Frail Elderly Syndrome, Older
Conditions
Keywords
screening tool, prevention, artificial intelligence, geriatric syndromes
Brief summary
Introduction: Due to the worldwide ageing population and the increasing prevalence of chronic illnesses and cognitive deficits, appropriate instruments for frailty early screening and assessment of geriatric syndromes are required. Digital tools based on artificial intelligence can help improve the evaluation and screening of health risks in the older adults. Objectives: This study aims to evaluate the effectiveness of the PRESAGE CARE questionnaire in screening geriatric syndromes among older adults and to assess its correlation with standardized test scores of the comprehensive geriatric assessment (CGA). Methods: A monocentric prospective observational study will be conducted between October 2023 and July 2024, on seniors who are in the day hospitalization service of Charles-Foix hospital in France. During the consultation with older adults, the investigators will ask the patient's home caregiver to use the PRESAGE CARE application to fill in the PRESAGE CARE questionnaire. Once the questionnaire will be completed on the application, the PRESAGE platform will automatically recorde specific risks for each senior. Subsequently, the investigators will retrieve the data from standardized tests conduct on the same day and enter the results for each patient into the platform. The primary outcome will be the correlation between standardized test scores and the risks computed by the Presage Care system for the risk of depression, the risk of undernutrition, the risk of cognitive impairment and the risk of loss of autonomy in activities of daily living. Secondary outcome will be the comparison of the Area Under the Median Curve (AUC), and the sensitivity, specificity and positive predictive values of artificial intelligence (AI)-based models on the Présage Care questionnaire.
Detailed description
During the consultation and the assessment of the older adult by the physicians and the nurses at the day hospitalization service of the hospital, patient's relative fill out PRESAGE CARE questionnaire. Then, the investigators retrieved the data from the standardized tests completed the same day by the health professionals and filled in the test results for each patient on the platform.
Interventions
DEVICEPresage care
A screening questionnaire completed by the family caregiver focused on functional and clinical autonomy of the patient (ie, activities of daily life), possible medical symptoms (eg, fatigue, falls, and pain), changes in behavior (eg, recognition and aggressiveness), and communication with relative or their surroundings.
Sponsors
Assistance Publique - Hôpitaux de Paris
Presage
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
SCREENING
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
65 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Inclusion criteria : 1. family members of individuals aged 65 and above, 2. who are present within the day hospitalization service of Charles-Foix Hospital 3. and able to provide informed consent. * Non inclusion criteria 1. refuse to give their consent to participate in the study, 2. or have cognitive impairments 3. or are occasional caregivers.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Correlationbetween standardized test scores and the corresponding risks identified by PRESAGE questionnaire. | 6 months | Kendall's correlation coefficient (Tau test) will be used to analyze the correlation between the risks calculated by the Présage Care algorithm and the results of the associated standardized scale. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| predictive performance of Presage Care questionnaire for health risk detection | 6 months | For secondary outcoumes, the Chi2 test will be used to compare the high-risk rate and the diagnostic rate for each risk. Presage Care risks will be represented as binary Yes/No variables (at risk/no risk) or as ordinal variables (different levels of risk). Scores on the standardized MNA, GDS15 and MMSE tests will be grouped into binary variables according to the threshold for each test described in the literature, with 0 = normal score. For the IADL and ADL scales, the threshold defined represents total autonomy (independence in each of the domains scored). Predictive performance will be assessed by the median area under the curve, sensitivity, specificity, positive predictive value, negative predictive value, Youden index, for each Presage risk with the associated scale score. Data analysis will be performed using R studio (version 2022.07.2+576) and Python software. |
Contacts
Primary ContactJacques-Henry Veyron, MS
Backup ContactCharlotte THERY, PhD
Outcome results
None listed