Accidental falls in hospital inpatients: evaluation of sensitivity and specificity of two risk assessment tools.

Link to article at PubMed

Related Articles

Accidental falls in hospital inpatients: evaluation of sensitivity and specificity of two risk assessment tools.

J Adv Nurs. 2010 Mar;66(3):690-6

Authors: Lovallo C, Rolandi S, Rossetti AM, Lusignani M

AIM: This paper is a report of a study comparing the effectiveness of two falls risk assessment tools (Conley Scale and Hendrich Risk Model) by using them simultaneously with the same sample of hospital inpatients. BACKGROUND: Different risk assessment tools are available in literature. However, neither recent critical reviews nor international guidelines on fall prevention have identified tools that can be generalized to all categories of hospitalized patients. METHOD: A prospective observational study was carried out in acute medical, surgical wards and rehabilitation units. From October 2007 to January 2008, 1148 patients were assessed with both instruments, subsequently noting the occurrence of falls. The sensitivity, specificity, positive and negative predictive values, and Receiver Operating Characteristics curves were calculated. RESULTS: The number of patients correctly identified with the Conley Scale (n = 41) was higher than with the Hendrich Model (n = 27). The Conley Scale gave sensitivity and specificity values of 69.49% and 61% respectively. The Hendrich Model gave a sensitivity value of 45.76% and a specificity value of 71%. Positive and negative predictive values were comparable. CONCLUSION: The Conley Scale is indicated for use in the medical sector, on the strength of its high sensitivity. However, since its specificity is very low, it is deemed useful to submit individual patients giving positive results to more in-depth clinical evaluation in order to decide whether preventive measures need to be taken. In surgical sectors, the low sensitivity values given by both scales suggest that further studies are warranted.

PMID: 20423404 [PubMed - indexed for MEDLINE]

Leave a Reply

Your email address will not be published. Required fields are marked *