A screening system for Michigan's home- and community-based long-term care programs
Related Experiment Video
Updated: Sep 30, 2026

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Brant E Fries1, Lisa R Shugarman, John N Morris
1Institute of Gerontology and School of Public Health, University of Michigan, Ann Arbor 48109-2007, USA. bfries@umich.edu
Purpose:
To develop a screening system for Michigan's MI Choice publicly funded home- and community-based services programs, to aid in identifying both individuals eligible for services and their most appropriate level of care (LOC).
Design And Methods:
Identify assessment items from the Minimum Data Set for Home Care (MDS-HC) assessment instrument that are predictive of five LOCs determined by expert care managers: nursing home, home care, intermittent personal care, homemaker, and information and referral (without services).
Results:
The algorithm based on approximately 30 client characteristics agrees with expert opinions substantially better (kappa =.62) than systems based on activities of daily living and instrumental activities of daily living only (kappa <.40).
Implications:
The screening algorithm can be used both over the telephone to identify clients who will not be fully assessed (as they are unlikely to receive services) and in person to recommend the appropriate LOC.
Related Concept Videos
Documentation in Long-Term and Home Healthcare Setting
Long-Term Care Facilities
Specialized Care Centers and Settings-I
Daycare centers
They provide several functions. Some facilities care for healthy newborns and children whose parents work, while others are medically focused and care for...
Continuing Care
Integrated Healthcare System
Restorative Care
Health Information Technology and Healthcare Information System
Health Information Technology, commonly called HIT, integrates advanced information systems and technology in healthcare settings. Its primary functions include:

