ABSTRACT
Aim
To compare the characteristics of participants who discontinued prematurely with those who remained in the ICOPE program (Integrated Care for Older PEople) in France and to compare completely adherent participants with partially adherent ones after 2 years of follow‐up.
Design
Retrospective observational study.
Methods
We analysed the data of older age people participating during 2 years from the ICOPE MONITOR database. The study compared the population that discontinued follow‐up with the population currently in follow‐up. Among the population in follow‐up, a comparison was made between the completely adherent and partially adherent populations.
Results
In total, 8672 participants had a follow‐up period of at least 2 years. After 2 years, three‐quarters of the participants were still in follow‐up with varying levels of adherence. Participants who discontinued follow‐up are older and had more compromised Step 1 levels across all domains of intrinsic capacity (IC). Partially adherent participants were older and generally more compromised in IC than completely adherent participants. Those participants least adherent to ICOPE presented higher declines in clinical parameters.
Conclusion
Among autonomous older age people, the most impaired in intrinsic capacity domains and aged participants were more likely to discontinue follow‐up, highlighting the need to focus efforts on this group. On the other hand, younger robust and healthier older age people represent a good target for ICOPE program, in terms of adherence and primary prevention.
Reported Method
EQUATOR guidelines: STROBE.
Patient of Public Contribution
No patient or public contribution.
Keywords: gerontology, ICOPE program, older people, primary care, quantitative approaches
Summary.
- What problem did the study address?
-
○Participants' adherence as the identification of profiles prematurely discontinuing their participation contributes to program improvement.
-
○
- What were the main findings?
-
○The most impaired participants are the least adherent, requiring closer follow‐up.
-
○
- Where and on whom will the research have an impact?
-
○This research will help all healthcare professionals who want to implement the ICOPE program worldwide.
-
○
1. Introduction
The WHO Integrated Care for Older People (ICOPE) program is designed for individuals aged 60 and over with the aim of promoting healthy ageing by preventing functional decline. This initiative also aims to empower individuals to take an active role in their own health promotion. This primary care preventive program relies on regular monitoring of intrinsic capacity (IC) (Tavassoli et al. 2021) and the early implementation of preventive actions when an impairment in IC is identified. The WHO defines IC as the combination of an individual's physical and mental capacities. Six domains define IC: cognition, psychology, mobility, nutrition, vision and hearing. IC interacts with an individual's environment and illnesses. Monitoring IC helps prevent IC decline through appropriate interventions. The success of this approach needs a strong participation and adherence from the participants (Tavassoli et al. 2022). This adherence to follow‐up can be viewed through the profiles of the participants: age, gender and results from screening and assessment data.
2. Background
2.1. ICOPE Program
The ICOPE program (WHO n.d.) innovates with a pragmatism program, through simple screening tools, accessible digital supports, a straightforward implementation process and the mobilisation of health carers. ICOPE is structured into 5 steps (screening, in‐depth assessment, care and prevention personalised plan, follow‐up and community involvement) and aims to be applicable in all countries (Sum et al. 2022). In France, policymakers closely monitor this large‐scale public health program given the demographic ageing of the French population, the scarcity of healthcare resources and the public health challenge of preventing dependency. These strengths enable a large number of persons to engage in this program, as demonstrated by its implementation in the Occitania region (France) (Tavassoli et al. 2022). The feasibility of ICOPE in real‐life settings has already been demonstrated on a large scale (Tavassoli et al. 2021). With the funding support of the French Ministry of Health and regional health agencies, the ICOPE program can be proposed to any older adults aged 60 and over. Step 1 of the screening process is carried out using the ICOPE MONITOR app, either through self‐assessment by the participant or a caregiver or by a healthcare professional (HCP) (mainly nurses) or another professional (such as a postal worker or social services agent for example). The results from Step 1 are analysed by a trained HCP using decision‐making algorithms and their expertise to determine the clinical relevance of any anomaly in the screening test. If the anomaly is new and/or clinically relevant, the HCP recommends Step 2 for further evaluation. Based on the results of the Step 2 assessment, the HCP offers the participant recommendations (Step 3) discussed with their General Practitioner (GP). These recommendations are made in relation to the clinical assessment, standard guidelines, and the participant's motivations. Step 4 of the program enables monitoring of the implementation of these recommendations and tracking of intrinsic capacity through a Step 1 screening every 6 months. All primary care HCPs involved are previously trained by the (…) Hospital team, and they are mostly nurses.
2.2. Interest in Adherence to the Program
Adherence to intervention is a major determinant of treatment success. Non‐adherence can reduce the benefit to the patient and can lead to substantial worsening of the diseases, functional decline, death and increased healthcare costs. Non‐adherence can also impact the healthcare system and generates doubts about the feasibility of implementing a recommendation. In the ICOPE framework, participants are invited to undergo a screening test (step 1) every 6 months to detect any risk of IC decline as early as possible and to take early action. Adherence to the program involves completing these follow‐ups regularly, either alone or with assistance; each time, the results are interpreted by a healthcare professional who can then propose the next steps in the process at the slightest clinically relevant alert. Acting early in the ageing trajectory helps prevent any decline and the onset of frailty or even dependency (Morley et al. 2013). The challenge is preventing IC impairment and falling into dependency, which is a known risk in terms of health complications, deterioration in quality of life and institutionalisation (Clegg et al. 2013) (Marengoni et al. 2009). In this context, a significant adherence to achieve clinical prevention objectives is expected. In ICOPE program, adherence to follow‐up can be impacted by the ability to use the ICOPE MONITOR application, perceived interest in tracking the evolution of their health condition and especially health status (Tajudeen et al. 2022). The three‐year experimentation allows us to now study the adherence of participants and identify profiles of participants who prematurely discontinue their follow‐up. These participants initially had curiosity about the program, but they likely did not understand the importance of continuing or do not find a response to their needs in the ICOPE program. Better communication about the ICOPE program directed at this population may be necessary. Ensuring that the target population identified in Step 1 adheres to the program over time, regardless of health status, is of practical interest. Understanding the characteristics of participants with impairments in at least one domain of intrinsic capacity but imperfectly adherent to ICOPE is the first step to better implement ICOPE. This would give the opportunity to enhance the program by adjusting the follow‐up procedure for potentially less or non‐adherent individuals.
2.3. The Study
Our research hypothesis is that the health status of participants is a significant determinant of their level of adherence to the ICOPE follow‐up. The specific objectives of this study are (1) to compare the characteristics of participants who prematurely discontinued their follow‐up with those who remained in the program; (2) to compare, within the followed population, the characteristics of participants completely adherent with those of partially adherent participants, defined by suboptimal follow‐up at 2 years.
3. Methods
3.1. Design
The study concerns participants whose data has been recorded in the ‘ICOPE MONITOR’ application between 01 January 2020 and 20 February, 2023, for participants with at least a follow‐up of 2 years ± 4 months since their enrolment in ICOPE program. This is an observational descriptive study. The deployment of ICOPE in France has been previously detailed (Tavassoli et al. 2022) (Tavassoli et al. 2021). To enable the implementation of the ICOPE program in France, the Gerontopole of the University Hospital of Toulousecreated an ICOPE MONITOR platform that includes digital tools (application and chatbot for the completion of Step 1, data management system for data collection, monitoring and management of the pathway by healthcare professionals). A team of trained nurses and doctors remotely manages Step 1 alerts, follows the participants and coordinates the care pathway. The complete ICOPE pathway can thus be offered to participants (steps 1–2–3–4) with close monitoring to anticipate and react quickly upon identification of impairment in one or more domains of IC. This monitoring involves the completion of Step 1 every 6–12 months through self‐assessment, by a healthcare professional or by other professional (like social services agents) and engaging the participants in the subsequent steps of the program based on their IC evolution.
3.2. Study Population
The data is collected at the ICOPE MONITOR digital platform. The eligibility criteria for ICOPE are an age of 60 years and over‐ and independency for activities of daily living. To have sufficient retrospective data for comparing populations in terms of adherence, a duration of 2 years ± 4 months after inclusion was selected. Among the studied population, participants who prematurely discontinued their follow‐up are considered as discontinuations.
3.2.1. Population in Discontinued Follow‐Up
The population in discontinued follow‐up completed the initial Step 1, underwent Step 2 (in‐depth assessment) and Step 3 (care and prevention plan), if required by Step 1 results, and then continued their follow‐up variably. However, these participants discontinued their follow‐up either at their request, due to a deterioration in their health, being lost to follow‐up, deceased or for another reason. The other participants are considered in ongoing follow‐up and are divided into two groups based on their level of adherence (completely or partially).
3.2.2. Completely Adherent Population
The choice was to adopt a restrictive definition where 100% of the program's actions had to be completed within the expected timeframe for participants to be considered completely adherent. This population enrolled in the program through the initial Step 1 screening, underwent Step 2 and step 3 (if required by Step 1 results), and completed their follow‐up Step 1 within the specified time, i.e., one Step 1 follow‐up between 6 and 12 months, a second between 12 and 18 months and finally a third between 18 and 24 months. A tolerance of an additional 4 months was allowed for the completion of these follow‐ups, making a total study period of 28 months (Supporting Information).
3.2.3. Partially Adherent Population
The partially adherent population enrolled in the program through the initial Step 1 identification, underwent Step 2 and Step 3 (if required by Step 1 results) but did not complete their follow‐up steps1 within the recommended protocol timelines (Supporting Information). The proposed pathway was incomplete with at least one of the follow‐ups missing within the specified time during the analysed period (24 months ± 4 months).
3.3. Study Variables
The population discontinuing the follow‐up is described by the reason for discontinuation and the average time of discontinuation occurrence.
3.3.1. Data Collection
The data collected at the initial Step 1 (Figure S1) includes: memory complaint, recall of 3 words and temporal orientation for cognition; weight loss and loss of appetite for nutrition; vision problems or eye diseases or use of antihypertensive or antidiabetic treatment and complaint of recent vision loss; whispered voice test and complaint of recent hearing loss; feelings of depression and loss of interest in doing things for mental health; timed 5‐chair rise test for mobility. The results of the Step 2 include: the number of comorbidities and medications, the number of people living alone and receiving home care, the score on the Mini‐Mental State Examination evaluating cognitive status (Kalafat et al. 2003) (MMSE from 0 to 30, with a higher score indicating better performance), the score on the Patient Health Questionnaire‐9 measuring the severity of depressive symptoms (Kroenke et al. 2001) (PHQ‐9 from 0 to 27, with a higher score indicating severe depression), the score on the Short Physical Performance Battery evaluating the level of functional performance (Guralnik et al. 1994) (from 0 to 12, with good performance SPPB ≥ 10/12, average performance between 7 and 9 and bad performance ≤ 6/12), the score on Fried's Frailty Phenotype (Fried et al. 2001) consisting of 5 criteria (participants are classified as robust with 0 criteria, pre‐frail with 1 or 2, and frail with 3 or more), the score on the Lawton Instrumental Activity of Daily Living Score (IADL from 0 to 8; the higher the score, the more independent the person is) (Graf 2008), the overall score on the Mini‐Nutritional Assessment MNA from 0 to 30 (from 24 to 30 indicates a normal nutritional status; from 17 to 23.5, an at‐risk status, and a score of less than 17 indicates poor nutritional status) (Lilamand et al. 2015), Body Mass Index (BMI), the number of participants with abnormal far‐vision and with abnormal near‐vision (by the WHO simple visual acuity scale) (WHO n.d.) and an assessment of HHIE‐S (Hearing Handicap Inventory—Screening for older people) for hearing from 0 to 40 (Duchêne et al. 2022).
3.3.2. Predictors
To predict non‐adherence and partial adherence, the predictors proposed are: the age, the gender, the proportion of initial Step 1 identifying at least one impairment in IC, the number of impaired domains, the presence of an alert in each IC domain, the duration of the timed 5‐chair rise test, the number of Step 2 assessments performed and the clinical data collected at the Step 2. The predictors were selected based on the data available in the ICOPE MONITOR database. The data collected in steps 1 and 2 of ICOPE program are from standardised validated tests.
3.3.3. Ethical Considerations
The ICOPE database complies with data confidentiality and security requirements in accordance with the General Data Protection Regulation (GDPR) and has been validated by the National Commission on Informatics and Liberties (CNIL) in 2017 (registration number 247169284 s, reference MMS/OSS/NDT171027). After evaluation and validation by the data protection officer and according to the GDPR, this study completed all the criteria, and it is registered in the register of data study of the Toulouse Hospital (number's register: RnIPH 2023–122) and covered by the MR‐004.
3.4. Statistical Analyses
Qualitative variables are presented by their frequencies and percentages. The distribution of quantitative variables is described by their mean and standard deviation or median and interquartile range, depending on the distribution of the variable. A comparison between the characteristics of participants who discontinued their follow‐up and those of the ongoing population is performed. Subsequently, the characteristics of the completely adherent population are compared to those of the partially adherent population. The tests used to compare these two populations include the Chi‐squared test or Fisher's exact test for frequency comparison, depending on the sample size. Quantitative variables are compared using the Student's t‐test or the Mann–Whitney test, depending on the variable distribution. A multivariate logistic regression model is conducted to study the factors associated with follow‐up discontinuation and partial adherence among participants in the ongoing population. A stepwise backward model is implemented with the variables of interest selected at a threshold of 0.20 in bivariate analyses, considering only the variables provided in Step 1. This is done due to clinical interest in identifying strategies to detect and improve participant adherence from Step 1 onward. The analyses for this study are performed using the Stata software package (StataCorp LP, College Station, TX, USA), version 14.2.
4. Results
Between 1 January 2020 and 20 February 2023, a total of 28,839 participants were enrolled in the ICOPE MONITOR program, with 27,082 participants meeting the eligibility criteria for the ICOPE pathway. Among them, 8672 participants had a follow‐up period of at least 2 years (+/−4 months) after enrolment and were included in the current study. Of these, 2176 (25.1%) participants discontinued their follow‐up, while 6496 (74.9%) participants continued their follow‐up. Within the ongoing follow‐up population, 2174 (33.5%) participants were completely adherent, and 4322 (66.5%) were partially adherent (Figure S2). Figure 1 illustrates the participants and the reasons for inclusion and exclusion, along with the number of participants who discontinued follow‐up and those who were completely or partially adherent. Supporting Information presents the pathway of participants considered completely adherent or partially adherent.
FIGURE 1.
Flowchart of the study.
4.1. Comparison of Participants Who Have Discontinued Follow‐Up and Participants Currently Under Follow‐Up
Among the 8672 participants in this study, 2176 participants have discontinued follow‐up (25.1%). The reasons include the participant's decision (n = 767, 35.3%), deterioration in the participant's health (n = 522, 24.0%), loss of contact with the participant (n = 508, 23.4%), participant's death (n = 174, 8.0%) or other reasons (n = 205, 9.4%). The discontinuation occurred on average after 11.8 months (±6.7). Table 1 compares the characteristics of the 2176 participants who have discontinued follow‐up with those of the 6496 participants currently under follow‐up. It appears that participants who have discontinued follow‐up are older than those currently under follow‐up and are significantly less often assessed by self‐assessment (p < 0.01). The results show a higher frequency of anomalies at Step 1 (p < 0.01), and the results are more often declined in all domains of IC except for vision. The results at Step 2 show that they are generally frailer and more often have impairments in cognitive, nutritional and mobility domains. The regression model (Table 2) shows that discontinuation of follow‐up is significantly associated with age and impairments in different IC domains. An increase in the number of alerts and having undergone Step 2 evaluation are significantly protective against discontinuation of follow‐up.
TABLE 1.
Comparison of participants who prematurely stopped the follow‐up vs. participants still in follow‐up.
| Steps 1, N = 8672 | |||
|---|---|---|---|
| Follow‐up discontinuation N = 2176 | Currently under follow‐up, N = 6496 | p | |
| Including by self‐screening, yes | 237 (10.9) | 839 (12.9) | 0.01 |
| Age in years* | 79.1 ± 8.5 | 76.4 ± 8.3 | < 0.01 |
| Age groups | |||
| 60–69 years | 330 (15.2) | 1435 (22.1) | < 0.01 |
| 70–79 years | 738 (33.9) | 2642 (40.7) | |
| 80 years and over | 1108 (50.9) | 2419 (37.2) | |
| Sex, female | 1261 (58.0) | 3948 (60.8) | 0.02 |
| First Step 1 | |||
| Step1 positive screening, yes | 2112 (97.1) | 6140 (94.5) | < 0.01 |
| Number of alerts per subject** | 3 (2–4) | 2 (2–4) | < 0.01 |
| Cognitive alert, yes | 1460 (67.1) | 3648 (56.2) | < 0.01 |
| Psychological alert, yes | 908 (41.7) | 2308 (35.5) | < 0.01 |
| Mobility alert, yes | 961 (44.2) | 2034 (31.3) | < 0.01 |
| Nutritional alert, yes | 533 (24.5) | 1011 (15.6) | < 0.01 |
| Visual alert, yes | 1512 (69.5) | 4513 (69.5) | 0.99 |
| Hearing alert, yes | 1260 (57.9) | 3223 (49.6) | < 0.01 |
| Chair rise in seconds* | 13.6 ± 6.1 | 12.7 ± 5.2 | < 0.01 |
| Completion of a Step 2 assessment | 184 (8.5) | 631 (9.7) | 0.08 |
| Steps 2, N = 815 | |||
|---|---|---|---|
| Follow‐up discontinuation N = 184 | Currently under follow‐up, N = 631 | p | |
| Number of comorbidities** | 3 (1–4) | 3 (2–4) | 0.3 |
| Number of treatments** | 5 (3–7) | 4 (2–5) | 0.009 |
| Living alone, yes | 83 (45.1) | 307 (48.7) | 0.4 |
| Presence of home caregivers, yes | 112 (60.9) | 318 (50.4) | 0.01 |
| Cognition | |||
| MMSE* | 23.3 ± 5.2 | 26.1 ± 3.3 | < 0.01 |
| Psychology | |||
| PHQ‐9** | 5 (1–10) | 4 (1–8) | 0.4 |
| Mobility | |||
| SPPB** | 8 (5–11) | 10 (7–12) | < 0.01 |
| SPPB performances | |||
| High | 69 (37.5) | 335 (55.1) | < 0.01 |
| Moderate | 44 (23.9) | 151 (23.9) | |
| Low | 65 (35.3) | 122 (19.3) | |
| Missing data | 6 (3.3) | 23 (3.7) | |
| Fried score ** | 2 (1–3) | 2 (1–3) | < 0.01 |
| Fried score status | |||
| Robust | 16 (8.7) | 129 (20.4) | < 0.01 |
| Pre‐frail | 87 (47.3) | 331 (52.5) | |
| Frail | 81 (44.0) | 164 (26.0) | |
| Missing data | 0 | 7 (1.1) | |
| IADL** | 6 (4–8) | 8 (6–8) | < 0.01 |
| Nutrition | |||
| MNA* | 23.0 ± 4.7 | 25.2 ± 3.4 | < 0.01 |
| Nutritional status of MNA | |||
| Normal | 97 (52.7) | 447 (70.8) | < 0.01 |
| At risk | 61 (33.2) | 126 (20.0) | |
| Malnutrition | 19 (10.3) | 17 (2.7) | |
| Missing data | 7 (3.8) | 41 (6.5) | |
| BMI in kg/m2 * | 25.3 ± 4.8 | 26.0 ± 4.9 | 0.06 |
| Vision | |||
| Distant vision, abnormal | 22 (12.0) | 48 (7.6) | 0.06 |
| Near vision, abnormal | 28 (15.2) | 53 (8.4) | 0.004 |
| Audition | |||
| HHIE‐S ** | 4 (0–14) | 4 (0–12) | 0.8 |
Note: Data is presented by their *mean ± SD ** median (IQR) otherwise n (%). Bold values indicate significant difference if p < 0.05.
Abbreviations: BMI, Body Mass Index; DM, missing data; HHIE‐S, Hearing Handicap Inventory for the Elderly—Screening (from 0 to 40); IADL, Instrumental Activity of Daily Living Score de Lawton (from 0 to 8); MMSE, Mini‐Mental Status Examination (from 0 to 30); MNA, Mini Nutritional Assessment (from 0 to 12); Phq‐9, Patient Health Questionnaire‐9 (from 0 to 27); SPPB, Short Physical Performance Battery (from 0 to 12).
TABLE 2.
Factors associated with premature follow‐up discontinuation, multivariate analysis (logistic regression) N = 6013.
| Odds ratio | p | IC 95%. | |
|---|---|---|---|
| Age in years | 1.03 | < 0.01 | [1.02; 1.04] |
| Number of alerts per subject | 0.87 | 0.04 | [0.75; 0.99] |
| Cognitive alert*, yes | 1.55 | < 0.01 | [1.27; 1.88] |
| Psychological alert*, yes | 1.26 | 0.02 | [1.04; 1.53] |
| Mobility alert*, yes | 1.59 | < 0.01 | [1.32; 1.93] |
| Nutritional alert*, yes | 1.67 | < 0.01 | [1.36; 2.05] |
| Hearing alert*, yes | 1.35 | 0.004 | [1.10; 1.65] |
| Completion of a Step 2 assessment*, yes | 0.63 | < 0.01 | [0.51; 0.78] |
The reference category is ‘no’.
4.2. Comparison Between Partially Adherent Participants and Completely Adherent Participants in the Population Currently Under Follow‐Up (N = 6496)
Table 3 compares the characteristics of 4322 partially adherent participants and 2174 completely adherent participants in the ongoing follow‐up. Partially adherent participants are significantly more often enrolled in the programme through self‐assessment and are older than completely adherent participants (p < 0.01). The results for Step 1 in the partially adherent population show a less significant frequency of impaired Step 1 (p = 0.003), but the results in various domains of IC are generally more altered in partially adherent individuals who less frequently underwent a Step 2 evaluation (p < 0.01). The results for Step 2 in partially adherent participants are generally more impaired than in completely adherent participants. The regression model (Table 4) reports that partial adherence is significantly associated with completing Step 1 through self‐assessment, higher age and a higher frequency of alerts in different domains. An increase in the number of alerts, an alert in vision and the completion of Step 2 evaluation are significantly protective against partial adherence.
TABLE 3.
Comparison of partially adherent participants with those completely adherent in the population of participants currently under follow‐up.
| Steps 1, N = 6496 | |||
|---|---|---|---|
| Partially adherent N = 4322 | Completely adherent N = 2174 | p | |
| Including by self‐screening, yes | 592 (13.7) | 247 (11.4) | 0.008 |
| Age in years* | 77.1 ± 8.2 | 74.3 ± 8.3 | < 0.01 |
| Age groups | |||
| 60–69 years | 801 (18.5) | 634 (29.2) | < 0.01 |
| 70–79 years | 1792 (41.5) | 850 (39.1) | |
| 80 years and over | 1729 (40.0) | 690 (31.7) | |
| Sex, female | 2609 (60.4) | 1339 (61.6) | 0.34 |
| Step 1 | |||
| Step 1 positive screening, yes | 4059 (93.9) | 2081 (95.7) | 0.003 |
| Number of alerts per subject ** | 3 (2–4) | 2 (2–3) | 0.02 |
| Cognitive alert, yes | 2494 (57.7) | 1154 (53.1) | < 0.01 |
| Psychological alert, yes | 1590 (36.8) | 718 (33.0) | 0.003 |
| Mobility alert, yes | 1512 (35.0) | 522 (24.0) | < 0.01 |
| Nutritional alert, yes | 739 (17.1) | 272 (12.5) | < 0.01 |
| Visual alert, yes | 2733 (63.2) | 1780 (81.9) | < 0.01 |
| Hearing alert, yes | 2196 (50.8) | 1027 (47.2) | 0.007 |
| Chair rise in seconds * | 13.2 ± 5.4 | 11.8 ± 4.7 | < 0.01 |
| Completion of a Step 2 assessment | 271 (6.3) | 360 (16.6) | < 0.01 |
| Steps 2, N = 631 | |||
|---|---|---|---|
| Partially adherent N = 271 | Completely adherent N = 360 | p | |
| Number of comorbidities** | 3 (2–4) | 2 (2–4) | 0.5 |
| Number of treatments** | 4 (2–6) | 4 (2–6) | 0.6 |
| Living alone, yes | 124 (45.8) | 183 (50.8) | 0.2 |
| Presence of home caregivers, yes | 140 (51.7) | 178 (49.4) | 0.6 |
| Cognition | |||
| MMSE* | 25.8 ± 3.8 | 26.4 ± 2.9 | 0.02 |
| Psychology | |||
| PHQ‐9** | 5 (2–8) | 4 (1–8) | 0.04 |
| Mobility | |||
| SPPB | 9 (6–11) | 10 (8–12) | 0.005 |
| SPPB Performances | |||
| High | 126 (46.5) | 209 (58.1) | 0.004 |
| Moderate | 63 (23.3) | 88 (24.4) | |
| Low | 67 (24.7) | 55 (15.3) | |
| Missing data | 15 (5.6) | 8 (2.2) | |
| Fried score** | 2 (1–3) | 2 (1–2) | 0.1 |
| Fried score status | |||
| Robust | 49 (18.1) | 80 (22.2) | 0.2 |
| Pre‐frail | 140 (51.7) | 191 (53.1) | |
| Frail | 79 (29.1) | 85 (23.6) | |
| Missing data | 3 (1.1) | 4 (1.1) | |
| IADL** | 8 (6–8) | 8 (7–8) | 0.001 |
| Nutrition | |||
| MNA* | 24.9 ± 3.4 | 25.3 ± 3.5 | 0.16 |
| Nutritional status MNA | |||
| Normal | 184 (67.9) | 263 (73.1) | 0.9 |
| At risk | 55 (20.3) | 71 (19.7) | |
| Malnutrition | 7 (2.6) | 10 (2.8) | |
| Missing data | 25 (9.2) | 16 (4.4) | |
| BMI in kg/m2* | 26.1 ± 4.7 | 25.9 ± 5.0 | 0.6 |
| Vision | |||
| Distant vision, abnormal | 19 (7.0) | 29 (8.6) | 0.4 |
| Near vision, abnormal | 20 (7.4) | 33 (8.6) | 0.004 |
| Audition | |||
| HHIE‐S** | 6 (0–14) | 4 (0–10) | 0.05 |
Note: Data is presented by their *mean ± SD; **median (IQR) otherwise n (%). Bold values indicate significant difference if p < 0.05.
Abbreviations: BMI, Body Mass Index; DM, missing data; HHIE‐S, Hearing Handicap Inventory for the Elderly—Screening (from 0 to 40); IADL, Instrumental Activity of Daily Living Score de Lawton (from 0 to 8); MMSE, Mini‐Mental Status Examination (from 0 to 30); MNA, Mini Nutritional Assessment (from 0 to 12); Phq‐9, Patient Health Questionnaire‐9 (from 0 to 27); SPPB, Short Physical Performance Battery (from 0 to 12).
TABLE 4.
Factors associated with partial adherence, multivariate analysis (logistic regression) N = 4451.
| Odds ratio | p | IC 95%. | |
|---|---|---|---|
| Self‐screening*, yes | 1.70 | < 0.01 | [1.29; 2.25] |
| Age in years | 1.03 | < 0.01 | [1.02; 1.04] |
| Number of alerts per subject | 0.73 | < 0.01 | [0.63; 0.84] |
| Cognitive alert*, yes | 1.49 | < 0.01 | [1.21; 1.84] |
| Psychological alert*, yes | 1.6 | < 0.01 | [1.30; 1.97] |
| Mobility alert*, yes | 2.19 | < 0.01 | [1.77; 2.71] |
| Nutritional alert*, yes | 1.75 | < 0.01 | [1.37; 2.24] |
| Visual alert*, yes | 0.51 | < 0.01 | [0.41; 0.65] |
| Completion of a Step 2 assessment*, yes | 0.25 | < 0.01 | [0.20; 0.31] |
The reference category is ‘no’.
5. Discussion
Our study focused on the associated factors of the 2‐year adherence to ICOPE program and the profiles of participants who discontinued follow‐up or showed partial adherence to the programme. After 2 years, three‐quarters of the participants are still under follow‐up with varying levels of adherence. Participants who discontinued follow‐up are older and have more altered Step 1 results than participants still in follow‐up for all domains of IC. Among participants still in follow‐up, partially adherent participants are older and generally more impaired in terms of IC than completely adherent participants. These results confirm our hypothesis that participants with the least adherence tend to have the most impaired clinical profiles.
5.1. Adherence Rate
Our study was able to analyse data within the framework of ‘the real‐world’ use of ICOPE and found a medium‐term follow‐up rate equivalent to that observed in clinical trials studying home‐based prevention actions, such as the Pro‐Age study (Stuck et al. 2007a), which reported a 1‐year follow‐up rate of 70.6% (Dapp et al. 2011). The Lucas cohort (Neumann et al. 2017) (Dapp et al. 2012) in Germany included 2580 participants and offered prevention actions to older adults after a health questionnaire at baseline and 1 year later. The authors described a complete adherence rate at 1 year of 36%, including the same participant profile as ICOPE. This adherence rate is higher than the complete adherence rate in our study, but ours is in a real‐world setting, and it covers a 2‐year follow‐up period. The high adherence to ICOPE, comparable to that observed in a clinical research setting, reinforces the operational nature of this public health prevention initiative.
5.2. Profile of Participants
Similar to ICOPE MONITOR's Step 1, adherence to the Health Risk Appraisal for Older Persons (HRA‐O) tool developed in the Pro‐Age study (Stuck et al. 2002a) is reduced in older subjects and those with perceived average or poor health (Stuck et al. 2007b). It is worth noting that for HRA‐O, participants reported difficulty responding to a lengthy questionnaire (Stuck et al. 2007b), while ICOPE's Step 1 screening tool is quick to administer. Tanja et al. described the preferences of 2498 older adults for different modalities of implementing a home‐based fall prevention programme (Dorresteijn et al. 2012). Individuals over 80 years old and those with more impaired health usually prefer individualised home‐based prevention actions compared to a digital tool (Dorresteijn et al. 2012). The use of connected tools is the least chosen by these population. These results are consistent with the clinical profile of the participants of ICOPE who are the least adherent to the follow‐up requiring the use of the ICOPE MONITOR app. Scientific literature (Nittas et al. 2019) on prevention actions through digital tools indicates that while age is not a barrier, having a positive prior experience with digital tools is a lever for better adherence. This could evolve rapidly in upcoming generations of older adults who will be more accustomed to new technologies. Overall, various studies on prevention in older adults show that the most impaired participants adhere to individualised and specific home‐based prevention actions, provided by a healthcare professional (Dapp et al. 2005) (Stuck et al. 2002b). This aligns with what ICOPE offers after performing Step 1. This reinforces the importance of offering this screening step to as many individuals as possible and then providing follow‐up for this target population that generates alerts at Step 1. Lear et al. (2021) evaluated the effectiveness of an internet‐based self‐management program versus conventional care on hospitalisations in patients with chronic diseases. The authors explained that the home‐based internet programme encourages collaboration and involvement of various healthcare professionals with the participants, in an average‐aged population of 70 years. The main reason for discontinuation in 35% of cases was the participant's desire to discontinue follow‐up. In the future, a study on the barriers to adherence and the individual reasons for non‐adherence should be conducted to improve the implementation of the program.
5.3. Implications to Nursing Practice
The ICOPE program was designed by the WHO for deployment in primary care settings (WHO n.d.). HCP, especially nurses, plays an important role due to their proximity to older people. In France, nurses have included and keep including the majority of participants in the program (Berbon et al. 2025), receiving funding in the framework program. In this study, the results, associated with non‐adherence, are easily accessible from Step 1, allowing for early detection and intervention by nurses targeting the population presenting this type of profile. The population at higher risk of non‐adherence must be closely monitored. This study helps identify this population and guide the efforts of community nurses towards the older people who are most at risk (older and frail individuals) in order to support them within the program. In fact, this population, which uses self‐assessment less frequently, should be able to benefit from the support of healthcare professionals. Nurses, in particular, play a crucial role in the prevention of functional decline due to their holistic and comprehensive view of the individual in the process of ageing.
5.4. Limitations
The limitations of our study lie in the challenges of a real‐world setting. There may be more lost follow‐ups, missing data or a lack of adherence to the studied action than in a clinical trial. Other factors that may influence adherence, such as level of empowerment of the intervention, socio‐cultural level or acceptability of digital tools, and support from formal and informal caregivers were not measured in this study. These are factors that could be the subject of future studies. However, the strength of our work lies in its large sample size. This is the first time that such a large population, completing the ICOPE program, can be described, providing useful elements to facilitate its implementation.
6. Conclusions
This study shows that the most impaired and older participants are at the highest risk of discontinuation of follow‐up and partial adherence to ICOPE. Caregivers and professionals participating in the programme must be aware that among participants, the older, sicker and likely to benefit the most from the intervention are also those at risk of leaving the program or being only partially adherent. Primary care HCP, especially nurses, contributes in adherence to follow‐up by providing targeted support to these populations. It is essential to focus efforts on increasing adherence to the program for this target population. Special attention should be given to them by reinforcing human support and personalisation of proposals and follow‐up. This work on participant adherence to ICOPE MONITOR can be repeated to ensure improvement in adherence with the deployment of the program in routine practice. On the other hand, our results suggest that younger robust and healthier older age people may represent a good target for ICOPE program, in terms of adherence and primary prevention. In fact, by also encouraging self‐assessment, ICOPE program will allow older age people to become actors of their own health in a participative preventing medicine in order to engage them in a healthy ageing process.
The ICOPE program has been incorporated into French law since April 2024 with the aim of a large national scale that will need to target the most relevant populations to follow. This will be achieved by prioritising communication efforts towards younger and robust individuals to empower them to manage follow‐up independently while directing HCP's attention to older, more vulnerable populations.
Disclosure
Database complies with data confidentiality and security requirements in accordance with the General Data Protection Regulation (GDPR) and has been validated by the National Commission on Informatics and Liberties (CNIL) in 2017 (registration number 247169284s, reference MMS/OSS/NDT171027). After evaluation and validation by the data protection officer and according to the GDPR, this study completed all the criteria, and it is registered in the register of data study of the Toulouse Hospital (number's register: RnIPH 2023–122) and covered by the MR‐004. There is a statistician on the author team and state which author.
Consent
The consent of all patients has been given.
Conflicts of Interest
The authors declare no conflicts of interest.
Peer Review
The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer‐review/10.1111/jan.16740.
Supporting information
Figure S1. Step1.
Figure S2. Pathway of participants considered completely adherent or partially adherent.
Acknowledgements
We thank all the healthcare professionals who participated in the ICOPE Care programme; the members of the Gerontopole of Toulouse especially those of the ‘Regional Team for Ageing and Prevention of Dependency’ (Isabelle Carrié, Justine de Kerimel, Céline Mathieu, Fanny Paris, Delphine Pennetier, Brune Rieunier, Alessia Robert‐Millocco) and the ‘ICOPE remote monitoring platform’ (Pascale Baby, Laure Bouchon, Marie Christine Cazes, Florence Da Costa, Laure Aldebert, Magali Poly, Charlene Seguela, Lay‐Nien Sephan); all the members of the Occitanie Territorial Teams of Ageing and Prevention of Dependency; and the project leaders of National Experimentation—Article 51 (Mutualité Française PACA, DAC Sante landes, Inter‐CPTS Haut‐Rhin, CPTS Haute‐Corrèze, Perigueux University Hospital, Filieris Sud, Civic Hospitals of Lyon, CPTS Grand Sud Réunion, InterURPS Pay de la Loire, CPTS Cerebellum, DAC 17, DAC 46, Clinique des Augustines, Brest University Hospital, Mutualité Française Bretagne).
Funding: This work was performed in the context of the IHU Health Age which received funding from the Agence Nationale de la Recherche as part of the France 2030 program (reference number: ANR‐23‐IAHU‐0011). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. The ICOPE‐Care program was supported by grants from the Occitania Regional Health Agency (Region Occitanie and Pyrénées‐Méditerranée; reference number 1901175), ICOPE National Experimentation ‐ Article 51 (Ministry of Solidarity and Health ‐ Order of July 19, 2022 ‐ NOR : SPRS2221913A), the European Regional Development Fund (project number MP0022856) and the Interreg Program V‐A Spain‐France‐Andorra (European Union) in the context of the APTITUDE project (reference EFA232/16).
Data Availability Statement
The data reported in this manuscript (text, tables, figures and the Supporting Information) will be shared in de‐identified form owing to privacy protections. Request for de‐identified data and a data dictionary will be evaluated by the ICOPE MONITOR data‐sharing committee, which can be contacted at the following address: tavassoli.n@chu‐toulouse.fr. Data will be made available for investigators upon request for a pre‐identified scientific purpose developed in a research proposal with sound methodology, subject to the approval of the appropriate ICOPE MONITOR committee; a data use agreement must also to be signed. Given the nature of the ICOPE MONITOR program (on‐going, real‐life data from healthcare services), the authors are unable to make the dataset publicly accessible at this time. In addition, according to the ICOPE MONITOR policy, all analyses using ICOPE MONITOR data (including the present study) must be evaluated and approved by the committee after submission of a comprehensive analysis proposal. Therefore, for researchers interested in accessing the data used for the current study, the same evaluation procedure must be followed.
References
- Berbon, C. , Takeda C., Balardy L., et al. 2025. “Implementing the WHO ICOPE Program in Clinical Practice: Three Years of Lessons From Monitoring 27,082 Participants Using the ICOPE Monitor Digital Tool.” Journals of Gerontology: Series A 80, no. 3: glae278. 10.1093/gerona/glae278. [DOI] [PubMed] [Google Scholar]
- Clegg, A. , Young J., Iliffe S., Rikkert M. O., and Rockwood K.. 2013. “Frailty in Elderly People.” Lancet 381, no. 9868: 752–762. 10.1016/S0140-6736(12)62167-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dapp, U. , Anders J., von Renteln‐Kruse W., Golgert S., Meier‐Baumgartner H. P., and Minder C. E.. 2012. “The Longitudinal Urban Cohort Ageing Study (LUCAS): Study Protocol and Participation in the First Decade.” BMC Geriatrics 12, no. 1: 35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dapp, U. , Anders J., von Renteln‐Kruse W., and Meier‐Baumgartner H. P.. 2005. “Active Health Promotion in Old Age: Methodology of a Preventive Intervention Programme Provided by an Interdisciplinary Health Advisory Team for Independent Older People.” Journal of Public Health 13, no. 3: 122–127. [Google Scholar]
- Dapp, U. , Anders J. A. M., von Renteln‐Kruse W., et al. 2011. “A Randomized Trial of Effects of Health Risk Appraisal Combined With Group Sessions or Home Visits on Preventive Behaviors in Older Adults.” Journals of Gerontology: Series A 66A, no. 5: 591–598. [DOI] [PubMed] [Google Scholar]
- Dorresteijn, T. A. C. , Rixt Zijlstra G. A., Van Eijs Y. J. J., Vlaeyen J. W. S., and Kempen G. I. J. M.. 2012. “Older People's Preferences Regarding Programme Formats for Managing Concerns About Falls.” Age and Ageing 41, no. 4: 474–481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Duchêne, J. , Billiet L., Franco V., and Bonnard D.. 2022. “Validation of the French Version of HHIE‐S (Hearing Handicap Inventory for the Elderly ‐ Screening) Questionnaire in French Over‐60 Year‐Olds.” European Annals of Otorhinolaryngology, Head and Neck Diseases 139, no. 4: 198–201. 10.1016/j.anorl.2021.11.003. [DOI] [PubMed] [Google Scholar]
- Fried, L. P. , Tangen C. M., Walston J., et al. 2001. “Frailty in Older Adults: Evidence for a Phenotype.” Journals of Gerontology. Series A, Biological Sciences and Medical Sciences 56, no. 3: M146–M157. [DOI] [PubMed] [Google Scholar]
- Graf, C. 2008. “The Lawton Instrumental Activities of Daily Living Scale.” American Journal of Nursing Research 108, no. 4: 52–62. quiz 62–3. [DOI] [PubMed] [Google Scholar]
- Guralnik, J. M. , Simonsick E. M., Ferrucci L., et al. 1994. “A Short Physical Performance Battery Assessing Lower Extremity Function: Association With Self‐Reported Disability and Prediction of Mortality and Nursing Home Admission.” Journals of Gerontology 49, no. 2: M85–M94. [DOI] [PubMed] [Google Scholar]
- Kalafat, M. , Hugonot‐Diener L., and Poitrenaud J.. 2003. “The Mini Mental State (MMS): French Standardization and Normative Data.” Revue de Neuropsychologie 13: 209–236. [Google Scholar]
- Kroenke, K. , Spitzer R. L., and Williams J. B.. 2001. “The PHQ‐9: Validity of a Brief Depression Severity Measure.” Journal of General Internal Medicine 16, no. 9: 606–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lear, S. A. , Norena M., Banner D., et al. 2021. “Assessment of an Interactive Digital Health–Based Self‐Management Program to Reduce Hospitalizations Among Patients With Multiple Chronic Diseases: A Randomized Clinical Trial.” JAMA Network Open 4, no. 12: e2140591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lilamand, M. , Kelaiditi E., Cesari M., et al. 2015. “Validation of the Mini Nutritional Assessment‐Short Form in a Population of Frail Elders Without Disability. Analysis of the Toulouse Frailty Platform Population in 2013.” Journal of Nutrition, Health & Aging 19, no. 5: 570. 4. [DOI] [PubMed] [Google Scholar]
- Marengoni, A. , Von Strauss E., Rizzuto D., Winblad B., and Fratiglioni L.. 2009. “The Impact of Chronic Multimorbidity and Disability on Functional Decline and Survival in Elderly Persons. A Community‐Based, Longitudinal Study.” Journal of Internal Medicine 265, no. 2: 288–295. 10.1111/j.1365-2796.2008.02017.x. [DOI] [PubMed] [Google Scholar]
- Morley, J. E. , Vellas B., van Kan G. A., et al. 2013. “Frailty Consensus: A Call to Action.” Journal of the American Medical Directors Association 14, no. 6: 392–397. 10.1016/j.jamda.2013.03.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Neumann, L. , Dapp U., von Renteln‐Kruse W., and Minder C. E.. 2017. “Health Promotion and Preventive Care Intervention for Older Community‐Dwelling People: Long‐Term Effects of a Randomised Controlled Trial (RCT) Within the LUCAS Cohort.” Journal of Nutrition, Health & Aging 21, no. 9: 1016–1023. [DOI] [PubMed] [Google Scholar]
- Nittas, V. , Lun P., Ehrler F., Puhan M. A., and Mütsch M.. 2019. “Electronic Patient‐Generated Health Data to Facilitate Disease Prevention and Health Promotion: Scoping Review.” Journal of Medical Internet Research 21, no. 10: e13320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stuck, A. E. , Egger M., Hammer A., Minder C. E., and Beck J. C.. 2002b. “Home Visits to Prevent Nursing Home Admission and Functional Decline in Elderly PeopleSystematic Review and Meta‐Regression Analysis.” Journal of the American Medical Association 287, no. 8: 1022–1028. [DOI] [PubMed] [Google Scholar]
- Stuck, A. E. , Elkuch P., Dapp U., et al. 2002a. “Feasibility and Yield of a Self‐Administered Questionnaire for Health Risk Appraisal in Older People in Three European Countries.” Age and Ageing 31, no. 6: 463–467. [DOI] [PubMed] [Google Scholar]
- Stuck, A. E. , Kharicha K., Dapp U., et al. 2007a. “The PRO‐AGE Study: An International Randomised Controlled Study of Health Risk Appraisal for Older Persons Based in General Practice.” BMC Medical Research Methodology 7: 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Stuck, A. E. , Kharicha K., Dapp U., et al. 2007b. “Development, Feasibility and Performance of a Health Risk Appraisal Questionnaire for Older Persons.” BMC Medical Research Methodology 7: 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sum, G. , Lau L. K., Jabbar K. A., et al. 2022. “The World Health Organization (WHO) Integrated Care for Older People (ICOPE) Framework: A Narrative Review on Its Adoption Worldwide and Lessons Learnt.” International Journal of Environmental Research and Public Health 20, no. 1: 154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tajudeen, F. P. , Bahar N., Tan M. P., Peer Mustafa M. B., Saedon N. I., and Jesudass J.. 2022. “Understanding User Requirements for a Senior‐Friendly Mobile Health Application.” Geriatrics (Basel) 7, no. 5: 110. 10.3390/geriatrics7050110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tavassoli, N. , P de S B., Berbon C., et al. 2022. “Implementation of the WHO Integrated Care for Older People (ICOPE) Programme in Clinical Practice: A Prospective Study.” Lancet Healthy Longevity 3, no. 6: e394–e404. [DOI] [PubMed] [Google Scholar]
- Tavassoli, N. , Piau A., Berbon C., et al. 2021. “Framework Implementation of the INSPIRE ICOPE‐CARE Program in Collaboration With the World Health Organization (WHO) in the Occitania Region.” Journal of Frailty & Aging 10, no. 2: 103–109. [DOI] [PubMed] [Google Scholar]
- WHO . n.d. Integrated Care for Older People: Guidelines on Community‐LEVEL Interventions to Manage Declines in Intrinsic Capacity. Accessed April 25, 2024. https://www.who.int/publications‐detail‐redirect/9789241550109. [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Step1.
Figure S2. Pathway of participants considered completely adherent or partially adherent.
Data Availability Statement
The data reported in this manuscript (text, tables, figures and the Supporting Information) will be shared in de‐identified form owing to privacy protections. Request for de‐identified data and a data dictionary will be evaluated by the ICOPE MONITOR data‐sharing committee, which can be contacted at the following address: tavassoli.n@chu‐toulouse.fr. Data will be made available for investigators upon request for a pre‐identified scientific purpose developed in a research proposal with sound methodology, subject to the approval of the appropriate ICOPE MONITOR committee; a data use agreement must also to be signed. Given the nature of the ICOPE MONITOR program (on‐going, real‐life data from healthcare services), the authors are unable to make the dataset publicly accessible at this time. In addition, according to the ICOPE MONITOR policy, all analyses using ICOPE MONITOR data (including the present study) must be evaluated and approved by the committee after submission of a comprehensive analysis proposal. Therefore, for researchers interested in accessing the data used for the current study, the same evaluation procedure must be followed.