Health and clinical informatics transforms medicine and health care by analyzing, designing, implementing, and evaluating systems and interventions to improve patient care, enhance access to care, advance individual and population health outcomes, and strengthen the clinician-patient relationship.
Among the areas of research expertise in health and clinical informatics include:
- Re-use of clinical data
- Text mining and natural language processing
- Information retrieval (search)
- People and organizational issues
- Health information exchange
- Telemedicine/telehealth
- Data quality
- Patient decision-making and decision aids
- Informatics for chronic disease management
- Electronic Health Record simulation for patient safety
Some grant-funded projects include:
Semi-structured Information Retrieval in Clinical Text for Cohort Identification
William Hersh, Steven Bedrick
The major goals of this project are to develop information retrieval techniques for cohort identification based on clinical text. A collaboration with Mayo Clinic, both sites use an extract of electronic health record data from 100,000 patients, including text reports, to identify the best approaches to correctly identifying patient cohorts for possible inclusion in 56 different clinical studies.
Funder: National Library of Medicine, National Institutes of Health
Text Mining Pipeline to Accelerate Systematic Reviews in Evidence-Based Medicine
Aaron Cohen
Systematic reviews are essential for determining which treatments and interventions are safe and effective. At present, systematic reviews are written largely by laborious manual methods. The proposed studies will reduce the time and effort needed to write systematic reviews, and thereby enhance evidence-based medicine and the incorporation of best practices into clinical care.
Funder: National Library of Medicine, National Institutes of Health
A Turn-Key EHR Simulation Program to Reduce Diagnostic Error in Ambulatory Care
Jeff Gold, Vishnu Mohan
In this proposal, we created and validated a turn-key library of EHR based simulations to improve diagnostic safety in ambulatory care which is both generalizable and scalable. After performing a problem analysis by using a combination of administrative data from the Pennsylvania Patient Safety Authority and claims data from The Doctors Company to identify diagnoses at risk for diagnostic error in ambulatory care and EHR use errors associated with said errors, we integrated these data to create a rubric to allow for the design of a comprehensive EHR based simulation library to study and reduce diagnostic error across the 5 major ambulatory specialties. We then developed simulations for 5 ambulatory specialties. and validated the ability of the simulation activities to serve as a training tool to fundamentally change EHR use patterns and reduce diagnostic error. Finally, this library, including all simulation materials, scripts and EHR charts were be coalesced into an online repository. We will further create a series of FHIR based applications which will facilitate “loading” of the simulated EHR charts into the major EHR systems. Thus, by the end of the funding period, we will have created a library of validated EHR base simulation exercises to reduce diagnostic error in ambulatory care and developed novel technology to facilitate widespread implementation.
Funder: Agency for Healthcare Research and Quality
Jeff Gold, Vishnu Mohan
In this proposal, we expanded our initial EHR simulation station to allow for robust usability analysis (through video eye-tracking and key logging) and create additional cases to expand both in the MICU and other ICU environments, determined whether repeated testing can improve EHR use and error recognition within the ICU, employed usability data to redesign EHR training and user interface and convened a consensus conference to establish the role for EHR simulation to improve use, education and design.
Funder: Agency for Healthcare Research and Quality
Jeff Gold, Vishnu Mohan
The goal of this award is to establish a Diagnostic Center of Excellence (DATAEYES) focused on identification of EHR contribution to DE, and use this information to deploy a suite of solutions to improve software, user and system. We will achieve this by using national data to create an informed taxonomy to be integrated into institution data collection tools, to facilitate institution-wide capture of EHR contributions to DE in Aim #1. We will then develop and validate these tools in Aim #2 and use this information, in combination with in situ workflow observations, to inform how, when and why the EHR is contributing to DE. This information will be used to create high- fidelity simulated EHR charts to facilitate both workflow specific training on EHR best practices and guide EHR redesign and monitor the impact of these interventions via EHR audit logs in Aim #3. The 3 centers participating (OHSU, Medstar Health, Brigham and Women's Hospital) will allow further ascertainment of the impact of both EHR vendors being studied (Cerner and Epic) and local workflow specific practices. We will then leverage our collaborations with patient safety organization and industry to disseminate these findings and the infrastructure developed at DATAEYES will serve as a core resource for the other DCE sites, allowing for rapid evaluation and prototyping of future EHR based solutions.
Funder: Agency for Healthcare Research and Quality
