SHM Converge Scientific Abstract ePoster Gallery-Reading the Words but Missing the Meaning: Rethinking Hospital Health-Literacy Screening

Pdf Summary

This prospective cross-sectional study examined whether the REALM-SF, a rapid hospital health literacy screen based on word recognition, accurately reflects patients’ understanding of medical terms. Among 466 hospitalized adults, participants were asked to read seven medical words aloud and then explain their meanings. Comprehension was scored from 0 to 7 using dictionary-based definitions, and the relationship between reading score and comprehension was analyzed with quasi-Poisson regression adjusted for age, race, gender, and ethnicity.<br /><br />The sample had a mean age of 66.8 years; 52.8% were female, 80.9% were White, and 7.5% were Hispanic/Latino. Word recognition was significantly associated with comprehension: each additional correctly read word was linked to a 29.7% higher expected comprehension score (IRR 1.297, p = .001). Female gender was also modestly associated with slightly better comprehension (IRR 1.103, p = .038), while age, race, and ethnicity were not significant after adjustment.<br /><br />However, the overall predictive value was poor. Although 81% of participants read all seven words correctly, only 4.5% could correctly define all of them. The model explained very little variability in comprehension (pseudo-R² = 0.035), showing that word recognition substantially overestimated actual medical understanding.<br /><br />The study concludes that commonly used rapid screening tools like the REALM-SF may miss many hospitalized adults with important comprehension deficits. The authors recommend that hospitals not rely solely on word-recognition screening and instead consider universal health literacy precautions, such as teach-back, or use assessment strategies that measure comprehension more directly to improve patient safety and health equity.

Asset Subtitle

Christopher L. Steele

Meta Tag

Author List Christopher L. Steele, Rose Karvandi

Category Research

Concept Medical Word Recognition

Concept Patient Comprehension

Concept Hospitalized Adults

Concept Predictive Value

Concept Health Literacy Screening

Distinguished Non-Finalist

Presenter Organization UConn Health

Presenting Author Christopher L. Steele

Track Other

Keywords

REALM-SF

health literacy

word recognition

medical comprehension

hospitalized adults

cross-sectional study

quasi-Poisson regression

teach-back

patient safety

health equity

Medical Word Recognition

Patient Comprehension

Hospitalized Adults

Predictive Value

Health Literacy Screening