Research experience is often important for academic and career development. This paper describes the implementation and impact of a training program for temporary research assistants (RAs) at an academic medical center. The program includes a 9-month didactic lecture series covering research and professional development skills, a Quality Improvement project focused on improving research processes, and manuscript writing. Overall, the program goals of increasing confidence, self-efficacy, job satisfaction, and well-being, as well as providing an opportunity for career exploration, were met. Thus, this program has the potential to support temporary RAs and enhance their early research experiences.

Clinical research is complex, and research-related terms can be challenging to understand. Clear, supportive communication with patients, potential study participants, and their caregivers must be prioritized by healthcare providers as well as investigators and their research teams. In clinical research, health literacy best practices support the ethical tenets of respect, justice, and beneficence. Plain language advances the understanding of informed consent documents, as well as comprehension of educational information, recruitment materials, study instructions, and study results summaries, among others. Further, a more collaborative research partnership is fostered when study participants are given understandable materials, while a lack of understanding can delay accrual and decrease adherence. We launched a pilot initiative to develop a consensus-driven, plain language clinical research glossary to promote clarity, consistency, and transparency across clinical research stakeholder groups. The resulting resource, described herein, is intended to be used widely to support a greater understanding of clinical research and empower study participants. Considerations for expansion are also discussed.

OBJECTIVES/GOALS: The Colorado Immersion Training in Community Engagement (CIT) provides experiential training in community-based participatory research (CBPR). We summarized evaluations from the first ten years on program outcomes and lessons learned to inform future programming and the field of community research. METHODS/STUDY POPULATION: CIT is a six-month program that includes didactic sessions, group readings and reflections and a one week immersion experience in a Colorado community experiencing health disparities. The Evaluation Center - University of Colorado Denver provides external evaluation services to the Colorado Clinical and Translational Science Institute (CCTSI). Evaluators used three primary methods to collect data to evaluate the program; Document review, interviews, and surveys. Evaluation data were analyzed using the framework of CBPR Principles and the CBPR conceptual logic model to understand potential shifts in researchers’ thinking and actions, as well as to identify the potential for longer-term impact. RESULTS/ANTICIPATED RESULTS: As of 2020, CIT trained 122 researchers. 25 CIT alumni went on to receive 33 CCTSI Pilot Grants. Out of these 33 grants, an additional $8,723,000 of external grant funding was awarded to continue projects. Community Research Liaisons are a crucial component of the program through their efforts to bridge academic researchers and communities. Survey results indicated that community partners reported high levels of satisfactions with the program and shared stories about the impact they have seen in their community. Common challenges for researchers remain 1) protected time and institutional value and, 2) funding opportunities. DISCUSSION/SIGNIFICANCE: CIT has grown CBPR researchers. Similar programs should consider contributing factors: establishing institutional value and funding for CBPR initiatives with protected time for developing and maintaining community relationships. CIT offers a crucial connection between academia and community.

OBJECTIVES/SPECIFIC AIMS: The aim of this study is to compare intestinal phosphorus absorption in healthy adults and moderate stage chronic kidney disease patients in the context of a controlled feeding study METHODS/STUDY POPULATION: Participants are 30-75 years old and include 10 healthy subjects and 10 moderate-staged CKD patients. Each subject pool will be enrolled in a 9-day study period including 7 days of controlled feeding of a 1500 mg phosphorus diet. Following the controlled feeding, two days of absorption tests will take place (oral and IV tests) utilizing radioisotopic phosphorus to calculate fractional absorption efficiency. RESULTS/ANTICIPATED RESULTS: Current enrollment has produced 7 total matched subject (current n = 14/20). Four of the 7 pairs of completed subjects are female and 3 of 7 are black. Preliminary kinetic modeling data from the first enrolled subject show a moderate CKD patient with fractional absorption of 0.375. With forthcoming analyses, we expect that this fractional absorption result will not be statistically different from this subject’s matched pair, nor will each groups average absorption be different from the other. Additionally, we expect absorption to be maintained even with changes in secondary outcomes measures in serum (FGF23, 1,25-dihydroxyvitamin D, parathyroid hormone, and total phosphorus) in CKD patients. DISCUSSION/SIGNIFICANCE OF IMPACT: Lack of statistical difference in fractional phosphorus absorption between gropus would support that intestinal phosphorus absorption is inappropriately normal in CKD patients compared to healthy adults, despite evidence of abnormal phosphorus homeostatic mechanisms. Future studies will consider the effect of dietary P restriction, the most common nutrition intervention in moderate stage CKD, on fractional absorption efficiency in CKD.

OBJECTIVES/SPECIFIC AIMS: The 2 primary objectives were to (i) insure that Scholars can effectively communicate the translational impact of their research to a lay audience and (ii) assess the benefits and efficacy of having community, as well as faculty members, judge the translational impact of KL2 Scholar’s poster presentations. An explicit secondary goal was to further the engagement of community members in CTSI-sponsored translational research. METHODS/STUDY POPULATION: CTSI’s Education, Community Engagement, Discovery and Translation, and Translational Workforce Development Cores created the translational impact questions and evaluation sheets. The Community Engagement and Office of Discovery and Translation recruited community judges from their respective networks and they were assigned to relevant studies. Scholars were provided with the judges scoring template in advance. After the Research Poster Session, the KL2 Scholars evaluated the quality of their presentations and the impact of having feedback from Community Judges. The Community Judges evaluated their perceived “added value” to the research presentations and their interactions with the Scholars. Both Scholars and judges completed evaluations of the poster presentation and judging process, performed on a 5-point Likert scale. RESULTS/ANTICIPATED RESULTS: KL2 Scholars felt that the community impact judges provided valuable feedback on their research (3.8/5) and were satisfied overall with the poster session (3.4/5). In evaluating their own presentations, Scholars tended to rate themselves higher (4.2–4.6/5) on the clarity of their translational impact presentations than the community judges rated the Scholars (4.1–4.2/5). Scholars also rated themselves somewhat higher in the quality of their dealing with any ethical issues and their dissemination plan (4.0/5) than the community judges (3.8/5). Judges were very positive and felt they brought value to the experience (4.2–4.4/5). DISCUSSION/SIGNIFICANCE OF IMPACT: Community judges added qualitative value to the Scholar presentations based on the Scholar and community judge evaluations and based on comparison based on prior year poster sessions. Documenting the degree of impact of the combination of this proscribed poster format and community-judging process awaits future assessment of Scholar presentations before and after the next annual poster presentation.

OBJECTIVES/GOALS: To build a multi-function health profile for older adult (>50 years) women with a history of incarceration; explore group differences by age [45-54, 55-64, 65+], race, and length of incarceration; compare with age-matched control with no history of incarceration; and identify relative contribution of life course risk and asset factors. METHODS/STUDY POPULATION: We will analyze data from the Health and Retirement Survey (HRS) (UMichigan Institute of Social Research). The sample: women >50 years who took the leave-behind psychosocial questionnaire. The sample includes women with history of incarceration (n = 118; 2.93%) and (n = 115; 2.55%), respectively, and in the control n = 4,021 women (2012) and n = 4,114 women (2014). We will use descriptive statistics to profile physical, functional, cognitive, and social health; bivariable tests to compare groups on age-related morbidity, multimorbidity, frailty, and 4-year mortality risk; measure within group differences by age strata and race; estimate GLMs for effects of life course risk on dependent variables in and between groups; and if data permit, test direct mediation by life course risks and indirect by life course assets. RESULTS/ANTICIPATED RESULTS: Our results will characterize health in 4 health domains of women over 50 who have a history of incarceration. In bivariable analysis, we expect significant differences between groups on the dependent variables. Based on previous study using this data set, life course and accumulated stress theory, and our own previous research, we hypothesize that women with incarceration history will have more and earlier cardiovascular disease, stroke, and multimorbidity, higher 4-year mortality risk, and more and earlier cognitive impairment. We anticipate significant contributors to aging-related health outcomes to include childhood challenge and trauma history and, for Black women, perceived racial discrimination. We anticipate education and social support will partially mediate relationships. DISCUSSION/SIGNIFICANCE: The CHARMS findings, based on large-sample, representative, longitudinal HRS survey data, will contribute a profile of multi-function health status, risk, and assets in older women with CLS involvement. The much-needed characterization of aging in the group will set the stage for future interventional study to guide shifts in clinical practice.

OBJECTIVES/GOALS: Over 30% of patients with rheumatoid arthritis (RA) exhibit fibromyalgianess, a symptom cluster associated with increased pain sensitivity. Up to half of RA patients use oral glucocorticoids (GCs) long-term despite their known, dose-dependent toxicity. We examined the association between fibromyalgianess and oral GC persistence in RA patients. METHODS/STUDY POPULATION: We used data from the Central Pain in Rheumatoid Arthritis (CPIRA) cohort to follow participants with active RA on oral prednisone who initiated a new disease-modifying anti-rheumatic drug. We measured fibromyalgianess using the Fibromyalgia Survey Questionnaire (FSQ), previously shown to correlate with key fibromyalgia features often superimposed upon RA. We stratified fibromyalgianess severity as follows: FSQ<8 low, 8-10 moderate, >10 high/very high. We defined GC persistence as GC use at 3 month followup visit. We assessed the association between baseline fibromyalgianess (exposure) and GC persistence at followup (outcome) using multiple logistic regression, adjusted for demographics, RA duration, serostatus, and inflammatory activity measured by swollen joint count and C reactive protein. RESULTS/ANTICIPATED RESULTS: Of 97 participants on prednisone at baseline, 65% were taking prednisone at follow-up. Fifty-seven percent of participants with low baseline fibromyalgianess had persistent GC use, compared to 84% with high or very high fibromyalgianess. After adjustment as outlined above, participants with high/very high baseline fibromyalgianess remained more likely to be on prednisone at follow-up, relative to those with low fibromyalgianess (OR 4.99 [95% CI 1.20 – 20.73]). DISCUSSION/SIGNIFICANCE: In this cohort of patients with active RA, high fibromyalgianess is associated with persistent GC use, independent of inflammatory activity. This finding suggests non-inflammatory pain related to fibromyalgianess may be misclassified as inflammatory pain related to RA disease activity.

OBJECTIVES/SPECIFIC AIMS: Acute care research is a unique area of clinical research that demands specialized skills, knowledge, and talents from empathetic professionals working in the field. Building off existing competencies for clinical research professionals, the Cincinnati Acute Care Research Council (ACRC) developed additional areas of competency for professionals working in the acute care research discipline. METHODS/STUDY POPULATION: Qualitative data obtained from job shadowing, clinical observations, and interviews were analyzed to understand the educational needs and desires of the acute care research workforce. We then utilized Bloom’s Taxonomy to build acute care research competencies that are measurable for job performance and build off of foundational clinical research professionals’ domains and competencies developed by the Joint Task Force of Clinical Trial Competency. RESULTS/ANTICIPATED RESULTS: Results suggest 35 special interest competencies for acute care clinical research professionals under 8 common domains set by the Joint Task Force of Clinical Trial Competency. Additionally an approved ACRC tactic, from actionable learnings through community assessments throughout 2017, is the creation of a Task Force made up of acute care research Principal Investigators and Clinical Research Directors to focus on the identified training and professional development obstacles in the clinical research enterprise. DISCUSSION/SIGNIFICANCE OF IMPACT: The competencies developed for acute care research should serve as guidelines for training a workforce prepared for the challenges of conducting research with each acute audience, as its own vulnerable population. These competencies will guide development of a multi-pronged program of professional development that will include new hire onboarding, new hire on-job training, and ongoing on-job training.

OBJECTIVES/GOALS: The A2CPS was funded by the National Institutes of Health (NIH) Common Fund to identify biomarkers and their collective biosignatures (combination of several biomarkers) that predict susceptibility or resilience to the development of chronic pain. METHODS/STUDY POPULATION: The A2CPS includes 2-Multisite Clinical Centers (10 recruitment sites and 6 data collection sites), 1-Clinical Coordinating Center, 1-Data Integration and Resource Center, 3-Omics Data Generation Centers, and representation from the NIH. The A2CPS will recruit a large cohort from 2 different surgical interventions, total knee arthroplasty (n?1800) and thoracotomy (n?1800). This observational study will collect candidate and exploratory biomarkers across the following domains: clinical pain, fatigue, function, sleep, psychosocial, genomics, proteomics, metabolomics, lipidomics, pain sensitivity, and brain imaging. Data will be collected before and up to 3 months after surgery to determine factors that predict chronic pain at 6 months. RESULTS/ANTICIPATED RESULTS: Recruitment started in 2021 following standard operating procedures and is ongoing at both Multisite Clinical Centers. The A2CPS will provide an example of collaborative, multidisciplinary efforts in establishing a data repository consisting of biopsychosocial markers that will be available to the research community to test novel hypotheses. This presentation will describe the conceptual design, study aims, biomarker selection, protocol standardization and study implementation for the A2CPS. An update on study progress and data completeness will be presented. Final results will be reported after study completion which is anticipated by 2024. DISCUSSION/SIGNIFICANCE: Identifying biomarkers and biosignatures that predict high- versus low-risk for the transition to chronic pain will inform future clinical trials, identify novel therapeutic targets, and advance personalized pain treatment strategies; ultimately transforming the prevention and treatment of chronic postsurgical pain.

ABSTRACT IMPACT: The perspectives and guidance from adolescents and young adults (AYA) reported in this study could inform the evidence-based development and delivery of mobile health (mHealth) interventions to improve the health of AYA with chronic diseases. OBJECTIVES/GOALS: To elicit advice from AYA with chronic healthcare needs regarding if and how mHealth interventions could effectively promote illness self-management skills. We selected this goal because including the user perspective from the beginning of the design process could lead to greater future adoption. METHODS/STUDY POPULATION: We purposively recruited AYA patients from a pediatric hospital with heterogeneous chronic illnesses to identify universal chronic disease views rather than condition-specific perspectives. We conducted qualitative face-to-face semi-structured interviews with (N = 19) AYA between 16 and 20 years old (63.2% Latinx; 21.1% Black; 10.5% White; 5.3% Multiracial). Using ATLAS.ti, three coders completed thematic analysis to inform a conceptual framework on how AYA believe mHealth interventions could promote the development and maintenance of self-management skills. Member checking was conducted over the phone to obtain participant feedback on themes to enhance the validity of qualitative results. RESULTS/ANTICIPATED RESULTS: Results suggest that AYA develop self-management skills through several strategies, including 1) getting organized, 2) ‘making it work for me’ and 3) keeping the ‘right’ mentality. AYA described developing these strategies through: 1) receiving social support, 2) accessing helpful tools and technologies, and 3) going through a maturation process. They provided recommendations for how mHealth interventions could improve this process, including: 1) ‘what’ recommendations, describing the content or active ingredients that should be included in mHealth interventions, and 2) ‘how’ recommendations, describing the technological aspects or style in which the interventions should be delivered. DISCUSSION/SIGNIFICANCE OF FINDINGS: The results suggest that an appealing mHealth intervention could increase the support for AYA patients to proactively acquire self-management skills, avoiding trial-and-error or uneven access to guidance. Improving self-management could prevent poor health outcomes and increase quality of life.

OBJECTIVES/SPECIFIC AIMS: Describe the framework for tier advancement of research professionals. Describe the various forms of assessments of competencies. How competencies are used to provide transparency into professional development opportunities. Discuss the results of the first tier advancement opportunity for research staff. METHODS/STUDY POPULATION: These processes were developed at Duke, an academic medical center with over 2000 active clinical research protocols and 300 new clinical trials per year. Roughly 500 employees are categorized into tiered classifications, allowing them opportunities for advancement through competency testing. Approximately 10% opted for tier testing, and their results will be shared. RESULTS/ANTICIPATED RESULTS: Competency assessments were developed for all 42 of Duke’s research professional competencies, some using 2 modalities of testing. Almost 12% of the research professionals classified in tiered positions opted to attempt the tier advancement process. Of those, 37 completed, and the vast majority reached their desired tier. Results by competency will be provided. DISCUSSION/SIGNIFICANCE OF IMPACT: The use of objectively assessed competencies is an important step in the development of a workforce. By (1) maintaining alignment with industry standards for competencies, (2) holding staff to a high bar, and (3) offering a consistent approach to career growth, Duke is working to develop and maintain a workforce that supports high quality research.

OBJECTIVES/SPECIFIC AIMS: This study aims to evaluate potentially traumatic events (PTEs) and its relationship with posttraumatic stress symptoms (PTSS), posttraumatic growth (PTG), and resilience in a sample of help-seeking individuals in Puerto Rico. METHODS/STUDY POPULATION: This is an analytic, cross sectional design. Adults receiving health services will participate in the study. Recruited participants will provide informed consent during a visit to a community mental health clinic or community hospital. They will complete a demographic document and four retrospective questionnaires about the variables of study. RESULTS/ANTICIPATED RESULTS: We expect that a high rate of potentially traumatic events (PTEs) is associated with an increased rate of posttraumatic stress symptoms (PTSS). We also expect that a high rate of PTSS is associated with an increased rate of posttraumatic growth (PTG). We expect that a high rate of resilience is associated with low rates of PTSS and PTG. DISCUSSION/SIGNIFICANCE OF IMPACT: This is a first step in the development of effective, clearly targeted interventions, specifically designed to treat negative effects, and also to facilitate positive change and resilience after PTE exposure.

OBJECTIVES/GOALS: Scaffolding aids learning by gradually removing assistance to encourage independence (Bickhard, 2013; Gross, 2015). Mightier is a commercially available biofeedback game that fosters emotion regulation (ER) skill practice through play. This study aims to evaluate the ability of Mightier to scaffold the learning of ER and reduce irritability. METHODS/STUDY POPULATION: Data were collected via online caregiver report. Inclusion criteria were age of child (<= 18 years) and study enrollment prior to the child engaging with Mightier. Children wear a heart rate (HR) monitor while playing games in the Mightier app library. As their HR increases, play becomes more difficult. Children can pause the game to use a scaffolded ER activity or regulate independently to return the game to normal difficulty levels. Caregivers were instructed to use the game ad libitum. Participants included caregivers of 195 children (Mage = 7.84 years old; range = 4-18 years old; 24.28% female; 74.57% male); the sample was predominantly White (56%). Caregivers completed the Affective Reactivity Index, a measure of their child’s irritability, before and after playing Mightier for 8-12 weeks. RESULTS/ANTICIPATED RESULTS: A first Wilcoxon Signed Rank Test revealed a significant reduction in the ratio of scaffolded cooldowns (using a guided activity to regulate) to total cooldowns on the first play day (Md =.50) versus the last play day (Md = .22), z = –6.51, p < .001. A second test revealed a statistically significant increase in the ratio of independent cooldowns (regulating on their own) to total cooldowns on the first play day (Md = .50) versus the last day of play (Md = .80), z = 6.34, p < .001. Given the significant, inverse relationships between scaffolded and independent ER on the first play day versus the last play day, further analyses will examine potential mediation and moderation effects of game engagement (play minutes, total cooldowns, scaffolded cooldowns, and independent cooldowns) on changes in irritability. DISCUSSION/SIGNIFICANCE: ER is vital for healthy development and protects against mental health challenges (LeBlanc etal., 2017). Across domains, scaffolded practice promotes learning (Vygotsky, 1978). Results reveal that scaffolded practice leads to independent ER during play. Future research should explore whether this pathway leads to independent ER outside of play.

OBJECTIVES/GOALS: Cannabis laws may impact cannabis and tobacco use, given high prevalence of co-use of these products among young adults (YAs). The objective of this study was to examine trends in YA any cannabis, blunt, cigarette, and cigar use from 2002-2018 in states that passed adult and medical use (AMU) or medical use only (MUO) cannabis laws during that time (N=16). METHODS/STUDY POPULATION: Using data from the National Survey on Drug Use and Health, we conducted a segmented regression analysis to calculate absolute percent change (APC) in past 30-day cannabis and tobacco use between time points. The National Cancer Institute’s Joinpoint software was used to also estimate points of inflection (Joinpoints) when the slope of a trend significantly (p<0.05) changes. Separate models were estimated for each state, with time as the independent variable measured in years. Up to three Joinpoints per model were allowed. The model with optimal Joinpoints was determined using a model selection criterion via a permutation test. Joinpoints and APCs were compared with key legalization dates to describe patterns within and across states with varying cannabis policies. RESULTS/ANTICIPATED RESULTS: Generally, the 16 states showed a steady decline in YA cigarette smoking over time, a slight decline in cigar smoking, and increases in cannabis and blunt use. AMU states had lower average 2018 prevalence of cigarette smoking than MUO states (18.3% vs. 21.5%) and higher cannabis use (32.3% vs. 21.3%). Cannabis use consistently increased following opening of MUO retail outlets. Generally, there appears to be a slight delay in cannabis use increases following AMU laws, and in some states temporary declines. For example, Washington experienced an initial decrease (-20.3%) following AMU passage (2012) then increase (+16.3%) after retail dispensaries opened in 2014. In AMU states, blunt use has surpassed cigar smoking, while in MUO states, the prevalence of blunt and cigar use is similar. DISCUSSION/SIGNIFICANCE: Introduction of cannabis laws are correlated with increases in YA cannabis and blunt use, with higher cannabis use in AMU states. Trends may also correlate other state political, economic, or social factors. Joinpoint regression can assess changes in a policy’s target behavior with no a priori assumptions regarding timing of policy effects.

Objectives/Goals: Multivariable statistical models are commonly used as clinical decision aids. However, more research examining real-world implementation of models is needed. The goal of this mixed methods study is to explore to what degree clinician uptake of risk models is influenced by perceptions of the potential risks and benefits of taking action on model output. Methods/Study Population: Mixed qualitative and quantitative data for the ’low-risk/low-reward’ scenario arise from an implementation trial of a electronic medical records-based (EMR) model to alert primary care providers of potential cognitive problems in patients ages 65 and older. We evaluated adoption of the EMR model based on provider actions (cognitive screening/referrals/medications/diagnoses) following Epic best practice alert (BPA), and acceptability of the models through semi-structured qualitative interviews. Data for the ’high-risk/high-reward’ scenario will be gathered from clinical notes documenting the use of statistical models to support elective surgical decision-making in scenarios where such models have been regularly used at our institution. Mixed methods analyses will be holistic. Results/Anticipated Results: Early quantitative analyses following implementation indicated that primary care providers largely did not adopt the BPA, but did take action (referrals, diagnoses, medications) when automated cognitive screening was performed in preparation for the appointment. Qualitative analyses of interview data to understand provider perspectives are ongoing. In the high-risk/high-reward scenario, we expect that providers will more frequently utilize statistical models for decision-making and report positive attitudes regarding their use, when compared to the low-risk/low-reward scenario. Discussion/Significance of Impact: Though used in practice with increasing frequency, the implementation of personalized statistical prediction models has not been widely studied. Understanding in what scenarios providers feel risk models are useful and actionable can guide decision-making for the development and implementation of future models.

Objectives/Goals: In this project, we set out to supplement our existing grant writing workshops with targeted, learner-centered, multimodal training. This method will assist us in moving toward a more equitable training landscape, reaching a wider variety of learners, by freely disseminating these resources. Methods/Study Population: To increase access and impact of training materials for our career development grant writing workshops, we restructured our pre-workshop training videos. We culled expert advice from lengthy recorded lectures into brief (less than 5 minute) how-to videos that target instruction to writing specific sections of an NIH K grant. We then coupled these how-to videos with easy-to-navigate, open access online courses that further illustrate best practices for writing key sections of NIH K grants. These resources were given to registered workshop attendees and made available through a public Canvas course, the Diamond portal, MICHR website, and U-M Innovation Partnerships to disseminate the materials through multiple channels. Results/Anticipated Results: Four online courses and complementary videos were developed over six months, each focusing on a specific section of the NIH K grant proposal. These resources provide targeted instruction for writing the Specific Aims, Candidate Background, Career Development Plan, Career Goals and Objectives, and Mentor Letter. Learners accessed all four of the online courses. Released in January 2024, we continue to gather data on whether learners believe their knowledge about writing successful K grants has increased after using the resources, if they believe the courses have prepared them to write the section of the grant covered, and whether learners would recommend the courses. We will analyze these results to better understand how learners are using and responding to these new resources. Discussion/Significance of Impact: These how-to videos and online courses provide targeted, learner-centered training and fill an important gap by meeting learners where they are and extending the impact of our training beyond our institution. Widely disseminating online interactive training resources is a model we are applying beyond grant writing to other projects.

OBJECTIVES/SPECIFIC AIMS: The prevention, management, and treatment of HIV, STDs, and HCV requires continuous training that reflects contemporary best-practice and innovative care models. In order to improve the NYS AIDS Institute’s comprehensive web-enabled training program, which enhances the capacity of a diverse healthcare workforce, a needs assessment (NA) of our community of practice (CoP) is needed to better understand their training needs, circumstances, and instructional modalities preferences. The goal of the assessment was to better understand our CoP’s preferences of online trainings, and as a result to develop a “responsive design” system that will enhance user’s learning experience thus improving patient care. METHODS/STUDY POPULATION: We developed and deployed an NA survey using REDCap. The instrument consisted in 27 questions related to providers’ preferences on receiving continuing educational training and their use of technologies, including mobile platforms, online modules, webinars, and telehealth. As part of the recruitment strategy, several resources were deployed over a 1-month recruitment period including sequential email blasts, website promotion, and assessment links included in newsletters and social media. Weekly reminders were also used to promote the participation from our CoP. RESULTS/ANTICIPATED RESULTS: A total of 310 respondents participated in the NA, with 85.8% from NYS. 177 were clinicians (20.5% MD, 2.9% PA, 17.3% NP, and 16.3% RN) and 133 nonclinical providers (case/care managers, social workers, public health professionals, coordinators/administrators, and other). The participants worked in hospitals, community health centers, substance use centers, private practices, and state/local health departments. More than 90% of respondents indicated that they preferred both live/in-person and online training, and participants most strongly indicated that they stayed up-to-date on current developments through CDC, the AIDS Institute, and conferences. More than 60% of respondents considered that receiving CE credit for the training was very important and 28% indicated they would use training materials in Spanish if offered. In terms of technology, over 80% of the respondents preferred computers, but more 50% also used mobile devices (computer at home 61.8%, computer at work 85%, tablet 29.9%, iPhone 20.9%, Android 16.6%, other device 2.3%). DISCUSSION/SIGNIFICANCE OF IMPACT: Accessing an online CoP provided a useful opportunity to assess training needs and preferences of clinical and nonclinical providers. Most providers indicated that they were primarily likely to use a work computer to complete online training or secondarily a home computer. With a significant portion of respondents indicating use of tablets, smartphones, and other devices, online training opportunities should be developed with responsive design to assure flexibility and access. In addition to online training, participants indicated that they also strongly valued live, in-person training. Offering training with CDC and the NYS AIDS Institute branding, in Spanish, together with offering continuing education credit, were all seen as desirable training elements. Accessing this online CoP helped streamline and target training priorities and logistics.

OBJECTIVES/GOALS: Diverse medication-based studies require longitudinal drug dose information. EHRs can provide such data, but multiple mentions of a drug in the same clinical note can yield conflicting dose. We aimed to develop statistical methods which address this challenge by predicting the valid dose in the event that conflicting doses are extracted. METHODS/STUDY POPULATION: We extracted dose information for two test drugs, tacrolimus and lamotrigine, from Vanderbilt EHRs using a natural language processing system, medExtractR, which was developed by our team. A random forest classifier was used to estimate the probability of correctness for each extracted dose on the basis of subject longitudinal dosing patterns and extracted EHR note context. Using this feasibility measure and other features such as a summary of subject dosing history, we developed several statistical models to predict the dose on the basis of the extracted doses. The models developed based on supervised methods included a separate random forest regression, a transition model, and a boosting model. We also considered unsupervised methods and developed a Bayesian hierarchical model. RESULTS/ANTICIPATED RESULTS: We compared model-predicted doses to physician-validated doses to evaluate model performance. A random forest regression model outperformed all proposed models. As this model is a supervised model, its utility would depend on availability of validated dose. Our preliminary result from a Bayesian hierarchical model showed that it can be a promising alternative although performing less optimally. The Bayesian hierarchical model would be especially useful when validated dose data are not available, as it was developed in unsupervised modeling framework and hence does not require validated dose that can be difficult and time consuming to obtain. We evaluated the feasibility of each method for automatic implementation in our drug dosing extraction and processing system we have been developing. DISCUSSION/SIGNIFICANCE OF IMPACT: We will incorporate the developed methods as a part of our complete medication extraction system, which will allow to automatically prepare large longitudinal medication dose datasets for researchers. Availability of such data will enable diverse medication-based studies with drastically reduced barriers to data collection.