Proposal for a New Predictive Scale for Recurrent Risk of Fall in a Cohort of Community-Dwelling Patients with Stroke

Inverted U-shaped relationship between Barthel Index Score and falls in Chinese non-bedridden patients: a cross-sectional study

BACKGROUND

Performing activities of daily living comprise an important risk factor for falls among non-bedridden stroke inpatients in rehabilitation departments.

OBJECTIVES

To explore the correlation between Barthel Index score and the occurrence of falls in non-bedridden stroke rehabilitation inpatients.

METHODS

In this cross-sectional study, information of patients grouped as non-bedridden patients by the Longshi Scale was collected.

RESULTS

A total of 3097 patients were included in this study, with a fall incidence of 10.43%. After adjusting covariates, the total score of Barthel Index and falls in non-bedridden inpatients after stroke presented an inverted U-shaped curve relationship, in which inflection point was 60. The effect sizes on the left and right sides of infection point were 1.02 (95%CI 1.00-1.04) and 0.97 (95%CI 0.96-0.99), respectively.

CONCLUSIONS

Non-bedridden stroke patients with moderate activities of daily living (ADL) capacity may be at particularly increased risk of falls in rehabilitation departments.

Reliability and validity of balance tests for community-dwelling older adults with musculoskeletal ambulation disability symptom complex

INTRODUCTION

Physical performance-based tests are useful indices for identifying and managing fall risk in community-dwelling older adults. However, measurement errors in these tests have not been examined. This study aimed to clarify the reliability and validity of balance tests in community-dwelling older adults with musculoskeletal ambulation disability symptom complex.

METHODS

This cross-sectional study used a test-retest design and included 52 participants from three medical facilities. The participants reported demographic and medical information using a questionnaire. Primary outcomes were the timed up-and-go test, five-times sit-to-stand test, one-leg standing test, and functional reach test results.

RESULTS

Significant correlations were found between each balance test and fear of falling and between the four balance tests. Each balance test demonstrated discriminant validity with intraclass correlation coefficients (1,1) ≥0.81. The minimal detectable change was 1.4 s, 1.6 s, 3.3 s, and 4.0 cm for the timed up-and-go test, five-time sit-to-stand test, one-leg standing test, and functional reach test, respectively.

CONCLUSION

Changes exceeding the minimal detectable change95 in balance tests can be used to judge improvement or decline in physical function, which may inform clinical decision-making. The minimal detectable change95 of balance tests should be used when implementing or modifying fall prevention programs.

Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking

BACKGROUND

Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy.

OBJECTIVE

Develop machine learning models using inertial sensors to objectively classify fall risk in stroke survivors. Determine optimal sensor configurations and clinical test protocols.

METHODS

21 stroke survivors performed balance, Timed Up and Go, 10 Meter Walk, and Sit-to-Stand tests with and without dual-tasking. A total of 8 motion sensors captured lower limb and trunk kinematics, and 92 spatiotemporal gait and clinical features were extracted. Supervised models-Support Vector Machine, Logistic Regression, and Random Forest-were implemented to classify high vs. low fall risk. Sensor setups and test combinations were evaluated.

RESULTS

The Random Forest model achieved 91% accuracy using dual-task balance sway and Timed Up and Go walk time features. Single thorax sensor models performed similarly to multi-sensor models. Balance and Timed Up and Go best-predicted fall risk.

CONCLUSION

Machine learning models using minimal inertial sensors during clinical assessments can accurately quantify fall risk in stroke survivors. Single thorax sensor setups are effective. Findings demonstrate a feasible objective fall screening approach to assist rehabilitation.

Brief composite mobility index predicts post-stroke fallers after hospital discharge

Introduction

Community-dwelling, ambulatory stroke survivors fall at very high rates in the first 3-6 months. Current inpatient clinical assessments for fall risk have inadequate predictive accuracy. We found that a pre-discharge obstacle-crossing test has excellent specificity (83%) but lacks acceptable sensitivity (67%) for identifying would-be fallers and non-fallers post discharge.

Hypothesis

We assessed the hypothesis that combining the obstacle-crossing test with other highly discriminatory fall risk factors would compensate for the obstacle test's fair sensitivity and yield an instrument with superior prediction accuracy.

Methods

45 ambulatory stroke survivors (60 ± 11 years old, 15 ± 11 days post stroke) being discharged home completed a battery of physical performance-based and self-reported measures 1-5 days prior to discharge. After discharge, participants were prospectively followed and classified as fallers (≥1 fall) or non-fallers at 3 months. Pre-discharge measures with the largest effect sizes for differentiating fallers and non-fallers were combined into a composite index. Several variations of the composite index were examined to optimize accuracy.

Results

A 4-item discharge composite index significantly predicted fall status at 3-months. The goodness of fit of the regression model was significantly better than the obstacle-crossing test alone, χ ²(1) = 6.036, p = 0.014. Furthermore, whereas the obstacle-crossing test had acceptable overall accuracy (AUC 0.78, 95% CI, 0.60-0.90), the composite index had excellent accuracy (AUC 0.85, 95% CI, 0.74-0.96). Combining the obstacle-crossing test with only the step test produced a model of equivalent accuracy (AUC 0.85, 95% CI, 0.73-0.96) and with better symmetry between sensitivity and specificity (0.71, 0.83) than the 4-item composite index (0.86, 0.67). This 2-item index was validated in an independent sample of n = 30 and with bootstrapping 1,000 samples from the pooled cohorts. The 4-item index was internally validated with bootstrapping 1,000 samples from the derivation cohort plus n = 9 additional participants.

Conclusion

This study provides convincing proof-of-concept that strategic aggregation of performance-based and self-reported mobility measures, including a novel and demanding obstacle-crossing test, can predict post-discharge fallers with excellent accuracy. Further instrument development is warranted to construct a brief aggregate tool that will be pragmatic for inpatient use and improve identification of future post-stroke fallers before the first fall.

A Systematic Review of Fall Risk Factors in Stroke Survivors: Towards Improved Assessment Platforms and Protocols

Background/Purpose: To prevent falling, a common incident with debilitating health consequences among stroke survivors, it is important to identify significant fall risk factors (FRFs) towards developing and implementing predictive and preventive strategies and guidelines. This review provides a systematic approach for identifying the relevant FRFs and shedding light on future directions of research. Methods: A systematic search was conducted in 5 popular research databases. Studies investigating the FRFs in the stroke community were evaluated to identify the commonality and trend of FRFs in the relevant literature. Results: twenty-seven relevant articles were reviewed and analyzed spanning the years 1995-2020. The results confirmed that the most common FRFs were age (21/27, i.e., considered in 21 out of 27 studies), gender (21/27), motion-related measures (19/27), motor function/impairment (17/27), balance-related measures (16/27), and cognitive impairment (11/27). Among these factors, motion-related measures had the highest rate of significance (i.e., 84% or 16/19). Due to the high commonality of balance/motion-related measures, we further analyzed these factors. We identified a trend reflecting that subjective tools are increasingly being replaced by simple objective measures (e.g., 10-m walk), and most recently by quantitative measures based on detailed motion analysis. Conclusion: There remains a gap for a standardized systematic approach for selecting relevant FRFs in stroke fall risk literature. This study provides an evidence-based methodology to identify the relevant risk factors, as well as their commonalities and trends. Three significant areas for future research on post stroke fall risk assessment have been identified: 1) further exploration the efficacy of quantitative detailed motion analysis; 2) implementation of inertial measurement units as a cost-effective and accessible tool in clinics and beyond; and 3) investigation of the capability of cognitive-motor dual-task paradigms and their association with FRFs.

Clinical characterization, disability, and mortality in people with strokes during 90 days

OBJECTIVES

to describe clinical characteristics and mortality of people with ischemic cerebrovascular accidents (strokes); to compare disability before the event and 90 days after.

METHODS

longitudinal study with 308 people hospitalized in Salvador-BA. Data collection took place from 03/2019 to 01/2020. Descriptive and inferential statistics were used.

RESULTS

mean age was 64.8 years, and National Institute of Health Stroke Scale score was 10.7. The median length of stay in the hospital was 11 days. Afro-descendants predominated (84%), elementary educational level (68.4%), income up to three minimum wages (89.1%), arrival within 4.5 hours of symptoms (57.9%) and admission to a specialized unit (71.8%). Prevalence of thrombolysis: 26%. The asymptomatic before the event category predominated (85.3%) as did the moderate/severe disability (41.5%) after 90 days. 19.7% of the sample evolved to death. Conclusions: the high mortality and disability generated by the event have implications for health management and care.

The Added Value of Combined Timed Up and Go Test, Walking Speed, and Grip Strength on Predicting Recurrent Falls in Chinese Community-dwelling Elderly

Purpose

To determine whether combined performance-based models could exert better predictive values toward discriminating community-dwelling elderly with high risk of any-falls or recurrent-falls.

Participants and Methods

This prospective cohort study included a total of 875 elderly participants (mean age: 67.10±5.94 years) with 513 females and 362 males, recruited from Hangu suburb area of Tianjin, China. All participants completed comprehensive assessments.

Methods

We documented information about sociodemographic information, behavioral characteristics and medical conditions. Three functional tests—timed up and go test (TUGT), walking speed (WS), and grip strength (GS) were used to create combined models. New onsets of any-falls and recurrent-falls were ascertained at one-year follow-up appointment.

Results

In total 200 individuals experienced falls over a one-year period, in which 66 individuals belonged to the recurrent-falls group (33%). According to the receiver operating characteristic curve (ROC), the cutoff points of TUGT, WS, and GS toward recurrent-falls were 10.31 s, 0.9467 m/s and 0.3742 kg/kg respectively. We evaluated good performance as “+” while poor performance as “–”. After multivariate adjustment, we found “TUGT >10.31 s” showed a strong correlation with both any-falls (adjusted odds ratio (OR)=2.025; 95% confidence interval (CI)=1.425–2.877) and recurrent-falls (adjusted OR=2.150; 95%CI=1.169–3.954). Among combined functional models, “TUGT >10.31 s, GS <0.3742 kg/kg, WS >0.9467 m/s” showed strongest correlation with both any-falls (adjusted OR=5.499; 95%CI=2.982–10.140) and recurrent-falls (adjusted OR=8.260; 95%CI=3.880–17.585). And this combined functional model significantly increased discriminating abilities on screening recurrent-fallers than a single test (C-statistics=0.815, 95%CI=0.782–0.884, P<0.001), while not better than a single test in predicting any-fallers (P=0.083).

Conclusion

Elderly people with poor TUGT performance, weaker GS but quicker WS need to be given high priority toward fall prevention strategies for higher risks and frequencies. Meanwhile, the combined “TUGT–, GS–, WS+” model presents increased discriminating ability and could be used as a conventional tool to discriminate recurrent-fallers in clinical practice.

External validation of the recurrent falls risk scale in community-dwelling stroke individuals

OBJECTIVE

To externally validate the Recurrent Fall Risk Scale (ReFR) in community-dwelling stroke survivors.

METHODS

Cohort of stroke survivors with independent gait ability recruited from a reference outpatient stroke clinic. Besides sociodemographic and clinical data, the following scales were used: Modified Barthel Index (mBI), ReFR scale and National Institutes of Health Stroke Scale (NIHSS). Participants were followed up for 12 months to record the incidence of falls. Accuracy of the ReFR scale was measured by the area under the ROC curve.

RESULTS

One hundred and thirteen individuals were recruited between April 2016 and November 2016: mean age 54 years (± 14), 55% women, median time since the last stroke 24 months (range 12 -48 months), posterior vascular territory affected in 35% of the sample. Median NIHSS was 3 (range 1 to 6), median mBI 49 (range 46-50), median ReFR 3 (range 2 to 5). During the follow-up period, 32 (33%) subjects had at least one fall and 18 (19%) were recurrent fallers (two or more falls). The accuracy of ReFR scale was 0.67 (95% CI = 0.54-0.79), p = 0.026.

CONCLUSION

This study externally validated the ReFR as a tool to predict recurrent falls in individuals after stroke.

Components of physical fitness as prognostic factors for falls in elderly female exercise practitioners: a longitudinal study

Abstract Introduction: Physical exercise is used as a strategy for the prevention of falls because it improves the physical fitness of older adults. Objective: To determine which components of physical fitness are predictors of falls in elderly female exercise practitioners. Method: Longitudinal, descriptive, comparative study. The components of physical fitness (upper and lower limb strength and flexibility, agility, aerobic endurance, and hand grip strength) and the occurrence or not of falls in the last 12 months were analyzed in 80 older women practitioners of exercise from 2013 to 2016. Descriptive analysis, ROC curve attributing cut-off points, and binary logistic regression for the prediction of falls were used. Results: The mean age of the participants was 67.46 years (SD=7.65). Over the years, a significant difference between elderly fallers and non-fallers was observed for right hand grip strength (2013 and 2015), left hand grip strength (2014 and 2015), lower limb flexibility (2013, 2014 and 2015), agility (2015), and aerobic endurance (2015). In adjusted analysis, older women with poor lower limb flexibility in 2013 had a higher risk of falls in 2016 (OR=4.98; 95%CI 1.12 - 22.1). Older women with poor flexibility in 2015 also had a higher risk of falls (OR= 6.08; 95%CI 1.06 - 34.67). Conclusion: Poor performance in the lower limb flexibility, right and left hand grip strength, agility and aerobic endurance tests was associated with falls and these components are predictors of future falls in elderly exercise practitioners.