Comparison of decision tree, support vector machines, and Bayesian network approaches for classification of falls in Parkinson’s disease

Being able to accurately predict the risk of falling is crucial in patients with Parkinson’s dis- ease (PD). This is due to the unfavorable effect of falls, which can lower the quality of life as well as directly impact on survival. Three methods considered for predicting falls are decision trees (DT), Bayesian networks (BN), and support vector machines (SVM). Data on a 1-year prospective study conducted at IHBI, Australia, for 51 people with PD are used. Data processing are conducted using rpart and e1071 packages in R for DT and SVM, con- secutively; and Bayes Server 5.5 for the BN. The results show that BN and SVM produce consistently higher accuracy over the 12 months evaluation time points (average sensitivity and specificity > 92%) than DT (average sensitivity 88%, average specificity 72%). DT is prone to imbalanced data so needs to adjust for the misclassification cost. However, DT provides a straightforward, interpretable result and thus is appealing for helping to identify important items related to falls and to generate fallers’ profiles.

Educators’ perspectives about how older hospital patients can engage in a falls prevention education programme: A qualitative process evaluation

- Objectives Falls are the most frequent adverse event reported in hospitals. Patient and staff education delivered by trained educators significantly reduced falls and injurious falls in an older rehabilitation population. The purpose of the study was to explore the educators’ perspectives of delivering the education and to conceptualise how the programme worked to prevent falls among older patients who received the education. - Design A qualitative exploratory study. - Methods Data were gathered from three sources: conducting a focus group and an interview (n=10 educators), written educator notes and reflective researcher field notes based on interactions with the educators during the primary study. The educators delivered the programme on eight rehabilitation wards for periods of between 10 and 40 weeks. They provided older patients with individualised education to engage in falls prevention and provided staff with education to support patient actions. Data were thematically analysed and presented using a conceptual framework. - Results Falls prevention education led to mutual understanding between staff and patients which assisted patients to engage in falls prevention behaviours. Mutual understanding was derived from the following observations: the educators perceived that they could facilitate an effective three-way interaction between staff actions, patient actions and the ward environment which led to behaviour change on the wards. This included engaging with staff and patients, and assisting them to reconcile differing perspectives about falls prevention behaviours. - Conclusions Individualised falls prevention education effectively provides patients who receive it with the capability and motivation to develop and undertake behavioural strategies that reduce their falls, if supported by staff and the ward environment.

Reducing falls in hospital through education to change patient and staff behaviour: A stepped wedge cluster randomised controlled effectiveness trial

Background: Falls remain the most frequent adverse event reported in hospitals, particularly geriatric rehabilitation wards. Randomised trials reducing fall injuries in hospitals have been elusive. Our previous randomised trial (n = 1206) demonstrated that multimedia education with physiotherapist falls educator support reduced falls among patients with higher cognition levels, but this benefit was offset by a potential increase in falls rates among patients with poor cognition. In the previous trial, hospital staff were blinded to the allocation of individual patients, and only delivered usual care.

Visual Impairment and Falls Risk in the Elderly

Vision is highly important for balance and gait and visual impairments are significantly associated with locomotion problems and falls in older people. There is now a large body of research linking falls and fall-related injuries with visual problems, some of which are easily remedied by surgery or refractive correction. However there is also evidence that the kind of refractive correction provided (in terms of single-vision or multifocal correction) can also have an effect on fall risk. This chapter provides an overview of the major findings in this area.