Associate Professor, Department of Biomedical Informatics
University of Pittsburgh School of Medicine
Department of Biomedical Informatics
My main area of research interest is improving the design and usability of electronic health records and health IT systems in general, using a novel paradigm together with modern web technical approaches. This is based on giving nonprogrammer clinicians more control via a drag/drop platform approach which allows them to create their own software and tools. It has promise to increase the software’s efficiency and cognitive support, fit to clinician ways of thinking, work contexts, public health emergencies, ease of use, and evolvability to meet future needs and different specialty and work contexts.
My other major areas of interest are using informatics for patient/consumer engagement, particularly in minority communities and the underserved, global health informatics, and analytics to improve healthcare.
Current Funded research
RECOVER project to understand Post-Acute Sequelae of Covid via Computable Phenotypes RECOVER project (RECOVERCovid.org). This is a national project which combines 14 million patient records from health systems around the country, to understand Post-Acute Sequelae of Covid (‘Long Covid’) through a variety of computational and clinical studies. Our area involves chart review for the formulation of computable phenotypes for eight conditions as well as studying the effects of pregnancy on PASC development, and a general PASC phenotype. The results will help us identify risks and factors contributing to the development of PASC, and treatment considerations.
Evaluating and Enhancing Health Information Technology for COVID-19 Response Workflow in a Specialized COVID-19 in a Medically Underserved Community RO1HS028220
Emergency response during the Covid pandemic affected institutions differently based on their resources, training, emergency preparation, and patient demographics as well as many other factors. We study the pandemic response of a small safety net hospital in a highly underserved and lower income patient population in part of New York City, which was made into a ‘Covid-only’ hospital for part of the pandemic time, compared with one of the larger networked hospitals in the same city. Our findings have implications for public health resource allocation, emergency response planning, and understanding the burden or reporting on institutions.
Phillips Respironics project to use R3 data to understand comorbidities and factors affecting sleep disorders. We applied several machine learning techniques to predict sleep disorders among the UPMC patient population by analysis of medical records from a specialized sleep clinic and the larger community.
Past major funded projects
Finding the Safer Way: Novel Interaction Design Approaches to Health IT Safety R01HS023708.
This grant studied the use of the novel ‘composable’ approach to EHR design to improve safety, efficiency, cognitive support, and other aspects of EHR design. Further information is available at our lab site: http://www.ehrab.org .