida State Inpatient Database, we will identify patients discharged with acute ischemic stroke from 2015-2019 and all non-federal facilities that performed ?1 ST procedure per year. We will then calculate the proportion of stroke patients who live within 20, 65, 115, 165 and 200 miles (correlating with estimated ground transport times of 1, 2, 3, 4 and 5 hours, respectively) of centers that perform ST using ArcGIS software and evaluate the impact of varying the threshold ST volume required by each facility on this proportion. We will then perform multiple two proportion z-tests to compare proportions of patients within driving distance over time. RESULTS/ANTICIPATED RESULTS: We hypothesize that over time, and particularly after the pivotal trials of ST were published in 2015, that the number of facilities that perform ST have increased over time, which would increase the geographic access to ST. We also hypothesize that since the Joint Commission set the minimum procedural volume for proceduralists at Thrombectomy-Capable and Comprehensive Stroke Center to 15 per year, this would work to increase regionalization and could work to decrease geographic access to ST. However, we hope to elucidate the net impact of the interplay between these two opposing factors on regionalization of care over time which is currently unclear. DISCUSSION/SIGNIFICANCE: Current ST volume thresholds have focused on technical proficiency but may impact regionalization of care and geographic access to ST. Since access to ST is time-sensitive, a data-driven approach and better coordination on a regional level may be necessary to ensure timely access to ST.

OBJECTIVES/GOALS: We developed institution-wide RCR training to include all required elements; support trainees to identify key ethical questions that arise in research; and to identify methods to identify a solution; and disseminate results. METHODS/STUDY POPULATION: In 2011, we participated in developing an experimental model of RCR training led by the University of Michigan. We continue to offer this training model to career development awardees in clinical and translational research at OHSU across the institution. Interactive discussion in faculty and trainee groups includes responsibilities of a researcher and a systematic process to address real world research ethics issues. Each participant identifies a key research issue they have encountered and presents a poster at the final session. We have tracked post-training assessment of participant confidence in ethical decision making and in the range of topics identified by participants. RESULTS/ANTICIPATED RESULTS: Since 2012, 227 scholars and trainees have participated in the program with 44 faculty mentors facilitating. We will describe the current curriculum as it has evolved over the past 12 years, presenting trainees with an approach to identify ethical challenges that arise in their research and identify approaches to find a practical solution. We will report on the specific challenges in research ethics identified by participants over this period and how they have evolved. We will also present pre- and post-training data about confidence in ethical decision making. DISCUSSION/SIGNIFICANCE: This approach to RCR training is well-received, has evolved over time, and has led to dissemination. Success is attributed to allowing choice in topics relevant to trainees and practicality of the stepwise approach that is transferable to any situation.

OBJECTIVES/SPECIFIC AIMS: Emergency department (ED) length of stay (LOS) is one of the important indicators of quality and efficiency of ED service delivery and is reported to be both cause and result of ED crowding. Increased ED LOS is associated with ED crowding, increases service cost and sometimes poor patient outcome. Substance abuse is one of the major determinants of morbidity, mortality and healthcare needs. Substance abuse may confound the healthcare and service needs of patients in the ED irrespective of primary purpose of their ED visit and may lengthen the ED LOS. The aim of this study was to evaluate the effect of patients’ demographic and clinical characteristics and of different patient-related activities such as screening brief intervention and referral to treatment (SBIRT) on the ED LOS of patients discharged from the ED with a diagnosis of substance abuse. METHODS/STUDY POPULATION: We conducted a retrospective data analysis of electronic health records. The study population included 26971 patients who visited our hospital ED between 2013 and 2017, had a history of substance abuse and were discharged from the ED. An accelerated failure time (AFT) model was used to analyze the influence of covariates on patient ED LOS. The predictor factors in the model included age, gender, ED arrival shift and weekday, diagnosis history of mental health and drug use, acuity triage level from 1 to 5, with 1 being worse severity, and whether any lab tests were ordered, SBIRT intervention and whether patient was homeless. The AFT model is an alternative to the Cox Proportional Hazard Ratio model, which directly models the log of ED LOS as a function of a vector of covariates. The model defines the increase or decrease in LOS with the changes in the covariate levels as an acceleration factor or time ratio (TR). RESULTS/ANTICIPATED RESULTS: The overall median ED LOS was 4 hours with IQR of 4.2 hours. The average age of the study population was 39.3 years, 58.6% of the patients were male and 57% where White; 63.4% had a history of drug use; 43% had a history of mental health issue, and 0.4% were homeless. In the analysis using the AFT model, increased age (a year increase, TR =1.01, p =0.008), female sex (TR=1.044, P<0.001), SBIRT (TR=1.525, P <0.001), history of mental health issue (TR=1.117, P<0.00), evening arrival (evening vs night, TR=1.04, p=0.006), history of drug use (drug vs alcohol only, TR=1.04, p=0.001), higher acuity (triage level 1 vs 5, TR=2.795, p <0.001) and homelessness (TR=1.073, P = 0.021) lengthened the ED LOS. In contrast, weekend arrival (TR=0.956, p=0.004) and day shift arrival (day vs night, TR=0.958, p=0.004) shortened the ED LOS. DISCUSSION/SIGNIFICANCE OF IMPACT: We identified gender, age, SBIRT, arrival shift, weekend arrival, mental health status, substance abuse, acuity level and homelessness to be significant predictor of ED LOS. The fact that SBIRT increased the LOS should be balanced with the advantages of engaging patients into substance use disorder treatment. Understanding the determinants of ED LOS in this population may provide useful information for physicians or patients to better anticipate an individual’s LOS and to help administrators plan the ED staffing and other resources mobilization.

OBJECTIVES/GOALS: Lack of a vehicle and the distant location of supermarkets influence the ease with which people can shop for healthy foods. The aim of this qualitative study is to understand how transportation impacts food access and food purchases of Baltimore residents who do not own an automobile. METHODS/STUDY POPULATION: The Lyft Grocery Access Program was piloted in Baltimore, MD from November 2019 through September 2020. Eligible households resided in Healthy Food Priority Areas, formerly known as food deserts, in south and west Baltimore and also did not own a vehicle. Enrolled households were offered discounted Lyft rides to select supermarkets. Participants for the present study will be purposively recruited via email and phone using contact information that was provided by enrolled households during the pilot program. Each in-depth interview will be conducted via Zoom and recorded, transcribed and analyzed for themes by two trained coders. Data collection and analysis will occur simultaneously. Data collection will cease once data saturation is reached and themes will be derived from the data. RESULTS/ANTICIPATED RESULTS: This study is in progress. Anticipated themes may relate to the food environment, transportation and food access. DISCUSSION/SIGNIFICANCE: Access to healthy foods is an important determinant of health, and how food access is impacted by broader aspects of daily living such as transportation will add to the food access literature. Findings may provide new insights that can help inform food policy and transportation planning in urban communities.

OBJECTIVES/SPECIFIC AIMS: Our objectives are to provide opportunities for graduate students, clinical interns, and postdoctoral fellows in traditional training programs to have immersive experiences in clinical research conduct from a CRP perspective. In addition, we aimed to address common causes of job dissatisfaction by providing professional development and networking opportunities for the existing CRP workforce. METHODS/STUDY POPULATION: In collaboration with the CTSA workforce development group, the Duke Office of Clinical Research hosted a site visit for 19 PhD scientists interested in nontraditional career pathways and a short lecture series on project management careers in clinical research. Additionally, we crafted specific clinical research training electives for 20 masters students and 10 dietetic interns. Finally, in collaboration with UNC-CH, we combined Research Professional Networks to provide a pilot joint professional development event for 109 CRPs from both schools. RESULTS/ANTICIPATED RESULTS: The number of Masters students enrolling in the CRP elective grew from 7 students in year 1 to 13 students currently enrolled. A retro-pre/postprogram adapted CRAI survey was issued following program completion. Students self-reported increases in competence across all 24 skills measured. Largest increases were seen in areas specific to CRP roles such as consenting patients, understanding the IRB, and reviewing key study documents. A baseline culture survey issued at the joint Duke/UNC CRP event garnered a 65% response rate and indicated that the principal gaps in professional training are in communications, teamwork, leadership, and professionalism. Moreover, respondents indicated that creating a sense of community and providing networking opportunities were the most important outcomes for future CRP collaborations. Future evaluations of both of these programs will allow us to tailor training to be most effective in strengthening our CRP workforce. DISCUSSION/SIGNIFICANCE OF IMPACT: These initiatives lay the groundwork for the development of a robust training pipeline into CRP careers. Future initiative will apply lessons learned toward creating internship programs aimed at improving diversity and inclusion within these careers. In addition, by addressing the professional development needs of the existing workforce, we create a sustainable environment for well-trained professionals. By evaluating these primary initiatives, we can better define the critical elements that must be included in CRP educational, development, and support programs and subsequently apply these to ultimately improve the clinical and translational research being conducted in academic research settings.

ABSTRACT IMPACT: A machine learning approach using electronic health records can combine descriptive, population-level factors of pressure injury outcomes. OBJECTIVES/GOALS: Pressure injuries cause 60,000 deaths and cost $26 billion annually in the US, but prevention is laborious. We used clinical data to develop a machine learning algorithm for predicting pressure injury risk and prescribe the timing of intervention to help clinicians balance competing priorities. METHODS/STUDY POPULATION: We obtained 94,745 electronic health records with 7,000 predictors to calibrate a predictive algorithm of pressure injury risk. Machine learning was used to mine features predicting changes in pressure injury risk; random forests outperformed neural networks, boosting and bagging in feature selection. These features were fit to multilevel ordered logistic regression to create an algorithm that generated empirical Bayes estimates informing a decision-rule for follow-up based on individual risk trajectories over time. We used cross-validation to verify predictive validity, and constrained optimization to select a best-fit algorithm that reduced the time required to trigger patient follow-up. RESULTS/ANTICIPATED RESULTS: The algorithm significantly improved prediction of pressure injury risk (p<0.001) with an area under the ROC curve of 0.60 compared to the Braden Scale, a traditional clinician instrument of pressure injury risk. At a specificity of 0.50, the model achieved a sensitivity of 0.63 within 2.5 patient-days. Machine learning identified categorical increases in risk when patients were prescribed vasopressors (OR=16.4, p<0.001), beta-blockers (OR=4.8, p<0.001), erythropoietin stimulating agents (OR=3.0, p<0.001), or were ordered a urinalysis screen (OR=9.1, p<0.001), lipid panel (OR=5.7, p<0.001) or pre-albumin panel (OR=2.0, p<0.001). DISCUSSION/SIGNIFICANCE OF FINDINGS: This algorithm could help hospitals conserve resources within a critical period of patient vulnerability for pressure injury not reimbursed by Medicare. Savings generated by this approach could justify investment in machine learning to develop electronic warning systems for many iatrogenic injuries.