Home care aides’ observations and machine learning algorithms for the prediction of visits to emergency departments by older community-dwelling individuals receiving home care assistance: A proof of concept study

Abstract

Background Older individuals receiving home assistance are at high risk for emergency visits and unplanned hospitalization. Anticipating their health difficulties could prevent these events. This study investigated the effectiveness of an at-home monitoring method using social workers’ observations to predict risk for 7- and 14-day emergency department (ED) visits. Methods This was a prospective cohort study of persons ≥75 years, living at home and receiving assistance from home care aides (HCA) at 6 French facilities. After each home visit, HCAs reported on participants’ functional status using a smartphone application that recorded 27 functional items about each participant (e.g., ability to stand, move, eat, mood, loneliness). We recorded ED visits. Finally, we used machine learning techniques (i.e., leveraging random forest predictors) to develop a 7- and 14-day predictive algorithm for the risk of ED visit. Results The study included 301 participants, and the HCA made 9,987 observations. Over the mean 10-month follow-up, 97 participants (32%) had at least one ED visit. Modeling techniques identified 9 contributory factors from the longitudinal records of the HCA and developed a predictive algorithm for the risk of ED visit. The predictive performance (i.e., the area under the ROC curve) was 0.70 at 7 days and 0.67 at 14 days. Interpretation For frail elders receiving in-home care, information on functional status collected by HCA helps predict the risk of ED visits 7 to 14 days in advance. A survey system for real-time identification of risks could be developed using this exploratory work.

Home care workers caring for adults with heart failure need better access to training and technology: A role for implementation science

“…Third, working conditions can be poor and isolating [16,17]. Despite this, studies have shown that HCWs generally like their jobs, find them meaningful, and can have a positive impact on patient outcomes, when utilized at their skill level [18,19]. Our findings also demonstrate how the principles of implementation science could be used to leverage technology and advance the role of HCWs in the context of HF care.…”

Prevention of AcuTe admIssioN algorithm (PATINA): study protocol of a stepped wedge randomized controlled trial

“…Even though it is still a novel area of research, prediction models developed to predict the risk of acute hospitalization among older community-dwelling citizens have been found to be effective in achieving predictive performance [ 18 – 20 ]. These models have primarily been developed based on health care data from electronic hospital records.…”

“…Most models are based on data from electronic hospital records and many have been found effective in predicting risk of acute admissions [ 13 – 19 ]. Only a few studies have focused on prediction models solely based on home care data [ 20 ], and very few studies have implemented and tested a prediction model in practice, mainly due to ethical and economic considerations [ 21 ], barriers among professionals, such as trust in the technology [ 22 ], and the fact that data on health and care is registered primarily for the use and support for health professionals, not for input to algorithms [ 23 ]. Thus, datasets are more often designed for retrospective analysis than for prospective use [ 24 ].…”

Real-world Implementation of an eHealth System Based on Artificial Intelligence Designed to Predict and Reduce Emergency Department Visits by Older Adults: Pragmatic Trial

“…Patient (or family)-reported outcome measure (PROM) systems benefit patients with chronic diseases by improving quality of life, reducing mortality, reducing ED visits, and hospitalizations [12][13][14]. In 2019, we conducted an observational cohort study, enrolling 301 older individuals who received regular visits by home aides (HAs); we developed a machine learning algorithm to predict the risk of emergency visits, with a prediction window of 7-14 days and a predictive performance (ie, the area under the receiver operating characteristic curve) of 0.70 after 7 days and 0.67 after 14 days [15]. This algorithm opens the possibility of mobilizing health professionals to intervene early in an acute illness or in the decompensation of a chronic illness before they lead to an ED visit and unplanned hospitalization.…”