This project takes an evidence-based, interdisciplinary approach to decreasing the number of patient falls and associated injuries through the development of an instrument to detect patients who are at risk for falling, interventions to prevent falls, and the utilization of the NDNQI to record and track patient falls.

[sample]

Each patient who fell during a 3-month period was matched with two patients of a similar age from his/her inpatient unit, indicating a similar diagnosis.

 [method]After a thorough review of current literature and existing falls-risk assessment instruments, we conducted a case-control study to identify variables that distinguished our patients who fell from those who did not. This initial list of variables was developed from factors commonly named in the evidence as well as those identified in clinical practice. From those results, we developed a falls-risk assessment instrument that included a measure of the patient’s ability to complete basic activities of daily living (Katz Index), which has been shown to be of use in predicting patients at risk for falls in a previous JCAHO case-control study.  Interrater reliability of the instrument was 84 percent.

The instrument was pilot-tested on four inpatient units: neurology, leukemia/lymphoma, head and neck, and hepatobiliary and gastric mixed tumor. Patients were assessed for falls risk at admission and twice daily during hospitalization. If a patient fell, a team responded to review the risk score and individual environmental and patient variables. Patients were again matched with two patients of a similar age in a prospective case control study. 

 [findings]

In the first case control, the charts of 219 patients, 73 cases and 146 controls, were reviewed, and statistical significance, through chi-square testing, was found in four variables: at <0.01, sensory deficits; and, at <.001, history of falls, presence of psychotropic or anticonvulsant medication. Odd-ratios calculations produced similar findings.

In the second case control, the charts of charts of 178 patients, 57 cases and 121 controls, were reviewed. Fall history was significant at a <0.01 level, and psychotropic medication was significant at <0.05. From the Katz Index, the variables “Needs assistance with toileting” and “Needs assistance with transferring” were significant at <0.05.[conclusions]
Based on statistical findings and clinical data collection, we developed a final falls-risk assessment instrument that relies on categories rather than points to assign risk. That is, the presence of any of the statistically or clinically significant variables puts a patient at high risk for falling. We will continue to assess patients at admission and twice daily and implement appropriate interventions based on the assessment.
[implications]
The new falls program will be rolled out before the end of 2006. With the new instrument we will introduce a streamlined falls tracking system that will utilize NDNQI to track falls and falls with injury. It is anticipated that a more consistent and accurate reporting and tracking system will lead to improved feedback to individual nursing units regarding their patient fall rates, and benchmarking abilities with other facilities similar to ours, including National Comprehensive Cancer Network institutions.