Date of Award
Projects: SPU Access Only
Doctor of Nursing Practice (DNP)
Evaluation of the Hester Davis Scale for Fall Risk Assessment in the Oncology Population
Mercedes Swiercek, BSN, RN, SPU FNP student
Background and Significance
Falls in the hospital remain a patient safety problem in today’s healthcare, estimating up to 1 million patients in the hospital fall each year and lead to longer hospital stays, increased medical costs, poor health outcomes and psychological distress (AHRQ, 2019; JCAHO, 2015). Current best practice guidelines for fall prevention in healthcare settings is to use a risk assessment tool to identify patients at high risk of falling such as the Hester Davis Scale (HDS) or the John Hopkins Fall Risk Assessment Tool (JHFRAT). However, not all fall risk assessment tools have the ability to correctly identify patients at risk for falls in specific acute care populations such as the oncology population.
Problem and Purpose Statements
After transitioning fall risk assessment scales from the JHFRAT to the HDS, the project site expressed concerns about the scoring and risk identification of the oncology population. The purpose of this DNP project was to determine if fall risk assessment using the HDS was better at identifying oncology patients at risk for fall than the JHFRAT, thereby decreasing fall rates and improving patient safety.
Guided by Havelock’s Theory of Change, a retrospective chart analysis in two phases was completed to evaluate oncology fallers and non-fallers for a matched cohort study. This project’s sample consisted of female and male adults 18 and older with a primary cancer diagnosis from two inpatient oncology units at the project site. For Phase 1, HDS total score and risk category was calculated by the project lead and JHFRAT total score and risk category was collected from previous nurse documentation. For Phase 2, HDS total score and risk category was collected from previous nurse documentation. Overall accuracy, error rate, sensitivity and specificity were calculated using 2x2 contingency tables replicating a study by Kaiser et al. (2020). Additionally, SPSS was used to determine correlation between the HDS and the JHFRAT in Phase 1 and to complete two-tailed t-tests of fallers and non-fallers in Phase 2. Secondary measures included fall risk factors for fallers and non-fallers identified by the HDS were recorded for both phases.
Phase 1: The overall accuracy of the HDS (cut score > 15) was 54.8% while the JHFRAT (cut score > 13) had an overall accuracy rate of 65.5%. Using the same cut scores, the HDS and JHFRAT had an error rate of 45.2% and 34.5% (respectively). The results of the sensitivity and specificity for the cut score of > 15 were calculated at 90.5% and 19.0% for the HDS. The sensitivity and specificity for a cut score of > 13 using the JHFRAT was 45.2% and 85.7% respectively. A Pearson’s correlation was computed to assess the relationship between the HDS and the JHFRAT. There was strong positive correlation between the two variables, r = 0.554, n = 84 and this relationship was significant (p < 0.001). There was a significant difference in the mean scores of both the HDS (fallers 23.88, non-fallers 18.38) (p > 0.001) and the JHFRAT (fallers 13, non-fallers 7.36) (p = 0.002).
Phase 2: There was a significant difference in means between fallers (mean = 18.6) and non-fallers (mean = 13.3) (t = -2.352, p = 0.025). Using the cut score of > 15, sensitivity was calculated to be 68.6% and specificity was 43.8%. Overall accuracy for the HDS was 56.3% and the error rate was 43.7% respectively.
Considering the overall accuracy and error rate, both the JHFRAT and the HDS failed to meet the standards to accurately identify oncology patients at risk for falls. Yet, the JHFRAT performed better than the HDS. Additionally, the HDS showed little variation in risk assessment as the majority of patients were scored high risk indicating the HDS was not as good as finding patients truly at risk for a fall.
The project lead recommended a change to the HDS high risk cut score (> 18) to increase predictive metrics. This cut score was chosen based on the lowest average mean calculated from both phases of analysis. To ensure sustainability of this project and facilitate an additional retrospective chart review, a summary of the analysis and copies of the data collection sheets with instructions for completion were provided to the project site. A second recommendation was to complete further analysis of fall risk factors designated by the HDS in the oncology population. The project lead will also present project findings to the Oncology Nursing Society and staff education days to support future fall risk projects in planning.
Not all fall risk assessment tools created for one population can be universally applied to other populations as there may be unique factors applicable only to a specific population such as the oncology population. For this reason, it is important to create risk assessment tools that accurately identify those at high risk to ensure the safety of these patients. Advanced clinical practitioners such as a DNP-prepared nurses will be a part of the team that help to manage the clinical situation of these patients. Using accurate tools will help guide the precautions and education provided to patients throughout the treatment process. Lastly, DNP-prepared nurses will not only be implementing evidence based-practice to promote patient safety such as implementation of new fall risk assessment tools but they will also be leading practice change when a problem is presented as this project demonstrated.
Agency for Healthcare Research and Quality [AHRQ]. (2019). Falls. https://psnet.ahrq.gov/primer/falls
Kaiser, J., Wills, N., Reilly, T., Pratt, J., Tumbleson, V., Niemeyer, M., & Mindling, G. (2020). A roadmap for practice-based evidence: Validation of the Hester Davis Fall Risk Scale. Journal of Nursing Care Quality, 0(0), 1-6. doi: 10.1097/NCQ.0000000000000503
The Joint Commission [JCAHO]. (2015). Preventing falls and fall-related injuries in health care facilities. Sentinel Event Alert, (55), 1-5. https://www.jointcommission.org/-/media/deprecated-unorganized/imported-assets/tjc/system-folders/topics-library/sea_55pdf.pdf?db=web&hash=53EE3CDCBD00C29C89B781C4F4CFA1D7Se
Swiercek, Mercedes, "Evaluation of the Hester Davis Scale for Fall Risk Assessment in the Oncology Population" (2022). Doctor of Nursing Practice (DNP) Scholarly Projects. 33.