Home Health Care Use and Outcomes After Coronary Artery Bypass Grafting Among Medicare Beneficiaries.

BACKGROUND: Home health care (HHC) has been increasingly used to improve care transitions and avoid poor outcomes, but there is limited data on its use and efficacy following coronary artery bypass grafting. The purpose of this study was to describe HHC use and its association with outcomes among Medicare beneficiaries undergoing coronary artery bypass grafting. METHODS: Retrospective analysis of 100% of Medicare fee-for-service files identified 77 331 beneficiaries undergoing coronary artery bypass grafting and discharged to home between July 2016 and December 2018. The primary exposure of HHC use was defined as the presence of paid HHC claims within 30 days of discharge. Hierarchical logistic regression identified predictors of HHC use and the percentage of variation in HHC use attributed to the hospital. Propensity-matched logistic regression compared mortality, readmissions, emergency department visits, and cardiac rehabilitation enrollment at 30 and 90 days after discharge between HHC users and nonusers. RESULTS: A total of 26 751 (34.6%) of beneficiaries used HHC within 30 days of discharge, which was more common among beneficiaries who were older (72.9 versus 72.5 years), male (79.4% versus 77.4%), White (90.2% versus 89.2%), and not Medicare-Medicaid dual eligible (6.7% versus 8.8%). The median hospital-level rate of HHC use was 31.0% (interquartile range, 13.7%-54.5%) and ranged from 0% to 94.2%. Nearly 30% of the interhospital variation in HHC use was attributed to the discharging hospital (intraclass correlation coefficient, 0.296 [95% CI, 0.275-0.318]). Compared with non-HHC users, those using HHC were less likely to have a readmission or emergency department visit, were more likely to enroll in cardiac rehabilitation, and had modestly higher mortality within 30 or 90 days of discharge. CONCLUSIONS: A third of Medicare beneficiaries undergoing coronary artery bypass grafting used HHC within 30 days of discharge, with wide interhospital variation in use and mixed associations with clinical outcomes and health care utilization.

Characterizing changes to older adults' care transition patterns from hospital to home care in the initial year of COVID-19.

BACKGROUND: Skilled home healthcare (HH) provided in-person care to older adults during the COVID-19 pandemic, yet little is known about the pandemic's impact on HH care transition patterns. We investigated pandemic impact on (1) HH service volume; (2) population characteristics; and (3) care transition patterns for older adults receiving HH services after hospital or skilled nursing facility (SNF) discharge. METHODS: Retrospective, cohort, comparative study of recently hospitalized older adults (≥ 65 years) receiving HH services after hospital or SNF discharge at two large HH agencies in Baltimore and New York City (NYC) 1-year pre- and 1-year post-pandemic onset. We used the Outcome and Assessment Information Set (OASIS) and service use records to examine HH utilization, patient characteristics, visit timeliness, medication issues, and 30-day emergency department (ED) visit and rehospitalization. RESULTS: Across sites, admissions to HH declined by 23% in the pandemic's first year. Compared to the year prior, older adults receiving HH services during the first year of the pandemic were more likely to be younger, have worse mental, respiratory, and functional status in some areas, and be assessed by HH providers as having higher risk of rehospitalization. Thirty-day rehospitalization rates were lower during the first year of the pandemic. COVID-positive HH patients had lower odds of 30-day ED visit or rehospitalization. At the NYC site, extended duration between discharge and first HH visit was associated with reduced 30-day ED visit or rehospitalization. CONCLUSIONS: HH patient characteristics and utilization were distinct in Baltimore versus NYC in the initial year of the COVID-19 pandemic. Study findings suggest some older adults who needed HH may not have received it, since the decrease in HH services occurred as SNF use decreased nationally. Findings demonstrate the importance of understanding HH agency responsiveness during public health emergencies to ensure older adults' access to care.

Application of a Human Factors and Systems Engineering Approach to Explore Care Transitions of Sepsis Survivors From Hospital to Home Health Care.

STUDY AIM: This study aims to describe the transition-in-care work process for sepsis survivors going from hospitals to home health care (HHC) and identify facilitators and barriers to enable practice change and safe care transitions using a human factors and systems engineering approach. BACKGROUND: Despite high readmission risk for sepsis survivors, the transition-in-care work process from hospitals to HHC has not been described. METHODS: We analyzed semi-structured needs assessment interviews with 24 stakeholders involved in transitioning sepsis survivors from two hospitals and one affiliated HHC agency participating in the parent implementation science study, I-TRANSFER. The qualitative data analysis was guided by the Systems Engineering Initiative for Patient Safety (SEIPS) framework to describe the work process and identify work system elements. RESULTS: We identified 31 tasks characterized as decision making, patient education, communication, information, documentation, and scheduling tasks. Technological and organizational facilitators lacked in HHC compared to the hospitals. Person and organization elements in HHC had the most barriers but few facilitators. Additionally, we identified specific task barriers that could hinder sepsis information transfer from hospitals to HHC. CONCLUSION: This study explored the complex transition-in-care work processes for sepsis survivors going from hospitals to HHC. We identified barriers, facilitators, and critical areas for improvement to enable implementation and ensure safe care transitions. A key finding was the sepsis information transfer deficit, highlighting a critical issue for future study. APPLICATION: We recommend using the SEIPS framework to explore complex healthcare work processes before the implementation of evidence-based interventions.

Feasibility, usability, and acceptability of psychoeducational videoconferencing interventions for informal caregivers: A systematic review of randomized controlled trials.

Psychoeducational videoconferencing interventions bypass traditional in-person barriers to attendance and are effective in improving caregiving skills, self-care, and wellness among informal caregivers. Information on their feasibility, usability, and acceptability from the caregivers' perspective is needed to inform future designs and developments. This systematic review follows PRISMA 2020 guidelines to integrate this information. Five databases were systematically searched for relevant randomized control trials published between January 2012 and December 2022. Reference lists were cross-checked for additional studies. Relevant studies were appraised and had their data extracted. This review contains 14 randomized controlled trials. Retention rates ranged from 55.56% to 100%, and major reasons for withdrawing include deteriorating patient health, lack of interest, and technical difficulties (feasibility). Caregivers found the videoconference technology usable, although participants in one intervention experienced poor connectivity and persistent technical issues (usability). Most caregivers were satisfied with videoconferencing interventions, found their content applicable to their situation, and appreciated their structure (acceptability). Those in videoconferencing group interventions were satisfied with small caregiver group sizes (acceptability). Adding respite care to interventions and incorporating short and regular videoconferencing sessions may improve feasibility. Ensuring small group sizes in videoconferencing group interventions and using participatory design may enhance acceptability. Advocacy is needed for employees identifying as informal caregivers to receive employer support and for quality connectivity within underserved areas. This may improve the feasibility and usability of interventions, allowing caregivers to receive the support they need. In future studies, power analyses and recruiting more caregivers may better assess feasibility.

Utilizing patient-nurse verbal communication in building risk identification models: the missing critical data stream in home healthcare.

BACKGROUND: In the United States, over 12 000 home healthcare agencies annually serve 6+ million patients, mostly aged 65+ years with chronic conditions. One in three of these patients end up visiting emergency department (ED) or being hospitalized. Existing risk identification models based on electronic health record (EHR) data have suboptimal performance in detecting these high-risk patients. OBJECTIVES: To measure the added value of integrating audio-recorded home healthcare patient-nurse verbal communication into a risk identification model built on home healthcare EHR data and clinical notes. METHODS: This pilot study was conducted at one of the largest not-for-profit home healthcare agencies in the United States. We audio-recorded 126 patient-nurse encounters for 47 patients, out of which 8 patients experienced ED visits and hospitalization. The risk model was developed and tested iteratively using: (1) structured data from the Outcome and Assessment Information Set, (2) clinical notes, and (3) verbal communication features. We used various natural language processing methods to model the communication between patients and nurses. RESULTS: Using a Support Vector Machine classifier, trained on the most informative features from OASIS, clinical notes, and verbal communication, we achieved an AUC-ROC = 99.68 and an F1-score = 94.12. By integrating verbal communication into the risk models, the F-1 score improved by 26%. The analysis revealed patients at high risk tended to interact more with risk-associated cues, exhibit more "sadness" and "anxiety," and have extended periods of silence during conversation. CONCLUSION: This innovative study underscores the immense value of incorporating patient-nurse verbal communication in enhancing risk prediction models for hospitalizations and ED visits, suggesting the need for an evolved clinical workflow that integrates routine patient-nurse verbal communication recording into the medical record.

The Role and Initiatives Led by the Sepsis Coordinator to Improve Sepsis Bundle Compliance and Care Across the Continuum.

A dedicated sepsis coordinator role at Penn Medicine Lancaster General Hospital led initiatives to improve sepsis core measure compliance by 40% during the course of 4 years with submission of all sepsis cases. Chart abstraction and analysis of noncompliant cases identified areas for improvement: early recognition education, order set revisions, documentation support, and the implementation of a nurse-driven 24/7 sepsis monitoring process. The cooperative work with Penn Medicine affiliates, sharing best practices, improves overall sepsis bundle compliance and transitions of care. Ongoing achievements acknowledge the value of building relationships and leading improvements through the collaborative efforts of interprofessional teams.

Social Risk Factors are Associated with Risk for Hospitalization in Home Health Care: A Natural Language Processing Study.

OBJECTIVE: This study aimed to develop a natural language processing (NLP) system that identified social risk factors in home health care (HHC) clinical notes and to examine the association between social risk factors and hospitalization or an emergency department (ED) visit. DESIGN: Retrospective cohort study. SETTING AND PARTICIPANTS: We used standardized assessments and clinical notes from one HHC agency located in the northeastern United States. This included 86,866 episodes of care for 65,593 unique patients. Patients received HHC services between 2015 and 2017. METHODS: Guided by HHC experts, we created a vocabulary of social risk factors that influence hospitalization or ED visit risk in the HHC setting. We then developed an NLP system to automatically identify social risk factors documented in clinical notes. We used an adjusted logistic regression model to examine the association between the NLP-based social risk factors and hospitalization or an ED visit. RESULTS: On the basis of expert consensus, the following social risk factors emerged: Social Environment, Physical Environment, Education and Literacy, Food Insecurity, Access to Care, and Housing and Economic Circumstances. Our NLP system performed "very good" with an F score of 0.91. Approximately 4% of clinical notes (33% episodes of care) documented a social risk factor. The most frequently documented social risk factors were Physical Environment and Social Environment. Except for Housing and Economic Circumstances, all NLP-based social risk factors were associated with higher odds of hospitalization and ED visits. CONCLUSIONS AND IMPLICATIONS: HHC clinicians assess and document social risk factors associated with hospitalizations and ED visits in their clinical notes. Future studies can explore the social risk factors documented in HHC to improve communication across the health care system and to predict patients at risk for being hospitalized or visiting the ED.

Predicting emergency department visits and hospitalizations for patients with heart failure in home healthcare using a time series risk model.

OBJECTIVES: Little is known about proactive risk assessment concerning emergency department (ED) visits and hospitalizations in patients with heart failure (HF) who receive home healthcare (HHC) services. This study developed a time series risk model for predicting ED visits and hospitalizations in patients with HF using longitudinal electronic health record data. We also explored which data sources yield the best-performing models over various time windows. MATERIALS AND METHODS: We used data collected from 9362 patients from a large HHC agency. We iteratively developed risk models using both structured (eg, standard assessment tools, vital signs, visit characteristics) and unstructured data (eg, clinical notes). Seven specific sets of variables included: (1) the Outcome and Assessment Information Set, (2) vital signs, (3) visit characteristics, (4) rule-based natural language processing-derived variables, (5) term frequency-inverse document frequency variables, (6) Bio-Clinical Bidirectional Encoder Representations from Transformers variables, and (7) topic modeling. Risk models were developed for 18 time windows (1-15, 30, 45, and 60 days) before an ED visit or hospitalization. Risk prediction performances were compared using recall, precision, accuracy, F1, and area under the receiver operating curve (AUC). RESULTS: The best-performing model was built using a combination of all 7 sets of variables and the time window of 4 days before an ED visit or hospitalization (AUC = 0.89 and F1 = 0.69). DISCUSSION AND CONCLUSION: This prediction model suggests that HHC clinicians can identify patients with HF at risk for visiting the ED or hospitalization within 4 days before the event, allowing for earlier targeted interventions.

Home Healthcare Patients With Distinct Psychological, Cognitive, and Behavioral Symptom Profiles and At-Risk Subgroup for Hospitalization and Emergency Department Visits Using Latent Class Analysis.

One-third of home healthcare patients are hospitalized or visit emergency departments during a 60-day episode of care. Among all risk factors, psychological, cognitive, and behavioral symptoms often remain underdiagnosed or undertreated in older adults. Little is known on subgroups of older adults receiving home healthcare services with similar psychological, cognitive, and behavioral symptom profiles and an at-risk subgroup for future hospitalization and emergency department visits. Our cross-sectional study used data from a large, urban home healthcare organization (n = 87,943). Latent class analysis was conducted to identify meaningful subgroups of older adults based on their distinct psychological, cognitive, and behavioral symptom profiles. Adjusted multiple logistic regression was used to understand the association between the latent subgroup and future hospitalization and emergency department visits. Descriptive and inferential statistics were conducted to describe the individual characteristics and to test for significant differences. The three-class model consisted of Class 1: "Moderate psychological symptoms without behavioral issues," Class 2: "Severe psychological symptoms with behavioral issues," and Class 3: "Mild psychological symptoms without behavioral issues." Compared to Class 3, Class 1 patients had 1.14 higher odds and Class 2 patients had 1.26 higher odds of being hospitalized or visiting emergency departments. Significant differences were found in individual characteristics such as age, gender, race/ethnicity, and insurance. Home healthcare clinicians should consider the different latent subgroups of older adults based on their psychological, cognitive, and behavioral symptoms. In addition, they should provide timely assessment and intervention especially to those at-risk for hospitalization and emergency department visits.

Development and Validation of the Hospital-to-Home-Health Transition Quality (H3TQ) Index: A Novel Measure to Engage Patients and Home Health Providers in Evaluating Hospital-to-Home Care Transition Quality.

BACKGROUND: Patients requiring skilled home health care (HH) after hospitalization are at high risk of adverse events. Human factors engineering (HFE) approaches can be useful for measure development to optimize hospital-to-home transitions. OBJECTIVE: To describe the development, initial psychometric validation, and feasibility of the Hospital-to-Home-Health-Transition Quality (H3TQ) Index to identify patient safety risks. METHODS: Development: A multisite, mixed-methods study at 5 HH agencies in rural and urban sites across the United States. Testing: Prospective H3TQ implementation on older adults' hospital-to-HH transitions. Populations Studied: Older adults and caregivers receiving HH services after hospital discharge, and their HH providers (nurses and rehabilitation therapists). RESULTS: The H3TQ is a 12-item count of hospital-to-HH transitions best practices for safety that we developed through more than 180 hours of observations and more than 80 hours of interviews. The H3TQ demonstrated feasibility of use, stability, construct validity, and concurrent validity when tested on 75 transitions. The vast majority (70%) of hospital-to-HH transitions had at least one safety issue, and HH providers identified more patient safety threats than did patients/caregivers. The most frequently identified issues were unsafe home environments (32%), medication issues (29%), incomplete information (27%), and patients' lack of general understanding of care plans (27%). CONCLUSIONS: The H3TQ is a novel measure to assess the quality of hospital-to-HH transitions and proactively identify transitions issues. Patients, caregivers, and HH providers offered valuable perspectives and should be included in safety reporting. Study findings can guide the design of interventions to optimize quality during the high-risk hospital-to-HH transition.

Uncovering hidden trends: identifying time trajectories in risk factors documented in clinical notes and predicting hospitalizations and emergency department visits during home health care.

OBJECTIVE: This study aimed to identify temporal risk factor patterns documented in home health care (HHC) clinical notes and examine their association with hospitalizations or emergency department (ED) visits. MATERIALS AND METHODS: Data for 73 350 episodes of care from one large HHC organization were analyzed using dynamic time warping and hierarchical clustering analysis to identify the temporal patterns of risk factors documented in clinical notes. The Omaha System nursing terminology represented risk factors. First, clinical characteristics were compared between clusters. Next, multivariate logistic regression was used to examine the association between clusters and risk for hospitalizations or ED visits. Omaha System domains corresponding to risk factors were analyzed and described in each cluster. RESULTS: Six temporal clusters emerged, showing different patterns in how risk factors were documented over time. Patients with a steep increase in documented risk factors over time had a 3 times higher likelihood of hospitalization or ED visit than patients with no documented risk factors. Most risk factors belonged to the physiological domain, and only a few were in the environmental domain. DISCUSSION: An analysis of risk factor trajectories reflects a patient's evolving health status during a HHC episode. Using standardized nursing terminology, this study provided new insights into the complex temporal dynamics of HHC, which may lead to improved patient outcomes through better treatment and management plans. CONCLUSION: Incorporating temporal patterns in documented risk factors and their clusters into early warning systems may activate interventions to prevent hospitalizations or ED visits in HHC.

iCare4Me for FTD: A pilot randomized study to improve self-care in caregivers of persons with frontotemporal degeneration.

Introduction: A tremendous burden is placed on frontotemporal degeneration (FTD) caregivers who sacrifice their own self-care to manage the functional impairments of their loved one, contributing to high levels of stress and depression. Health coaching provides support for coping with stress while fostering self-care behaviors. We report on preliminary evidence for efficacy of a virtual health coach intervention aimed at increasing self-care. Methods: Thirty-one caregivers of persons with behavioral variant FTD (bvFTD) were assigned randomly to an intervention group, which included 10 coaching sessions over 6 months plus targeted health information or the control group receiving standard care augmented with the health information. Caregiver self-care (primary outcome), stress, depression, coping, and patient behavioral symptoms were collected at enrollment and 3 and 6 months. Change over time was evaluated between the intervention and control groups using linear mixed-effects models. Results: There was a significant group-by-time interaction for self-care monitoring (t58 = 2.37, p = 0.02 and self-care confidence (t58 = 2.32, p = 0.02) on the Self-Care Inventory, demonstrating that caregivers who received the intervention improved their self-care over time. Behavioral symptoms were reduced in bvFTD patients whose caregivers received the intervention (t54 = -2.15, p = 0.03). Discussion: This randomized controlled trial (RCT) shows promise for health coaching as a way to increase support that is urgently needed to reduce poor outcomes in FTD caregivers.

RWJF Future of Nursing Scholars experience and recommendations: Focus group results at final convening.

BACKGROUND: The Robert Wood Johnson Foundation launched the Future of Nursing Scholars program to support nurses to complete PhDs in 3 years in schools across the United States. PURPOSE: To explore why scholars participated in the program and to articulate challenges and facilitators to successful completion of their doctoral degrees. METHOD: Thirty-one scholars representing 18 different schools participated in focus groups at a convening in January 2022. FINDINGS: Scholars identified that funding and planned length of degree completion were important factors in their choosing the accelerated program. Mentorship, networking, and support were identified as facilitators to program completion with the tight timeline of three years noted as a challenge. DISCUSSION: Accelerated students require adequate resources including access to data, mentoring, and financing to overcome challenges presented by accelerated PhD training programs. Cohort models provide support and clarity of expectations for both students and mentors is critical.

Pandemic-Related Changes in Technology Use Among a Sample of Previously Hospitalized Older Adult New Yorkers: Observational Study.

BACKGROUND: The COVID-19 pandemic increased the importance of technology for all Americans, including older adults. Although a few studies have indicated that older adults might have increased their technology use during the COVID-19 pandemic, further research is needed to confirm these findings, especially among different populations, and using validated surveys. In particular, research on changes in technology use among previously hospitalized community-dwelling older adults, especially those with physical disability, is needed because older adults with multimorbidity and hospital associated deconditioning were a population greatly impacted by COVID-19 and related distancing measures. Obtaining knowledge regarding previously hospitalized older adults' technology use, before and during the pandemic, could inform the appropriateness of technology-based interventions for vulnerable older adults. OBJECTIVE: In this paper, we 1) described changes in older adult technology-based communication, technology-based phone use, and technology-based gaming during the COVID-19 pandemic, compared to before the COVID-19 pandemic and 2) tested whether technology use moderated the association between changes in in-person visits and well-being, controlling for covariates. METHODS: Between December 2020 and January 2021 we conducted a telephone-based objective survey with 60 previously hospitalized older New Yorkers with physical disability. We measured technology-based communication through three questions pulled from the National Health and Aging Trends Study COVID-19 Questionnaire. We measured technology-based smart phone use and technology-based video gaming through the Media Technology Usage and Attitudes Scale. We used paired t tests and interaction models to analyze survey data. RESULTS: This sample of previously hospitalized older adults with physical disability consisted of 60 participants, 63.3% of whom identified as female, 50.0% of whom identified as White, and 63.8% of whom reported an annual income of $25,000 or less. This sample had not had physical contact (such as friendly hug or kiss) for a median of 60 days and had not left their home for a median of 2 days. The majority of older adults from this study reported using the internet, owning smart phones, and nearly half learned a new technology during the pandemic. During the pandemic, this sample of older adults significantly increased their technology-based communication (mean difference=.74, P=.003), smart phone use (mean difference=2.9, P=.016), and technology-based gaming (mean difference=.52, P=.030). However, this technology use during the pandemic did not moderate the association between changes in in-person visits and well-being, controlling for covariates. CONCLUSIONS: These study findings suggest that previously hospitalized older adults with physical disability are open to using or learning technology, but that technology use might not be able to replace in-person social interactions. Future research might explore the specific components of in-person visits that are missing in virtual interactions, and if they could be replicated in the virtual environment, or through other means.

Palliative Care across Settings: Perspectives from Inpatient, Primary Care, and Home Health Care Providers and Staff.

BACKGROUND: Early introduction of palliative care can improve patient-centered outcomes for older adults with complex medical conditions. However, identifying the need for and introducing palliative care with patients and caregivers is often difficult. We aim to identify how and why a multi-setting approach to palliative care discussions may improve the identification of palliative care needs and how to facilitate these conversations. METHODS: Descriptive qualitative study to inform the development and future pilot testing of a model to improve recognition of, and support for, unmet palliative care needs in home health care (HHC). Thematic analysis of semi-structured interviews with providers across inpatient (n = 11), primary care (n = 17), and HHC settings (n = 10). RESULTS: Four key themes emerged: 1) providers across settings can identify palliative care needs using their unique perspectives of the patient's care, 2) identifying palliative care needs is challenging due to infrequent communication and lack of shared information between providers, 3) importance of identifying a clinical lead of patient care who will direct palliative care discussions (primary care provider), and 4) importance of identifying a care coordination lead (HHC) to bridge communication among multi-setting providers. These themes highlight a multi-setting approach that would improve the frequency and quality of palliative care discussions. CONCLUSIONS: A lack of structured communication across settings is a major barrier to introducing and providing palliative care. A novel model that improves communication and coordination of palliative care across HHC, inpatient and primary care providers may facilitate identifying and addressing palliative care needs in medically complex older adults.

Home Health Care Workers' Interactions with Medical Providers, Home Care Agencies, and Family Members for Patients with Heart Failure.

BACKGROUND: Despite providing frequent care to heart failure (HF) patients, home health care workers (HHWs) are generally considered neither part of the health care team nor the family, and their clinical observations are often overlooked. To better understand this workforce's involvement in care, we quantified HHWs' scope of interactions with clinicians, health systems, and family caregivers. METHODS: Community-partnered cross-sectional survey of English- and Spanish-speaking HHWs who cared for a HF patient in the last year. The survey included 6 open-ended questions about aspects of care coordination, alongside demographic and employment characteristics. Descriptive statistics were performed. RESULTS: Three hundred ninety-one HHWs employed by 56 unique home care agencies completed the survey. HHWs took HF patients to a median of 3 doctor appointments in the last year with 21.9% of them taking patients to ≥ 7 doctor appointments. Nearly a quarter of HHWs reported that these appointments were in ≥ 3 different health systems. A third of HHWs organized care for their HF patient with ≥ 2 family caregivers. CONCLUSIONS: HHWs' scope of health-related interactions is large, indicating that there may be novel opportunities to leverage HHWs' experiences to improve health care delivery and patient care in HF.

"Second set of eyes:" Family caregivers and post-acute home health care during the COVID-19 pandemic.

OBJECTIVES: This study aimed to explore how the COVID-19 pandemic shaped the experiences of family caregivers of older adults who were hospitalized with COVID-19 and discharged to post-acute, skilled home health care (HHC) services. METHODS: Thirty semi-structured interviews with family caregivers of older adults who received services from a large, not-for-profit HHC agency following hospitalization with COVID-19 infection were conducted between March-July 2021 and analyzed using thematic analysis. RESULTS: During the pandemic, family caregivers encountered societal and institutional barriers to assisting older adults across post-acute care transitions. These barriers included hospital visitation restrictions as well as difficulties accessing community-based resources and medical equipment. Despite limitations and delays in HHC services, many family caregivers identified post-acute HHC, delivered in-person or via telehealth, as important to addressing care gaps for older adults, as well as their own needs for training and support during the pandemic. CONCLUSIONS: Policies intended to reduce the spread of COVID-19 introduced new challenges for caregivers during HHC. However, HHC agencies and their staff adapted within this context to provide a needed bridge of support.

Machine learning applied to electronic health record data in home healthcare: A scoping review.

OBJECTIVE: Despite recent calls for home healthcare (HHC) to integrate informatics, the application of machine learning in HHC is relatively unknown. Thus, this study aimed to synthesize and appraise the literature describing the application of machine learning to predict adverse outcomes (e.g., hospitalization, mortality) using electronic health record (EHR) data in the HHC setting. Our secondary aim was to evaluate the comprehensiveness of predictors used in the machine learning algorithms guided by the Biopsychosocial Model. METHODS: During March 2022 we conducted a literature search in four databases: PubMed, Embase, CINAHL, and Scopus. Inclusion criteria were 1) describing services provided in the HHC setting, 2) applying machine learning algorithms to predict adverse outcomes, defined as outcomes related to patient deterioration, 3) using EHR data and 4) focusing on the adult population. Predictors were mapped to the Biopsychosocial Model. A risk of bias analysis was conducted using the Prediction Model Risk Of Bias Assessment Tool. RESULTS: The final sample included 20 studies. Eighteen studies used predictors from standardized assessments integrated in the EHR. The most common outcome of interest was hospitalization (55%), followed by mortality (25%). Psychological predictors were frequently excluded (35%). Tree based algorithms were most frequently applied (75%). Most studies demonstrated high or unclear risk of bias (75%). CONCLUSION: Future studies in HHC should consider incorporating machine learning algorithms into clinical decision support systems to identify patients at risk. Based on the Biopsychosocial model, psychological and interpersonal characteristics should be used along with biological characteristics to enhance risk prediction. To facilitate the widespread adoption of machine learning, stakeholders should encourage standardization in the HHC setting.

The identification of clusters of risk factors and their association with hospitalizations or emergency department visits in home health care.

AIMS: To identify clusters of risk factors in home health care and determine if the clusters are associated with hospitalizations or emergency department visits. DESIGN: A retrospective cohort study. METHODS: This study included 61,454 patients pertaining to 79,079 episodes receiving home health care between 2015 and 2017 from one of the largest home health care organizations in the United States. Potential risk factors were extracted from structured data and unstructured clinical notes analysed by natural language processing. A K-means cluster analysis was conducted. Kaplan-Meier analysis was conducted to identify the association between clusters and hospitalizations or emergency department visits during home health care. RESULTS: A total of 11.6% of home health episodes resulted in hospitalizations or emergency department visits. Risk factors formed three clusters. Cluster 1 is characterized by a combination of risk factors related to "impaired physical comfort with pain," defined as situations where patients may experience increased pain. Cluster 2 is characterized by "high comorbidity burden" defined as multiple comorbidities or other risks for hospitalization (e.g., prior falls). Cluster 3 is characterized by "impaired cognitive/psychological and skin integrity" including dementia or skin ulcer. Compared to Cluster 1, the risk of hospitalizations or emergency department visits increased by 1.95 times for Cluster 2 and by 2.12 times for Cluster 3 (all p < .001). CONCLUSION: Risk factors were clustered into three types describing distinct characteristics for hospitalizations or emergency department visits. Different combinations of risk factors affected the likelihood of these negative outcomes. IMPACT: Cluster-based risk prediction models could be integrated into early warning systems to identify patients at risk for hospitalizations or emergency department visits leading to more timely, patient-centred care, ultimately preventing these events. PATIENT OR PUBLIC CONTRIBUTION: There was no involvement of patients in developing the research question, determining the outcome measures, or implementing the study.