Click on the image for more information!
Strike Gold in your 2018 Research:
Text Mining Workshop
Text Mining and Natural Language Processing let researchers to turn text into data. Learn new techniques to mine clinical notes, social media and other rich sources of information to advance your research agenda, strengthen grant applications and open new doors to generating patient value.
D2V is excited to offer a free, two-part workshop tailored to both PI’s and analysts:
Morning Session (8:00-12:00): Designed for PI’s, the morning session will provide an overview of how to apply text mining in your research and learn about the results you can expect from the practice.
Afternoon Session (1:00-5:00): A practical skills workshop for analysts to learn about commonly-used tools and techniques in text mining, and participate in hands-on exercises.
8:00-12:00 | 1:00-5:00
May 15, 2018
All study materials will be provided
Education 2 North | University of Anschutz Medical Campus
For more information or to reserve a space, please email Lisa Sandy at firstname.lastname@example.org.
Truth vs. Truthiness in Clinical Data: Drawing Sensible Conclusions from Retrospective Analysis
Mark Weiner, Assistant Dean of Informatics
Temple University School of Medicine
Thursday, April 12 | 1:00 - 2:00 PM
Education 2 North, Room 1202
New sources of Big Data combined with increasingly sophisticated analytical techniques hold the promise of addressing critical problems in healthcare, particularly the relationship between exposures and clinical outcomes. However, the data that is fed into sophisticated analytical methods is typically refined from its raw sources in ways that can influence the results. This talk discusses the idiosyncrasies inherent to the collection, processing and transformation of raw clinical data into an analytical data set, some sensible approaches to address the idiosyncrasies and the many factors one needs to consider in interpreting the results of analyses.
Nudge Units to Improve the Delivery of Health Care
Mitesh Patel, Director
Penn Medicine Nudge Unit
Wednesday, April 11 | 3:30 - 4:30 PM
Education 2 North, Room 2302
Nudges are subtle changes to choice architecture that can have outsized effects on our behavior. The Penn Medicine Nudge Unit is the world's first behavioral design team embedded within the operations of a health care system. The team takes a systematic approach to designing, implementing, and evaluating how nudges can improve health care values and patient outcomes. This presentation will describe the Nudge Unit approach and discuss examples of how nudges and incentives can be used to effectively change clinician and patient behavior.
From Clinical Assessments to Population Data: Vision to Reality
Florence Pirce Grant Professor of Community Health, Brown University
Senior Health Scientist, Providence Veterans Affairs Medical Center
Tuesday, March 6 | 12:00 - 1:00 PM
Education 2 North, Room 1304
Sidney Katz believed that structuring clinical assessments determines the quality of information and this was reflected in the 1987 Nursing Home Reform Act, which mandated all residents be comprehensively assessed using a standardized instrument. Over the last 30 years these assessments, now computerized and linked to Medicare and Medicaid claims, have transformed our knowledge of long term care populations, services and policies. Enlightened facilities stratify risk, target treatment and conduct quality improvement interventions. Researchers link assessment and utilization data, longitudinally tracking patients and create facility, county, state and year aggregates for public use. These data facilitate policy evaluation, pharmaco-epidemiology and comparative effectiveness studies. Most recently, pragmatic cluster randomized trials of quality innovations are undertaken, rivaling research being done by hospital systems. This lecture documents a career shaping this transformation of knowledge and practice, revealing the great opportunities that remain.
Personalizing Medicine and Advancing Systems Physiology Using Mechanistic Learning and Clinical Data
David Albers, Assistant Professor of Biomedical Informatics
Thursday, January 18 | 12:00 - 1:00 PM
Education 2 South, Room 2201
This talk will focus on introducing and demonstrating how to integrate physiologic mechanisms and machine learning to advance health care and physiological and biological understanding. While my collaborators and I use this approach in a variety of contexts, this talk will focus on our work with type 2 diabetes. The computational engine we use is data assimilation (DA), therefore, I will start by discussing what DA is, what it can do, how we use it, and where I see it taking us. Given the task of forecasting blood glucose using sparse, self-monitoring data, I will then demonstrate how DA can be used to define phenotypes, aid in chronic disease self-management, better understand pathophysiology, and also to help understand why prediction of physiology is difficult. I will then delve into some of the methodology and modeling challenges that we must face to advance clinical and scientific applications. I will finish with a portrait of the potential that this approach has for furthering personalized medical treatment as well as our understanding of physiology and pathophysiology.
Analytics Approaches for Strategic, Operational & Clinical Decision-Making in Healthcare
Çağlar Çağlayan, PhD Candidate in Operations Research
Georgia Institute of Technology
Tuesday, December 12 | 12:00 - 1:00 PM
Education 2 North, Room 1206
With the availability of massive datasets, data-driven analytics approaches offer great promise in healthcare research and practice. This seminar will showcase the use of some of these methods by providing cases in studying disease progression and control and analyzing healthcare systems.
The first part of the talk will focus on breast cancer screening policies in the high-risk population. The performance of mammography is not satisfactory for women at high-risk for breast cancer. Other technologies such as ultrasound and MRI might address some of the limitations of mammography. Currently, there is no consensus on optimal use of these technologies. Çağlayan and colleagues developed a novel optimization model to identify optimal, practical and cost-effective screening policies. He will present their main findings and discuss the policy implications of their results on the current practice.
The second part of the talk will present a novel method to optimize staffing levels in complex healthcare delivery systems such as Emergency Department (ED). Keeping waiting times at acceptable levels is not only important for patient satisfaction but also a patient safety concern in EDs. However, optimizing physician staffing levels is not an easy task for a system with unscheduled time-varying arrivals, medium-to-long service times, multiple patient classes and multiple treatment stages. In this talk, Çağlayan will
introduce their novel network model to tackle the physician-staffing problem in EDs.
If time permits, the talk will also briefly cover the use of multi-state survival analysis models to study the clinical course of diseases with multiple intermediate- or end- points.
Improving Delivery of Palliative Care in Oncology: Identifying Gaps, Understanding Determinants, and Informing Scalable Practice Change
Devon Check, Delivery Science Fellow
Kaiser Permanente Northern California, Division of Research
Thursday, December 7 | 12:00 - 1:00 PM
Education 2 North, Room 2102
Palliative care is increasingly recognized as a critical component of high-quality cancer care, particularly for patients with advanced disease. However, evidence-based palliative care practices – including strategies for managing disease- and treatment-related symptoms – are not always implemented in routine oncology practice. I will discuss my experience conducting research related to improving delivery of evidence-based palliative care among patients with cancer. Specifically, I will describe my prior studies, which have identified gaps in care delivery; my current projects, which focus on understanding determinants of these gaps; and my plans for future research, which will leverage implementation science methods to support equitable, population-based palliative care delivery for patients with cancer or other serious illnesses.
The Snow System: The Norwegian Practice-Based Research Network IT Infrastructure
Johan Gustav Bellika, Professor
Norwegian Centre for E-Health Research
Thursday, November 30 | 12:00 - 1:00 PM
Education 2 North, Room 1202
Primary care offices in Norway normally use in-house standalone EHR systems, which therefor represent a totally decentralized and heterogeneous data resource with no IT support for research. A consequence of this situation is that it becomes very time consuming for the GPs to find patients that are eligible for clinical studies. Performing research in primary care in Norway have for this reason been time consuming and with a high risk of failure. The Snow system solve this problem by utilizing distributed computations to produce lists of eligible patients from the local EHR systems for each GP that have consented to participate in a research project. The Snow system will perform research data handling to minimize the workload put on GPs to participate in research and provide the researchers with fast access to study data.
Parsing Big Data B.S. for Fun and ProfitJames King, Principal Data Scientist, U.S. Department of Defense
Monday, July 17 | 12:00 - 1:00 PM
Education 2 North, Room 2302
Do you have a strong desire to monetize your synergies and provide value-added, cross-platform solutions? Do you yearn to foster flexible, paradigm-shifting strategic incentivization schemes for mission-critical core competencies?
Neither do we.
In this interactive conversation, James King will discuss data science and machine learning concepts in plain, useful language. You'll get a better sense of which methods might be useful in your current projects and you'll see a fun selection of other people's "gee whiz" results. IT spirits permitting, you'll also see the live training and operation of a machine learning model that "sings" the blues.
Medical Maximizing vs. Minimizing: A Framework for Understanding Patient Values and Preferences for Healthcare
Laura Scherer, Assistant Professor of Psychology, University of Missouri
Thursday, May 11 | 12:00 - 1:00 PM
Education 2 North, Room 1206
There is growing recognition that some medical interventions are not always helpful, and can sometimes be harmful, and it follows that just because we can do something about a health problem does not always mean that we should. As result, guidelines increasingly recommend shared decision making and incorporation of patient preferences into healthcare decisions, and therefore a critical question becomes how patients decide to take action and do something, versus not, when it comes to their health.
This talk will present evidence that some people can be described as “medical maximizers” who prefer an active approach to healthcare, whereas others are “medical minimizers” who prefer watch-and-wait approaches. Maximizers may frequently seek medical care, potentially at significant personal and financial costs, and with little gain in terms of health outcomes. By contrast, minimizers may tend to underutilize medical resources, with potentially negative health outcomes. I will discuss the development of a 10-item scale to measure this individual difference, as well as evidence that it uniquely predicts a range of healthcare utilization outcomes. The measure also strongly predicts patient preferences for interventions from cancer screening to end of life care. Finally, some preliminary data suggest that this individual difference plays a role in patient satisfaction with their physician. I will discuss the potential utility of this measure for identifying and addressing over- and under-utilization, developing targeted informational interventions, and fostering shared decision making.
Deliberative Practices in the Health Arena: Meanings, Methods, and Challenges to Diversity, Inclusivity, and Equity
Erika Blacksher, PhD
is Associate Professor and Director of Undergraduate Studies in the Department of Bioethics & Humanities at the University of Washington. Blacksher will discuss the evolving meaning of public deliberation, what distinguishes it from participatory processes more generally, and challenges to its methods and goals. Drawing on the AHRQ-funded Community Forum Study and her own work, Blacksher will explore challenges that strike at public deliberation's most fundamental goal—to be a democratizing force in public policy.