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.

Prevalence and predictors of incident ADRD diagnosis following a Medicare home health episode.

INTRODUCTION: Home health (HH) may be an important source of care for those with early-stage/undiagnosed Alzheimer's Disease and Related Dementias (ADRD), but little is known regarding prevalence or predictors of incident ADRD diagnosis following HH. METHODS: Using 2010-2012 linked Master Beneficiary Summary File (MBSF) and HH assessment data for 40,596 Medicare HH patients, we model incident ADRD diagnosis within 1 year of HH via multivariable logistic regression. RESULTS: Among HH patients without diagnosed ADRD, 10% received an incident diagnosis within 1 year. In adjusted models, patients were three times more likely to receive an incident ADRD diagnosis if they had HH clinician-reported impaired overall cognition (compared to patients without reported impairment) and twice as likely if they were community-referred (compared to hospital-referred patients). DISCUSSION: There is a pressing need to develop tailored HH clinical pathways and protect access to community-referred HH to support community-living older adults with early-stage/undiagnosed ADRD.

Clinical notes: An untapped opportunity for improving risk prediction for hospitalization and emergency department visit during home health care.

BACKGROUND/OBJECTIVE: Between 10 and 25% patients are hospitalized or visit emergency department (ED) during home healthcare (HHC). Given that up to 40% of these negative clinical outcomes are preventable, early and accurate prediction of hospitalization risk can be one strategy to prevent them. In recent years, machine learning-based predictive modeling has become widely used for building risk models. This study aimed to compare the predictive performance of four risk models built with various data sources for hospitalization and ED visits in HHC. METHODS: Four risk models were built using different variables from two data sources: structured data (i.e., Outcome and Assessment Information Set (OASIS) and other assessment items from the electronic health record (EHR)) and unstructured narrative-free text clinical notes for patients who received HHC services from the largest non-profit HHC organization in New York between 2015 and 2017. Then, five machine learning algorithms (logistic regression, Random Forest, Bayesian network, support vector machine (SVM), and Naïve Bayes) were used on each risk model. Risk model performance was evaluated using the F-score and Precision-Recall Curve (PRC) area metrics. RESULTS: During the study period, 8373/86,823 (9.6%) HHC episodes resulted in hospitalization or ED visits. Among five machine learning algorithms on each model, the SVM showed the highest F-score (0.82), while the Random Forest showed the highest PRC area (0.864). Adding information extracted from clinical notes significantly improved the risk prediction ability by up to 16.6% in F-score and 17.8% in PRC. CONCLUSION: All models showed relatively good hospitalization or ED visit risk predictive performance in HHC. Information from clinical notes integrated with the structured data improved the ability to identify patients at risk for these emergent care events.

Detecting Language Associated With Home Healthcare Patient's Risk for Hospitalization and Emergency Department Visit.

BACKGROUND: About one in five patients receiving home healthcare (HHC) services are hospitalized or visit an emergency department (ED) during a home care episode. Early identification of at-risk patients can prevent these negative outcomes. However, risk indicators, including language in clinical notes that indicate a concern about a patient, are often hidden in narrative documentation throughout their HHC episode. OBJECTIVE: The aim of the study was to develop an automated natural language processing (NLP) algorithm to identify concerning language indicative of HHC patients' risk of hospitalizations or ED visits. METHODS: This study used the Omaha System-a standardized nursing terminology that describes problems/signs/symptoms that can occur in the community setting. First, five HHC experts iteratively reviewed the Omaha System and identified concerning concepts indicative of HHC patients' risk of hospitalizations or ED visits. Next, we developed and tested an NLP algorithm to identify these concerning concepts in HHC clinical notes automatically. The resulting NLP algorithm was applied on a large subset of narrative notes (2.3 million notes) documented for 66,317 unique patients ( n = 87,966 HHC episodes) admitted to one large HHC agency in the Northeast United States between 2015 and 2017. RESULTS: A total of 160 Omaha System signs/symptoms were identified as concerning concepts for hospitalizations or ED visits in HHC. These signs/symptoms belong to 31 of the 42 available Omaha System problems. Overall, the NLP algorithm showed good performance in identifying concerning concepts in clinical notes. More than 18% of clinical notes were detected as having at least one concerning concept, and more than 90% of HHC episodes included at least one Omaha System problem. The most frequently documented concerning concepts were pain, followed by issues related to neuromusculoskeletal function, circulation, mental health, and communicable/infectious conditions. CONCLUSION: Our findings suggest that concerning problems or symptoms that could increase the risk of hospitalization or ED visit were frequently documented in narrative clinical notes. NLP can automatically extract information from narrative clinical notes to improve our understanding of care needs in HHC. Next steps are to evaluate which concerning concepts identified in clinical notes predict hospitalization or ED visit.

Surviving COVID-19 After Hospital Discharge: Symptom, Functional, and Adverse Outcomes of Home Health Recipients.

BACKGROUND: Little is known about recovery from coronavirus disease 2019 (COVID-19) after hospital discharge. OBJECTIVE: To describe the home health recovery of patients with COVID-19 and risk factors associated with rehospitalization or death. DESIGN: Retrospective observational cohort. SETTING: New York City. PARTICIPANTS: 1409 patients with COVID-19 admitted to home health care (HHC) between 1 April and 15 June 2020 after hospitalization. MEASUREMENTS: Covariates and outcomes were obtained from the mandated OASIS (Outcome and Assessment Information Set). Cox proportional hazards models were used to estimate the hazard ratio (HR) of risk factors associated with rehospitalization or death. RESULTS: After an average of 32 days in HHC, 94% of patients were discharged and most achieved statistically significant improvements in symptoms and function. Activity-of-daily-living dependencies decreased from an average of 6 (95% CI, 5.9 to 6.1) to 1.2 (CI, 1.1 to 1.3). Risk for rehospitalization or death was higher for male patients (HR, 1.45 [CI, 1.04 to 2.03]); White patients (HR, 1.74 [CI, 1.22 to 2.47]); and patients with heart failure (HR, 2.12 [CI, 1.41 to 3.19]), diabetes with complications (HR, 1.71 [CI, 1.17 to 2.52]), 2 or more emergency department visits in the past 6 months (HR, 1.78 [CI, 1.21 to 2.62]), pain daily or all the time (HR, 1.46 [CI, 1.05 to 2.05]), cognitive impairment (HR, 1.49 [CI, 1.04 to 2.13]), or functional dependencies (HR, 1.09 [CI, 1.00 to 1.20]). Eleven patients (1%) died, 137 (10%) were rehospitalized, and 23 (2%) remain on service. LIMITATIONS: Care was provided by 1 home health agency. Information on rehospitalization and death after HHC discharge is not available. CONCLUSION: Symptom burden and functional dependence were common at the time of HHC admission but improved for most patients. Comorbid conditions of heart failure and diabetes, as well as characteristics present at admission, identified patients at greatest risk for an adverse event. PRIMARY FUNDING SOURCE: No direct funding.

Improving transitions and outcomes of sepsis survivors (I-TRANSFER): a type 1 hybrid protocol.

BACKGROUND: This protocol is based on home health care (HHC) best practice evidence showing the value of coupling timely post-acute care visits by registered nurses and early outpatient provider follow-up for sepsis survivors. We found that 30-day rehospitalization rates were 7 percentage points lower (a 41% relative reduction) when sepsis survivors received a HHC nursing visit within 2 days of hospital discharge, at least 1 more nursing visit the first week, and an outpatient provider follow-up visit within 7 days compared to those without timely follow-up. However, nationwide, only 28% of sepsis survivors who transitioned to HHC received this timely visit protocol. The opportunity exists for many more sepsis survivors to benefit from timely home care and outpatient services. This protocol aims to achieve this goal.  METHODS: Guided by the Consolidated Framework for Implementation Research, this Type 1 hybrid pragmatic study will test the effectiveness of the Improving Transitions and Outcomes of Sepsis Survivors (I-TRANSFER) intervention compared to usual care on 30-day rehospitalization and emergency department use among sepsis survivors receiving HHC. The study design includes a baseline period with no intervention, a six-month start-up period followed by a one-year intervention period in partnership with five dyads of acute and HHC sites. In addition to the usual care/control periods from the dyad sites, additional survivors from national data will serve as control observations for comparison, weighted to produce covariate balance. The hypotheses will be tested using generalized mixed models with covariates guided by the Andersen Behavioral Model of Health Services. We will produce insights and generalizable knowledge regarding the context, processes, strategies, and determinants of I-TRANSFER implementation. DISCUSSION: As the largest HHC study of its kind and the first to transform this novel evidence through implementation science, this study has the potential to produce new knowledge about the impact of timely attention in HHC to alleviate symptoms and support sepsis survivor's recovery at home. If effective, the impact of this intervention could be widespread, improving the quality of life and health outcomes for a growing, vulnerable population of sepsis survivors. A national advisory group will assist with widespread results dissemination.

Effect of Nurse Practitioner Interventions on Hospitalizations in the Community Transitions Intervention Trial.

BACKGROUND: Despite improvements in hypertension treatment in the United States, Black and Hispanic individuals experience poor blood pressure control and have worse hypertension-related outcomes compared to Whites. OBJECTIVE: The aim of the study was to determine the effect on hospitalization of supplementing usual home care (UHC) with two hypertension-focused transitional care interventions-one deploying nurse practitioners (NPs) and the other NPs plus health coaches. METHODS: We examined post hoc the effect of two hypertension-focused NP interventions on hospitalizations in the Community Transitions Intervention trial-a three-arm, randomized controlled trial comparing the effectiveness of (a) UHC with (b) UHC plus a 30-day NP transitional care intervention or (c) UHC plus NP plus 60-day health coach intervention. RESULTS: The study comprised 495 participants: mean age = 66 years; 57% female; 70% Black, non-Hispanic; 30% Hispanic. At the 3- and 12-month follow-up, all three groups showed a significant decrease in the average number of hospitalizations compared to baseline. The interventions were not significantly different from UHC. CONCLUSION: The results of this post hoc analysis show that, during the study period, decreases in hospitalizations in the intervention groups were comparable to those in UHC, and deploying NPs provided no detectable value added. Future research should focus on testing ways to optimize UHC services.

Does Early Follow-Up Improve the Outcomes of Sepsis Survivors Discharged to Home Health Care?

BACKGROUND: There is little evidence to guide the care of over a million sepsis survivors following hospital discharge despite high rates of hospital readmission. OBJECTIVE: We examined whether early home health nursing (first visit within 2 days of hospital discharge and at least 1 additional visit in the first posthospital week) and early physician follow-up (an outpatient visit in the first posthospital week) reduce 30-day readmissions among Medicare sepsis survivors. DESIGN: A pragmatic, comparative effectiveness analysis of Medicare data from 2013 to 2014 using nonlinear instrumental variable analysis. SUBJECTS: Medicare beneficiaries in the 50 states and District of Columbia discharged alive after a sepsis hospitalization and received home health care. MEASURES: The outcomes, protocol parameters, and control variables were from Medicare administrative and claim files and the home health Outcome and Assessment Information Set (OASIS). The primary outcome was 30-day all-cause hospital readmission. RESULTS: Our sample consisted of 170,571 mostly non-Hispanic white (82.3%), female (57.5%), older adults (mean age, 76 y) with severe sepsis (86.9%) and a multitude of comorbid conditions and functional limitations. Among them, 44.7% received only the nursing protocol, 11.0% only the medical doctor protocol, 28.1% both protocols, and 16.2% neither. Although neither protocol by itself had a statistically significant effect on readmission, both together reduced the probability of 30-day all-cause readmission by 7 percentage points (P=0.006; 95% confidence interval=2, 12). CONCLUSIONS: Our findings suggest that, together, early postdischarge care by home health and medical providers can reduce hospital readmissions for sepsis survivors.

Usability Evaluation of a Dashboard for Home Care Nurses.

The introduction of electronic health records has produced many challenges for clinicians. These include integrating technology into clinical workflow and fragmentation of relevant information across systems. Dashboards, which use visualized data to summarize key patient information, have the potential to address these issues. In this article, we outline a usability evaluation of a dashboard designed for home care nurses. An iterative design process was used, which consisted of (1) contextual inquiry (observation and interviews) with two home care nurses; (2) rapid feedback on paper prototypes of the dashboard (10 nurses); and (3) usability evaluation of the final dashboard prototype (20 nurses). Usability methods and assessments included observation of nurses interacting with the dashboard, the system usability scale, and the Questionnaire for User Interaction Satisfaction short form. The dashboard prototype was deemed to have high usability (mean system usability scale, 73.2 [SD, 18.8]) and was positively evaluated by nurse users. It is important to ensure that technology solutions such as the one proposed in this article are designed with clinical users in mind, to meet their information needs. The design elements of the dashboard outlined in this article could be translated to other electronic health records used in home care settings.

Accuracy of electronically reported "meaningful use" clinical quality measures: a cross-sectional study.

BACKGROUND: The federal Electronic Health Record Incentive Program requires electronic reporting of quality from electronic health records, beginning in 2014. Whether electronic reports of quality are accurate is unclear. OBJECTIVE: To measure the accuracy of electronic reporting compared with manual review. DESIGN: Cross-sectional study. SETTING: A federally qualified health center with a commercially available electronic health record. PATIENTS: All adult patients eligible in 2008 for 12 quality measures (using 8 unique denominators) were identified electronically. One hundred fifty patients were randomly sampled per denominator, yielding 1154 unique patients. MEASUREMENTS: Receipt of recommended care, assessed by both electronic reporting and manual review. Sensitivity, specificity, positive and negative predictive values, positive and negative likelihood ratios, and absolute rates of recommended care were measured. RESULTS: Sensitivity of electronic reporting ranged from 46% to 98% per measure. Specificity ranged from 62% to 97%, positive predictive value from 57% to 97%, and negative predictive value from 32% to 99%. Positive likelihood ratios ranged from 2.34 to 24.25 and negative likelihood ratios from 0.02 to 0.61. Differences between electronic reporting and manual review were statistically significant for 3 measures: Electronic reporting underestimated the absolute rate of recommended care for 2 measures (appropriate asthma medication [38% vs. 77%; P < 0.001] and pneumococcal vaccination [27% vs. 48%; P < 0.001]) and overestimated care for 1 measure (cholesterol control in patients with diabetes [57% vs. 37%; P = 0.001]). LIMITATION: This study addresses the accuracy of the measure numerator only. CONCLUSION: Wide measure-by-measure variation in accuracy threatens the validity of electronic reporting. If variation is not addressed, financial incentives intended to reward high quality may not be given to the highest-quality providers. PRIMARY FUNDING SOURCE: Agency for Healthcare Research and Quality.

Using Natural Language Processing to Identify Home Health Care Patients at Risk for Diagnosis of Alzheimer's Disease and Related Dementias.

This study aimed to: (1) validate a natural language processing (NLP) system developed for the home health care setting to identify signs and symptoms of Alzheimer's disease and related dementias (ADRD) documented in clinicians' free-text notes; (2) determine whether signs and symptoms detected via NLP help to identify patients at risk of a new ADRD diagnosis within four years after admission. This study applied NLP to a longitudinal dataset including medical record and Medicare claims data for 56,652 home health care patients and Cox proportional hazard models to the subset of 24,874 patients admitted without an ADRD diagnosis. Selected ADRD signs and symptoms were associated with increased risk of a new ADRD diagnosis during follow-up, including: motor issues; hoarding/cluttering; uncooperative behavior; delusions or hallucinations; mention of ADRD disease names; and caregiver stress. NLP can help to identify patients in need of ADRD-related evaluation and support services.

Electronic health records and ambulatory quality of care.

CONTEXT: The US Federal Government is investing up to $29 billion in incentives for meaningful use of electronic health records (EHRs). However, the effect of EHRs on ambulatory quality is unclear, with several large studies finding no effect. OBJECTIVE: To determine the effect of EHRs on ambulatory quality in a community-based setting. DESIGN: Cross-sectional study, using data from 2008. SETTING: Ambulatory practices in the Hudson Valley of New York, with a median practice size of four physicians. PARTICIPANTS: We included all general internists, pediatricians and family medicine physicians who: were members of the Taconic Independent Practice Association, had patients in a data set of claims aggregated across five health plans, and had at least 30 patients per measure for at least one of nine quality measures selected by the health plans. INTERVENTION: Adoption of an EHR. MAIN OUTCOME MEASURES: We compared physicians using EHRs to physicians using paper on performance for each of the nine quality measures, using t-tests. We also created a composite quality score by standardizing performance against a national benchmark and averaging standardized performance across measures. We used generalized estimation equations, adjusting for nine physician characteristics. KEY RESULTS: We included 466 physicians and 74,618 unique patients. Of the physicians, 204 (44 %) had adopted EHRs and 262 (56 %) were using paper. Electronic health record use was associated with significantly higher quality of care for four of the measures: hemoglobin A1c testing in diabetes, breast cancer screening, chlamydia screening, and colorectal cancer screening. Effect sizes ranged from 3 to 13 percentage points per measure. When all nine measures were combined into a composite, EHR use was associated with higher quality of care (sd 0.4, p = 0.008). CONCLUSIONS: This is one of the first studies to find a positive association between EHRs and ambulatory quality in a community-based setting.

The effects of electronic prescribing by community-based providers on ambulatory medication safety.

BACKGROUND: US federal policies are incentivizing use of electronic prescribing (e-prescribing) to improve safety. However, little is known about e-prescribing's actual impact on medication safety over time. A study was conducted to assess the effect of implementing a commercial electronic health record (EHR) with e-prescribing on rates and types of prescribing errors. Understanding safety effects from e-prescribing will be important as providers increasingly e-prescribe. METHODS: Prescriptions written by 20 community-based primary care providers in the Hudson Valley region of New York from November 2008 to November 2009 were retrospectively studied. All providers adopted a commercial EHR with robust clinical decision support and extensive technical support to aid in prescribing. Errors were identified by standardized prescription and chart review. RESULTS: Some 1,629 prescriptions were analyzed at three months postimplementation, and 1,738 prescriptions were analyzed at one year postimplementation. Use of e-prescribing resulted in relatively low error rates (6.0 errors per 100 prescriptions). These rates were sustained over time but without further improvement (6.0 versus 4.5 errors per 100 prescriptions, p = .15). Antibiotics were the class of medications most frequently involved (12.7% of overall errors), and direction errors were most common (24% of errors). CONCLUSIONS: This study is the first, as far as known, to quantitatively evaluate prescribing errors early after EHR implementation and after sustained use among community-based primary care providers. Relatively low rates of errors with e-prescribing were found early and after prolonged use. Extensive support for providers before, during, and after implementation may mitigate potential safety threats from implementation of an EHR system and result in sustained safety benefits over the long-term.

Covid-19: Home Health Aides' Perceived Preparedness and Self-Reported Availability for Work: Six Month Survey Results.

Throughout the COVID-19 pandemic New York City home health aides continuously provided care, including to patients actively infected or recovering from COVID-19. Analyzing survey data from 1316 aides, we examined factors associated with perceptions of how well their employer prepared them for COVID-19 and their self-reported availability for work (did they "call out" more than usual). Organizational work environment and COVID-19-related supports were predominant predictors of self-reported perceptions of preparedness. Worker characteristics and COVID-19-related stressors were predominant predictors of self-reported availability. Mental distress, satisfaction with employer communications, and satisfaction with supervisor instructions were significantly associated with both outcomes. The study uniquely describes self-reported perceptions of preparedness and availability as two separate worker outcomes potentially modifiable by different interventions. Better public health emergency training and adequate protective equipment may increase aides' perceived preparedness; more household supports could facilitate their availability. More effective employer communications and mental health initiatives could potentially improve both outcomes. Industry collaboration and systemic changes in federal, state, and local policies should enhance intervention impacts.

HomeADScreen: Developing Alzheimer's disease and related dementia risk identification model in home healthcare.

BACKGROUND: More than 50 % of patients with Alzheimer's disease and related dementia (ADRD) remain undiagnosed. This is specifically the case for home healthcare (HHC) patients. OBJECTIVES: This study aimed at developing HomeADScreen, an ADRD risk screening model built on the combination of HHC patients' structured data and information extracted from HHC clinical notes. METHODS: The study's sample included 15,973 HHC patients with no diagnosis of ADRD and 8,901 patients diagnosed with ADRD across four follow-up time windows. First, we applied two natural language processing methods, Word2Vec and topic modeling methods, to extract ADRD risk factors from clinical notes. Next, we built the risk identification model on the combination of the Outcome and Assessment Information Set (OASIS-structured data collected in the HHC setting) and clinical notes-risk factors across the four-time windows. RESULTS: The top-performing machine learning algorithm attained an Area under the Curve = 0.76 for a four-year risk prediction time window. After optimizing the cut-off value for screening patients with ADRD (cut-off-value = 0.31), we achieved sensitivity = 0.75 and an F1-score = 0.63. For the first-year time window, adding clinical note-derived risk factors to OASIS data improved the overall performance of the risk identification model by 60 %. We observed a similar trend of increasing the model's overall performance across other time windows. Variables associated with increased risk of ADRD were "hearing impairment" and "impaired patient ability in the use of telephone." On the other hand, being "non-Hispanic White" and the "absence of impairment with prior daily functioning" were associated with a lower risk of ADRD. CONCLUSION: HomeADScreen has a strong potential to be translated into clinical practice and assist HHC clinicians in assessing patients' cognitive function and referring them for further neurological assessment.

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.

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.

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.

Clinical and Demographic Profiles of Home Care Patients With Alzheimer's Disease and Related Dementias: Implications for Information Transfer Across Care Settings.

Home health care (HHC) clinicians serving individuals with Alzheimer's disease and related dementias (ADRD) do not always have information about the person's ADRD diagnosis, which may be used to improve the HHC plan of care. This retrospective cohort study examined characteristics of 56,652 HHC patients with varied documentation of ADRD diagnoses. Data included clinical assessments and Medicare claims for a 6-month look-back period and 4-year follow-up. Nearly half the sample had an ADRD diagnosis observed in the claims either prior to or following the HHC admission. Among those with a prior diagnosis, 63% did not have it documented on the HHC assessment; the diagnosis may not have been known to the HHC team or incorporated into the care plan. Patients with ADRD had heightened risk for adverse outcomes (e.g., urinary tract infection and aspiration pneumonia). Interoperable data across health care settings should include ADRD-specific elements about diagnoses, symptoms, and risk factors.

Risk of Rehospitalization or Emergency Department Visit is Significantly Higher for Patients who Receive Their First Home Health Care Nursing Visit Later than 2 Days After Hospital Discharge.

OBJECTIVES: This study explored the association between the timing of the first home health care nursing visits (start-of-care visit) and 30-day rehospitalization or emergency department (ED) visits among patients discharged from hospitals. DESIGN: Our cross-sectional study used data from 1 large, urban home health care agency in the northeastern United States. SETTING/PARTICIPANTS: We analyzed data for 49,141 home health care episodes pertaining to 45,390 unique patients who were admitted to the agency following hospital discharge during 2019. METHODS: We conducted multivariate logistic regression analyses to examine the association between start-of-care delays and 30-day hospitalizations and ED visits, adjusting for patients' age, race/ethnicity, gender, insurance type, and clinical and functional status. We defined delays in start-of-care as a first nursing home health care visit that occurred more than 2 full days after the hospital discharge date. RESULTS: During the study period, we identified 16,251 start-of-care delays (34% of home health care episodes), with 14% of episodes resulting in 30-day rehospitalization and ED visits. Delayed episodes had 12% higher odds of rehospitalization or ED visit (OR 1.12; 95% CI: 1.06-1.18) compared with episodes with timely care. CONCLUSIONS AND IMPLICATIONS: The findings suggest that timely start-of-care home health care nursing visit is associated with reduced rehospitalization and ED use among patients discharged from hospitals. With more than 6 million patients who receive home health care services across the United States, there are significant opportunities to improve timely care delivery to patients and improve clinical outcomes.