Evaluation of the Predictive Accuracy of the interRAI Falls Clinical Assessment Protocol, Scott Fall Risk Screen, and a Supplementary Falls Risk Assessment Tool Used in Residential Long-Term Care: A Retrospective Cohort Study

Development and External Validation of a Machine Learning-based Fall Prediction Model for Nursing Home Residents: A Prospective Cohort Study

OBJECTIVES

To develop and externally validate a machine learning-based fall prediction model for ambulatory nursing home residents. The focus is on predicting fall occurrences within 6 months after baseline assessment through a binary classification task, aiming to provide staff with an effective and user-friendly fall-risk assessment tool.

DESIGN

Prospective cohort study.

SETTING AND PARTICIPANTS

A total of 864 older residents living in 4 nursing homes between May 2022 and March 2023 in China.

METHODS

Potential fall-risk predictors were collected through in-person interviews and assessments of anthropometric and physical function. Participants were followed for 6 months, with falls recorded by trained nurses. Seven machine learning algorithms, including Logistic Regression (LR), Gradient Boosting Machine (GBM), eXtreme Gradient Boosting (XGBoost), Random Forest (RF), Support Vector Machine (SVM), Neural Networks (NN), and Decision Tree (DT), were used to develop prediction models. Performance was evaluated using the area under the receiver operating characteristic curve (ROC-AUC) and Precision-Recall curve (PR-AUC), with calibration assessed via a calibration curve. Feature importance was visualized using SHapley Additive exPlanations (SHAP).

RESULTS

The 6 selected predictors were balance, grip strength, fatigue, fall history, age, and comorbidity. The ROC-AUC for the models ranged from 0.710 to 0.750, PR-AUC from 0.415 to 0.473, sensitivity from 0.704 to 0.914, and specificity from 0.511 to 0.687 in the validation cohort. The LR model was converted into a nomogram.

CONCLUSIONS AND IMPLICATIONS

The machine learning-based fall-prediction models effectively identified nursing home residents at high risk of falls. The developed nomogram can be integrated into clinical practice to enhance fall risk assessment protocols, ultimately improving patient safety and care in nursing homes.

A study on the falls factors among the older adult with cognitive impairment based on large-sample data

Introduction

This study explored the correlative factors of falls among the older adult with cognitive impairment, to provide distinct evidence for preventing falls in the older adult with cognitive impairment compared with the general older adult population.

Methods

This study was based on a cross-sectional survey, with an older adult population of 124,124 was included. The data was sourced from the Elderly Care Unified Needs Assessment for Long-Term Care Insurance in Shanghai. Binary and multivariable logistic regression analyses were conducted sequentially on the correlative factors of falls. Multivariable logistic regression was performed on variables that were significant, stratified by cognitive function levels.

Results

The incidence of fall in the past 90 days was 17.67% in this study. Specific variables such as gender (male), advanced age (≥80), residence with a elevator (or lift), mild or moderate disability, quality of sleep (acceptable/poor) were negatively correlated with falls, while higher education level, living alone, residence with indoor steps, unclean and untidy living environment, MCI or dementia, chronic diseases, restricted joints, impaired vision, and the use of diaper were positively correlative factors of falls. Comparing with older adult with normal cognitive functions, older adult with dementia faced a higher risk of falling due to accessibility barrier in the residence. For general older adults, less frequency of going outside and poor social interactions were positively correlated with falls, while for older adult with cognitive impairments, going outside moderately (sometimes) was found positively correlated with falls. Older adults with cognitive impairments have increased fall risks associated with chronic diseases, restricted joints, and the use of diaper. The risk of falling escalated with the greater number of chronic diseases.

Discussion

For older adult with cognitive impairments, it is advisable to live with others. Additionally, creating an accessible living environment and maintaining the cleanness and tidiness can effectively reduce the risk of falls, particularly for those with MCI or dementia. Optimal outdoor activity plans should be developed separately based on the cognitive function of older adults. Older adult with dementia who have comorbidities should be paid special attention in fall prevention compared to the general older adult population.

Counting what counts: assessing quality of life and its social determinants among nursing home residents with dementia

BACKGROUND

Maximizing quality of life (QoL) is a major goal of care for people with dementia in nursing homes (NHs). Social determinants are critical for residents' QoL. However, similar to the United States and other countries, most Canadian NHs routinely monitor and publicly report quality of care, but not resident QoL and its social determinants. Therefore, we lack robust, quantitative studies evaluating the association of multiple intersecting social determinants with NH residents' QoL. The goal of this study is to address this critical knowledge gap.

METHODS

We will recruit a random sample of 80 NHs from 5 Canadian provinces (Alberta, British Columbia, Manitoba, Nova Scotia, Ontario). We will stratify facilities by urban/rural location, for-profit/not-for-profit ownership, and size (above/below median number of beds among urban versus rural facilities in each province). In video-based structured interviews with care staff, we will complete QoL assessments for each of ~ 4,320 residents, using the DEMQOL-CH, a validated, feasible tool for this purpose. We will also assess resident's social determinants of QoL, using items from validated Canadian population surveys. Health and quality of care data will come from routinely collected Resident Assessment Instrument - Minimum Data Set 2.0 records. Knowledge users (health system decision makers, Alzheimer Societies, NH managers, care staff, people with dementia and their family/friend caregivers) have been involved in the design of this study, and we will partner with them throughout the study. We will share and discuss study findings with knowledge users in web-based summits with embedded focus groups. This will provide much needed data on knowledge users' interpretations, usefulness and intended use of data on NH residents' QoL and its health and social determinants.

DISCUSSION

This large-scale, robust, quantitative study will address a major knowledge gap by assessing QoL and multiple intersecting social determinants of QoL among NH residents with dementia. We will also generate evidence on clusters of intersecting social determinants of QoL. This study will be a prerequisite for future studies to investigate in depth the mechanisms leading to QoL inequities in LTC, longitudinal studies to identify trajectories in QoL, and robust intervention studies aiming to reduce these inequities.

Factors Associated with Falls in Canadian Long Term Care Homes: a Retrospective Cohort Study

Background

Half of Canadians living in long-term care (LTC) homes will fall each year resulting in consequences to independence, quality of life, and health. The objective in this study was to analyze factors that contribute to, or are protective against, falls in Canadian LTC homes.

Methods

We analyzed of a retrospective cohort of a stratified random sample of Canadian LTC homes in Western Canada from 2011-2017. We accessed variables from the RAI-MDS 2.0 to assess the association of the dependent variable "fall within the last 31-180 days" with multiple independent factors, using generalized estimating equation models.

Results

A total of 28,878 LTC residents were analyzed. Factors found to increase the odds of falling were other fractures (OR 3.64 [95% confidence interval; CI 3.27, 4.05]), hip fractures (OR 3.58 [3.27, 3.93]), moderately impaired cognitive skills (OR 2.45 [2.28, 2.64]), partial support to balance standing (OR 2.44 [2.30, 2.57]), wandering (OR 2.31 [2.18, 2.44]).

Conclusion

A range of factors identified were associated with falls for people living in LTC homes. Individual physical ability represented the largest group of independent factors contributing to falls. Residents who experience any fracture or an acute change in behaviour, mobility, or activities of daily living (ADL) should be considered at increased risk of falls.

Developing the Patient Falls Risk Report: A Mixed-Methods Study on Sharing Falls-Related Clinical Information from Home Care with Primary Care Providers

If interRAI home care information were shared with primary care providers, care provision and integration could be enhanced. The objective of this study was to co-develop an interRAI-based clinical information sharing tool (i.e., the Patient Falls Risk Report) with a sample of primary care providers. This mixed-methods study employed semi-structured interviews to inform the development of the Patient Falls Risk Report and online surveys based on the System Usability Scale instrument to test its usability. Most of the interview sample (n = 9) believed that the report could support patient care by sharing relevant and actionable falls-related information. However, criticisms were identified, including insufficient detail, clarity, and support for shared care planning. After incorporating suggestions for improvement, the survey sample (n = 27) determined that the report had excellent usability with an overall usability score of 83.4 (95% CI = 78.7-88.2). By prioritizing the needs of end-users, sustainable interRAI interventions can be developed to support primary care.

Falls prediction using the nursing home minimum dataset

OBJECTIVE

The purpose of the study was to develop and validate a model to predict the risk of experiencing a fall for nursing home residents utilizing data that are electronically available at the more than 15 000 facilities in the United States.

MATERIALS AND METHODS

The fall prediction model was built and tested using 2 extracts of data (2011 through 2013 and 2016 through 2018) from the Long-term Care Minimum Dataset (MDS) combined with drug data from 5 skilled nursing facilities. The model was created using a hybrid Classification and Regression Tree (CART)-logistic approach.

RESULTS

The combined dataset consisted of 3985 residents with mean age of 77 years and 64% female. The model's area under the ROC curve was 0.668 (95% confidence interval: 0.643-0.693) on the validation subsample of the merged data.

DISCUSSION

Inspection of the model showed that antidepressant medications have a significant protective association where the resident has a fall history prior to admission, requires assistance to balance while walking, and some functional range of motion impairment in the lower body; even if the patient exhibits behavioral issues, unstable behaviors, and/or are exposed to multiple psychotropic drugs.

CONCLUSION

The novel hybrid CART-logit algorithm is an advance over the 22 fall risk assessment tools previously evaluated in the nursing home setting because it has a better performance characteristic for the fall prediction window of ≤90 days and it is the only model designed to use features that are easily obtainable at nearly every facility in the United States.