OBJECTIVES/SPECIFIC AIMS: Our objectives with this project are to engage communities through technology creating a communication channel with affected communities and stakeholders about mosquito-borne illness, vector control and environmental health risk. Furthermore, engaging communities to electronically map ecological risks that impact mosquito-borne illness with the goal of creating a mobile application that will work as an ecological surveillance against mosquito proliferation and potential mosquito population reduction, and finally pilot test and evaluate potential benefits in communities where the application was used. METHODS/STUDY POPULATION: We propose a methodology to perform formative community work that will underscore a distributed, democratized ecological surveillance through an integration of multidimensional health behavior theories that address the challenges of ZIKV in Culebra, a marginalized island community off the coast of the main island of Puerto Rico. Using participatory design, we will develop, test, and evaluate users’ experiences towards mobile applications using qualitative (interviews) and quantitative (survey) methodologies. A mobile application with the capacity of mapping, use of social-media, crowdsourcing, and photo-voice in a dynamic and simple way will allow community members to alert “hot-zone” locations to the stakeholders interested in creating ecological action in their community. This multidimensional concept integrates explanatory and prospective approaches and will generate systematic short-term solutions for mosquito control and long-term solutions providing the necessary tools for community empowerment. RESULTS/ANTICIPATED RESULTS: Our proposed design will facilitate better understanding of the interactions between community members and socio-environmental determinants of mosquito-borne diseases. Furthermore, our proposed project will not only facilitate communication among members of a community, but also it will provide a platform for engagement and empowerment, establishing a change in the preventive paradigm of how communities face the negative impacts of micro-ecologies that surround them. DISCUSSION/SIGNIFICANCE OF IMPACT: Our proposed community collaboratory mHealth tool mZAP! (Zonas, Accion y Proteccion) will address the lack of community participation efforts against mosquito-borne diseases contributed simultaneously by the disengagement and disempowerment of community members. mZAP! will serve as an innovative tool to engage marginalized and communities made vulnerable in Puerto Rico. This approach should be successful as Puerto Rico is one of the most digitally connected countries in Latin America, with high mobile phone usage rates and social media use. Using mZAP!, communities will report and map breeding sites, use social media and crowd sensing, targeting against powerful tools against mosquito ecologies in their own environments. This application could result in an effective way to change the paradigms for public health approaches to use Information Communications Technologies (ICTs) to empower communities.

OBJECTIVES/SPECIFIC AIMS: This study sought to develop a mHealth application which was capable of predicting the spread of infectious diseases during the height of the Ebola outbreak in Lagos, Nigeria. Following the success of this primary task, the research then sought to understand behavioral health issues which are indicative of chronic diseases, such as sedentary behaviors and where they occur at a geospatial level in real-time. The results of this study are now being used to develop a larger scale 500 person study in Rochester, NY, USA. METHODS/STUDY POPULATION: During a 3-month period individuals were asked to install a mobile health application known as Node onto the their android device. Consent was done remotely, individuals were recruited through the Lagos University Teaching Hospital, Nigeria Institute of Medical Research, and the University of Lagos. Participants were paid 50 USD/month for each month of study completion, while continuous location data was collected in addition to survey information about participants. RESULTS/ANTICIPATED RESULTS: During the study period 70 individuals enrolled, using this data we were able to create network based models which indicated that diseases were more likely to spread at the beginning of the week, and also indicated who would be most susceptible to being patient zero. In phase 2 we have started to look at behavioral patterns to determine the risk of chronic disease among our study population, by examining their human mobility patterns, since we can determine average sleep patterns, activity patterns using machine learning classifiers, and time spent in traffic—all of which we can visualize in a real-time geospatial manner with higher objectivity than traditional mechanisms for data collection. DISCUSSION/SIGNIFICANCE OF IMPACT: In developing countries, using Nigeria as our example most chronic disease and household studies only enroll a few thousand participants for a country numbering 150 million plus. Using our rapidly available application we were able within 1 week to enroll 70 participants on 1 year of funding, this creates a framework for larger scale public health studies which can be done in developing countries and also demonstrates the value in mHealth which can both answer questions of infectious disease and chronic diseases at the same time. Our results indicate that at an infectious disease level in city environments diseases may be prevented by targeting events early in the week. While at a chronic disease level the lack of reliable power results in less sedentary behavior as individuals seek locations to charge phones, while those with more stable western-like lifestyles have started to exhibit the conditions which cause such outcomes as obesity, which has begun to rise in developing countries. Ultimately, these results serve as a staging point to launch a more wide scale study both in the United States and Nigeria within the year, now that feasibility has been established.

OBJECTIVES/GOALS: Diabetic ketoacidosis (DKA) is a life-threatening complication of diabetes. Though largely preventable, DKA is one of the most common acute complications of diabetes. In the US, rates of DKA hospitalization and associated costs have been increasing over the past two decades. METHODS/STUDY POPULATION: In this study, we used the Kentucky Statewide Inpatient Database (2010-2019) and the Nationwide Readmission Database (2010-2018) to explore variation in rates of DKA hospitalization across key sociodemographic subgroups (age, sex, race/ethnicity, rural/urban, insurance coverage, and county-level poverty index) and identify clinical predictors of DKA hospitalization. The primary outcome was hospitalization with a first diagnosis of DKA identified using ICD-9 and -10 codes. Crude rates were calculated using state- and county-level population estimates obtained from the US Census Bureau and are presented as the total number of events per 10,000 people. Regression models will be used to examine the associations between DKA hospitalization and clinical predictors. RESULTS/ANTICIPATED RESULTS: In Kentucky, from 2010-2019, rates of DKA hospitalization increased by 45% (from 65.5 to 94.8 per 100,000). The largest variation was observed by age, race/ethnicity, and insurance. In those aged 15-44, rates of DKA hospitalization were three times higher than rates in the youngest (<15) and oldest (>75) groups (>130 vs <45 per 100,000). Non-Hispanic Blacks experienced rates of DKA hospitalization that were 2x higher than rates observed in non-Hispanic Whites (183.9 vs 92.6 per 100,000). Those covered by Medicaid had the highest rates of DKA hospitalization (171.3 vs 32.4 per 100,000 in commercially insured). Small, but consistent, disparities were observed in rural vs urban counties and higher poverty rates. Predictors of DKA hospitalization are being examined in the Nationwide Readmission Database. DISCUSSION/SIGNIFICANCE: Our findings underscore significant variation in DKA risk across key sociodemographic subgroups and will examine and confirm previously identified clinical predictors of DKA. Because DKA is largely preventable, identifying individuals at higher risk and targeting interventions and services to these individuals may help reduce DKA rates.

OBJECTIVES/GOALS:

1. 1. analyze urinary protein exosome content pattern before and during DASH diet.

2. 2. characterize urine electrolyte changes associated with changes in protein profiles, and hormonal changes before/after DASH diet.

3. 3. analyze the association of these changes to the DASH-related BP response.

METHODS/STUDY POPULATION: In this proof of concept study, hypertension stage 1 volunteers will receive a DASH based menu during 14 consecutive days of elective admission to the RU research hospital. Participants will complete a food frequency questionnaire (VioScreen) with a bionutritionist. Throughout the intervention period, participants will be assessed for blood pressure, plasma renin and aldosterone, and 24 hour urines for electrolytes, creatinine, protein, albumin and first morning urine collected for exosomes. Exosome analysis will be performed by a commercial lab. Proteome analysis will be conducted in the RU Mass-spectrometry service. RESULTS/ANTICIPATED RESULTS: The causal pathway we will elucidate hypothesizes that: 1) changes in diet affect blood electrolytes, and through these, aldosterone. 2) Aldosterone alters the expression of specific transporter proteins in the renal tubule; protein expression will be reflected in the urine exosome. 3) These transporters affect the excretion of electrolytes, as reflected by urinary ratio of sodium (Na) to Potassium (K). During consumption of the Western diet, the Na/K ratio is approximately 2-2.5, whereas we expect the urinary sodium/potassium ratio to be <1, when the participant is eating a DASH based diet. DISCUSSION/SIGNIFICANCE OF IMPACT: This assay provides a clinical tool to assess dietary adherence, and the project will provide insights into the mechanism whereby DASH reduces blood pressure.

OBJECTIVES/SPECIFIC AIMS: Chronic obstructive pulmonary disease (COPD) is a leading cause of both hospitalizations and readmissions in the United States, and about 1 in 5 hospitalized patients with COPD will be readmitted within 30 days. COPD-focused self-management programs are frequently used to help patients better manage their symptoms and prevent hospitalization. However, while the majority of patients with COPD have at least one comorbidity, most trials of COPD self-management programs either excluded patients with significant comorbidities or did not analyze the impact of comorbidities on patient outcomes. Using data from the BREATHE trial of a COPD self-management program, this study aims to determine if patient post-intervention outcomes differ based on the intensity and type of patient comorbidities. METHODS/STUDY POPULATION: In total, 240 patients hospitalized for COPD were randomly assigned to either a comprehensive self-management intervention or usual transitional care. Primary outcomes for this trial were the number of COPD-related hospitalizations and emergency department visits at 6 months and changes in COPD-specific quality of life. To determine whether patient comorbidities modify the effect of the self-management intervention on readmission and quality of life outcomes, we will compare patient outcomes across groups stratified by comorbidity burden (Charlson Comorbidity Index) and type (baseline diagnosis of congestive heart failure, diabetes, and depression). In addition, we will use regression analysis with interaction terms to test for interaction between comorbidity burden/type and intervention assignment. RESULTS/ANTICIPATED RESULTS: We hypothesize that the effect of the self-management intervention will differ in patients with greater comorbidity burden due to competing medical demands for patients with multimorbidity. DISCUSSION/SIGNIFICANCE OF IMPACT: The results of this study will help clinicians better target disease-specific self-management programs to the groups of patients with COPD who are likely to receive the greatest benefit from this type of intervention.

Objectives/Goals: This study aims to evaluate diversity of participants in GLP-1 T2DKM clinical research with regard to sex, race, and ethnicity by using results data available through the ClinicalTrials.gov database. Sample population estimates for studies were calculated using the 2020 Census and compared within groups with respect to sex, race, and ethnicity. Methods/Study Population: The public ClinicalTrials.gov database was searched for interventional studies with GLP1 inclusion as treatment (n = 2,397). This search was then filtered to studies where results were reported (n = 772). From these studies, 466 studies focused on type 2 diabetes as a condition and thus became the analysis dataset. Participant and protocol information for these 466 studies were obtained from the clinical trials transformation initiative (CTTI) as an AACT data download. Observed to expected ratios were calculated for each subgroup-based population estimates from the 2020 Census and using the baseline counts of participants for studies where sex, race, and ethnicity were provided. In addition to within group comparisons, study characteristics (e.g. phase) were included in models to assess influence of covariates. Results/Anticipated Results: Of the 466 studies, 430 (92%) reported sex, 171 (37%) reported race, and 145 (31%) reported Hispanic ethnicity. Among those found to be underrepresented in studies (defined as a ratio < 1): females (mean  =  0.89, median  =  0.92); Black/African Americans (mean  =  0.88, median  =  0.39). Hispanic or Latinos mean ratio was 1.16 (95% CL: 0.97, 1.35) but had the least available data. When including covariates in the models, there were statistically significant differences in ratios with respect to sex as females had significantly lower odds compared to males (ratio > =  1), with the odds being about 21% of those for males. With respect to race, Black or African American individuals had significantly lower odds (about 32% of those of White individuals) (ratio > =  1). Discussion/Significance of Impact: This study reveals significant underrepresentation of females (mean ratio 0.89) and Black/African Americans (mean ratio 0.88) in clinical trials for GLP-1 drugs in type 2 diabetes. These disparities highlight the need for more inclusive research to ensure diverse populations benefit equally from medical advancements.

OBJECTIVES/GOALS: This study examined patient perceptions on the benefits, barriers, and facilitators of conventional and complementary/behavioral pain strategies that can be offered in the ED setting including physical therapy, mindfulness, acupuncture, and yoga. METHODS/STUDY POPULATION: We conducted and recorded semi-structured interviews with 30 patients who presented to the ED with musculoskeletal pain. Interviews focused on patients’ perceptions of NP pain treatments, barriers/facilitators to utilization, and recommendations that would promote engagement. A hierarchical coding system was developed and refined using the interview guide, the Theory of Planned Behavior, and preliminary review of the transcripts. The iterative process of developing the coding system allowed us to identify preliminary themes. RESULTS/ANTICIPATED RESULTS: Patients believe education on pain and the mind-body connection would give a sense of pain control. Likely barriers to engaging at the ED include lights, noise, interruptions, and uncertainty of their medical status. Post-discharge NP treatment barriers are financial and logistical. Engagement can be facilitated by a desire to avoid opioids, familiarity with meditation practices, and consistent positive communication with the health care team. Patients desire evidence on effectiveness, including testimonials, and suggested NP techniques should be introduced early, with written materials used for post-discharge referral. Patients prefer in-person treatment but would appreciate a virtual option. The initial session should occur in-person to build trust and facilitate virtual session engagement. DISCUSSION/SIGNIFICANCE: Patients are willing to engage in nonpharmacological pain treatment, however the unpredictable ED environment, uncertainty of their medical status, and financial and time constraints are significant barriers.

OBJECTIVES/SPECIFIC AIMS: To assess the usability and adoption of RxUniverse, a novel platform that enables health care providers to directly disseminate proven, evidence-based mobile health apps to patients. METHODS/STUDY POPULATION: Among 5 pilot clinical sites, 40 physicians and front-line providers consisting of medical assistants and receptionists were trained on the RxUniverse platform. They were instructed on the platform’s purpose, were shown a demonstration of the functionality, and were observed in a trial process of prescribing an app. Specific implementation plans were designed with the help of the clinic staff in order to best fit in with their present workflows. The well-validated System Usability Score (SUS) was used to assess the usability of the platform. Prescriptions of 100 relevant app prescriptions within a 8-week pilot period was set as the adoption goal. RESULTS/ANTICIPATED RESULTS: Within the pilot period, greater than 2000 apps were prescribed across all users. Of the 40 providers trained on the RxUniverse platform, 26 prescribed >5 apps during the trial period. Of these 26 individuals, 18 prescribed >20 apps, 14 prescribed >50 apps, and 5 prescribed >80 apps; 58% of users reported frequent use (weekly or daily) of the platform. In total, 19 responses were received for the SUS survey. The RxUniverse platform received a usability score of 82%. DISCUSSION/SIGNIFICANCE OF IMPACT: As the pace of innovation continues to accelerate, health care providers will need to quickly integrate new digital-based tools into their workflows, and patients will need to be able to easily and readily access these tools. RxUniverse provides the necessary mechanisms, user-friendly interface, and EHR integration functionality to accomplish this. The total number of apps prescribed surpassed 2000, which far exceeded the initial target of 100 apps. The platform also scored an 82% on the SUS, which is considered an “A” by industry standards. By comparison, other health apps considered to have to be in the highest-rating groups have reported scores of 77.5% and an overall average of 68% among all systems. These outcomes demonstrate the high adoption and usability of the RxUniverse platform, an important platform that can be used to prescribe the latest technologies directly to patients.

Objectives/Goals: Inclusive physical activity (PA) interventions could improve trans and gender diverse (TGD) adolescents’ PA levels, perceived social support from peers and adults, and mental health, but no stakeholder-informed interventions exist. We describe parents’ impressions of TGD adolescent children’s PA experiences and advice for a PA intervention. Methods/Study Population: We conducted individual Zoom interviews with parents of 13- to 20-year-old TGD adolescents (n = 15). All parents were recruited from a gender healthcare clinic. All children self-identified as TGD, an umbrella term including people who have a gender identity different from social expectations for their assigned sex. We asked questions regarding children’s current and historical participation in PA, parents’ perceptions of barriers to PA for their TGD children, and parents’ desired intervention components. After each interview, parent participants were compensated $100 for their time. We analyzed interview transcripts, focusing on insights to incorporate in an intervention. This study was approved by the University’s Institutional Review Board. Results/Anticipated Results: Parents shared rich histories of their children’s PA participation, compounding barriers to PA, numerous benefits of PA, and a range of preferences for program activities and inclusive practices (e.g., safety protocols, training for adult leaders). They also emphasized the need for TGD youth to build social connectedness through PA. Discussion/Significance of Impact: We gathered concrete advice from parents on creating a PA intervention, which we will use to build a social PA program that meets TGD adolescents’ needs. Addressing health disparities and improving PA, social support, and mental health among TGD adolescents will require such stakeholder input to improve upon existing PA and sports opportunities.

OBJECTIVES/GOALS: The Title V Project at the Medical Sciences Campus aims to expand the knowledge in Clinical and Translational Research (CTR) and diversify the CTR workforce throughout Puerto Rico. A Pilot Project Program (PiP) offers research training for Undergraduate students (UgS), Graduate Students (GS), and Undergraduate program Faculty (UgF). METHODS/STUDY POPULATION: Since 2021, the Title V Project has established a rolling application process to which researchers from any scientific background related to CTR in all post-secondary institutions in Puerto Rico may submit research proposals. These are peer-reviewed considering the following criteria: the research team composition must include UgS, GS, and UgF; the primary researcher’s expertise; the significance of the proposed topic related to Puerto Rico’s health problems; and the research plan’s quality. In addition, proposals must include a career plan for student and faculty members to participate in further training in CTR-related topics, such as scientific communication and statistical analyses, also offered through the Title V program. RESULTS/ANTICIPATED RESULTS: Twelve (12) PiP research teams showcase diversity in research areas with representation from Nursing, Occupational Therapy, Cancer, Cell Biology, Microbiology, Anatomy and Medical Images, Electro-Chemistry, Anatomy, and Physiology. In addition, Pip’s team members represent eleven (11) different institutions across seven different geographical areas, whose complete profiles we delineate in the presentation. Teams have the participation of twelve (12) primary researchers, five (5) mentors, twelve (12) UgF, seventeen (17) UGs, four (4) medical students in different stages, and nine (9) GS. We will present the composition, research topics, development, and participants' feedback. DISCUSSION/SIGNIFICANCE: The PiP program has been instrumental in organizing interdisciplinary and interinstitutional research teams. It has proven to be an effective strategy for fostering inclusion, diversity, and equity in CTR and promotes the practice of team science. Teams' research responds to health issues in this Hispanic population.

OBJECTIVES/SPECIFIC AIMS: Familial partial lipodystrophy (FPLD) is an inherited, rare syndrome characterized by selective absence of adipose tissue from extremities which is associated with severe insulin resistance, and metabolic dyslipidemia (with hypertriglyceridemia, and low HDL) Typically, 30%–50% of patients with FPLD demonstrate a pathogenic variant in Lamin A (LMNA) gene that is associated with inherited cardiomyopathy and arrhythmia syndromes. We inquired the prevalence of having abnormal ECGs and echocardiograms in FPLD and whether there is a difference in evaluated parameters with respect to genotype. METHODS/STUDY POPULATION: We conducted a retrospective review of an established a cohort of 58 patients (age range: 12–71, M/F 8/50) with FPLD. Demographic characteristics, genotype, fasting triglyceride, hemoglobin A1c, LDL, and HDL levels were collected; ECGs and echocardiograms were also interrogated. RESULTS/ANTICIPATED RESULTS: Out of 58 patients, 22 (38 %) displayed a pathogenic variant in the LMNA gene. In total, 71% of patients (41/58) had an abnormal ECG and echocardiogram; 40% (23/58) of the patients displayed an arrhythmia on the ECGs (13 in the patients with LMNA variants and 10 in the non-LMNA group). The likelihood of having an arrhythmia was significantly higher in the patients with LMNA variants versus those without (odds ratio of 3.4, CI: 1.1–10.6). DISCUSSION/SIGNIFICANCE OF IMPACT: The overall prevalence of abnormal ECHO and/or ECG is high at 45/58 (78 %) in FPLD. Patients with LMNA variants have a 3.4 times increased risk of developing cardiac arrhythmias compared to those without. We recommend vigilant, monitoring for cardiac disease in FPLD and for arrhythmias in patients with FPLD and LMNA variants.

OBJECTIVES/GOALS: Precisely, the goal of the device is to initiate a friction force between the delivery system and the arterial vessel wall to both assure immediate stent deployment and prevent axial advancement of the stent-anchoring wire. METHODS/STUDY POPULATION: A prototype was constructed and its effectiveness of applying a friction force to a vessel wall was tested ex vivo using an LRX Plus Materials Testing Machine. Afterwards, the experimental performance of the device was compared to that of a finite element simulated model. RESULTS/ANTICIPATED RESULTS: The device demonstrated the ability to apply a friction force to the vessel wall to meet its objective. However, experimental values were consistently greater than those gathered from the simulation. Since the force prescribed by the device is minimal, future work includes increasing the force capabilities of the device and defining force requirements. DISCUSSION/SIGNIFICANCE OF IMPACT: Upon further development and testing, this device can be implemented into endovascular neurosurgery to improve occlusion rates of intracranial aneurysms and reduce patient risk during these operations. CONFLICT OF INTEREST DESCRIPTION: I am pursuing intellectual property on this invention. I was careful not to describe the invention in too much detail in my abstract submission for this reason. This research is my thesis work, and I placed on one year embargo on it before it is published to give us time to sort out IP. I would like to be considered for inclusion in Translational Science 2020 if I am able to get IP on this work before publishing, which I expect will be the case. I have every intention of obtaining IP before the conference in April 2020.

Objectives/Goals: This project, at the intersection of psychological and neuroimaging science, aims to develop and validate an open, transparent, and reproducible workflow for measuring functional brain activity using functional near-infrared spectroscopy (fNIRS), advancing methodological rigor and translational potential in human neuroscience. Methods/Study Population: A Python-based workflow using MNE was designed to improve reproducibility in fNIRS data acquisition. The workflow includes the following: (1) electromagnetic digitizer accuracy testing and distortion correction; (2) integration of head shape and source-detector location data; (3) visualization tools for verifying optode placement accuracy before data collection; and (4) implementation entirely in an open-source software to promote accessibility and transparency. Results/Anticipated Results: Difference in data quality, spatial accuracy, and reproducibility between standard fNIRS practice and this open workflow will be systematically evaluated. Discussion/Significance of Impact: This project advances translation in neuroimaging by embedding rigor and transparency into all stages of the workflow. By openly sharing software tools and validation methods, the workflow enables independent replication, fosters methodological transparency, and reduces reliance on proprietary systems.

The Clinical and Translational Science Award (CTSA) Program recognizes that advancing diversity, equity, inclusion, and accessibility (DEIA) requires moving beyond statements of commitment to transformative actions. In 2021, the CTSA Program created a Task Force (TF) to initiate work in support of structural and transformational initiatives that advance DEIA for the consortium and its individual hubs. We describe the process of forming the expertise-driven (DEIA) TF and our activities to date. We 1) developed and adopted the DEIA Learning Systems Framework to guide our approach; 2) defined a set of recommendations across four focus areas (Institutional; Programmatic; Community-Centered; and Social, Cultural, Environmental); and 3) designed and disseminated a survey to capture the CTSA Program’s baseline demographic, community, infrastructural, and leadership diversity. The CTSA Consortium also elevated the TF to a standing Committee to extend our understanding, development, and implementation of DEIA approaches to translational and clinical science. These initial steps provide a foundation for collectively fostering environment that support DEIA across the research continuum.

Access to local, population specific, and timely data is vital in understanding factors that impact population health. The impact of place (neighborhood, census tract, and city) is particularly important in understanding the Social Determinants of Health. The University of Rochester Medical Center’s Clinical and Translational Science Institute created the web-based tool RocHealthData.org to provide access to thousands of geographically displayed publicly available health-related datasets. The site has also hosted a variety of locally curated datasets (eg., COVID-19 vaccination rates and community-derived health indicators), helping set community priorities and impacting outcomes. Usage statistics (available through Google Analytics) show returning visitors with a lower bounce rate (leaving a site after a single page access) and spent longer at the site than new visitors. Of the currently registered 1033 users, 51.7% were from within our host university, 20.1% were from another educational institution, and 28.2% identified as community members. Our assessments indicate that these data are useful and valued across a variety of domains. Continuing site improvement depends on new sources of locally relevant data, as well as increased usage of data beyond our local region.

OBJECTIVES/GOALS: Physical therapy (PT) is key for treating functional decline that inpatients experience but is a constrained resource in hospital settings. The Activity Measure Post-Acute Care (AM-PAC) score is a mobility measurement tool that has been used to define misallocation of PT. We aim to optimize PT referrals using AM-PAC-based clinical decision support . METHODS/STUDY POPULATION: We conducted a prospective study of patients admitted to University of Chicago Medical Center. AM-PAC scores were assessed by nursing staff every 12 hours. Clinical decision support was designed using validated AM-PAC cutoffs (> 18, a predictor of discharge to home). The tool was embedded in hospital medicine note templates, requiring providers to indicate PT referral status based on current AM-PAC scores. The primary outcome, unskilled consult , was defined as PT referral for patients with AM-PAC > 18. Data were collected for one year prior to implementation and one year after implementation for intervention (hospital medicine) and control (general internal medicine) services. Difference in differences analysis was used to assess the association between the intervention and unskilled consults. RESULTS/ANTICIPATED RESULTS: Between October 2018 and March 2021, 18,241 admissions were eligible for the study. Compared to preintervention, there was a lower rate of referral to PT for patients with high AM-PAC mobility scores in the post-intervention period [18.5% vs 16.6%; X2(1) = 7.02; p < 0.01]. In the postintervention time period, the control group experienced a 2.6% increase in unskilled consults while the intervention group experienced a 2.3% decrease, a difference in differences of 4.9% (95% CI -0.07–-0.03 for difference in differences) controlling for age sex, race, LOS, and change in mobility. Compared to preintervention, there was no statistically significant difference in mean change in mobility score post-intervention for either group. DISCUSSION/SIGNIFICANCE: Our results suggest that clinical decision support can decrease unskilled PT consults. Many functionally independent patients can mobilize with nursing or other mobilization staff. Hospitals should consider mobility score-based decision support to prioritize PT for impaired and at-risk patients.