Correction: Machine Learning Model for Predicting Mortality Risk in Patients With Complex Chronic Conditions: Retrospective Analysis.

[This corrects the article DOI: 10.2196/52782.].

Developing new hospital-at-home models based on Comprehensive Geriatric Assessment: Implementation recommendations by the Working Group on Hospital-at-Home and Community Geriatrics of the Catalan Society of Geriatrics and Gerontology.

Last decade, the Government of Catalonia have urged an integrated care strategy for planning the care model to older populations living with frailty, multimorbidity and advanced illnesses. Based on international evidence that was reviewed by a group of experts from the Catalan Society of Gerontology and Geriatrics, we summarised some recommendation to adapt hospital-at-home care to older populations in our system. We defined Comprehensive Geriatric Assessment (CGA) hospital-at-home (HaH) as a specialised home hospitalisation service formed by interdisciplinary teams, characterised by using the clinical methodology of CGA, and by adapting geriatric units' protocols for the provision of person-centred care at home. Main benefits of CGA-HaH in these populations are: response to heath crises according to individualised care plans based on the situational diagnosis carried out by Primary Care teams; provision of a comprehensive health and social approach tailored to the complexity of cases and situations; and adaptation of multipurpose hospitalisation, by working on different person-centred care, aspects, such as caregivers support on care provision, focusing on function or home adaptation.

Machine Learning Model for Predicting Mortality Risk in Patients With Complex Chronic Conditions: Retrospective Analysis.

Background: The health care system is undergoing a shift toward a more patient-centered approach for individuals with chronic and complex conditions, which presents a series of challenges, such as predicting hospital needs and optimizing resources. At the same time, the exponential increase in health data availability has made it possible to apply advanced statistics and artificial intelligence techniques to develop decision-support systems and improve resource planning, diagnosis, and patient screening. These methods are key to automating the analysis of large volumes of medical data and reducing professional workloads. Objective: This article aims to present a machine learning model and a case study in a cohort of patients with highly complex conditions. The object was to predict mortality within the following 4 years and early mortality over 6 months following diagnosis. The method used easily accessible variables and health care resource utilization information. Methods: A classification algorithm was selected among 6 models implemented and evaluated using a stratified cross-validation strategy with k=10 and a 70/30 train-test split. The evaluation metrics used included accuracy, recall, precision, F1-score, and area under the receiver operating characteristic (AUROC) curve. Results: The model predicted patient death with an 87% accuracy, recall of 87%, precision of 82%, F1-score of 84%, and area under the curve (AUC) of 0.88 using the best model, the Extreme Gradient Boosting (XGBoost) classifier. The results were worse when predicting premature deaths (following 6 months) with an 83% accuracy (recall=55%, precision=64% F1-score=57%, and AUC=0.88) using the Gradient Boosting (GRBoost) classifier. Conclusions: This study showcases encouraging outcomes in forecasting mortality among patients with intricate and persistent health conditions. The employed variables are conveniently accessible, and the incorporation of health care resource utilization information of the patient, which has not been employed by current state-of-the-art approaches, displays promising predictive power. The proposed prediction model is designed to efficiently identify cases that need customized care and proactively anticipate the demand for critical resources by health care providers.

Drug exposure may have a substantial influence on COVID-19 prognosis among residents of long-term care facilities: an exploratory analysis.

OBJECTIVES: To explore the association between drug exposure and SARS-CoV-2 prognosis among elderly people living in long-term care facilities (LTC) DESIGN: We carried out a cross-sectional study among old people living in LTC that had a proven SARS-CoV-2 infection, including socio-demographic data, comorbidities and drug intake at the moment of the diagnosis. The study was focused on ACE2 inhibitors, ARA-II blockers, inhaled bronchodilators, oral corticoids, platelet antiaggregants, oral anti-coagulants, statins and Vitamin D. RESULTS: 1 306 individuals were included, with a mean age of 86.7 years, and 72.3% were females. The case fatality rate was 24.4%. Among the studied exposures platelet antiaggregants were the most prevalent (24.7%). After adjusting for propensity score, the intake of inhaled corticoids (OR 0.73; p=0.03) and statins (OR 0.65; p=0.03) were found to be protective factors of death, whereas ACE2 inhibitor showed an almost significant association (OR 0.73, p=0.07). CONCLUSIONS: Considering the high prevalence of drug intake among elderly people, drug exposure may be an important Covid-19 disease modifier in LTC residents and should be considered when exploring prognostic risk factors associated to Covid-19.

A retrospective cohort study of risk factors for mortality among nursing homes exposed to COVID-19 in Spain.

Long-term care (LTC) facilities have shown remarkably high mortality rates during the coronavirus disease 2019 (COVID-19) outbreak in many countries1, and different risk factors for mortality have been identified in this setting2-5. Using facilities as the unit of analysis, we investigated multiple variables covering facility characteristics and socioeconomic characteristics of the geographic location to identify risk factors for excess mortality from a comprehensive perspective. Furthermore, we used a clustering approach to detect patterns in datasets and generate hypotheses regarding potential relationships between types of nursing homes and mortality trends. Our retrospective analysis included 167 nursing homes providing LTC to 8,716 residents during the COVID-19 outbreak in Catalonia (northeast Spain). According to multiple regression analysis, COVID-19-related and overall mortality at the facility level were significantly associated with a higher percentage of patients with complex diseases, lower scores on pandemic preparedness measures and higher population incidence of COVID-19 in the surrounding population. When grouping nursing homes into eight clusters based on common features, we found higher mortality rates in four clusters, mainly characterized by a higher proportion of residents with complex chronic conditions or advanced diseases, lower scores on pandemic preparedness, being located in rural areas and larger capacity, respectively.

[Hospital-at-home Integrated Care Programme tailored to older patients with disabling acute processes: identification of prognostic factors]. / Hospitalización Domiciliaria Integral para la atención a pacientes mayores con procesos agudos discapacitantes: factores predictivos de éxito terapéutico.

INTRODUCTION: Several authors have demonstrated the efficacy of different hospital-at-home strategies in older patients. The identification of prognostic factors is key for improving the targeting process of candidates. METHODS: We performed an analysis of a cohort of older patients attended due to disabling health crises (medical, orthopaedics, or stroke) by a hospital-at-home scheme developed in an integrated care institution over a 5-year period. Main outcomes were: health crisis resolution (discharge to Primary Care); functional resolution (relative functional gain ≥35%), and their combined variable. A logistic regression analysis was performed, including clinical variables from Comprehensive Geriatric Assessment at admission to detect factors related to favourable outcomes. RESULTS: A total of 484 patients were included. The main characteristics were: age 84.4 (6.7), female gender 69%, baseline Barthel score 74.2 (22.6), family-private caregiver/nursing home 82%/18%, referral from hospital wards/emergency department-community in 55%/45%. The main results (for selected processes medical/orthopaedics/stroke) were: health crisis resolution 71.7/87.5/77.6%; functional resolution 72.1/84.9/73.5%; favourable crisis resolution (health crisis resolution with functional resolution) 67.1/81.6/67.3%. Favourable crisis resolution was associated with [OR (95%CI)]: orthopaedic as main diagnosis [2.00 (1.22-3.29)], Barthel score at admission higher than 40 points [2.00 (1.18-3.38)], and the absence of pressure ulcers at admission [2.80 (1.68-4.65)]. CONCLUSIONS: Patients presenting with an orthopaedic diagnosis, not having severe disability at admission, and not having pressure ulcers at admission could obtain better results on favourable crisis resolution. Suffering cognitive impairment or delirium, or being institutionalised, was not found related with less favourable results.

Does the presence of pressure ulcers entail "high risk of negative outcomes" in geriatric rehabilitation? Results from a retrospective cohort study.

AIM: A retrospective cohort study was performed in order to evaluate the prevalence of pressure ulcers (PrUs) in older patients admitted to a geriatric rehabilitation unit of a postacute care hospital and to investigate the impact of the presence of PrUs on clinical outcomes of the rehabilitation process. METHODS: We studied 668 post-acute patients consecutively attended, from January 2010 to December 2011. The effect of having PrUs at admission was evaluated based on its impact on outcomes: final destination, functional status, mortality and length of stay in the rehabilitation unit. RESULTS: PrUs prevalence at admission was 16%. Patients with PrUs were older, more disabled and had more complex conditions, including malnutrition and cognitive impairment. In the bivariate analysis, we found patients with PrUs at admission had worst final outcome (%): discharge home (69.2 vs.82.5), discharge long term care setting (14 vs.6.4), discharge acute care (8.4 vs.6.2) and death (8.4 vs.4.8); p < 0.001, and worst Barthel Index score at discharge 57 (SD 34.1) vs.83 (SD 33.6); p < 0.001, with longer length of stay in the unit 61 (SD 42.3) vs.53 (SD 37.1); p 0.004. In the multivariate analysis, PrUs presence was found as one of the variables with significant association to no return to home. Finally, a negative association between PrUs at admission and functional gain at discharge of the postacute unit was identified. CONCLUSIONS: PrUs were prevalent and had negative impact on clinical outcomes of our geriatric unit, as discharge destination, functional gain and Length of Stay, in vulnerable patients.

Hospital-at-home integrated care programme for older patients with orthopaedic conditions: Early community reintegration maximising physical function.

OBJECTIVE: Older citizens with orthopaedic conditions need specialised care for the facilitation of early community reintegration and restitution of physical function. We introduced a new community care programme as an alternative to usual hospital rehabilitation for orthopaedic patients. STUDY DESIGN: This was an observational study of a cohort of older orthopaedic patients attending a hospital-at-home integrated care programme (HHU), compared with a contemporary cohort of users of a geriatric rehabilitation unit (GRU) in the urban area of Badalona, Catalonia, Spain. MAIN OUTCOMES MEASURES: Functional gain at discharge was measured using the Barthel Index (BI). Other outcomes were: length of intervention (days), rehabilitation efficiency and discharge destination. RESULTS: Over the 2 years of the study we assessed 270 patients (69 at HHU; 201 at GRU). We found no significant differences in baseline characteristics between HHU and GRU groups-mean (IQR) or % age 83 (79-87) vs. 84 (79-88), cognitive impairment 27.5% vs. 24.9%, functional decline 40 (31-48) vs. 43 (32-58). Overall, we found no statistically significant differences between HHU and GRU groups on functional gain: 35 (22-45) vs. 32 (18-46), and discharge home 85.5% vs. 86.1%. Length of intervention was shorter in the HHU group, 43 (32-56) vs. 57 (44-81); p<0.01, for hip fracture patients. In a multivariate analysis, the adjusted mean difference in rehabilitation efficiency between HHU and GRU groups in the hip fracture subgroup was 0.27 (0.09 to 0.46); p=0.004. CONCLUSIONS: This hospital-at-home service obtained similar clinical results to the usual hospital-based rehabilitation care, and for hip fracture patients attending that service, rehabilitation efficiency was better.

A critical review of Early Supported Discharge for stroke patients: from evidence to implementation into practice.

After an acute stroke, a multidimensional approach based on multidisciplinary work and rehabilitation is required in order to promote functional independence and social reinsertion and to maintain medical stability. These activities are usually developed in the hospital setting as a continuum of the acute phase, but hospitalization is resource consuming and resources are limited. Early Support Discharge strategies base postacute care and rehabilitation at home after an early discharge planning and represent possible alternatives to conventional hospitalization. Recent evidence suggests that Early Supported Discharge might be superior to hospitalization from both the clinical-functional and the economic viewpoints. Moreover, home-based rehabilitation might potentiate important determinants of effectiveness, such as patient's motivation and goal-directed rehabilitation. However, hitherto produced evidence and recommendations show a number of limitations related to the organization models, the inclusion/exclusion criteria, and the questionable applicability of results to any healthcare setting worldwide. In this article, we critically review different methodological and organizational aspects of the available studies. For example in the definition of the target population, based mainly on residual disability and medical stability, we suggest that other relevant aspects, such as premorbid functional status, cognitive function, and previous institutionalization, should be better defined. Focusing on the outcomes, we suggest that, besides strong outcomes such as global functioning, surrogate outcomes, such as physical function, could help to refine the specific interventions. Finally, considering that the majority of studies were conducted in northern Europe, further studies are needed to test the implementation of Early Supported Discharge in different regions.

[Hospital-at-home in older patients: a scoping review on opportunities of developing comprehensive geriatric assessment based services]. / Hospitalización domiciliaria en el paciente anciano: revisión de la evidencia y oportunidades de la geriatría.

This scoping review focused on the opportunity of developing new hospital-at-home schemes in our health systems adapted to older patients with complex conditions due to acute illness. A review was conducted on articles including, randomized controlled trials, systematic reviews and meta-analysis in PubMed and Cochrane Library, from January 1990 to July 2013. Search terms were: hospital-at-home, Early Supported Discharge, hospital in the home and home hospitalization. An analysis was performed to include: the intervention model (admission avoidance or early discharge), age, diagnosis, main inclusion criteria and intervention characteristics (disciplines involved, duration of intervention, main outcomes and objectives). It is concluded that there are several models of hospital-at-home care, with favorable clinical outcomes. The majority of teams in our country focused on acute health care in the less elderly with chronic diseases. Other schemes based on comprehensive geriatric assessment and interdisciplinary teams specialized in complex interventions are also highlighted. The development of comprehensive geriatric assessment based hospital-at-home care by teams led by geriatricians is an opportunity to develop alternatives to conventional hospitalization interventions tailored to older patients.