The Syncope-Falls Index: a tool for predicting risk of syncope and complex falls in the older adult based on cumulative health deficits

Abstract

Background Syncope is aetiologically diverse and associated with adverse outcomes; in older people, there is clinical overlap with complex falls presentations (i.e. recurrent, unexplained, and/or injurious). Aim To formulate an index to predict future risk of syncope and falls in the Irish Longitudinal Study on Ageing (TILDA). Design/Methods Using the frailty index methodology, we selected, from TILDA Wave 1 (2010), 40 deficits that might increase risk of syncope and falls. This syncope-falls index (SYFI) was applied to TILDA Wave 1 participants aged 65 and over, who were divided into three risk groups (low, intermediate, high) based on SYFI tertiles. Multivariate logistic regression models were used to investigate, controlling for age and sex, how SYFI groups predicted incident syncope, complex falls and simple falls occurring up to Wave 4 of the study (2016). Results At Wave 1, there were 3499 participants (mean age 73, 53% women). By Wave 4, of the remaining 2907 participants, 185 (6.4%) had reported new syncope, 1077 (37.0%) complex falls, and 218 (7.5%) simple falls. The risk of both syncope and complex falls increased along the SYFI groups (high risk group: OR 1.88 [1.26–2.80], P = 0.002 for syncope; 2.22 [1.82–2.72], P < 0.001 for complex falls). No significant relationship was identified between SYFI and simple falls. Conclusion The 6-year incidences of falls and syncope were high in this cohort. SYFI could help identify older adults at risk of syncope and complex falls, and thus facilitate early referral to specialist clinics to improve outcomes.

Using Explainable AI (XAI) for the Prediction of Falls in the Older Population

“…A total of 46 features were used. The majority of the input features were taken from the previously investigated Syncope-Falls Index (SYFI), a 40-deficit frailty index which was derived from the simple arithmetic sum of 40 health deficits selected from TILDA Wave 1 which was based on their clinically postulated likelihood of increasing the risk of syncope and falls within the older population [6]. Age and sex, which were not originally included in the SYFI, were also included as input features in the present work.…”

“…Table 1 summarises the 46 input features that were utilised in the present study. Details of the 40 SYFI deficits have been described elsewhere [6]. Falls were classified into the following types: simple, complex, and syncope.…”

“…This strategy mirrors the UK's 2013 National Institute for Health and Care Excellence (NICE) guidelines in that they also recommend a multifactorial risk assessment, ideally within a specialised falls service, after an older person has experienced a fall or demonstrated abnormalities in their gait and balance [4]. This prompt, multifaceted assessment is particularly important when falls are recurrent, unexplained, and/or injurious because these types of falls tend to have worse clinical outcomes [5,6].…”

“…A previous falls prediction effort that was based on the simple arithmetic sum of individual health deficits (frailty index) offered significant results but with low degree of explainability [6]. However, by implementing XAI, new insights may be obtained through us being able to understand which features are the most important, and how they influence the prediction outcome.…”

“…In this work, we utilised the large, publicly available data resource from The Irish Longitudinal Study on Ageing (TILDA) to implement Random Forests and XAI techniques for the prediction of different types of falls and analyse their contributory factors using input features that included those of a previously investigated frailty index [6].…”