Projects

Our Current Projects

COACH - Collaboration Oriented Approach to Controlling High blood pressure

One of our projects is the development of a patient-facing clinical decision support (CDS) tool for hypertension. This work is funded by AHRQ as a two-year U18 grant, the main goal of which is to understand how best to support patients as they address hypertension, focusing on their goals, experiences, motivation, and engagement with the health care system, and to build a FHIR-based application that embodies those best practices while facilitating adherence to clinical practice guidelines.

Specific Aims:

  1. With the help of patients and an interdisciplinary research team, translate multiple hypertension guidelines and protocols into Clinical Quality Language (CQL) query modules that are flexible and evidence-based.
  2. Leveraging previous work, build CDS artifacts and a FHIR application that can elicit inputs and provide guidance to both patients and health care teams, validating against a database of patients with hypertension.
  3. Evaluate this application with appropriate patients and health care teams to learn its potential impact in assisting tailored decision making, refining the CDS artifacts and disseminating.

Funding Agency: AHRQ
Collaborators: University of Missouri-Columbia, Vanderbilt University Medical Center

Implementation of an Electronic Care Plan for People with Multiple Chronic Conditions (eCarePlan)

The Agency for Healthcare Research and Quality (AHRQ) and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) aim to develop and pilot test an electronic care planning application that will improve care coordination for people with multiple chronic conditions (MCC). The application will have patient- and clinician-facing elements.

OHSU, in collaboration with RTI International and EMI Advisors, will design and implement a series of pilot tests to assess the usefulness of the application for both patients and clinicians. By investigating whether the app effectively facilitates standardized data collection and sharing across clinical and community settings and systems, this study will determine whether the app has the potential to improve the way complex care is coordinated.

Care plans support a personalized way to document an individual’s health care goals. They are crucial tools in addressing the health needs of people with MCC. Electronic care plans enable clinicians to view relevant information anytime, anywhere, and enable patients to access their health information, directly. When both patients and clinicians see and interact with the same information in real time, both clinical and nonclinical needs are more likely to be addressed.

Funding Agency: AHRQ
Collaborators: RTI International, EMI Consultants

Bridge2AI – Voice as a Biomarker of Health (Skills & Workforce Development)

Funded by the NIH Common Fund’s Bridge to Artificial Intelligence (Bridge2AI) program, this multi-institution project aims to integrate voice as a biomarker of health. The project will generate new data sets using voice samples and AI/ML to identify mood disorders, respiratory diseases, voice disorders, neurological disorders, sleep disorders, and much more. The program aims to build a large-scale, diverse human voice database using privacy-preserving AI, giving medical teams a new tool for diagnosing conditions associated with voice alterations.

OHSU is partnered with Washington University to lead the Bridge2AI Skills & Workforce Development (SWD) module. SWD is focused on bridging expertise across people in the biomedical and behavioral research domains to develop an AI/machine-learning research workforce. This project will equip current and future workforces with the skills to correctly use and evaluate responsible biomedical AI models capable of assisting in the screening, diagnosis, and treatment of a broad range of diseases while maintaining patient autonomy.

Funding Agency: NIH
Collaborators: University of South Florida, Washington University, and several other institutions

Multiple Chronic COnditions: MultiPle dAta SouRcEs (MC COMPARE)

CMP is collaborating with researchers from Brigham and Women’s Health (BWH) to provide an interoperable tool that makes it easy to gather health information from multiple sources for people with multiple chronic conditions (MCC). This trial is significant as it addresses the issue of clinical data fragmentation that hinders the effectiveness of interventions for people with MCC. The study population for this multi-site research includes those enrolled in ongoing trials at OHSU and at BWH.

Our study aims are multiple: First, we will adapt and implement the eCarePlan tool to automatically extract data from the two partner studies. We will focus on unrecognized chronic conditions, partner study variables, partner study outcomes, health behaviors, social determinants of health, and adverse event outcomes. We will use optical character recognition (OCR) and natural language processing (NLP) to extract and harmonize the data from electronic free text notes and scanned documents. In addition, we will use data harmonization framework to produce pragmatic guidelines for those seeking to integrate the health data and will measure the impact for patients with MCCs, especially in understanding the balance of treatment benefits and harm. Lastly, we will replicate the outcome measurement methods of the two partner studies to explore the impact of using the eCarePlan on measurement of clinical outcomes and adverse events.

These enhanced data sets will help us to understand the benefit of FHIR to improve the accuracy of clinical trial results and decision making for older adults. Funding for this 3-year trial is provided by the National Institutes of Health (NIH).

Funding Agency: NIH
Collaborators: Brigham & Women’s Health

Person-Centered Care Planning for Persons with Multiple Chronic Conditions (PCCP4P)

The overall purpose of task order is to conduct foundational research in pursuit of the Agency for Healthcare Research and Quality’s (AHRQ) larger, long-term goal of promoting person-centered care planning (PCCP) as standard practice for persons with Multiple Chronic Conditions (MCC). PCCP holds great promise for delivering high quality, high value, coordinated healthcare for persons living with MCC that is driven by their goals, preferences, and values. However, many questions remain about the resources, conditions, processes, and implementation strategies required for healthcare systems and providers to deliver on this promise fully and routinely. In partnership with subject matter experts, policymakers, and healthcare leaders, OHSU will conduct the following to develop a roadmap for transforming care for people living with MCCs:

  1. Gather information on the current state of PCCP, including models in use, their scale, implementation barriers, and potential solutions.
  2. Identify promising models of PCCP for further development, testing, and implementation, including digital solutions.
  3. Determine the organizational, policy, payment, technology, cost, and resource requirements for implementing PCCP in diverse healthcare settings.
  4. Identify research priorities, strategies, recommendations, and next steps to make PCCP a routine practice in caring for individuals with MCC.

Funding Agency: AHRQ
Collaborators: OHSU Family Medicine, ORPRN, School of Nursing, Internal Medicine; University of Wisconsin-Madison, University of Oklahoma Health Science Center, University of California-San Francisco

Patient Centered Primary Care Home (PCPCH) Web System

Our team hosts and maintains the Patient Centered Primary Care Home (PCPCH) recognition web application. Initially implemented in January 2014, we now host, update, upgrade and enhance the recognition web application.

Funding Agency: Oregon Health Authority

PCPCH Website

Interprofessional Care Access Network (ICAN)

The Interprofessional Care Access Network (ICAN) system is a web-based application developed by the CMP team in collaboration with OHSU's School of Nursing to provide the ability for authorized School of Nursing personnel to remotely view, add, and modify data in customizable forms for individuals participating in associated program(s).  This system replaces paper-based collection methods.  The structure of these forms is easily updatable to provide for future needs, while maintaining previous structures on historical forms for backward compatibility.  Includes basic reporting and querying logic to generate exportable data sets.

Funding: OHSU School of Nursing

Collaborators: OHSU School of Nursing

Past Projects

Sixty-four Oregon practices completed all four years of the Comprehensive Primary Care initiative at the close of 2016 after achieving significant improvements in quality and reducing utilization of care. In response, the Center for Medicare and Medicaid Innovation (CMMI) of the Center for Medicare and Medicaid Services (CMS) launched an advanced model of this initiative, Comprehensive Primary Care Plus (CPC+), in January 2017 with 156 enrolled Oregon practices supported by 15 local payers. CPC+ is the largest national primary care payment and clinical practice redesign initiative in U.S. history. Nationally, the model includes 54 payers that support 2,891 primary care practices, consisting of 13,090 clinicians, and serving more than 1.76 million Medicare, millions of Medicaid, and a large number of commercially-insured beneficiaries across 14 regions. CMS will reopen solicitation for practices and payers beginning in January 2018, allowing practices in up to 10 new regions to apply, accepting a maximum of 5,500 practices across rounds one and two. Payers in new regions as well as new payers in existing regions will be welcome to apply.

The regional learning network in Oregon is led by Dr. David Dorr from Care Management Plus (CMP) and Dr. Ronald Stock from the Oregon Rural Practice-based Research Network (ORPRN). CMP and ORPRN provided technical assistance and targeted facilitation to the CPC Classic practices over the past four years. Practices were successful in adding care management staff, pharmacists, behavioral health programs and providers, and patient and family advisory councils. Oregon practices also reduced hospitalizations rates by 9.2% and 30-day unplanned readmissions rates by 6.7% amongst Medicare beneficiaries from one year prior to the initiative through March 2016. Additionally, all practices were successful in achieving quality benchmarks in 2015 and qualified for regional shared savings.

CPC+ practices, organized into Tracks 1 and 2, will continue to receive national learning content, as well as technical assistance and tailored facilitation from the regional learning network team. The CPC+ aims focus around the five key Comprehensive Primary Care functions, which are access and continuity, care management, comprehensiveness and coordination, patient and caregiver engagement, and planned care and population health. CPC+ practices had an opportunity to engage in conversations around these functions at the first CPC+ Oregon In-Person Learning Session held on April 11th at The Armory in Portland, OR. The session was attended by 335 attendees representing 130 CPC+ practices. Foundations were laid for future work around empanelment and access, risk stratification, implementation of patient and family advisory councils, as well as collaborations between practices and payers.

CMP and ORPRN will continue to build upon our relationships with Oregon practices, and to utilize our combined expertise in practice facilitation, health information technology support, risk stratification, and practice transformation to continue to support the practices in CPC+. Oregon primary care practices have been at the forefront of healthcare quality change and implementation of value-based payment methodologies, and will continue to light the path forward.

Funded by Centers for Medicare and Medicaid Services with a contract from The Lewin Group

Read more at the CPC+ blog.

CPC+ Website

This project connects the Oregon Health Authority with OHSU experts in health informatics to assist with various aspects of clinical quality measure reporting, including but not limited to technical assistance, data review, and providing consultation on clinical quality metrics registry planning.

Funded by Oregon Health Authority

CQMR Website

This project aims to improve primary care practices’ ability to successfully risk stratify their patient populations, in order to tailor interventions to current and future high-risk high-need patients. This mixed-methods study involves a quantitative analysis of factors that affect the validity and usefulness of risk stratification, and a qualitative analysis of risk stratification approaches and the potential for their improvement. The outcome of the project is intended to be a practical approach to risk stratification implementation aimed first at ambulatory practices, and second, at those designing and evaluating programs with risk stratification as a core component of their methodology. The qualitative aspect of this study is being co-led by Dr. Deborah Cohen within the OHSU Department of Family Medicine.

Funded by The Commonwealth Fund

The ONC Health IT Curriculum provides a national resource that gives community colleges and other institutions of higher education access to a wealth of curricular materials on health information technology (HIT). The materials were originally developed to train HIT workforce roles that could help hospitals and providers to achieve meaningful use of the electronic health record (EHR) and qualify for incentive payments. The objective of our current efforts is to update the materials, especially in the emerging topical areas, since the original project ended in 2012, and to expand the curriculum to train a larger and more diverse workforce. With the expertise of Dr. William Hersh, Principal Investigator of the OHSU ONC HIT Curriculum Development Center; Dr. David Dorr, an expert in population management, care coordination, and HIT; and several other faculty, we are updating and expanding these materials, and currently working on training a more diverse group of incumbent health care workers who develop, implement, and support HIT, including clinical informaticians, health information management specialists, IT specialists, clinical champions, care coordinators, and panel managers.

Funded by Office of the National Coordinator

In 2014, faculty from the Knight Cardiovascular Institute and the Department of Medical Informatics & Clinical Epidemiology (DMICE) formed a new collaboration with the goal to elevate the institute’s capabilities to analyze data and information essential to improving cardiovascular health outcomes and reducing cardiovascular health care costs. The main goals are to address top-priority quality measure needs, develop expertise in the areas of data requirements, sources and normalization, expand data analytics and reporting capabilities, use data to make better informed institutional decisions, and establish a process to advance research agendas at the intersection of cardiovascular medicine and informatics. This is a collaboration with the Informatics Discovery Lab within DMICE.

Funded by the Knight Cardiovascular Institute

Our Specific Aims (with RE-AIM factors) are to:

  1. Identify, recruit and conduct baseline assessments in 320 small and medium primary care practices across the geographically contiguous region of WA, OR and ID. (REACH)
  2. Provide comprehensive external practice support to build QI capacity within these practices. (IMPLEMENTATION & ADOPTION)
  3. Disseminate and support the adoption of PCOR findings relevant to the ABCS quality measures. (EFFECTIVENESS)
  4. Conduct a rigorous evaluation of the effectiveness of providing external practice support to implement PCOR findings and improve ABCS measures. (EFFECTIVENESS)
  5. Assess the sustainability of changes made in QI capacity and ABCS improvements and develop a model of dissemination and primary care practice support infrastructure. (MAINTENANCE)

Funded by the Agency for Healthcare Research and Quality

H2N Website

The overall goal of the IQ-MAPLE project is to improve the quality of care provided to patients with several heart, lung and blood conditions by facilitating more accurate and complete problem list documentation. In the first aim, we will design and validate a series of problem inference algorithms, using rule-based techniques on structured data in the EHR and NLP on unstructured data. Both of these techniques will yield candidate problems that the patient is likely to have, and the results will be integrated. In Aim 2, we will design CDS interventions in the EHRs of the four study sites to alert physicians when a candidate problem is detected that is missing from the patient’s problem list – the clinician will then be able to accept the alert and add the problem, override the alert, or ignore it entirely. In Aim 3, we will conduct a randomized trial and evaluate the effect of the problem list alert on three endpoints: alert acceptance, problem list addition rate, and clinical quality. This study is a collaboration with Brigham and Women's Hospital, with Dr. Adam Wright as the Principal Investigator. Other investigator sites include Holy Spirit Hospital and Vanderbilt University.

Funded by the National Heart, Lung, and Blood Institute

This study is in collaboration with the OHSU Department of Family Medicine, with Dr. Deborah Cohen as Principal Investigator. We will examine the nature of cognition, task distribution, and clinical work in various health care delivery settings, and will attempt to address current knowledge gaps regarding our understanding of health care providers' information needs and health care decision making processes, both individually and collectively. The aims of this project are to:

  1. Identify clinicians and clinical care teams' information and decision making needs when coordinating care for complex patients
  2. Identify their information needs regarding patients' social determinants of health
  3. Identify design principles and develop health IT tools to meet these needs.

Funded by the Agency for Healthcare Research and Quality

The focus of the NIH Big Data to Knowledge (BD2K) program is to support the research and development of innovative and transforming approaches and tools to maximize and accelerate the integration of Big Data and data science into biomedical research. The OHSU Department of Medical Informatics has several funded BD2K grants that are focused on enhancing data science training in biomedical research, as well as supporting the development of new methods. The OHSU BD2K Skills Courses have provided a series of training opportunities for a variety of learners. This training initiative is led by David Dorr, Melissa Haendel, and Shannon McWeeney. The individual offerings of the course are intended to progress from novice to expert.

Funded by the National Library of Medicine at National Institutes of Health

BD2K Website

TOPMED is a two-arm, cluster randomized controlled trial, investigating how to transform primary care clinics in order to achieve high value elements (HVEs). Both arms received quality improvement guidance, including external practice facilitation, IT-based milestone reporting, and financial incentives based on self-selected QI goals. Intervention clinics were directed to select their quality improvement goals from a list of HVEs, which have been demonstrated in the literature and by stakeholders to improve patient outcomes and lower costs. Results include patient satisfaction, utilization, costs, and improvement in quality measures. This study had practice facilitation support provided by the Oregon Rural Practice-based Research Network.

Funded by the Gordon and Betty Moore Foundation

The Comprehensive Primary Care (CPC) initiative is a large innovative effort to redesign care delivery and provider payment from the Centers for Medicare and Medicaid Services’ (CMS) Center for Medicare and Medicaid Innovation (CMMI). This four-year multi-payer initiative launched in October 2012 to strengthen primary care. Through this initiative, CMS has leveraged resources from roughly 30 payers in four states (Arkansas, Colorado, New Jersey and Oregon) and three regions (New York’s Capital District and Hudson Valley, Ohio and Kentucky’s Cincinnati-Dayton region, and Oklahoma’s Greater Tulsa region) to support transformation through enhanced payments, data feedback, and learning activities in 492 primary care practices serving approximately 315,000 Medicare beneficiaries and more than 2.5 million patients.

CMS offered population-based care management fees and shared savings opportunities to support these five core comprehensive primary care functions:

  • providing care management for those at greatest risk
  • improving health care access and continuity
  • planning care for chronic conditions and preventive care
  • coordinating care across the medical neighborhood
  • patient and caregiver engagement. using health information technology to support population health

The initiative is testing whether focus on these functions, along with the continuous use of data to guide improvement, and the use of health information technology to support population health, can achieve improved care, better health for populations, and lower costs.

Participating CPC practices receive curriculum developed by CMMI learning and diffusion experts, as well as targeted coaching and practice facilitation from a regional technical assistance (TA) team. Regional TA teams provide direct assistance and develop and implement a local curriculum using three modalities: a series of group meetings (in-person or virtual); a virtual webinar series; and specific goals for monitoring study milestones. The TA team in Oregon is led by Dr. David Dorr, with additional supporting faculty, Dr. LJ Fagnan, Dr. Evan Saulino, Dr. Cherie Brunker, and Dr. Ronald Stock as part of the leadership team.

Funded by Centers for Medicare and Medicaid Services with a contract from TMF Health Quality Institute

CPC Website

The goal of this project is improvement of risk stratification methods and predictions of outcomes for patients with multiple chronic conditions. Specifically, there are three aims: 1) to compare accuracy and completeness of diagnoses through manual validation of a “gold standard” patient-diagnosis dataset; 2) to explore whether the modest predictive validity of standard risk scores is due to variability in the data source and validity; and 3) to create optimized risk prediction models based on the knowledge about comorbidity data gained in the first two aims using advanced machine learning algorithms and data transformation approaches.

Funded by the Agency for Healthcare Research and Quality

The Network comprises a group of experts in geriatrics and primary care with the vision to transform PCMHs to recognize, facilitate, encourage, and ultimately improve the care of older adults and their caregivers. Over a two-year period, the Network will identify and refine selected topics of critical importance to the field of geriatrics in PCMHs. Three pilot projects have been planned for 2015-2016 based on an initial evaluation of needs.

Funded by the Gerontological Society of America

PCMH Website