Combining Items From 3 Federally Mandated Assessments Using Rasch Measurement to Reliably Measure Cognition Across Postacute Care Settings

Abstract

Objective: To combine items from the Functional Independence Measure, Minimum Data Set (MDS) 2.0, and the Outcome and Assessment Information Set (OASIS)-B to reliably measure cognition across postacute care settings and facilitate future studies of patient cognitive recovery.

Design: Rasch analysis of data from a prospective, observational cohort study.

Setting: Postacute care inclusive of inpatient rehabilitation facilities, skilled nursing facilities, and home health agencies.

Participants: Patients (N=147) receiving rehabilitation services.

Interventions: Not applicable.

Main outcome measures: Functional Independence Measure, MDS 2.0, and the OASIS-B.

Results: Six cognition items demonstrated good construct validity with no misfitting items, unidimensionality, good precision (person separation reliability, 0.95), and an item hierarchy that reflected a clinically meaningful continuum of cognitive challenge.

Conclusions: This is the first attempt to combine the cognition items from the 3 historically, federally mandated assessments to create a common metric for cognition. These 6 items could be adopted as standardized patient assessment data elements to improve cognitive assessment across postacute care settings.

Keywords: Cognition; Outcome measures; Rehabilitation; Stroke; Subacute care.