UMass Chan Medical School
Advancing Malnutrition Measurement Using Linked EHR and MDS Data
Real World Data Source
LTC Data Cooperative
Dr. Yuan is an epidemiologist and assistant professor in the Department of Population and Quantitative Health
Sciences at UMass Chan Medical School. Her research uses real-world data, including the Minimum Data Set
(MDS) and Medicare claims, to examine trajectories and heterogeneity of frailty, cognitive impairment and other
aging-related conditions among older adults. She also studies medication safety and effectiveness in national
nursing home cohorts. Building on this foundation, her research aims to develop clinically meaningful, datadriven
approaches to improve identification of aging-related conditions among older adults in long-term care
settings and generate evidence to improve care.
Malnutrition is a common but often underrecognized problem among older adults in nursing homes and is
associated with frailty, hospitalization, and mortality. While the Minimum Data Set (MDS) captures indicators of
nutritional risk, its structured assessments may not fully reflect how risk evolves over time. This project will
leverage electronic health record (EHR) data from the Long-Term Care (LTC) Data Cooperative to develop and
evaluate EHR-derived indicators that capture nuanced clinical context and improve identification of
malnutrition risk in routine nursing home care. This award will provide Dr. Yuan with training and experience
using LTC Data Cooperative EHR data linked with MDS and Medicare data to: (1) Assess the availability and
documentation patterns of EHR-derived malnutrition risk indicators; (2) Evaluate concordance between EHRderived
indicators and MDS-based malnutrition measures; and (3) Examine the construct and predictive
validity of these indicators using nutrition orders and health outcomes. By generating foundational evidence on
the availability, validity, and clinical relevance of EHR-derived malnutrition indicators, this project will clarify
whether these measures complement existing MDS assessments and support earlier identification of nursing
home residents at elevated malnutrition risk. These findings will inform the use of EHR-derived indicators in
studies and initiatives aimed at improving nutrition-related care in long-term care settings. This work will lay
the foundation for future research developing and validating EHR-based measures for aging-related conditions
in long-term care.