Feasibility Study for Improving the Relevance of Diagnostic Proposals for an Artificial Intelligence Software in the Elderly Population. (Intel@Med-Fais)

The ageing of the population is accompanied by the problem of chronic pathologies and sometimes heavy dependence, requiring admission to Nursing Homes (NHs). Approximately 660,000 people currently live in NHs in France. One out of 3 NHs does not have a coordinating doctor, even though the law requires it, and access to care in these NHs is very unequal nationally and especially in the Limousin and Dordogne regions. Some Hospices may find themselves in a situation where there is no coordinating doctor and difficult access to General Practitioners (GPs) visiting a large area.

This inequality of access to care results in a difference in care that can go as far as a loss of opportunity for residents who are immediately transferred to the emergency department (ED) with a risk of iatrogeny or delirium once in the ED or a risk of inappropriate hospitalization.

Residents are hospitalized:

• when the latter could have been avoided because the health care team, not knowing what attitude to adopt, prioritized hospitalization
• Late because the resident waited for the attending physician to come, which resulted in a worsening of symptoms.

The arrival of Artificial Intelligence (AI) is an opportunity to find new models of care organization that can mitigate medical desertification but also develop advanced practices in gerontology. For example, nurses will be able to intervene at a first level for early detection, better triage and early management of certain pathologies.

The "MEDVIR society" AI, developed by a French company, is a medical decision support system with Artificial Intelligence and offers pre-diagnosis based on the information collected (medical and surgical history, concomitant treatments and symptoms). MEDVIR is a diagnostic aid tool and does not replace the doctor who remains at the end of the chain, the final decision-maker.

Before research is conducted to integrate this technology into routine care, it is important to validate the diagnostic relevance of AI in the elderly, as it has been validated in the general population.

This pilot feasibility study will then enable us to methodologically dimension a future project to evaluate the efficiency of this new care system in the management of elderly patients in medical deserts in France.

Study Overview

• Status: Completed
• Conditions: Artificial Intelligence; Geriatrics; Diagnostic Proposals Elderly
• Intervention / Treatment: Diagnostic test: intel@med-feasibility

• Study Type: Observational
• Enrollment (Actual): 18

Contacts and Locations

Study Locations

• France
  • Isle, France, 87170
    • Ehpad Des Bayles
  • Limoges, France, 87000
    • Ehpad Le Roussillon

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 65 years and older (Older Adult)
• Accepts Healthy Volunteers: N/A
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Resident in nursing home presenting a health problem that requires the call of his/her attending physician.

Description

Inclusion Criteria:

• Patient aged 65 or over
• Patient living in one of the two NHs tests
• Patient with a functional complaint or abnormal symptoms involving the call of a physician
• Patient or his legal representative who has not expressed his opposition to the collection of his medical and personal data
• Patient affiliated to social security

Exclusion Criteria:

• End-of-life patient
• Patient with a clear vital emergency according to the physician
• Chronic aphasic patient

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

AI

Diagnostic test: intel@med-feasibility; Initial evaluation by the Nurse using the AI tool which enters into MEDVIR the patient's symptoms or functional complaints and comorbidities and is complemented by a telemedicine solution for data transmission to the remote tele-expert physician located in a regulatory center or health center. The remote doctor (a geriatrician from "Prevention Care for Elderly Unit (UPSAV) platform" of the Clinical Gerontology Division of the Limoges University Hospital) analyses the data collected by the Nurse and establishes a symptom severity criterion and a diagnosis with the help of the AI technology. Both the geriatrician and the NHs nurse are not aware of the proposals made by AI. The geriatrician can nevertheless initiate a visit to the resident's place of residence in order to verify the information necessary to establish the diagnosis and thus improve the relevance of the algorithm. For the resident, this study doesn't interfere with the usual care.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
AI diagnostic proposals	Number of AI diagnostic proposals in adequacy with the medical diagnosis in the month of the study	1 month

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
severity diagnoses	Number of severity diagnoses proposed by the AI solution versus medical diagnosis of the remote geriatrician over the month of the study	1 month
satisfaction survey	Analysis of the satisfaction survey filled in by the users (NHs nurses, participants, NHs Directors and geriatricians)	1 month

Collaborators and Investigators

• Sponsor: University Hospital, Limoges

Study record dates

Study Major Dates

• Study Start (Actual): December 21, 2019
• Primary Completion (Actual): September 22, 2020
• Study Completion (Actual): September 22, 2020

Study Registration Dates

• First Submitted: January 23, 2020
• First Submitted That Met QC Criteria: January 23, 2020
• First Posted (Actual): January 27, 2020

Study Record Updates

• Last Update Posted (Actual): June 21, 2022
• Last Update Submitted That Met QC Criteria: June 14, 2022
• Last Verified: October 1, 2021

More Information

Terms related to this study

• Keywords: Artificial Intelligence; Geriatrics; diagnostic proposals elderly
• Other Study ID Numbers: 87RI19_0026 (Intel@Med-Faisa)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No