Citation

Latt MD, Loh KF, Ge L, Hepworth A. Australas. J. Ageing 2016; 35(3): 167-173.

Affiliation

Department of Aged Chronic Care and Rehabilitation, Sydney Local Health District, Sydney, New South Wales, Australia.

Copyright

(Copyright © 2016, Australian Council on the Ageing, Publisher John Wiley and Sons)

DOI

10.1111/ajag.12256

PMID

26991034

Abstract

AIM: We examined the validity of the Ontario Modified STRATIFY (OM) (St Thomas's Risk Assessment Tool in Falling Elderly Inpatients), The Northern Hospital Modified STRATIFY (TNH) and STRATIFY in predicting falls in an acute aged care unit.

METHOD: Data were collected prospectively from 217 people presenting consecutively and falls identified during hospitalisation.

RESULTS: Sensitivities of OM (80.0, 95% confidence interval (CI) 58.4 to 91.9%), TNH (85, CI 64.0 to 94.8%) and STRATIFY (80.0, CI 58.4 to 91.0%) were similar. The STRATIFY had higher specificity (61.4, CI 54.5 to 67.9%) than OM (37.1, CI 30.6 to 44.0%) and TNH (51.3, CI 44.3 to 58.2%). Accuracy (percentage of patients correctly classified as 'faller' or 'non-faller') was higher using STRATIFY (63.1, CI 56.5 to 69.3%) and TNH (54.4, CI 47.8 to 61.0%) than with OM (41.0, CI 34.7 to 47.7%, P < 0.0001).

CONCLUSION: Screening tools have limited accuracy in identifying patients at high risk of falls.

© 2016 AJA Inc.

Language: en