Establishing the Reliability and Validity of a Fall Risk Assessment Tool for Acute Psychiatric Inpatients

Fu, Shiu-Yun K.

Monday, 31 October 2011

Shiu-Yun K. Fu, PhD, RN
Department of Nursing, Fu Jen Catholic University, New Taipei City, Taiwan
Hei-Jiuan Liang, MSN, RN
Department of Nursing, Beitou Armed Forces Hospital, New Taipei City, Taiwan
Venne-Shiang Huang, MA
Department of Mathematics, Fu Jen Catholic University, New Taipei City, Taiwan

Learning Objective 1: The learner will be able to understand the valid localized fall risk assessment tool and how accommodate the cultural and social context.

Learning Objective 2: The learner will be able to have an awareness of the importance of risk predictions of fall for psychiatric inpatients in their quality of care.

This study was based on Irvin's (1999) Fall Risk Assessment Tool studies in the US. The purpose of the research was (1) to verify the localized categories of fall risk among acute psychiatric inpatients; (2) to examine the sensitivity and specificity of the localized tool and (3) to investigate the important risk factors of fall within Taiwan. This study had employed purposive sampling strategies within two specialized teaching hospitals of psychiatry, both located in Northern Taiwan. 224 patients were involved in the study. The data analysis uses ROC curve analysis and multivariable logistic regression models.

The research results revealed that the subjects of the dual-stage case-disclosures have most frequently been diagnosed to have suffered from schizophrenia and that 60.3% to 64.7% of those who have fallen were males. It was established that the localized assessment tool score was 8 on the seventh day and that the sensitivity of the high-risk fall assessment measured 88% with specificity reaching 70% (AUC=.79) as being the optimal point of score. With analysis and predictions from multivariable logistic regression, the Odds Ratio (OR) of fall under the adjustments and between factors would influence the “Fall” thus the dependant variables had been obtained.

In conclusion, the four respective categories of variables are those of mental confusion, gait and balance, frequency of elimination and the general environment of the ward. These categories which were investigated in this study can be used as an important tool for risk prediction by medical and nursing clinical staff in the prevention of fall amongst their patients.

See more of: Poster Session II
See more of: Oral Paper & Poster: Evidence-Based Practice Contest