Opinion versus practice regarding the use of rehabilitation services in home care: an investigation using machine learning algorithms

Estimation of service length with the machine learning algorithms and neural networks for patients who receiving home health care

The main purpose of the study is to develop an estimation model using machine learning algorithms and to ensure the effective and efficient implementation of home health care service planning in hospitals with these algorithms. The necessary approvals for the study were obtained. The data set was created by obtaining patient data (except for data such as Turkish Republic identification number) from 14 hospitals providing Home Health Care Services in the city of Diyarbakır. The data set was subjected to necessary pre-processing and descriptive statistics were applied. For the estimation model, Decision Tree, Random Forest and Multi-layer Perceptron Neural Network algorithms were used. It was found that the number of days of home health care service, which the patients received, varied depending on their age and gender. It was observed that the patients were generally in the disease groups that required Physiotherapy and Rehabilitation treatments. It was determined that the length of service for patients can be predicted with a high reliability rate (Multi-Layer Model Acc: 90.4%, Decision Tree Model Acc: 86.4%, Random Forest Model Acc: 88.5%) using machine learning algorithms. In the light of the findings and data patterns obtained in the study, it is thought that effective and efficient planning will be made in terms of health management. In addition, it is believed that estimating the average length of service for patients will contribute to strategic planning of human resources for health, and to reducing medical consumables, drugs and hospital expenses.

Recent functional decline and outpatient follow-up after hospital discharge: a cohort study

BACKGROUND

Functional decline is common following acute hospitalization and is associated with hospital readmission, institutionalization, and mortality. People with functional decline may have difficulty accessing post-discharge medical care, even though early physician follow-up has the potential to prevent poor outcomes and is integral to high-quality transitional care. We sought to determine whether recent functional decline was associated with lower rates of post-discharge physician follow-up, and whether this association changed during the COVID-19 pandemic, given that both functional decline and COVID-19 may affect access to post-discharge care.

METHOD

We conducted a retrospective cohort study using health administrative data from Ontario, Canada. We included patients over 65 who were discharged from an acute care facility during March 1st, 2019 - January 31st, 2020 (pre-COVID-19 period), and March 1st, 2020 - January 31st, 2021 (COVID-19 period), and who were assessed for home care while in hospital. Patients with and without functional decline were compared. Our primary outcome was any physician follow-up visit within 7 days of discharge. We used propensity score weighting to compare outcomes between those with and without functional decline.

RESULTS

Our study included 21,771 (pre-COVID) and 17,248 (COVID) hospitalized patients, of whom 15,637 (71.8%) and 12,965 (75.2%) had recent functional decline. Pre-COVID, there was no difference in physician follow-up within 7 days of discharge (Functional decline 45.0% vs. No functional decline 44.0%; RR = 1.02, 95% CI 0.98-1.06). These results did not change in the COVID-19 period (Functional decline 51.1% vs. No functional decline 49.4%; RR = 1.03, 95% CI 0.99-1.08, Z-test for interaction p = 0.72). In the COVID-19 cohort, functional decline was associated with having a 7-day physician virtual visit (RR 1.15; 95% CI 1.08-1.24) and a 7-day physician home visit (RR 1.64; 95% CI 1.10-2.43).

CONCLUSIONS

Functional decline was not associated with reduced 7-day post-discharge physician follow-up in either the pre-COVID-19 or COVID-19 periods. In the COVID-19 period, functional decline was positively associated with 7-day virtual and home-visit follow-up.

Using machine learning for the personalised prediction of revision endoscopic sinus surgery

BACKGROUND

Revision endoscopic sinus surgery (ESS) is often considered for chronic rhinosinusitis (CRS) if maximal conservative treatment and baseline ESS prove insufficient. Emerging research outlines the risk factors of revision ESS. However, accurately predicting revision ESS at the individual level remains uncertain. This study aims to examine the prediction accuracy of revision ESS and to identify the effects of risk factors at the individual level.

METHODS

We collected demographic and clinical variables from the electronic health records of 767 surgical CRS patients ≥16 years of age. Revision ESS was performed on 111 (14.5%) patients. The prediction accuracy of revision ESS was examined by training and validating different machine learning models, while the effects of variables were analysed using the Shapley values and partial dependence plots.

RESULTS

The logistic regression, gradient boosting and random forest classifiers performed similarly in predicting revision ESS. Area under the receiving operating characteristic curve (AUROC) values were 0.744, 0.741 and 0.730, respectively, using data collected from the baseline visit until six months after baseline ESS. The length of time during which data were collected improved the prediction performance. For data collection times of 0, 3, 6 and 12 months after baseline ESS, AUROC values for the logistic regression were 0.682, 0.715, 0.744 and 0.784, respectively. The number of visits before or after baseline ESS, the number of days from the baseline visit to the baseline ESS, patient age, CRS with nasal polyps (CRSwNP), asthma, non-steroidal anti-inflammatory drug exacerbated respiratory disease and immunodeficiency or suspicion of it all associated with revision ESS. Patient age and number of visits before baseline ESS carried non-linear effects for predictions.

CONCLUSIONS

Intelligent data analysis identified important predictors of revision ESS at the individual level, such as the frequency of clinical visits, patient age, Type 2 high diseases and immunodeficiency or a suspicion of it.

Patterns of home care assessment and service provision before and during the COVID-19 pandemic in Ontario, Canada

OBJECTIVE

The objective was to compare home care episode, standardised assessment, and service patterns in Ontario's publicly funded home care system during the first wave of the COVID-19 pandemic (i.e., March to September 2020) using the previous year as reference.

STUDY DESIGN AND SETTING

We plotted monthly time series data from March 2019 to September 2020 for home care recipients in Ontario, Canada. Home care episodes were linked to interRAI Home Care assessments, interRAI Contact Assessments, and home care services. Health status measures from the patient's most recent interRAI assessment were used to stratify the receipt of personal support, nursing, and occupational or physical therapy services. Significant level and slope changes were detected using Poisson, beta, and linear regression models.

RESULTS

The March to September 2020 period was associated with significantly fewer home care admissions, discharges, and standardised assessments. Among those assessed with the interRAI Home Care assessment, significantly fewer patients received any personal support services. Among those assessed with either interRAI assessment and identified to have rehabilitation needs, significantly fewer patients received any therapy services. Among patients receiving services, patients received significantly fewer hours of personal support and fewer therapy visits per month. By September 2020, the rate of admissions and services had mostly returned to pre-pandemic levels, but completion of standardised assessments lagged behind.

CONCLUSION

The first wave of the COVID-19 pandemic was associated with substantial changes in Ontario's publicly funded home care system. Although it may have been necessary to prioritise service delivery during a crisis situation, standardised assessments are needed to support individualised patient care and system-level monitoring. Given the potential disruptions to home care services, future studies should examine the impact of the pandemic on the health and well-being of home care recipients and their caregiving networks.

Application Scenarios for Artificial Intelligence in Nursing Care: Rapid Review

Background

Artificial intelligence (AI) holds the promise of supporting nurses’ clinical decision-making in complex care situations or conducting tasks that are remote from direct patient interaction, such as documentation processes. There has been an increase in the research and development of AI applications for nursing care, but there is a persistent lack of an extensive overview covering the evidence base for promising application scenarios.

Objective

This study synthesizes literature on application scenarios for AI in nursing care settings as well as highlights adjacent aspects in the ethical, legal, and social discourse surrounding the application of AI in nursing care.

Methods

Following a rapid review design, PubMed, CINAHL, Association for Computing Machinery Digital Library, Institute of Electrical and Electronics Engineers Xplore, Digital Bibliography & Library Project, and Association for Information Systems Library, as well as the libraries of leading AI conferences, were searched in June 2020. Publications of original quantitative and qualitative research, systematic reviews, discussion papers, and essays on the ethical, legal, and social implications published in English were included. Eligible studies were analyzed on the basis of predetermined selection criteria.

Results

The titles and abstracts of 7016 publications and 704 full texts were screened, and 292 publications were included. Hospitals were the most prominent study setting, followed by independent living at home; fewer application scenarios were identified for nursing homes or home care. Most studies used machine learning algorithms, whereas expert or hybrid systems were entailed in less than every 10th publication. The application context of focusing on image and signal processing with tracking, monitoring, or the classification of activity and health followed by care coordination and communication, as well as fall detection, was the main purpose of AI applications. Few studies have reported the effects of AI applications on clinical or organizational outcomes, lacking particularly in data gathered outside laboratory conditions. In addition to technological requirements, the reporting and inclusion of certain requirements capture more overarching topics, such as data privacy, safety, and technology acceptance. Ethical, legal, and social implications reflect the discourse on technology use in health care but have mostly not been discussed in meaningful and potentially encompassing detail.

Conclusions

The results highlight the potential for the application of AI systems in different nursing care settings. Considering the lack of findings on the effectiveness and application of AI systems in real-world scenarios, future research should reflect on a more nursing care–specific perspective toward objectives, outcomes, and benefits. We identify that, crucially, an advancement in technological-societal discourse that surrounds the ethical and legal implications of AI applications in nursing care is a necessary next step. Further, we outline the need for greater participation among all of the stakeholders involved.

Developing and validating models for predicting nursing home admission using only RAI-HC instrument data

OBJECTIVE

In recent years research has identified important predictors for nursing home admission (NHA). However, as far as we know, the previous risk models use complex variable sets from many sources and the output is a single risk value. The objective of this study was to develop an NHA risk model with a variable set from single data source and richer output information.

METHODS

In this study, we developed a model selecting variables only from the RAI-HC (Resident Assessment Instrument - Home Care) system. Furthermore, we used principal component analysis and K-means clustering to target proper interventions for high-risk clients.

RESULTS

The performance of the model was close to the complex previous model (recall [Formula: see text] vs. [Formula: see text] and specificity [Formula: see text] vs. [Formula: see text]). For the risk clients, three intervention clusters (deficiency in physical functionality, deficiency in cognitive functionality and depression and mood disorders) were found.

CONCLUSION

The NHA risk model and intervention clusters are important because they enable the identification of proper interventions for the right clients. The fact that the model with RAI-HC data alone was accurate enough simplifies the integration of the NHA risk model into practice because it uses data from one system and the algorithm can be integrated easily into the source system.

Implementation of a Standardized Comprehensive Assessment Tool in France: A Case Using the InterRAI Instruments

BACKGROUND

The improvement of quality of care requires a standardized and comprehensive assessment tool but implementation is challenging.

PURPOSE

We have reported on the development of the interRAI instruments in France from the onset to the mandatory use at the national level. We also have identified in the literature and in practices, incentives and barriers for the implementation of this integrated clinical information system in long term care.

RESULTS

Three periods in the interRAI instruments development were identified over the last twenty years. The first one was a research approach about improving quality of long term care. The second one was an experimental clinical use into an integrated care model with case management. The third one was a call for tenders issued by a French national agency, and the choice to use the interRAI-HC (Home Care) for all case managers. The main incentives and barriers that were identified include the national context, the target population, the providers involved and the impact on their practice, the interRAI instrument characteristics, training and leadership.

CONCLUSION

This historical overview of the development of interRAI instruments in France gives health care organizations pertinent information to guide the implementation of a standardized and comprehensive assessment tool.

Attitudes About Cognitive Screening: A Survey of Home Care Physical Therapists

BACKGROUND AND PURPOSE

Home care physical therapists (PTs) are inconsistent regarding cognitive screening, possibly because screening tools were traditionally considered the domain of other home care disciplines, or because therapists perceive their dementia training to be inadequate. A cross sectional study was designed to survey home care therapists' attitudes and beliefs about the management of persons with dementia and to find out whether any specific cognitive tools or measures are currently used.

METHODS

A 5-point Likert-type survey was administered to home care PTs via an online survey. Three state home care associations and individual home care agencies agreed to share the survey link. The survey was also made available to American Physical Therapy Association members through the Home Health and Geriatric Section listservs.

RESULTS AND DISCUSSION

Two hundred fifty-one PTs opened the survey and 233 completed the survey. Respondents included 180 females and 53 males. Seventy-four had a bachelor's degree (BS), 53 held a master's degree (MS), 104 had achieved a doctor of physical therapy (DPT) or doctor of philosophy (PhD) degree, and 2 did not provide this information. Significant differences were found between those with the highest doctoral degrees and those with master's or bachelor's degrees (P = .01) regarding whether they were qualified to screen (strongly agree, agree) for cognitive deficits. Therapists with the highest degrees also attended continuing education for dementia training more than those with less formal education (P = .042.) Gender differences were found in 2 questions regarding positive outcomes (P = .010 and .42); for both questions, males were more likely to believe that dementia has a negative impact. Eighty-seven percent indicated that PTs are qualified (strongly agree, agree), but only 53% said that they possess the necessary skills (strongly agree, agree) to perform cognitive screens. Specialty certification revealed significant differences in several of the questions. No significance was found for any question regarding years of practice or years in home care. The Mini-Mental State Examination and the Clock Drawing Test were most frequently cited among PTs who conduct cognitive screening.

CONCLUSIONS

Physical therapists recognize that they are qualified to perform cognitive screening but may need additional training to utilize cognitive findings to enhance interventions and outcomes in home care. More research is needed to determine which screens are most relevant for therapist use and to examine the effect of cognitive screening on therapy outcomes.

Development and validation of classifiers and variable subsets for predicting nursing home admission

Background

In previous years a substantial number of studies have identified statistically important predictors of nursing home admission (NHA). However, as far as we know, the analyses have been done at the population-level. No prior research has analysed the prediction accuracy of a NHA model for individuals.

Methods

This study is an analysis of 3056 longer-term home care customers in the city of Tampere, Finland. Data were collected from the records of social and health service usage and RAI-HC (Resident Assessment Instrument - Home Care) assessment system during January 2011 and September 2015. The aim was to find out the most efficient variable subsets to predict NHA for individuals and validate the accuracy. The variable subsets of predicting NHA were searched by sequential forward selection (SFS) method, a variable ranking metric and the classifiers of logistic regression (LR), support vector machine (SVM) and Gaussian naive Bayes (GNB). The validation of the results was guaranteed using randomly balanced data sets and cross-validation. The primary performance metrics for the classifiers were the prediction accuracy and AUC (average area under the curve).

Results

The LR and GNB classifiers achieved 78% accuracy for predicting NHA. The most important variables were RAI MAPLE (Method for Assigning Priority Levels), functional impairment (RAI IADL, Activities of Daily Living), cognitive impairment (RAI CPS, Cognitive Performance Scale), memory disorders (diagnoses G30-G32 and F00-F03) and the use of community-based health-service and prior hospital use (emergency visits and periods of care).

Conclusion

The accuracy of the classifier for individuals was high enough to convince the officials of the city of Tampere to integrate the predictive model based on the findings of this study as a part of home care information system. Further work need to be done to evaluate variables that are modifiable and responsive to interventions.

Developing a Referral Protocol for Community-Based Occupational Therapy Services in Taiwan: A Logistic Regression Analysis

Because resources for long-term care services are limited, timely and appropriate referral for rehabilitation services is critical for optimizing clients' functions and successfully integrating them into the community. We investigated which client characteristics are most relevant in predicting Taiwan's community-based occupational therapy (OT) service referral based on experts' beliefs. Data were collected in face-to-face interviews using the Multidimensional Assessment Instrument (MDAI). Community-dwelling participants (n = 221) ≥ 18 years old who reported disabilities in the previous National Survey of Long-term Care Needs in Taiwan were enrolled. The standard for referral was the judgment and agreement of two experienced occupational therapists who reviewed the results of the MDAI. Logistic regressions and Generalized Additive Models were used for analysis. Two predictive models were proposed, one using basic activities of daily living (BADLs) and one using instrumental ADLs (IADLs). Dementia, psychiatric disorders, cognitive impairment, joint range-of-motion limitations, fear of falling, behavioral or emotional problems, expressive deficits (in the BADL-based model), and limitations in IADLs or BADLs were significantly correlated with the need for referral. Both models showed high area under the curve (AUC) values on receiver operating curve testing (AUC = 0.977 and 0.972, respectively). The probability of being referred for community OT services was calculated using the referral algorithm. The referral protocol facilitated communication between healthcare professionals to make appropriate decisions for OT referrals. The methods and findings should be useful for developing referral protocols for other long-term care services.