Evaluation of three fall-risk assessment tools in an acute care setting

J Adv Nurs. 2007 Nov;60(4):427-35. doi: 10.1111/j.1365-2648.2007.04419.x.

Abstract

Aim: This paper is a report of a study to evaluate the validity of three fall-risk assessment tools to identify patients at high risk for falls.

Background: Patient falls make up 38% of all adverse events occurring in hospital settings, and may result in physical injury and undesirable emotional and financial outcomes. No single fall-risk assessment tool has been conclusively validated.

Method: The Morse Fall Scale, St Thomas Risk Assessment Tool in Falling Elderly Inpatients, and Hendrich II Fall Risk Model were validated in inter-rater reliability and validity studies in 2003. This included assessment of the probability of disagreement, kappa-values, sensitivity, specificity, positive predictive values and negative predictive values of the assessment tools with the associated 95% CI.

Findings: One hundred and forty-four patients were recruited for the inter-rater reliability study. The probabilities of disagreement were between 2.8% and 9.7%, and 95% CI for all tools ranged from 1.1% to 15.7%. The kappa-values were all higher than 0.80. In the validity study, 5489 patients were recruited to observe 60 falls. The Morse Fall Scale at a cutoff score of 25 and Heindrich II Fall Risk Model at a cutoff score of 5 had strong sensitivity values of 88% and 70%, respectively. However, in comparison with the Morse Fall Scale (specificity = 48.3%), only the Heindrich II Fall Risk Model had a more acceptable level of specificity (61.5%).

Conclusion: The Heindrich II Fall Risk Model is potentially useful in identifying patients at high risk for falls in acute care facilities.

Publication types

  • Evaluation Study

MeSH terms

  • Accidental Falls / prevention & control*
  • Accidental Falls / statistics & numerical data
  • Adult
  • Aged
  • Aged, 80 and over
  • Factor Analysis, Statistical
  • Female
  • Geriatric Assessment / methods
  • Geriatric Assessment / statistics & numerical data
  • Humans
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Prospective Studies
  • Risk Assessment
  • Risk Management / methods*