Retrospective evaluation of the world falls guidelines-algorithm in older adults

Area of Science:

• Gerontology
• Public Health
• Epidemiology

Background:

• The World Falls Guidelines (WFG) algorithm stratifies patients into low, intermediate, and high fall risk categories.
• Evaluating different operationalizations of the WFG algorithm is crucial for optimizing its clinical utility.
• Comparison with existing fall screening tools like the AGS/BGS algorithm and 3KQ is necessary to determine the most effective approach.

Purpose of the Study:

• To assess the predictive performance of various operationalizations of the World Falls Guidelines (WFG) algorithm for falls in older adults.
• To compare the WFG algorithm's effectiveness against other established fall screening tools.
• To determine the optimal components for operationalizing the WFG algorithm to maximize fall risk prediction.

Main Methods:

• A population-based cohort of 1509 adults aged 65 years and older from the Longitudinal Aging Study Amsterdam was analyzed.
• Falls were ascertained over a 1-year follow-up period using fall calendars.
• Screening tool items were retrospectively operationalized using baseline data, with proxies employed when necessary.

Main Results:

• The WFG algorithm demonstrated sensitivities ranging from 30.9% to 48.0% and specificities from 77.0% to 88.2%.
• Operationalizing the WFG algorithm with the 3KQ tool instead of fall history improved sensitivity but reduced specificity.
• The WFG algorithm showed comparable predictive performance to the AGS/BGS algorithm and fall history alone, outperforming the 3KQ alone in specificity.

Conclusions:

• The WFG algorithm is effective for identifying patients at risk of falls, especially when integrated with the 3KQ.
• The Clinical Frailty Scale (CFS) and Fried's frailty criteria can be used interchangeably within the WFG algorithm's operationalization.
• While the 3KQ alone offers higher sensitivity, the WFG algorithm provides a more balanced predictive performance with defined risk categories.