Predict Fall Risk Research Articles

Predicting future fallers in Parkinson’s disease using kinematic data over a period of 5 years

Parkinson’s disease (PD) increases fall risk, leading to injuries and reduced quality of life. Accurate fall risk assessment is crucial for effective care planning. Traditional assessments are subjective and time-consuming, while recent assessment methods based on wearable sensors have been limited to 1-year follow-ups. This study investigated whether a short sensor-based assessment could predict falls over up to 5 years. Data from 104 people with PD without prior falls were collected using six wearable sensors during a 2-min walk and a 30-s postural sway task. Five machine learning classifiers analysed the data. The Random Forest classifier performed best, achieving 78% accuracy (AUC = 0.85) at 60 months. Most models showed excellent performance at 24 months (AUC > 0.90, accuracy 84–92%). Walking and postural variability measures were key predictors. Adding clinicodemographic data, particularly age, improved model performance. Wearable sensors combined with machine learning can effectively predict fall risk, enhancing PD management and prevention strategies.

Research on the Elderly Fitness Fall Risk Prediction Model Based on Decision Tree

As the global aging process intensifies, the issue of falls has become a common health risk in the daily lives of the elderly, especially in physical fitness activities. This article establishes a fall risk prediction model using a decision tree algorithm based on fitness data of the elderly and analyzes the relationship between various health factors and fall risk. The study found that features such as heart rate variability, blood sugar levels, and exercise duration have significant impacts on fall prediction. Through the analysis of the ROC curve and PR curve of the model, the results show that the model performs well in the prediction accuracy and risk identification of fall events. The research in this article not only provides a predictive tool for the safety of the elderly in fitness but also provides a theoretical basis and practical guidance for preventing falls in the elderly and formulating personalized safe exercise plans.

An Electronic Medical Record-Based Prognostic Model for Inpatient Falls: Development and Internal-External Cross-Validation.

Effective fall prevention interventions in hospitals require appropriate allocation of resources early in admission. To address this, fall risk prediction tools and models have been developed with the aim to provide fall prevention strategies to patients at high risk. However, fall risk assessment tools have typically been inaccurate for prediction, ineffective in prevention, and time-consuming to complete. Accurate, dynamic, individualized estimates of fall risk for admitted patients using routinely recorded data may assist in prioritizing fall prevention efforts. The objective of this study was to develop and validate an accurate and dynamic prognostic model for inpatient falls among a cohort of patients using routinely recorded electronic medical record data. We used routinely recorded data from 5 Australian hospitals to develop and internally-externally validate a prediction model for inpatient falls using a Cox proportional hazards model with time-varying covariates. The study cohort included patients admitted during 2018-2021 to any ward, with no age restriction. Predictors used in the model included admission-related administrative data, length of stay, and number of previous falls during the admission (updated every 12 hours up to 14 days after admission). Model calibration was assessed using Poisson regression and discrimination using the area under the time-dependent receiver operating characteristic curve. There were 1,107,556 inpatient admissions, 6004 falls, and 5341 unique fallers. The area under the time-dependent receiver operating characteristic curve was 0.899 (95% CI 0.88-0.91) at 24 hours after admission and declined throughout admission (eg, 0.765, 95% CI 0.75-0.78 on the seventh day after admission). Site-dependent overestimation and underestimation of risk was observed on the calibration plots. Using a large dataset from multiple hospitals and robust methods to model development and validation, we developed a prognostic model for inpatient falls. It had high discrimination, suggesting the model has the potential for operationalization in clinical decision support for prioritizing inpatients for fall prevention. Performance was site dependent, and model recalibration may lead to improved performance.

Predicting Fall Risk in Elderly Individuals: A Comparative Analysis of Machine Learning Models Using Patient Characteristics, Functional Balance Tests, and Computerized Dynamic Posturography.

Predicting Fall Risk in Elderly Individuals: A Comparative Analysis of Machine Learning Models Using Patient Characteristics, Functional Balance Tests, and Computerized Dynamic Posturography.

Malnutrition and Fall Risk in Older Adults: A Comprehensive Assessment Across Different Living Situations.

Malnutrition among older adults is associated with numerous adverse effects, including increased morbidity, mortality, prolonged hospital stays, and a heightened risk of falls. This study aims to investigate the prevalence of malnutrition in different groups of older adults using the F-MNA, anthropometry, and s-albumin and the association between nutritional status and fall risk. A total of 228 participants aged 60 years and older were divided into three groups: (1) patients in an internal medicine ward, (2) individuals living in family homes, and (3) residents of care homes. Disease profiles, nutritional status (assessed using the F-MNA and SNAQ), body composition, fall risk, and biochemical markers were evaluated. The results indicated the highest prevalence of malnutrition among hospitalized individuals. Fall risk was associated with age, calf circumference, the F-MNA, the SNAQ, serum albumin levels, residence in a care home, comorbidities, and the number of medications taken daily. Regression analysis revealed that age, calf circumference, and residence in a care home were independent predictors of fall risk in older adults. Older adults are at significant risk of malnutrition, with the risk notably increasing during hospitalization and long-term stays in care homes. Hospitalized individuals had the poorest nutritional status and were at significant risk of further weight loss, underscoring the importance of post-discharge care and rehabilitation.

Can Timed Up and Go Test discriminate the risk of falling in patients with Multiple Sclerosis with low to moderate impairment?

Purpose The aim of the study was to discriminate the risk of falling in patients with Multiple Sclerosis with low to moderate impairment who do not use walking aid using the Timed Up and Go (TUG) Test. Methods This study was planned as a prospective descriptive study with 204 patients. Falls were recorded in three-month prospective diaries, and the balance was evaluated using the TUG Test. Results By using three-month prospective diaries, the mean age of 109 faller patients and 95 non-faller patients was 35.87 ± 8.35 and 35.35 ± 9.39 years, respectively. The faller group’s mean TUG Test score was 9.46 ± 4.14, whereas the non-faller group’s was 7.74 ± 2.02 s. The TUG Test has an area under the receiver operating characteristic curve value of 0.666 in discriminating fallers from non-fallers (0.592–0.740, %95 Cl). To identify fallers, the TUG Test score of ≥7.85 s has a sensitivity of 0.65 and a specificity of 0.65. Conclusion As a result of the present study, the TUG Test was found to have moderate sensitivity and specificity in predicting the risk of falling. It is not a sufficient stand-alone measure for fall risk prediction, so it is recommended that this test be included as part of a comprehensive falls screening programme.

Vision Care as A Strategy to Prevent Falls Among People with Moderate or Severe Intellectual Disability in the Hostel Setting in Hong Kong

Background: Factors contributing to the higher fall risk among people with intellectual disabilities (PIDs) are complex due to their discrete patterns of multimorbidity. Visual impairment, such as cataract, was common at old age and could be a crucial risk factor. Given the insufficient evidence regarding this relationship, the present study aimed at investigating this issue in a hostel setting in Hong Kong. Methods: This study was conducted in four hostels which served people with moderate to severe intellectual disability. Health data of residents collected in the year of 2022-23 was utilized for the current analysis. Data included fall risk, assessed by Morse Fall Scale, as well as other health conditions (osteoporosis, osteoarthritis, cataract) and demographics (age and gender) among residents. Results: The sample consisted of 199 residents (85 were males and 114 females), with an age range of 22 to 76 years. Around 40% (85; 42.7%) of them aged above 45 years old. Among the whole sample, cataract was the most common health condition and its prevalence reached 27.14%. Those who aged 45-year or above were 4.61 times (95%CI 2.09-11.07) more likely to have cataract. Bivariate analysis results showed that presence of fall risk was associated with older age (above 45 years old; OR 2.38; 95%CI 1.28-4.49), diagnosis of cataract (OR 3.3; 95%CI 1.71-6.33) and osteoarthritis (OR 12.68; 95%CI 1.70-564.75). Logistic regression analysis further illustrated that cataract (p = 0.04) remained as a significant predictor of fall risk after controlling age, gender, osteoarthritis and osteoporosis in the model. Conclusion: Our data showed that presence of cataract diagnosis was significantly associated with higher fall risk among PIDs in hostel setting. Given cataract was a prevalent condition especially among aging PIDs, early screening and intervention could be crucial components of fall prevention strategy in a hostel setting.

Older Adult Fall Risk Prediction with Deep Learning and Timed Up and Go (TUG) Test Data.

Falls are a major health hazard for older adults; therefore, in the context of an aging population, predicting the risk of a patient suffering falls in the near future is of great impact for health care systems. Currently, the standard prospective fall risk assessment instrument relies on a set of clinical and functional mobility assessment tools, one of them being the Timed Up and Go (TUG) test. Recently, wearable inertial measurement units (IMUs) have been proposed to capture motion data that would allow for the building of estimates of fall risk. The hypothesis of this study is that the data gathered from IMU readings while the patient is performing the TUG test can be used to build a predictive model that would provide an estimate of the probability of suffering a fall in the near future, i.e., assessing prospective fall risk. This study applies deep learning convolutional neural networks (CNN) and recurrent neural networks (RNN) to build such predictive models based on features extracted from IMU data acquired during TUG test realizations. Data were obtained from a cohort of 106 older adults wearing wireless IMU sensors with sampling frequencies of 100 Hz while performing the TUG test. The dependent variable is a binary variable that is true if the patient suffered a fall in the six-month follow-up period. This variable was used as the output variable for the supervised training and validations of the deep learning architectures and competing machine learning approaches. A hold-out validation process using 75 subjects for training and 31 subjects for testing was repeated one hundred times to obtain robust estimations of model performances At each repetition, 5-fold cross-validation was carried out to select the best model over the training subset. Best results were achieved by a bidirectional long short-term memory (BLSTM), obtaining an accuracy of 0.83 and AUC of 0.73 with good sensitivity and specificity values.

Predicting falls with ultrasound, physical parameters or fall-risk questions among older adults: A prospective cohort study

BackgroundFalls are a significant issue among older adults, leading to morbidity and mortality. Screening for fall risk in the ED is crucial but challenging due to time limitations and patient conditions. Sarcopenia, characterized by muscle loss, is associated with increased fall risk, and ultrasound has been proposed as a non-invasive tool to measure muscle mass in this context. MethodsThis prospective cohort study enrolled 174 older adults from an urban teaching hospital's EDOU, assessing muscle mass via POCUS, grip strength, Timed Up and Go (TUG) test, and fall risk using the STEADI toolkit. The patients were followed up over six months to assess if they had fallen or not. ResultsFollow-up identified 37 participants (21 %) as patients who fell. There was no significant association between POCUS-measured muscle mass, grip strength, or TUG test performance with future falls. In contrast, STEADI questionnaire responses demonstrated significant differences between patients who fell and did not fall, suggesting its potential utility in predicting fall risk in this population. ConclusionThe simpler tool, the STEADI questionnaire, may offer more practicality in screening fall risk compared to complex ultrasound measurements or physical performance tests among the older in ED.

Feasibility of Fall-Risk Detection in Older Adults: Real-World Use of Sensor Data With Machine Learning.

To use machine learning techniques with sensor data to predict fall risk in older adults aging in place. We tested the feasibility of using anomaly detection on a dataset comprising 315 days of continuous unobtrusive sensor data obtained from a single participant to predict fall risk within a 10-day window. Predictions were validated with performance metrics, including accuracy, F1 score, and receiver operating characteristic-area under curve (ROC-AUC), using actual falls documented in the electronic health record. The model resulted with accuracy = 0.96 (95% confidence interval [CI] [0.94, 0.99]), F1 = 0.78 (95% CI [0.73, 0.83]), and ROC-AUC = 0.89 (95% CI [0.85, 0.93]). The application of anomaly detection on sensor data may provide a timely and valid indication of fall risk in older adults within a 10-day window. Further research and validation are warranted to confirm these findings and expand the scope of application in the domain of older adult care and health care support. [Journal of Gerontological Nursing, 50(10), 7-10.].

212P Identifying biomarkers for prediction of fall-risk in patients with inclusion body myositis (IBM)

212P Identifying biomarkers for prediction of fall-risk in patients with inclusion body myositis (IBM)

Fall risk prediction using temporal gait features and machine learning approaches.

Falls have been acknowledged as a major public health issue around the world. Early detection of fall risk is pivotal for preventive measures. Traditional clinical assessments, although reliable, are resource-intensive and may not always be feasible. This study explores the efficacy of artificial intelligence (AI) in predicting fall risk, leveraging gait analysis through computer vision and machine learning techniques. Data was collected using the Timed Up and Go (TUG) test and JHFRAT assessment from MMU collaborators and augmented with a public dataset from Mendeley involving older adults. The study introduces a robust approach for extracting and analyzing gait features, such as stride time, step time, cadence, and stance time, to distinguish between fallers and non-fallers. Two experimental setups were investigated: one considering separate gait features for each foot and another analyzing averaged features for both feet. Ultimately, the proposed solutions produce promising outcomes, greatly enhancing the model's ability to achieve high levels of accuracy. In particular, the LightGBM demonstrates a superior accuracy of 96% in the prediction task. The findings demonstrate that simple machine learning models can successfully identify individuals at higher fall risk based on gait characteristics, with promising results that could potentially streamline fall risk assessment processes. However, several limitations were discovered throughout the experiment, including an insufficient dataset and data variation, limiting the model's generalizability. These issues are raised for future work consideration. Overall, this research contributes to the growing body of knowledge on fall risk prediction and underscores the potential of AI in enhancing public health strategies through the early identification of at-risk individuals.

Identification of medication-related fall risk in adults and older adults admitted to hospital: A machine learning approach

The study aimed to develop and validate, through machine learning, a fall risk prediction model related to prescribed medications specific to adults and older adults admitted to hospital. A case-control study was carried out in a tertiary hospital, involving 9,037 adults and older adults admitted to hospital in 2016. The variables were analyzed using the algorithms: logistic regression, naive bayes, random forest and gradient boosting. The best model presented an area under the curve = 0.628 in the older adult subgroup, compared to an area under the curve (AUC) = 0.776 in the adult subgroup. A specific model was developed for this sample. The gradient boosting model presented the best performance in the sample of older adults (AUC = 0.71). Models developed to predict the risk of falls based on medications specifically aimed at older adults presented better performance in relation to models developed in the total population studied.

The prediction model of fall risk for the elderly based on gait analysis

IntroductionEarly screening and identification are crucial for fall prevention, and developing a new method to predict fall risk in the elderly can address the current lack of objectivity in assessment tools.MethodsA total of 132 elderly individuals over 80 years old residing in some nursing homes in Shanghai were selected using a convenient sampling method. Fall history information was collected, and gait data during a 10-meter walk were recorded. Logistic regression was employed to establish the prediction model, and a nomogram was used to assess the importance of the indicators. The Bootstrap method was utilized for internal validation of the model, while the verification set was used for external validation. The predictive performance of the model was evaluated using the area under the ROC curve, calibration curve, and decision curve analysis (DCA) to assess clinical benefits.ResultsThe incidence of falls in the sample population was 36.4%. The Tinetti Gait and Balance Test (TGBT) score (OR = 0.832, 95% CI: 0.734,0.944), stride length (OR = 0.007, 95% CI: 0.000,0.104), difference in standing time (OR = 0.001, 95% CI: 0.000,0.742), and mean stride time (OR = 0.992, 95% CI:0.984,1.000) were identified as significant factors. The area under the ROC curve was 0.878 (95% CI: 0.805, 0.952), with a sensitivity of 0.935 and specificity of 0.726. The Brier score was 0.135, and the Hosmer-Lemeshow test (χ2 = 10.650, P = 0.222) indicated a good fit and calibration of the model.ConclusionThe TGBT score, stride length, difference in standing time, and stride time are all protective factors associated with fall risk among the elderly. The developed risk prediction model demonstrates good discrimination and calibration, providing valuable insights for early screening and intervention in fall risk among older adults.

Leveraging feature selection for enhanced fall risk prediction in elderly using gait analysis

There is no effective fall risk screening tool for the elderly that can be integrated into clinical practice. Developing a system that can be easily used in primary care services is a current need. Current studies focus on the use of multiple sensors or activities to achieve higher accuracy. However, multiple sensors and activities reduce the availability of these systems. This study aims to develop a system to perform fall prediction for the elderly by using signals recorded from a single sensor during a short-term activity. A total of 168 features in the time and frequency domains were created using acceleration signals obtained from 71 elderly people. The features were weighted based on the ReliefF algorithm, and the artificial neural networks model was developed using the most important features. The best classification result was obtained using the 17 most important features of those weighted for K = 20 nearest neighbors. The highest accuracy was 82.2% (82.9% Sensitivity, 81.6% Specificity). The partially high accuracy obtained in our study shows that falling can be detected early with a sensor and a simple activity by determining the right features and can be easily applied in the assessment of the elderly during routine follow-ups.Graphical

Assessing muscle quality as a key predictor to differentiate fallers from non-fallers in older adults

Falling is an important public health issue because of its prevalence and severe consequences. Evaluating muscle performance is important when assessing fall risk. The study aimed to identify factors [namely muscle capacity (strength, quality, and power) and spatio-temporal gait attributes] that best discriminate between fallers and non-fallers in older adults. The hypothesis is that muscle quality, defined as the ratio of muscle strength to muscle mass, is the best predictor of fall risk. 184 patients were included, 81% (n = 150) were women and the mean age was 73.6 ± 6.83years. We compared body composition, mean grip strength, spatio-temporal parameters, and muscle capacity of fallers and non-fallers. Muscle quality was calculated as the ratio of maximum strength to fat-free mass. Mean handgrip and power were also controlled by fat-free mass. We performed univariate analysis, logistic regression, and ROC curves. The falling patients had lower muscle quality, muscle mass-controlled power, and mean weighted handgrip than the non-faller. Results showing that lower muscle quality increases fall risk (effect size = 0.891). Logistic regression confirmed muscle quality as a significant predictor (p < .001, OR = 0.82, CI [0.74; 0.89]). ROC curves demonstrated muscle quality as the most predictive factor of falling (AUC = 0.794). This retrospective study showed that muscle quality is the best predictor of fall risk, above spatial and temporal gait parameters. Our results underscore muscle quality as a clinically meaningful assessment and may be a useful complement to other assessments for fall prevention in the aging population.

Prognosis of fall risk in home care clients: A noninvasive approach using survival analysis

Abstract Aim The present study has designed, implemented, and evaluated a machine learning model that can predict fall risk and fall occurrence in community-dwelling elderly based on their home care usage. Subjects and methods A dataset consisting of 2542 weekly home care records for 1499 citizens (59% female, 41% male) with a mean age of 77 years (SD 10 years) was collected from a large municipality in Denmark. The data were recorded between January 1, 2021, and December 31, 2021. The dataset was divided into two cohorts. Subsequently, five machine learning-based survival analysis models were trained and evaluated using cross-validation. Results The CoxBoost model showed the best discriminative performance with a mean 0.64 (95% CI 0.57–0.72) Harrell’s concordance index, indicating better ranking than chance-level by 14% on average. However, the model could not accurately predict when the next fall would take place. Conclusions The proposed method enables professionals to assess individual fall risk by using home care records from an Electronic Health Record (EHR) system. This facilitates the initiation of targeted fall-prevention programs for those at highest risk. Additionally, it is expected that a risk-based approach can lead to a lower number needed to treat (NNT), indicating greater effectiveness of health interventions.

Tandem gait step-width increases more rapidly in more severely affected people with Parkinson's disease

IntroductionTandem gait performance reportedly predicts fall risk in people with Parkinson's disease (PwPD) and help distinguish PwPD from atypical parkinsonism. In a cross-sectional study, we previously showed that tandem gait step-width widens with increasing Hoehn and Yahr (H&Y) staging scores. In this longitudinal study, we aimed to determine if progression in tandem gait deficits is dependent on disease severity in PwPD. MethodsParticipants underwent an instrumented tandem gait measurement every 6 months for at least 2 years. The mean and variability of 4 tandem gait parameters were calculated at each visit: step-width, step-length, step-time, and step-velocity. The change in these parameters over time for 3 H&Y groups (stage 1, 2 and 2.5+) compared to aging controls was determined using a random coefficients regression model. The annual percent change in tandem gait parameters was correlated with initial disease features using Kendall's τB. Results66 participants were analyzed (46 PD, 20 controls). Mean step-width increased over time in an H&Y stage-dependent manner, with H&Y 2 and H&Y 2.5+ experiencing increases of 6% and 10% per year (p = 0.001 and 0.024 respectively). Annual percent-change in step-width was correlated with initial motor Unified Parkinson's Disease Rating Scale (UPDRS) scores (Kendall's τB = 0.229), total UPDRS scores (τB = 0.249), H&Y scores (τB = 0.266) and inversely correlated with Montreal Cognitive Assessment (MoCA) scores (τB = −0.209; ps ≤ 0.019). ConclusionTandem gait step-width widens over time more rapidly in more severely affected PD patients. These results suggest that tandem gait should be routinely clinically evaluated and considered in the management of imbalance in PwPD.

Investigating Eye Movement and Postural Stability Relationships Using Mobile Eye-Tracking and Posturography: A Cross-Sectional Study.

Vision and eye movements play a crucial role in maintaining postural stability. This study investigated the relationship between eye movements and postural control in healthy adults using mobile eye-tracking technology and posturography. Forty healthy participants underwent assessments of eye movements using a mobile eye-tracking system and postural stability using Tetrax posturography under various sensory conditions. Pearson correlation coefficients were computed to examine associations between eye movement parameters and postural control indices. Significant correlations were found between eye movement parameters and postural stability indices. Faster and more consistent horizontal eye movements were associated with better postural stability (r = -0.63, p < 0.05). Eye movement speed variability positively correlated with weight distribution indices under normal eyes open (r = 0.65, p < 0.05) and closed (r = 0.59, p < 0.05) conditions. Coordination of horizontal and vertical eye movements positively correlated with postural control (r = 0.69, p < 0.01). Negative correlations were observed between eye movement coordination and Fourier indices in various frequency bands (p < 0.05) and the stability index under different head positions (p < 0.05). The findings provide insights into sensory integration mechanisms underlying balance maintenance and highlight the importance of integrated sensory processing in postural stability. Eye movement assessments have potential applications in balance evaluation and fall risk prediction.

Fall Risk Screening in Older Adult Patients on Chronic Opioid Therapy: A Quality Improvement Project.

Many older adults prescribed opioid pain medications may be at increased risk of falls. As a quality improvement (QI) initiative, the University of Utah Sugar House clinic initiated a 9-month fall risk screening pilot for older adult patients on chronic opioids. This QI project sought to determine the feasibility of adding screening to a busy clinical practice, examine risk of falls in this patient cohort, and examine whether there were significant clinical and demographic differences between the patients who did or did not receive screening. We observed whether conventionally understood fall risk factors, including higher opioid doses, concurrent benzodiazepine prescription, and age, correlate with high fall risk. After the screening initiative, we determined that it was possible to significantly increase fall risk screening rates with prompts in Electronic Health Record (EHR). This cohort's fall risk screening increased from 18% to 33%. In the patients who were screened, 43% were found to be at high risk of falls. We did not see a correlation with high fall risk and patients taking higher doses of opioids or concurrent benzodiazepine prescriptions. These findings emphasize the need for consistent screening in primary care because review of the medication list alone is not a reliable predictor of fall risk.