4 April 2019 in BMC Medical Informatics and Decision Making
This paper presents a portable phenotyping system that is capable of integrating both rule-based and statistical machine learning based approaches. Our system utilizes UMLS to extract clinically relevant features from the unstructured text and then facilitates portability across different institutions and data systems by incorporating OHDSI’s OMOP Common Data Model (CDM) to standardize necessary data elements. Our system can also store the key components of rule-based systems (e.g., regular expression matches) in the format of OMOP CDM, thus enabling the reuse, adaptation and extension of many existing rule-based clinical NLP systems. We experimented with our system on the corpus from i2b2’s Obesity Challenge as a pilot study. Our system facilitates portable phenotyping of obesity and its 15 comorbidities based on the unstructured patient discharge summaries, while achieving a performance that often ranked among the top 10 of the challenge participants. Our system of standardization enables a consistent application of numerous rule-based and machine learning based classification techniques downstream across disparate datasets which may originate across different institutions and data systems.
15 November 2020 in Journal of Interprofessional Care
The lack of a proper system for ongoing open interprofessional communication among care providers increases miscommunications and medical errors. Seamless access to patient information is important for care providers to prevent miscommunication and improve patient safety. A shared understanding of the information needs of different care providers in an interprofessional team is lacking. Our purpose is to identify care providers’ information needs from the perspective of different professions for communication, shared understanding about the patient, and decision-making. We conducted semi-structured interviews with 10 subject matter experts representing eight professions, including dentistry, dietetics, medicine, nursing, occupational therapy, pharmacy, physical therapy, and social work in a 465-bed academic hospital at a large urban Midwestern city. We used an in-house rounding tool presenting physicians’ information needs and a hypothetical patient scenario to collect participants’ feedback. Interview notes were coded using direct content analysis. We identified 22 additional essential data elements for an interprofessional rounding tool. We categorized those into six domains: discharge-related, social determinants of health, hospital safety, nutrition, interprofessional situation awareness, and patient history. A well-designed validated rounding tool that includes an interprofessional team of care providers’ information needs could improve communication, care planning, and decision-making among them.