Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial
BackgroundThe proportion of very old adults in the population is increasing, representing a significant challenge. Due to their vulnerability, there is a higher frequency of unplanned hospitalizations in this population, leading to adverse events. Digital tools based on artif...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2024-12-01
|
| Series: | Journal of Medical Internet Research |
| Online Access: | https://www.jmir.org/2024/1/e55460 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850066427366604800 |
|---|---|
| author | Jacques-Henri Veyron François Deparis Marie Noel Al Zayat Joël Belmin Charlotte Havreng-Théry |
| author_facet | Jacques-Henri Veyron François Deparis Marie Noel Al Zayat Joël Belmin Charlotte Havreng-Théry |
| author_sort | Jacques-Henri Veyron |
| collection | DOAJ |
| description |
BackgroundThe proportion of very old adults in the population is increasing, representing a significant challenge. Due to their vulnerability, there is a higher frequency of unplanned hospitalizations in this population, leading to adverse events. Digital tools based on artificial intelligence (AI) can help to identify early signs of vulnerability and unfavorable health events and can contribute to earlier and optimized management.
ObjectiveThis study aims to report the implementation in assisted living facilities of an innovative monitoring system (Presage Care) for predicting the short-term risk of emergency hospitalization. We describe its use and assess its performance.
MethodsAn uncontrolled multicenter intervention study was conducted between March and August 2022 in 7 assisted living facilities in France that house very old and vulnerable adults. The monitoring system was set up to provide alerts in cases of a high risk of emergency hospitalization. Nurse assistants (NAs) at the assisted living facilities used a smartphone app to complete a questionnaire on the functional status of the patients, comprising electronic patient-reported outcome measures (ePROMs); these were analyzed in real time by a previously designed machine learning algorithm. This remote monitoring of patients using ePROMs allowed notification of a coordinating nurse or a coordinating NA who subsequently informed the patient’s nurses or physician. The primary outcomes were the acceptability and feasibility of the monitoring system in the context and confirmation of the effectiveness and efficiency of AI in risk prevention and detection in practical, real-life scenarios. The secondary outcome was the hospitalization rate after alert-triggered interventions.
ResultsIn this study, 118 of 194 (61%) eligible patients were included who had at least 1 follow-up visit. A total of 38 emergency hospitalizations were documented. The system generated 92 alerts for 47 of the 118 (40%) patients. Of these 92 alerts, 46 (50%) led to 46 health care interventions for 14 of the 118 (12%) patients and resulted in 4 hospitalizations. The other 46 of the 92 (50%) alerts did not trigger a health care intervention and resulted in 25 hospitalizations (P<.001). Almost all hospitalizations were associated with a lack of alert-triggered interventions (P<.001). System performance to predict hospitalization had a high specificity (96%) and negative predictive value (99.4%).
ConclusionsThe Presage Care system has been implemented with success in assisted living facilities. It was well accepted by coordinating nurses and performed well in predicting emergency hospitalizations. However, its use by NAs was less than expected. Overall, the system performed well in terms of performance and clinical impact in this setting. Nevertheless, further work is needed to improve the moderate use rate by NAs.
Trial RegistrationClinicalTrials.gov NCT05221697; https://clinicaltrials.gov/study/NCT05221697 |
| format | Article |
| id | doaj-art-bf85e5d4116844fbbe2f4e2fba1d9d4c |
| institution | DOAJ |
| issn | 1438-8871 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | Journal of Medical Internet Research |
| spelling | doaj-art-bf85e5d4116844fbbe2f4e2fba1d9d4c2025-08-20T02:48:45ZengJMIR PublicationsJournal of Medical Internet Research1438-88712024-12-0126e5546010.2196/55460Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic TrialJacques-Henri Veyronhttps://orcid.org/0000-0002-3262-266XFrançois Deparishttps://orcid.org/0009-0009-5880-4638Marie Noel Al Zayathttps://orcid.org/0009-0008-5929-1136Joël Belminhttps://orcid.org/0000-0003-1630-9582Charlotte Havreng-Théryhttps://orcid.org/0000-0003-1247-5268 BackgroundThe proportion of very old adults in the population is increasing, representing a significant challenge. Due to their vulnerability, there is a higher frequency of unplanned hospitalizations in this population, leading to adverse events. Digital tools based on artificial intelligence (AI) can help to identify early signs of vulnerability and unfavorable health events and can contribute to earlier and optimized management. ObjectiveThis study aims to report the implementation in assisted living facilities of an innovative monitoring system (Presage Care) for predicting the short-term risk of emergency hospitalization. We describe its use and assess its performance. MethodsAn uncontrolled multicenter intervention study was conducted between March and August 2022 in 7 assisted living facilities in France that house very old and vulnerable adults. The monitoring system was set up to provide alerts in cases of a high risk of emergency hospitalization. Nurse assistants (NAs) at the assisted living facilities used a smartphone app to complete a questionnaire on the functional status of the patients, comprising electronic patient-reported outcome measures (ePROMs); these were analyzed in real time by a previously designed machine learning algorithm. This remote monitoring of patients using ePROMs allowed notification of a coordinating nurse or a coordinating NA who subsequently informed the patient’s nurses or physician. The primary outcomes were the acceptability and feasibility of the monitoring system in the context and confirmation of the effectiveness and efficiency of AI in risk prevention and detection in practical, real-life scenarios. The secondary outcome was the hospitalization rate after alert-triggered interventions. ResultsIn this study, 118 of 194 (61%) eligible patients were included who had at least 1 follow-up visit. A total of 38 emergency hospitalizations were documented. The system generated 92 alerts for 47 of the 118 (40%) patients. Of these 92 alerts, 46 (50%) led to 46 health care interventions for 14 of the 118 (12%) patients and resulted in 4 hospitalizations. The other 46 of the 92 (50%) alerts did not trigger a health care intervention and resulted in 25 hospitalizations (P<.001). Almost all hospitalizations were associated with a lack of alert-triggered interventions (P<.001). System performance to predict hospitalization had a high specificity (96%) and negative predictive value (99.4%). ConclusionsThe Presage Care system has been implemented with success in assisted living facilities. It was well accepted by coordinating nurses and performed well in predicting emergency hospitalizations. However, its use by NAs was less than expected. Overall, the system performed well in terms of performance and clinical impact in this setting. Nevertheless, further work is needed to improve the moderate use rate by NAs. Trial RegistrationClinicalTrials.gov NCT05221697; https://clinicaltrials.gov/study/NCT05221697https://www.jmir.org/2024/1/e55460 |
| spellingShingle | Jacques-Henri Veyron François Deparis Marie Noel Al Zayat Joël Belmin Charlotte Havreng-Théry Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial Journal of Medical Internet Research |
| title | Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial |
| title_full | Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial |
| title_fullStr | Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial |
| title_full_unstemmed | Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial |
| title_short | Postimplementation Evaluation in Assisted Living Facilities of an eHealth Medical Device Developed to Predict and Avoid Unplanned Hospitalizations: Pragmatic Trial |
| title_sort | postimplementation evaluation in assisted living facilities of an ehealth medical device developed to predict and avoid unplanned hospitalizations pragmatic trial |
| url | https://www.jmir.org/2024/1/e55460 |
| work_keys_str_mv | AT jacqueshenriveyron postimplementationevaluationinassistedlivingfacilitiesofanehealthmedicaldevicedevelopedtopredictandavoidunplannedhospitalizationspragmatictrial AT francoisdeparis postimplementationevaluationinassistedlivingfacilitiesofanehealthmedicaldevicedevelopedtopredictandavoidunplannedhospitalizationspragmatictrial AT marienoelalzayat postimplementationevaluationinassistedlivingfacilitiesofanehealthmedicaldevicedevelopedtopredictandavoidunplannedhospitalizationspragmatictrial AT joelbelmin postimplementationevaluationinassistedlivingfacilitiesofanehealthmedicaldevicedevelopedtopredictandavoidunplannedhospitalizationspragmatictrial AT charlottehavrengthery postimplementationevaluationinassistedlivingfacilitiesofanehealthmedicaldevicedevelopedtopredictandavoidunplannedhospitalizationspragmatictrial |