OBJECTIVES: The incidence of falling in older adults has remained unchanged over the past decades, despite evidence-based prevention initiatives. Therefore, it is appropriate to reflect on the current screening approach for preventive initiatives. The objective of this study was to determine whether the multifactorial algorithm proposed by Lusardi et al. (2017) exhibits superior predictive validity compared to the currently employed algorithm by the Belgian National Institute for Health and Disability Insurance (NIHDI).
METHODS: The current study includes a secondary analysis of data collected from a falls-related study in the Department of Rehabilitation Sciences at Ghent University to compare the predictive validity of the two algorithms. Sensitivity, specificity, positive and negative predictive value and area under the curve (AUC) were calculated to ascertain which algorithm is more accurate.
RESULTS: The database included a total of 94 community-dwelling older adults (mean age 76 years ±7.4, 35% male). Thirty-nine participants experienced at least one fall in the 8 month follow up. Lusardi's approach has a higher sensitivity score (89.7% compared to 10.3%) and negative predictive value (89.9% compared to 61.1%), but a lower specificity score (61.8% compared to 100%) and positive predictive value (62.2% compared to 100%) than the NIHDI approach. The AUC is 0.76 for Lusardi's approach and 0.55 for the NIHDI approach.
CONCLUSION: The use of the multifactorial algorithm proposed by Lusardi et al. may be significant and more accurate in identifying adults at risk to falls. Further research is needed particularly with a larger, more heterogenous group of older adults.